cellular-infrastructure
/
osmo-msc
mirror of https://gerrit.osmocom.org/osmo-msc
9
0
Fork 0
Code Issues Releases Wiki Activity
osmo-msc/tests/msc_vlr/msc_vlr_test_call.err

1792 lines
131 KiB
Plaintext
Raw Normal View History

use osmo_init_logging2() with proper talloc ctx Since the logging allocations now also show up in the root context report, some tests need adjusted talloc checks. In msc_vlr_tests, also output the number of talloc blocks before tests are started to show that the number didn't change after the tests. Change-Id: Iae07ae60230c7bab28e52b5df97fa3844778158e
2018-03-30 21:04:04 +00:00
full talloc report on 'msgb' (total 0 bytes in 1 blocks)
make gsup ipa name configurable in osmo-msc.cfg Add a 'ipa-name' VTY command which overrides the default IPA name used by the MSC. This is a prerequisite for inter-MSC handover. Related: OS#3355 Change-Id: I317d6c59f77e92fbb2b875a83dc0ec2fa5cb6006
2018-12-06 11:06:59 +00:00
talloc_total_blocks(tall_bsc_ctx) == 13
use osmo_init_logging2() with proper talloc ctx Since the logging allocations now also show up in the root context report, some tests need adjusted talloc checks. In msc_vlr_tests, also output the number of talloc blocks before tests are started to show that the number didn't change after the tests. Change-Id: Iae07ae60230c7bab28e52b5df97fa3844778158e
2018-03-30 21:04:04 +00:00
msc_vlr_tests: revert IMSI parameter and test nr output Three recently merged commits take the msc_vlr_tests in a wrong direction. The IMSI is usually encoded in the hex streams. The rationale behind hex streams is that it is a) easily copied from a wireshark trace and b) exactly the bytes as sent by an actual phone. It is hard to parameterize the IMSI because we would have to employ our encoding functions, which I intentionally want to keep out of the loop here. The test number should not appear in the normal test output, so that adding a test or changing their order does not affect expected output for following tests. The nr is simply for manual invocation, only seen when invoked with -v. Revert - "VLR tests: always print test parameters" b0a4314911140b1599cccfc8171fcdab4cd9bfab. - "Expand VLR tests" d5feadeee8dd24f991df2892d6bcf0be8b0cf707. - "Move IMSI into test parameters" 093300d141c300651954473d73138b72de04d931. Change-Id: Ie1b49237746751021da88f6f07bbb9f780d077c9
2018-03-01 23:40:58 +00:00
===== test_call_mo
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
- Total time passed: 0.000000 s
- Location Update request causes a GSUP Send Auth Info request to HLR
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_MM_LOC_UPD_REQUEST
new conn
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn{RAN_CONN_S_NEW}: Allocated
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF unknown: MSC conn use + compl_l3 == 1 (0x1: compl_l3)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DRLL Dispatching 04.08 message GSM48_MT_MM_LOC_UPD_REQUEST (0x5:0x8)
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_NEW}: Updated ID
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DMM LOCATION UPDATING REQUEST: MI(IMSI)=901700000010650 type=NORMAL
DMM LU/new-LAC: 23/23
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_IDLE}: Allocated
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_IDLE}: is child of RAN_conn(LU:901700000010650)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_IDLE}: rev=R99 net=UTRAN Auth+Ciph
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_IDLE}: Received Event VLR_ULA_E_UPDATE_LA
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DREF VLR subscr unknown usage increases to: 1
DVLR set IMSI on subscriber; IMSI=901700000010650 id=901700000010650
DVLR New subscr, IMSI: 901700000010650
DREF VLR subscr IMSI:901700000010650 usage increases to: 2
DREF VLR subscr IMSI:901700000010650 usage decreases to: 1
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_IDLE}: vlr_loc_upd_node1()
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_IDLE}: state_chg to VLR_ULA_S_WAIT_AUTH
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: Allocated
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: is child of vlr_lu_fsm(LU:901700000010650)
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: Received Event VLR_AUTH_E_START
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DVLR GSUP tx: 08010809710000000156f0
GSUP --> HLR: OSMO_GSUP_MSGT_SEND_AUTH_INFO_REQUEST: 08010809710000000156f0
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: state_chg to VLR_SUB_AS_NEEDS_AUTH_WAIT_AI
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_NEW}: Received Event RAN_CONN_E_COMPLETE_LAYER_3
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_NEW}: state_chg to RAN_CONN_S_AUTH_CIPH
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF IMSI:901700000010650: MSC conn use - compl_l3 == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_AUTH_CIPH}: Awaiting results for Auth+Ciph, overruling event RAN_CONN_E_UNUSED
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
lu_result_sent == 0
- from HLR, rx _SEND_AUTH_INFO_RESULT; VLR sends Auth Req to MS
<-- GSUP rx OSMO_GSUP_MSGT_SEND_AUTH_INFO_RESULT: 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
DVLR GSUP rx 511: 0a010809710000000156f00362201039fa2f4e3d523d8619a73b4f65c3e14d21049b36efdf2208059a4f668f6fbe39231027497388b6cb044648f396aa155b95ef2410f64735036e5871319c679f4742a75ea125108704f5ba55f30000d2ee44b22c8ea9192708e229c19e791f2e4103622010c187a53a5e6b9d573cac7c74451fd46d210485aa31302208d3d50a000bf04f6e23101159ec926a50e98c034a6b7d7c9f418d2410df3a03d9ca5335641efc8e36d76cd20b25101843a645b98d00005b2d666af46c45d927087db47cf7f81e4dc703622010efa9c29a9742148d5c9070348716e1bb210469d5f9fb22083df176f0c29f1a3d2310eb50e770ddcc3060101d2f43b6c2b884241076542abce5ff9345b0e8947f4c6e019c2510f9375e6d41e1000096e7fe4ff1c27e392708706f996719ba609c03622010f023d5a3b24726e0631b64b3840f82532104d570c03f2208ec011be8919883d62310c4e58af4ba43f3bcd904e16984f086d724100593f65e752e5cb7f473862bda05aa0a2510541ff1f077270000c5ea00d658bc7e9a27083fd26072eaa2a04d036220102f8f90c780d6a9c0c53da7ac57b6707e2104b072446f220823f39f9f425ad6e6231065af0527fda95b0dc5ae4aa515cdf32f2410537c3b35a3b13b08d08eeb28098f45cc25104bf4e564f75300009bc796706bc6574427080edb0eadbea94ac2
DREF VLR subscr IMSI:901700000010650 usage increases to: 2
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_NEEDS_AUTH_WAIT_AI}: Received Event VLR_AUTH_E_HLR_SAI_ACK
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DVLR SUBSCR(IMSI:901700000010650) Received 5 auth tuples
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_NEEDS_AUTH_WAIT_AI}: state_chg to VLR_SUB_AS_WAIT_RESP
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_WAIT_RESP}: got auth tuple: use_count=1 key_seq=0 -- will use UMTS AKA (is_r99=yes, at->vec.auth_types=0x3)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
- sending UMTS Auth Request for IMSI:901700000010650: tuple use_count=1 key_seq=0 auth_types=0x3 and...
- ...rand=39fa2f4e3d523d8619a73b4f65c3e14d
- ...autn=8704f5ba55f30000d2ee44b22c8ea919
- ...expecting res=e229c19e791f2e41
DREF VLR subscr IMSI:901700000010650 usage decreases to: 1
<-- GSUP rx OSMO_GSUP_MSGT_SEND_AUTH_INFO_RESULT: vlr_gsupc_read_cb() returns 0
auth_request_sent == 1
lu_result_sent == 0
- MS sends Authen Response, VLR accepts and sends SecurityModeControl
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_MM_AUTH_RESP
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF IMSI:901700000010650: MSC conn use + dtap == 1 (0x2: dtap)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DRLL Dispatching 04.08 message GSM48_MT_MM_AUTH_RESP (0x5:0x14)
cosmetic: gsm48_rx_mm_auth_resp(): log 'UMTS AUTH', not 'R99 AUTH' Change-Id: Iba43c685cbe238d96175267e9cc954b2f2f3e7fc
2018-03-10 03:51:39 +00:00
DMM IMSI:901700000010650: MM UMTS AUTHENTICATION RESPONSE (res = e229c19e791f2e41)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_WAIT_RESP}: Received Event VLR_AUTH_E_MS_AUTH_RESP
cosmetic: vlr_auth_fsm: log RAN and size along with SRES/RES Change-Id: Ib0f9f573ffac2302fbd3ee28f48ccd8fce5fe286
2018-03-10 03:22:50 +00:00
DVLR SUBSCR(IMSI:901700000010650) AUTH on UTRAN received RES: e229c19e791f2e41 (8 bytes)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DVLR SUBSCR(IMSI:901700000010650) AUTH established UMTS security context
refactor VLR FSM result handling Instead of keeping separate enums for FSM results and translating between those and the actual 04.08 reject causes that will ultimately reach the MS, just pass enum gsm48_reject_value cause codes around everywhere. Collapse some VLR *_timeout() and *_cancel() api to just *_cancel() with a gsm48 cause arg. (Hopefully) improve a few reject causes, but otherwise just aim for more transparent decisions on which cause value is used, for future fixes of returned causes. Depends: I6661f139e68a498fb1bef10c266c2f064b72774a (libosmocore) Change-Id: I27bf8d68737ff1f8dc6d11fb1eac3d391aab0cb1
2018-04-06 00:57:51 +00:00
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_WAIT_RESP}: Authentication terminating with result PASSED
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_WAIT_RESP}: state_chg to VLR_SUB_AS_AUTHENTICATED
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Removing from parent vlr_lu_fsm(LU:901700000010650)
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Freeing instance
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Deallocated
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_AUTH}: Received Event VLR_ULA_E_AUTH_RES
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_AUTH}: vlr_loc_upd_post_auth()
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_AUTH}: Set Ciphering Mode
msc_vlr_tests: improve cipher mode coverage Actually call msc_vlr_set_ciph_mode() and wrap away a_iface_tx_cipher_mode() and ranap_iu_tx_sec_mode_cmd(). Hence we'll see decisions and errors in msc_vlr_set_ciph_mode() as well. Change-Id: Id23bc245d4b5707edcd27c44db272fbb211bf9bd
2018-03-02 00:50:09 +00:00
DMM -> SECURITY MODE CONTROL IMSI:901700000010650
- sending SecurityModeControl for UE ctx 42 send_ck=0 new_key=1
- ...ik=27497388b6cb044648f396aa155b95ef
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_AUTH}: state_chg to VLR_ULA_S_WAIT_CIPH
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF IMSI:901700000010650: MSC conn use - dtap == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_AUTH_CIPH}: Awaiting results for Auth+Ciph, overruling event RAN_CONN_E_UNUSED
msc_vlr_tests: clearly separate Ciph Mode from Security Mode checking Clearly distinguish between Ciphering Mode Command on GERAN and Security Mode Control on UTRAN. Cosmetic: explicitly verify the key strings in the testing code (not only in the expected output). Change-Id: Ica93ed06c4c63dc6768736d25231de8068001114
2018-03-10 01:06:47 +00:00
security_mode_ctrl_sent == 1
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
lu_result_sent == 0
- MS sends SecurityModeControl acceptance, VLR accepts and sends GSUP LU Req to HLR
DMM <- SECURITY MODE COMPLETE IMSI:901700000010650
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_CIPH}: Received Event VLR_ULA_E_CIPH_RES
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_CIPH}: vlr_loc_upd_post_ciph()
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DIUCS IMSI:901700000010650: tx CommonID 901700000010650
- Iu Common ID --RAN_UTRAN_IU--> MS (IMSI=901700000010650)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_CIPH}: vlr_loc_upd_node_4()
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_CIPH}: state_chg to VLR_ULA_S_WAIT_HLR_UPD
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_INIT}: Allocated
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_INIT}: is child of vlr_lu_fsm(LU:901700000010650)
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_INIT}: Received Event UPD_HLR_VLR_E_START
GSUP client: send CN domain IE on LU request Give the HLR a chance to send us updated subscriber data by indicating the CN domain to be Circuit Switched, only during a LU Request GSUP message. Adjust msc_vlr_tests to expect the added GSUP CN domain IE to indicate CS, i.e. append '280102'. Related: OS#3601 Change-Id: I0c2d33fbfdb4728e480679120d06b7f3a2ccfd76
2018-09-28 00:41:39 +00:00
DVLR GSUP tx: 04010809710000000156f0280102
GSUP --> HLR: OSMO_GSUP_MSGT_UPDATE_LOCATION_REQUEST: 04010809710000000156f0280102
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_INIT}: state_chg to UPD_HLR_VLR_S_WAIT_FOR_DATA
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
gsup_tx_confirmed == 1
lu_result_sent == 0
- HLR sends _INSERT_DATA_REQUEST, VLR responds with _INSERT_DATA_RESULT
<-- GSUP rx OSMO_GSUP_MSGT_INSERT_DATA_REQUEST: 10010809710000000156f00804032443f2
DVLR GSUP rx 17: 10010809710000000156f00804032443f2
DREF VLR subscr IMSI:901700000010650 usage increases to: 2
DVLR IMSI:901700000010650 has MSISDN:42342
VLR: log subscriber update * move log helpers to generic header * log subscriber update It's handy for troubleshooting issues with subscriber update via GSUP from HLR. Change-Id: I1958aeeb3ea99831c7e2c5ee9a6b59834baf4520
2017-12-27 12:23:44 +00:00
DVLR SUBSCR(MSISDN:42342) VLR: update for IMSI=901700000010650 (MSISDN=42342, used=2)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DVLR GSUP tx: 12010809710000000156f0
GSUP --> HLR: OSMO_GSUP_MSGT_INSERT_DATA_RESULT: 12010809710000000156f0
DREF VLR subscr MSISDN:42342 usage decreases to: 1
<-- GSUP rx OSMO_GSUP_MSGT_INSERT_DATA_REQUEST: vlr_gsupc_read_cb() returns 0
lu_result_sent == 0
- HLR also sends GSUP _UPDATE_LOCATION_RESULT
<-- GSUP rx OSMO_GSUP_MSGT_UPDATE_LOCATION_RESULT: 06010809710000000156f0
DVLR GSUP rx 11: 06010809710000000156f0
DREF VLR subscr MSISDN:42342 usage increases to: 2
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_HLR_UPD}: Received Event VLR_ULA_E_HLR_LU_RES
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_WAIT_FOR_DATA}: Received Event UPD_HLR_VLR_E_UPD_LOC_ACK
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_WAIT_FOR_DATA}: state_chg to UPD_HLR_VLR_S_DONE
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_DONE}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_DONE}: Removing from parent vlr_lu_fsm(LU:901700000010650)
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_DONE}: Freeing instance
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_DONE}: Deallocated
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_HLR_UPD}: Received Event VLR_ULA_E_UPD_HLR_COMPL
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_HLR_UPD}: state_chg to VLR_ULA_S_WAIT_LU_COMPL
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_INIT}: Allocated
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_INIT}: is child of vlr_lu_fsm(LU:901700000010650)
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_INIT}: Received Event LU_COMPL_VLR_E_START
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_INIT}: state_chg to LU_COMPL_VLR_S_WAIT_SUB_PRES
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_WAIT_SUB_PRES}: Received Event LU_COMPL_VLR_E_SUB_PRES_COMPL
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_WAIT_SUB_PRES}: lu_compl_vlr_new_tmsi()
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_WAIT_SUB_PRES}: state_chg to LU_COMPL_VLR_S_WAIT_TMSI_CNF
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
- sending LU Accept for MSISDN:42342, with TMSI 0x03020100
DREF VLR subscr MSISDN:42342 usage decreases to: 1
<-- GSUP rx OSMO_GSUP_MSGT_UPDATE_LOCATION_RESULT: vlr_gsupc_read_cb() returns 0
lu_result_sent == 1
- a LU Accept with a new TMSI was sent, waiting for TMSI Realloc Compl
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
llist_count(&net->ran_conns) == 1
ran_conn_is_accepted() == false
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
- MS sends TMSI Realloc Complete
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_MM_TMSI_REALL_COMPL
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + dtap == 1 (0x2: dtap)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DRLL Dispatching 04.08 message GSM48_MT_MM_TMSI_REALL_COMPL (0x5:0x1b)
DMM TMSI Reallocation Completed. Subscriber: MSISDN:42342
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_LU_COMPL}: Received Event VLR_ULA_E_NEW_TMSI_ACK
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_WAIT_TMSI_CNF}: Received Event LU_COMPL_VLR_E_NEW_TMSI_ACK
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DREF VLR subscr MSISDN:42342 usage increases to: 2
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_WAIT_TMSI_CNF}: state_chg to LU_COMPL_VLR_S_DONE
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_LU_COMPL}: Received Event VLR_ULA_E_LU_COMPL_SUCCESS
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_DONE}: Terminating (cause = OSMO_FSM_TERM_PARENT)
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_DONE}: Removing from parent vlr_lu_fsm(LU:901700000010650)
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_DONE}: Freeing instance
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_DONE}: Deallocated
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_LU_COMPL}: state_chg to VLR_ULA_S_DONE
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_ACCEPTED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_AUTH_CIPH}: state_chg to RAN_CONN_S_ACCEPTED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_ACCEPTED}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_ACCEPTED}: state_chg to RAN_CONN_S_RELEASING
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + release == 2 (0x102: dtap,release)
refactor subscr_conn and subscr_conn_fsm de-/alloc Refactor: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. 2. Add separate AUTH_CIPH state to the FSM. 3. Use conn->use_count to trigger conn release. 4. Add separate RELEASING state to the FSM. 5. Add rate counters for each of the three Complete Layer 3 types. Details: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. Historically, a gsm_subscriber_connection was allocated in libbsc land, and only upon Complete Layer 3 did libmsc add the fsm instance. After splitting openbsc.git into a separate osmo-msc, this is no longer necessary, hence: Closely tie gsm_subscriber_connection allocation to the subscr_conn_fsm instance: talloc the conn as a child of the FSM instance, and discard the conn as soon as the FSM terminates. 2. Add separate AUTH_CIPH state to the FSM. Decoding the Complete Layer 3 message is distinctly separate from waiting for the VLR FSMs to conclude. Use the NEW state as "we don't know if this is a valid message yet", and the AUTH_CIPH state as "evaluating, don't release". A profound effect of this: should we for any odd reason fail to leave the FSM's NEW state, the conn will be released right at the end of msc_compl_l3(), without needing to trigger release in each code path. 3. Use conn->use_count to trigger conn release. Before, the FSM itself would hold a use count on the conn, and hence we would need to ask it whether it is ready to release the conn yet by dispatching events, to achieve a use_count decrement. Instead, unite the FSM instance and conn, and do not hold a use count by the FSM. Hence, trigger an FSM "UNUSED" event only when the use_count reaches zero. As long as use counts are done correctly, the FSM will terminate correctly. These exceptions: - The new AUTH_CIPH state explicitly ignores UNUSED events, since we expect the use count to reach zero while evaluating Authentication and Ciphering. (I experimented with holding a use count by AUTH_CIPH onenter() and releasing by onleave(), but the use count and thus the conn are released before the next state can initiate transactions that would increment the use count again. Same thing for the VLR FSMs holding a use count, they should be done before we advance to the next state. The easiest is to simply expect zero use count during the AUTH_CIPH state.) - A CM Service Request means that even though the MSC would be through with all it wants to do, we shall still wait for a request to follow from the MS. Hence the FSM holds a use count on itself while a CM Service is pending. - While waiting for a Release/Clear Complete, the FSM holds a use count on itself. 4. Add separate RELEASING state to the FSM. If we decide to release for other reasons than a use count reaching zero, we still need to be able to wait for the msc_dtap() use count on the conn to release. (An upcoming patch will further use the RELEASING state to properly wait for Clear Complete / Release Complete messages.) 5. Add rate counters for each of the three Complete Layer 3 types. Besides LU, also count CM Service Request and Paging Response acceptance/rejections. Without these counters, only very few of the auth+ciph outcomes actually show in the counters. Related: OS#3122 Change-Id: I55feb379e176a96a831e105b86202b17a0ffe889
2018-03-30 22:02:14 +00:00
DREF VLR subscr MSISDN:42342 usage increases to: 3
DREF VLR subscr MSISDN:42342 usage decreases to: 2
- Iu Release --RAN_UTRAN_IU--> MS
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - dtap == 1 (0x100: release)
properly receive BSSMAP Clear Complete and Iu Release Complete When sending a BSSMAP Clear or Iu Release, do not immediately discard the conn, but wait until a BSSMAP Clear Complete / Iu Release Complete has been received. Hence we will no longer show in the log that an incoming Release/Clear Complete belongs to an unknown subscriber, but will still be around to properly log the release. Related: OS#3122 Change-Id: Ie4c6aaba3866d6e5b98004e8870a215e8cf8ffc1
2018-04-01 18:55:54 +00:00
iu_release_sent == 1
- LU was successful, and the conn has already been closed
- RNC sends Iu Release Complete
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - release == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_RELEASING}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_RELEASING}: state_chg to RAN_CONN_S_RELEASED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_RELEASED}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_DONE}: Terminating (cause = OSMO_FSM_TERM_PARENT)
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_DONE}: Removing from parent RAN_conn(LU:901700000010650)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_DONE}: fsm_lu_cleanup called with cause OSMO_FSM_TERM_PARENT
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_DONE}: Freeing instance
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_DONE}: Deallocated
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DRLL MSISDN:42342: Freeing RAN connection
refactor subscr_conn and subscr_conn_fsm de-/alloc Refactor: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. 2. Add separate AUTH_CIPH state to the FSM. 3. Use conn->use_count to trigger conn release. 4. Add separate RELEASING state to the FSM. 5. Add rate counters for each of the three Complete Layer 3 types. Details: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. Historically, a gsm_subscriber_connection was allocated in libbsc land, and only upon Complete Layer 3 did libmsc add the fsm instance. After splitting openbsc.git into a separate osmo-msc, this is no longer necessary, hence: Closely tie gsm_subscriber_connection allocation to the subscr_conn_fsm instance: talloc the conn as a child of the FSM instance, and discard the conn as soon as the FSM terminates. 2. Add separate AUTH_CIPH state to the FSM. Decoding the Complete Layer 3 message is distinctly separate from waiting for the VLR FSMs to conclude. Use the NEW state as "we don't know if this is a valid message yet", and the AUTH_CIPH state as "evaluating, don't release". A profound effect of this: should we for any odd reason fail to leave the FSM's NEW state, the conn will be released right at the end of msc_compl_l3(), without needing to trigger release in each code path. 3. Use conn->use_count to trigger conn release. Before, the FSM itself would hold a use count on the conn, and hence we would need to ask it whether it is ready to release the conn yet by dispatching events, to achieve a use_count decrement. Instead, unite the FSM instance and conn, and do not hold a use count by the FSM. Hence, trigger an FSM "UNUSED" event only when the use_count reaches zero. As long as use counts are done correctly, the FSM will terminate correctly. These exceptions: - The new AUTH_CIPH state explicitly ignores UNUSED events, since we expect the use count to reach zero while evaluating Authentication and Ciphering. (I experimented with holding a use count by AUTH_CIPH onenter() and releasing by onleave(), but the use count and thus the conn are released before the next state can initiate transactions that would increment the use count again. Same thing for the VLR FSMs holding a use count, they should be done before we advance to the next state. The easiest is to simply expect zero use count during the AUTH_CIPH state.) - A CM Service Request means that even though the MSC would be through with all it wants to do, we shall still wait for a request to follow from the MS. Hence the FSM holds a use count on itself while a CM Service is pending. - While waiting for a Release/Clear Complete, the FSM holds a use count on itself. 4. Add separate RELEASING state to the FSM. If we decide to release for other reasons than a use count reaching zero, we still need to be able to wait for the msc_dtap() use count on the conn to release. (An upcoming patch will further use the RELEASING state to properly wait for Clear Complete / Release Complete messages.) 5. Add rate counters for each of the three Complete Layer 3 types. Besides LU, also count CM Service Request and Paging Response acceptance/rejections. Without these counters, only very few of the auth+ciph outcomes actually show in the counters. Related: OS#3122 Change-Id: I55feb379e176a96a831e105b86202b17a0ffe889
2018-03-30 22:02:14 +00:00
DREF VLR subscr MSISDN:42342 usage decreases to: 1
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_RELEASED}: Freeing instance
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_RELEASED}: Deallocated
llist_count(&net->ran_conns) == 0
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DREF VLR subscr MSISDN:42342 usage increases to: 2
vsub != NULL == 1
strcmp(vsub->imsi, IMSI) == 0
vsub->lac == 23
DREF VLR subscr MSISDN:42342 usage decreases to: 1
---
- after a while, a new conn sends a CM Service Request. VLR responds with Auth Req, 2nd auth vector
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_MM_CM_SERV_REQ
new conn
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn{RAN_CONN_S_NEW}: Allocated
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF unknown: MSC conn use + compl_l3 == 1 (0x1: compl_l3)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DRLL Dispatching 04.08 message GSM48_MT_MM_CM_SERV_REQ (0x5:0x24)
DMM <- CM SERVICE REQUEST serv_type=0x08 MI(IMSI)=901700000010650
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_NEW}: Updated ID
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_INIT}: Allocated
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_INIT}: is child of RAN_conn(CM_SERVICE_REQ:901700000010650)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_INIT}: rev=R99 net=UTRAN Auth+Ciph
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_INIT}: Received Event PR_ARQ_E_START
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DREF VLR subscr MSISDN:42342 usage increases to: 2
DREF VLR subscr MSISDN:42342 usage increases to: 3
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_INIT}: proc_arq_vlr_fn_post_imsi()
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_INIT}: state_chg to PR_ARQ_S_WAIT_AUTH
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: Allocated
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: is child of Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650)
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: Received Event VLR_AUTH_E_START
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: state_chg to VLR_SUB_AS_WAIT_RESP
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_WAIT_RESP}: got auth tuple: use_count=1 key_seq=1 -- will use UMTS AKA (is_r99=yes, at->vec.auth_types=0x3)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
- sending UMTS Auth Request for MSISDN:42342: tuple use_count=1 key_seq=1 auth_types=0x3 and...
- ...rand=c187a53a5e6b9d573cac7c74451fd46d
- ...autn=1843a645b98d00005b2d666af46c45d9
- ...expecting res=7db47cf7f81e4dc7
DREF VLR subscr MSISDN:42342 usage decreases to: 2
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_NEW}: Received Event RAN_CONN_E_COMPLETE_LAYER_3
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_NEW}: state_chg to RAN_CONN_S_AUTH_CIPH
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - compl_l3 == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_AUTH_CIPH}: Awaiting results for Auth+Ciph, overruling event RAN_CONN_E_UNUSED
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
cm_service_result_sent == 0
auth_request_sent == 1
- needs auth, not yet accepted
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
ran_conn_is_accepted() == false
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
- MS sends Authen Response, VLR accepts and sends SecurityModeControl
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_MM_AUTH_RESP
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + dtap == 1 (0x2: dtap)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DRLL Dispatching 04.08 message GSM48_MT_MM_AUTH_RESP (0x5:0x14)
cosmetic: gsm48_rx_mm_auth_resp(): log 'UMTS AUTH', not 'R99 AUTH' Change-Id: Iba43c685cbe238d96175267e9cc954b2f2f3e7fc
2018-03-10 03:51:39 +00:00
DMM MSISDN:42342: MM UMTS AUTHENTICATION RESPONSE (res = 7db47cf7f81e4dc7)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_WAIT_RESP}: Received Event VLR_AUTH_E_MS_AUTH_RESP
cosmetic: vlr_auth_fsm: log RAN and size along with SRES/RES Change-Id: Ib0f9f573ffac2302fbd3ee28f48ccd8fce5fe286
2018-03-10 03:22:50 +00:00
DVLR SUBSCR(MSISDN:42342) AUTH on UTRAN received RES: 7db47cf7f81e4dc7 (8 bytes)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DVLR SUBSCR(MSISDN:42342) AUTH established UMTS security context
refactor VLR FSM result handling Instead of keeping separate enums for FSM results and translating between those and the actual 04.08 reject causes that will ultimately reach the MS, just pass enum gsm48_reject_value cause codes around everywhere. Collapse some VLR *_timeout() and *_cancel() api to just *_cancel() with a gsm48 cause arg. (Hopefully) improve a few reject causes, but otherwise just aim for more transparent decisions on which cause value is used, for future fixes of returned causes. Depends: I6661f139e68a498fb1bef10c266c2f064b72774a (libosmocore) Change-Id: I27bf8d68737ff1f8dc6d11fb1eac3d391aab0cb1
2018-04-06 00:57:51 +00:00
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_WAIT_RESP}: Authentication terminating with result PASSED
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_WAIT_RESP}: state_chg to VLR_SUB_AS_AUTHENTICATED
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Removing from parent Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650)
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Freeing instance
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Deallocated
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_AUTH}: Received Event PR_ARQ_E_AUTH_RES
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_AUTH}: _proc_arq_vlr_node2()
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_AUTH}: Set Ciphering Mode
msc_vlr_tests: improve cipher mode coverage Actually call msc_vlr_set_ciph_mode() and wrap away a_iface_tx_cipher_mode() and ranap_iu_tx_sec_mode_cmd(). Hence we'll see decisions and errors in msc_vlr_set_ciph_mode() as well. Change-Id: Id23bc245d4b5707edcd27c44db272fbb211bf9bd
2018-03-02 00:50:09 +00:00
DMM -> SECURITY MODE CONTROL MSISDN:42342
- sending SecurityModeControl for UE ctx 42 send_ck=0 new_key=1
- ...ik=1159ec926a50e98c034a6b7d7c9f418d
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_AUTH}: state_chg to PR_ARQ_S_WAIT_CIPH
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - dtap == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_AUTH_CIPH}: Awaiting results for Auth+Ciph, overruling event RAN_CONN_E_UNUSED
msc_vlr_tests: clearly separate Ciph Mode from Security Mode checking Clearly distinguish between Ciphering Mode Command on GERAN and Security Mode Control on UTRAN. Cosmetic: explicitly verify the key strings in the testing code (not only in the expected output). Change-Id: Ica93ed06c4c63dc6768736d25231de8068001114
2018-03-10 01:06:47 +00:00
security_mode_ctrl_sent == 1
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
cm_service_result_sent == 0
- MS sends SecurityModeControl acceptance, VLR accepts; above Ciphering is an implicit CM Service Accept
DMM <- SECURITY MODE COMPLETE MSISDN:42342
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_CIPH}: Received Event PR_ARQ_E_CIPH_RES
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_CIPH}: _proc_arq_vlr_node2_post_ciph()
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DIUCS MSISDN:42342: tx CommonID 901700000010650
- Iu Common ID --RAN_UTRAN_IU--> MS (IMSI=901700000010650)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_CIPH}: _proc_arq_vlr_node2_post_vlr()
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_CIPH}: _proc_arq_vlr_post_pres()
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_CIPH}: _proc_arq_vlr_post_trace()
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_CIPH}: _proc_arq_vlr_post_imei()
refactor VLR FSM result handling Instead of keeping separate enums for FSM results and translating between those and the actual 04.08 reject causes that will ultimately reach the MS, just pass enum gsm48_reject_value cause codes around everywhere. Collapse some VLR *_timeout() and *_cancel() api to just *_cancel() with a gsm48 cause arg. (Hopefully) improve a few reject causes, but otherwise just aim for more transparent decisions on which cause value is used, for future fixes of returned causes. Depends: I6661f139e68a498fb1bef10c266c2f064b72774a (libosmocore) Change-Id: I27bf8d68737ff1f8dc6d11fb1eac3d391aab0cb1
2018-04-06 00:57:51 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_CIPH}: proc_arq_fsm_done(PASSED)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_CIPH}: state_chg to PR_ARQ_S_DONE
refactor VLR FSM result handling Instead of keeping separate enums for FSM results and translating between those and the actual 04.08 reject causes that will ultimately reach the MS, just pass enum gsm48_reject_value cause codes around everywhere. Collapse some VLR *_timeout() and *_cancel() api to just *_cancel() with a gsm48 cause arg. (Hopefully) improve a few reject causes, but otherwise just aim for more transparent decisions on which cause value is used, for future fixes of returned causes. Depends: I6661f139e68a498fb1bef10c266c2f064b72774a (libosmocore) Change-Id: I27bf8d68737ff1f8dc6d11fb1eac3d391aab0cb1
2018-04-06 00:57:51 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_DONE}: Process Access Request result: PASSED
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_ACCEPTED
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + cm_service == 1 (0x8: cm_service)
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_AUTH_CIPH}: state_chg to RAN_CONN_S_ACCEPTED
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_ACCEPTED}: ran_conn_fsm_has_active_transactions: still awaiting first request after a CM Service Request
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
cm_service_result_sent == 0
---
- a call is initiated
- SETUP gets forwarded to MNCC
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_CC_SETUP
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + dtap == 2 (0xa: dtap,cm_service)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DRLL Dispatching 04.08 message GSM48_MT_CC_SETUP (0x3:0x5)
cosmetic: msc_vlr_tests: enable CC logging in debug There are a number of bad failures in CC teardown handling we're solving. It helps to see CC logging in the msc_vlr_tests. Change-Id: I56ac269d46b48b6b85efad81c4d2343bfc41ea90
2018-03-12 22:59:37 +00:00
DCC Unknown transaction ID 8, creating new trans.
DCC (ti 08 sub MSISDN:42342 callref 80000001) New transaction
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DREF VLR subscr MSISDN:42342 usage increases to: 3
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + trans_cc == 3 (0x1a: dtap,cm_service,trans_cc)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DMM MSISDN:42342: rx msg GSM48_MT_CC_SETUP: received_cm_service_request changes to false
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - cm_service == 2 (0x12: dtap,trans_cc)
gsm_04_08_cc: Add global guard timer for MNCC The external MNCC handler may hang indefinitely in cases where the remote end of the MNCC ceases to work properly. Add a global guard timer to make sure the call reaches ACTIVE state. Change-Id: I7375d1e17cd746aac4eadfe1e587e82cf1630d3d Related: OS#3599
2018-10-10 15:00:49 +00:00
DCC (sub 42342) starting guard timer with 180 seconds
cosmetic: msc_vlr_tests: enable CC logging in debug There are a number of bad failures in CC teardown handling we're solving. It helps to see CC logging in the msc_vlr_tests. Change-Id: I56ac269d46b48b6b85efad81c4d2343bfc41ea90
2018-03-12 22:59:37 +00:00
DCC (ti 08 sub MSISDN:42342) new state NULL -> INITIATED
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DCC Subscriber MSISDN:42342 (42342) sends SETUP to 123
DMNCC transmit message MNCC_SETUP_IND
cosmetic: msc_vlr_tests: enable CC logging in debug There are a number of bad failures in CC teardown handling we're solving. It helps to see CC logging in the msc_vlr_tests. Change-Id: I56ac269d46b48b6b85efad81c4d2343bfc41ea90
2018-03-12 22:59:37 +00:00
DCC Sending 'MNCC_SETUP_IND' to MNCC.
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
MSC --> MNCC: callref 0x80000001: MNCC_SETUP_IND
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_ACCEPTED}: Received Event RAN_CONN_E_COMMUNICATING
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_ACCEPTED}: state_chg to RAN_CONN_S_COMMUNICATING
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - dtap == 1 (0x10: trans_cc)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
- MNCC says that's fine
DMNCC receive message MNCC_CALL_PROC_REQ
gsm_04_08_cc: Add global guard timer for MNCC The external MNCC handler may hang indefinitely in cases where the remote end of the MNCC ceases to work properly. Add a global guard timer to make sure the call reaches ACTIVE state. Change-Id: I7375d1e17cd746aac4eadfe1e587e82cf1630d3d Related: OS#3599
2018-10-10 15:00:49 +00:00
DCC (sub 42342) stopping pending guard timer
DCC (sub 42342) starting guard timer with 180 seconds
cosmetic: msc_vlr_tests: enable CC logging in debug There are a number of bad failures in CC teardown handling we're solving. It helps to see CC logging in the msc_vlr_tests. Change-Id: I56ac269d46b48b6b85efad81c4d2343bfc41ea90
2018-03-12 22:59:37 +00:00
DCC (ti 08 sub 42342) Received 'MNCC_CALL_PROC_REQ' from MNCC in state 1 (INITIATED)
DCC (ti 08 sub MSISDN:42342) new state INITIATED -> MO_CALL_PROC
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DMSC msc_tx 2 bytes to MSISDN:42342 via RAN_UTRAN_IU
- DTAP --RAN_UTRAN_IU--> MS: GSM48_MT_CC_CALL_PROC: 8302
- DTAP matches expected message
MS <--Call Assignment-- MSC: subscr=MSISDN:42342 callref=0x80000001
- Total time passed: 1.000023 s
- The other call leg got established (not shown here), MNCC tells us so
DMNCC receive message MNCC_ALERT_REQ
gsm_04_08_cc: Add global guard timer for MNCC The external MNCC handler may hang indefinitely in cases where the remote end of the MNCC ceases to work properly. Add a global guard timer to make sure the call reaches ACTIVE state. Change-Id: I7375d1e17cd746aac4eadfe1e587e82cf1630d3d Related: OS#3599
2018-10-10 15:00:49 +00:00
DCC (sub 42342) stopping pending guard timer
DCC (sub 42342) starting guard timer with 180 seconds
cosmetic: msc_vlr_tests: enable CC logging in debug There are a number of bad failures in CC teardown handling we're solving. It helps to see CC logging in the msc_vlr_tests. Change-Id: I56ac269d46b48b6b85efad81c4d2343bfc41ea90
2018-03-12 22:59:37 +00:00
DCC (ti 08 sub 42342) Received 'MNCC_ALERT_REQ' from MNCC in state 3 (MO_CALL_PROC)
DCC (ti 08 sub MSISDN:42342) new state MO_CALL_PROC -> CALL_DELIVERED
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DMSC msc_tx 2 bytes to MSISDN:42342 via RAN_UTRAN_IU
- DTAP --RAN_UTRAN_IU--> MS: GSM48_MT_CC_ALERTING: 8301
- DTAP matches expected message
DMNCC receive message MNCC_SETUP_RSP
gsm_04_08_cc: Add global guard timer for MNCC The external MNCC handler may hang indefinitely in cases where the remote end of the MNCC ceases to work properly. Add a global guard timer to make sure the call reaches ACTIVE state. Change-Id: I7375d1e17cd746aac4eadfe1e587e82cf1630d3d Related: OS#3599
2018-10-10 15:00:49 +00:00
DCC (sub 42342) stopping pending guard timer
DCC (sub 42342) starting guard timer with 180 seconds
cosmetic: msc_vlr_tests: enable CC logging in debug There are a number of bad failures in CC teardown handling we're solving. It helps to see CC logging in the msc_vlr_tests. Change-Id: I56ac269d46b48b6b85efad81c4d2343bfc41ea90
2018-03-12 22:59:37 +00:00
DCC (ti 08 sub 42342) Received 'MNCC_SETUP_RSP' from MNCC in state 4 (CALL_DELIVERED)
DCC starting timer T313 with 30 seconds
DCC (ti 08 sub MSISDN:42342) new state CALL_DELIVERED -> CONNECT_IND
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DMSC msc_tx 2 bytes to MSISDN:42342 via RAN_UTRAN_IU
- DTAP --RAN_UTRAN_IU--> MS: GSM48_MT_CC_CONNECT: 8307
- DTAP matches expected message
- Total time passed: 2.000046 s
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_CC_CONNECT_ACK
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + dtap == 2 (0x12: dtap,trans_cc)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DRLL Dispatching 04.08 message GSM48_MT_CC_CONNECT_ACK (0x3:0xf)
cosmetic: msc_vlr_tests: enable CC logging in debug There are a number of bad failures in CC teardown handling we're solving. It helps to see CC logging in the msc_vlr_tests. Change-Id: I56ac269d46b48b6b85efad81c4d2343bfc41ea90
2018-03-12 22:59:37 +00:00
DCC stopping pending timer T313
DCC (ti 08 sub MSISDN:42342) new state CONNECT_IND -> ACTIVE
gsm_04_08_cc: Add global guard timer for MNCC The external MNCC handler may hang indefinitely in cases where the remote end of the MNCC ceases to work properly. Add a global guard timer to make sure the call reaches ACTIVE state. Change-Id: I7375d1e17cd746aac4eadfe1e587e82cf1630d3d Related: OS#3599
2018-10-10 15:00:49 +00:00
DCC (sub 42342) stopping pending guard timer
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DMNCC transmit message MNCC_SETUP_COMPL_IND
cosmetic: msc_vlr_tests: enable CC logging in debug There are a number of bad failures in CC teardown handling we're solving. It helps to see CC logging in the msc_vlr_tests. Change-Id: I56ac269d46b48b6b85efad81c4d2343bfc41ea90
2018-03-12 22:59:37 +00:00
DCC Sending 'MNCC_SETUP_COMPL_IND' to MNCC.
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
MSC --> MNCC: callref 0x80000001: MNCC_SETUP_COMPL_IND
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_COMMUNICATING}: Received Event RAN_CONN_E_COMMUNICATING
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - dtap == 1 (0x10: trans_cc)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
---
- RTP stream goes ahead, not shown here.
- Total time passed: 125.000091 s
---
- Call ends
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_CC_DISCONNECT
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + dtap == 2 (0x12: dtap,trans_cc)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DRLL Dispatching 04.08 message GSM48_MT_CC_DISCONNECT (0x3:0x25)
cosmetic: msc_vlr_tests: enable CC logging in debug There are a number of bad failures in CC teardown handling we're solving. It helps to see CC logging in the msc_vlr_tests. Change-Id: I56ac269d46b48b6b85efad81c4d2343bfc41ea90
2018-03-12 22:59:37 +00:00
DCC (ti 08 sub MSISDN:42342) new state ACTIVE -> DISCONNECT_IND
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DMNCC transmit message MNCC_DISC_IND
cosmetic: msc_vlr_tests: enable CC logging in debug There are a number of bad failures in CC teardown handling we're solving. It helps to see CC logging in the msc_vlr_tests. Change-Id: I56ac269d46b48b6b85efad81c4d2343bfc41ea90
2018-03-12 22:59:37 +00:00
DCC Sending 'MNCC_DISC_IND' to MNCC.
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
MSC --> MNCC: callref 0x80000001: MNCC_DISC_IND
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_COMMUNICATING}: Received Event RAN_CONN_E_COMMUNICATING
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - dtap == 1 (0x10: trans_cc)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DMNCC receive message MNCC_REL_REQ
gsm_04_08_cc: Add global guard timer for MNCC The external MNCC handler may hang indefinitely in cases where the remote end of the MNCC ceases to work properly. Add a global guard timer to make sure the call reaches ACTIVE state. Change-Id: I7375d1e17cd746aac4eadfe1e587e82cf1630d3d Related: OS#3599
2018-10-10 15:00:49 +00:00
DCC (sub 42342) starting guard timer with 180 seconds
cosmetic: msc_vlr_tests: enable CC logging in debug There are a number of bad failures in CC teardown handling we're solving. It helps to see CC logging in the msc_vlr_tests. Change-Id: I56ac269d46b48b6b85efad81c4d2343bfc41ea90
2018-03-12 22:59:37 +00:00
DCC (ti 08 sub 42342) Received 'MNCC_REL_REQ' from MNCC in state 12 (DISCONNECT_IND)
DCC starting timer T308 with 10 seconds
DCC (ti 08 sub MSISDN:42342) new state DISCONNECT_IND -> RELEASE_REQ
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DMSC msc_tx 2 bytes to MSISDN:42342 via RAN_UTRAN_IU
- DTAP --RAN_UTRAN_IU--> MS: GSM48_MT_CC_RELEASE: 832d
- DTAP matches expected message
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_CC_RELEASE_COMPL
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + dtap == 2 (0x12: dtap,trans_cc)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DRLL Dispatching 04.08 message GSM48_MT_CC_RELEASE_COMPL (0x3:0x2a)
cosmetic: msc_vlr_tests: enable CC logging in debug There are a number of bad failures in CC teardown handling we're solving. It helps to see CC logging in the msc_vlr_tests. Change-Id: I56ac269d46b48b6b85efad81c4d2343bfc41ea90
2018-03-12 22:59:37 +00:00
DCC stopping pending timer T308
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DMNCC transmit message MNCC_REL_CNF
cosmetic: msc_vlr_tests: enable CC logging in debug There are a number of bad failures in CC teardown handling we're solving. It helps to see CC logging in the msc_vlr_tests. Change-Id: I56ac269d46b48b6b85efad81c4d2343bfc41ea90
2018-03-12 22:59:37 +00:00
DCC Sending 'MNCC_REL_CNF' to MNCC.
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
MSC --> MNCC: callref 0x80000001: MNCC_REL_CNF
MS <--Call Release-- MSC: subscr=MSISDN:42342 callref=0x0
cosmetic: msc_vlr_tests: enable CC logging in debug There are a number of bad failures in CC teardown handling we're solving. It helps to see CC logging in the msc_vlr_tests. Change-Id: I56ac269d46b48b6b85efad81c4d2343bfc41ea90
2018-03-12 22:59:37 +00:00
DCC (ti 08 sub MSISDN:42342) new state RELEASE_REQ -> NULL
gsm_04_08_cc: Add global guard timer for MNCC The external MNCC handler may hang indefinitely in cases where the remote end of the MNCC ceases to work properly. Add a global guard timer to make sure the call reaches ACTIVE state. Change-Id: I7375d1e17cd746aac4eadfe1e587e82cf1630d3d Related: OS#3599
2018-10-10 15:00:49 +00:00
DCC (sub 42342) stopping pending guard timer
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DREF VLR subscr MSISDN:42342 usage decreases to: 2
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - trans_cc == 1 (0x2: dtap)
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_COMMUNICATING}: Received Event RAN_CONN_E_COMMUNICATING
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - dtap == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_COMMUNICATING}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_COMMUNICATING}: state_chg to RAN_CONN_S_RELEASING
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + release == 1 (0x100: release)
refactor subscr_conn and subscr_conn_fsm de-/alloc Refactor: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. 2. Add separate AUTH_CIPH state to the FSM. 3. Use conn->use_count to trigger conn release. 4. Add separate RELEASING state to the FSM. 5. Add rate counters for each of the three Complete Layer 3 types. Details: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. Historically, a gsm_subscriber_connection was allocated in libbsc land, and only upon Complete Layer 3 did libmsc add the fsm instance. After splitting openbsc.git into a separate osmo-msc, this is no longer necessary, hence: Closely tie gsm_subscriber_connection allocation to the subscr_conn_fsm instance: talloc the conn as a child of the FSM instance, and discard the conn as soon as the FSM terminates. 2. Add separate AUTH_CIPH state to the FSM. Decoding the Complete Layer 3 message is distinctly separate from waiting for the VLR FSMs to conclude. Use the NEW state as "we don't know if this is a valid message yet", and the AUTH_CIPH state as "evaluating, don't release". A profound effect of this: should we for any odd reason fail to leave the FSM's NEW state, the conn will be released right at the end of msc_compl_l3(), without needing to trigger release in each code path. 3. Use conn->use_count to trigger conn release. Before, the FSM itself would hold a use count on the conn, and hence we would need to ask it whether it is ready to release the conn yet by dispatching events, to achieve a use_count decrement. Instead, unite the FSM instance and conn, and do not hold a use count by the FSM. Hence, trigger an FSM "UNUSED" event only when the use_count reaches zero. As long as use counts are done correctly, the FSM will terminate correctly. These exceptions: - The new AUTH_CIPH state explicitly ignores UNUSED events, since we expect the use count to reach zero while evaluating Authentication and Ciphering. (I experimented with holding a use count by AUTH_CIPH onenter() and releasing by onleave(), but the use count and thus the conn are released before the next state can initiate transactions that would increment the use count again. Same thing for the VLR FSMs holding a use count, they should be done before we advance to the next state. The easiest is to simply expect zero use count during the AUTH_CIPH state.) - A CM Service Request means that even though the MSC would be through with all it wants to do, we shall still wait for a request to follow from the MS. Hence the FSM holds a use count on itself while a CM Service is pending. - While waiting for a Release/Clear Complete, the FSM holds a use count on itself. 4. Add separate RELEASING state to the FSM. If we decide to release for other reasons than a use count reaching zero, we still need to be able to wait for the msc_dtap() use count on the conn to release. (An upcoming patch will further use the RELEASING state to properly wait for Clear Complete / Release Complete messages.) 5. Add rate counters for each of the three Complete Layer 3 types. Besides LU, also count CM Service Request and Paging Response acceptance/rejections. Without these counters, only very few of the auth+ciph outcomes actually show in the counters. Related: OS#3122 Change-Id: I55feb379e176a96a831e105b86202b17a0ffe889
2018-03-30 22:02:14 +00:00
DREF VLR subscr MSISDN:42342 usage increases to: 3
DREF VLR subscr MSISDN:42342 usage decreases to: 2
- Iu Release --RAN_UTRAN_IU--> MS
properly receive BSSMAP Clear Complete and Iu Release Complete When sending a BSSMAP Clear or Iu Release, do not immediately discard the conn, but wait until a BSSMAP Clear Complete / Iu Release Complete has been received. Hence we will no longer show in the log that an incoming Release/Clear Complete belongs to an unknown subscriber, but will still be around to properly log the release. Related: OS#3122 Change-Id: Ie4c6aaba3866d6e5b98004e8870a215e8cf8ffc1
2018-04-01 18:55:54 +00:00
- RNC sends Iu Release Complete
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - release == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_RELEASING}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_RELEASING}: state_chg to RAN_CONN_S_RELEASED
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_RELEASED}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_DONE}: Terminating (cause = OSMO_FSM_TERM_PARENT)
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_DONE}: Removing from parent RAN_conn(CM_SERVICE_REQ:901700000010650)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_DONE}: Freeing instance
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_DONE}: Deallocated
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DRLL MSISDN:42342: Freeing RAN connection
refactor subscr_conn and subscr_conn_fsm de-/alloc Refactor: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. 2. Add separate AUTH_CIPH state to the FSM. 3. Use conn->use_count to trigger conn release. 4. Add separate RELEASING state to the FSM. 5. Add rate counters for each of the three Complete Layer 3 types. Details: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. Historically, a gsm_subscriber_connection was allocated in libbsc land, and only upon Complete Layer 3 did libmsc add the fsm instance. After splitting openbsc.git into a separate osmo-msc, this is no longer necessary, hence: Closely tie gsm_subscriber_connection allocation to the subscr_conn_fsm instance: talloc the conn as a child of the FSM instance, and discard the conn as soon as the FSM terminates. 2. Add separate AUTH_CIPH state to the FSM. Decoding the Complete Layer 3 message is distinctly separate from waiting for the VLR FSMs to conclude. Use the NEW state as "we don't know if this is a valid message yet", and the AUTH_CIPH state as "evaluating, don't release". A profound effect of this: should we for any odd reason fail to leave the FSM's NEW state, the conn will be released right at the end of msc_compl_l3(), without needing to trigger release in each code path. 3. Use conn->use_count to trigger conn release. Before, the FSM itself would hold a use count on the conn, and hence we would need to ask it whether it is ready to release the conn yet by dispatching events, to achieve a use_count decrement. Instead, unite the FSM instance and conn, and do not hold a use count by the FSM. Hence, trigger an FSM "UNUSED" event only when the use_count reaches zero. As long as use counts are done correctly, the FSM will terminate correctly. These exceptions: - The new AUTH_CIPH state explicitly ignores UNUSED events, since we expect the use count to reach zero while evaluating Authentication and Ciphering. (I experimented with holding a use count by AUTH_CIPH onenter() and releasing by onleave(), but the use count and thus the conn are released before the next state can initiate transactions that would increment the use count again. Same thing for the VLR FSMs holding a use count, they should be done before we advance to the next state. The easiest is to simply expect zero use count during the AUTH_CIPH state.) - A CM Service Request means that even though the MSC would be through with all it wants to do, we shall still wait for a request to follow from the MS. Hence the FSM holds a use count on itself while a CM Service is pending. - While waiting for a Release/Clear Complete, the FSM holds a use count on itself. 4. Add separate RELEASING state to the FSM. If we decide to release for other reasons than a use count reaching zero, we still need to be able to wait for the msc_dtap() use count on the conn to release. (An upcoming patch will further use the RELEASING state to properly wait for Clear Complete / Release Complete messages.) 5. Add rate counters for each of the three Complete Layer 3 types. Besides LU, also count CM Service Request and Paging Response acceptance/rejections. Without these counters, only very few of the auth+ciph outcomes actually show in the counters. Related: OS#3122 Change-Id: I55feb379e176a96a831e105b86202b17a0ffe889
2018-03-30 22:02:14 +00:00
DREF VLR subscr MSISDN:42342 usage decreases to: 1
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_RELEASED}: Freeing instance
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_RELEASED}: Deallocated
llist_count(&net->ran_conns) == 0
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DREF freeing VLR subscr MSISDN:42342
msc_vlr_tests: revert IMSI parameter and test nr output Three recently merged commits take the msc_vlr_tests in a wrong direction. The IMSI is usually encoded in the hex streams. The rationale behind hex streams is that it is a) easily copied from a wireshark trace and b) exactly the bytes as sent by an actual phone. It is hard to parameterize the IMSI because we would have to employ our encoding functions, which I intentionally want to keep out of the loop here. The test number should not appear in the normal test output, so that adding a test or changing their order does not affect expected output for following tests. The nr is simply for manual invocation, only seen when invoked with -v. Revert - "VLR tests: always print test parameters" b0a4314911140b1599cccfc8171fcdab4cd9bfab. - "Expand VLR tests" d5feadeee8dd24f991df2892d6bcf0be8b0cf707. - "Move IMSI into test parameters" 093300d141c300651954473d73138b72de04d931. Change-Id: Ie1b49237746751021da88f6f07bbb9f780d077c9
2018-03-01 23:40:58 +00:00
===== test_call_mo: SUCCESS
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
full talloc report on 'msgb' (total 0 bytes in 1 blocks)
make gsup ipa name configurable in osmo-msc.cfg Add a 'ipa-name' VTY command which overrides the default IPA name used by the MSC. This is a prerequisite for inter-MSC handover. Related: OS#3355 Change-Id: I317d6c59f77e92fbb2b875a83dc0ec2fa5cb6006
2018-12-06 11:06:59 +00:00
talloc_total_blocks(tall_bsc_ctx) == 13
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
msc_vlr_tests: revert IMSI parameter and test nr output Three recently merged commits take the msc_vlr_tests in a wrong direction. The IMSI is usually encoded in the hex streams. The rationale behind hex streams is that it is a) easily copied from a wireshark trace and b) exactly the bytes as sent by an actual phone. It is hard to parameterize the IMSI because we would have to employ our encoding functions, which I intentionally want to keep out of the loop here. The test number should not appear in the normal test output, so that adding a test or changing their order does not affect expected output for following tests. The nr is simply for manual invocation, only seen when invoked with -v. Revert - "VLR tests: always print test parameters" b0a4314911140b1599cccfc8171fcdab4cd9bfab. - "Expand VLR tests" d5feadeee8dd24f991df2892d6bcf0be8b0cf707. - "Move IMSI into test parameters" 093300d141c300651954473d73138b72de04d931. Change-Id: Ie1b49237746751021da88f6f07bbb9f780d077c9
2018-03-01 23:40:58 +00:00
===== test_call_mt
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
- Total time passed: 0.000000 s
- Location Update request causes a GSUP Send Auth Info request to HLR
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_MM_LOC_UPD_REQUEST
new conn
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn{RAN_CONN_S_NEW}: Allocated
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF unknown: MSC conn use + compl_l3 == 1 (0x1: compl_l3)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DRLL Dispatching 04.08 message GSM48_MT_MM_LOC_UPD_REQUEST (0x5:0x8)
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_NEW}: Updated ID
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DMM LOCATION UPDATING REQUEST: MI(IMSI)=901700000010650 type=NORMAL
DMM LU/new-LAC: 23/23
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_IDLE}: Allocated
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_IDLE}: is child of RAN_conn(LU:901700000010650)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_IDLE}: rev=R99 net=UTRAN Auth+Ciph
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_IDLE}: Received Event VLR_ULA_E_UPDATE_LA
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DREF VLR subscr unknown usage increases to: 1
DVLR set IMSI on subscriber; IMSI=901700000010650 id=901700000010650
DVLR New subscr, IMSI: 901700000010650
DREF VLR subscr IMSI:901700000010650 usage increases to: 2
DREF VLR subscr IMSI:901700000010650 usage decreases to: 1
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_IDLE}: vlr_loc_upd_node1()
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_IDLE}: state_chg to VLR_ULA_S_WAIT_AUTH
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: Allocated
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: is child of vlr_lu_fsm(LU:901700000010650)
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: Received Event VLR_AUTH_E_START
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DVLR GSUP tx: 08010809710000000156f0
GSUP --> HLR: OSMO_GSUP_MSGT_SEND_AUTH_INFO_REQUEST: 08010809710000000156f0
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: state_chg to VLR_SUB_AS_NEEDS_AUTH_WAIT_AI
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_NEW}: Received Event RAN_CONN_E_COMPLETE_LAYER_3
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_NEW}: state_chg to RAN_CONN_S_AUTH_CIPH
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF IMSI:901700000010650: MSC conn use - compl_l3 == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_AUTH_CIPH}: Awaiting results for Auth+Ciph, overruling event RAN_CONN_E_UNUSED
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
lu_result_sent == 0
- from HLR, rx _SEND_AUTH_INFO_RESULT; VLR sends Auth Req to MS
<-- GSUP rx OSMO_GSUP_MSGT_SEND_AUTH_INFO_RESULT: 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
DVLR GSUP rx 511: 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
DREF VLR subscr IMSI:901700000010650 usage increases to: 2
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_NEEDS_AUTH_WAIT_AI}: Received Event VLR_AUTH_E_HLR_SAI_ACK
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DVLR SUBSCR(IMSI:901700000010650) Received 5 auth tuples
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_NEEDS_AUTH_WAIT_AI}: state_chg to VLR_SUB_AS_WAIT_RESP
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_WAIT_RESP}: got auth tuple: use_count=1 key_seq=0 -- will use UMTS AKA (is_r99=yes, at->vec.auth_types=0x3)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
- sending UMTS Auth Request for IMSI:901700000010650: tuple use_count=1 key_seq=0 auth_types=0x3 and...
- ...rand=39fa2f4e3d523d8619a73b4f65c3e14d
- ...autn=8704f5ba55f30000d2ee44b22c8ea919
- ...expecting res=e229c19e791f2e41
DREF VLR subscr IMSI:901700000010650 usage decreases to: 1
<-- GSUP rx OSMO_GSUP_MSGT_SEND_AUTH_INFO_RESULT: vlr_gsupc_read_cb() returns 0
auth_request_sent == 1
lu_result_sent == 0
- MS sends Authen Response, VLR accepts and sends SecurityModeControl
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_MM_AUTH_RESP
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF IMSI:901700000010650: MSC conn use + dtap == 1 (0x2: dtap)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DRLL Dispatching 04.08 message GSM48_MT_MM_AUTH_RESP (0x5:0x14)
cosmetic: gsm48_rx_mm_auth_resp(): log 'UMTS AUTH', not 'R99 AUTH' Change-Id: Iba43c685cbe238d96175267e9cc954b2f2f3e7fc
2018-03-10 03:51:39 +00:00
DMM IMSI:901700000010650: MM UMTS AUTHENTICATION RESPONSE (res = e229c19e791f2e41)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_WAIT_RESP}: Received Event VLR_AUTH_E_MS_AUTH_RESP
cosmetic: vlr_auth_fsm: log RAN and size along with SRES/RES Change-Id: Ib0f9f573ffac2302fbd3ee28f48ccd8fce5fe286
2018-03-10 03:22:50 +00:00
DVLR SUBSCR(IMSI:901700000010650) AUTH on UTRAN received RES: e229c19e791f2e41 (8 bytes)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DVLR SUBSCR(IMSI:901700000010650) AUTH established UMTS security context
refactor VLR FSM result handling Instead of keeping separate enums for FSM results and translating between those and the actual 04.08 reject causes that will ultimately reach the MS, just pass enum gsm48_reject_value cause codes around everywhere. Collapse some VLR *_timeout() and *_cancel() api to just *_cancel() with a gsm48 cause arg. (Hopefully) improve a few reject causes, but otherwise just aim for more transparent decisions on which cause value is used, for future fixes of returned causes. Depends: I6661f139e68a498fb1bef10c266c2f064b72774a (libosmocore) Change-Id: I27bf8d68737ff1f8dc6d11fb1eac3d391aab0cb1
2018-04-06 00:57:51 +00:00
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_WAIT_RESP}: Authentication terminating with result PASSED
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_WAIT_RESP}: state_chg to VLR_SUB_AS_AUTHENTICATED
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Removing from parent vlr_lu_fsm(LU:901700000010650)
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Freeing instance
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Deallocated
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_AUTH}: Received Event VLR_ULA_E_AUTH_RES
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_AUTH}: vlr_loc_upd_post_auth()
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_AUTH}: Set Ciphering Mode
msc_vlr_tests: improve cipher mode coverage Actually call msc_vlr_set_ciph_mode() and wrap away a_iface_tx_cipher_mode() and ranap_iu_tx_sec_mode_cmd(). Hence we'll see decisions and errors in msc_vlr_set_ciph_mode() as well. Change-Id: Id23bc245d4b5707edcd27c44db272fbb211bf9bd
2018-03-02 00:50:09 +00:00
DMM -> SECURITY MODE CONTROL IMSI:901700000010650
- sending SecurityModeControl for UE ctx 42 send_ck=0 new_key=1
- ...ik=27497388b6cb044648f396aa155b95ef
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_AUTH}: state_chg to VLR_ULA_S_WAIT_CIPH
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF IMSI:901700000010650: MSC conn use - dtap == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_AUTH_CIPH}: Awaiting results for Auth+Ciph, overruling event RAN_CONN_E_UNUSED
msc_vlr_tests: clearly separate Ciph Mode from Security Mode checking Clearly distinguish between Ciphering Mode Command on GERAN and Security Mode Control on UTRAN. Cosmetic: explicitly verify the key strings in the testing code (not only in the expected output). Change-Id: Ica93ed06c4c63dc6768736d25231de8068001114
2018-03-10 01:06:47 +00:00
security_mode_ctrl_sent == 1
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
lu_result_sent == 0
- MS sends SecurityModeControl acceptance, VLR accepts and sends GSUP LU Req to HLR
DMM <- SECURITY MODE COMPLETE IMSI:901700000010650
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_CIPH}: Received Event VLR_ULA_E_CIPH_RES
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_CIPH}: vlr_loc_upd_post_ciph()
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DIUCS IMSI:901700000010650: tx CommonID 901700000010650
- Iu Common ID --RAN_UTRAN_IU--> MS (IMSI=901700000010650)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_CIPH}: vlr_loc_upd_node_4()
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_CIPH}: state_chg to VLR_ULA_S_WAIT_HLR_UPD
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_INIT}: Allocated
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_INIT}: is child of vlr_lu_fsm(LU:901700000010650)
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_INIT}: Received Event UPD_HLR_VLR_E_START
GSUP client: send CN domain IE on LU request Give the HLR a chance to send us updated subscriber data by indicating the CN domain to be Circuit Switched, only during a LU Request GSUP message. Adjust msc_vlr_tests to expect the added GSUP CN domain IE to indicate CS, i.e. append '280102'. Related: OS#3601 Change-Id: I0c2d33fbfdb4728e480679120d06b7f3a2ccfd76
2018-09-28 00:41:39 +00:00
DVLR GSUP tx: 04010809710000000156f0280102
GSUP --> HLR: OSMO_GSUP_MSGT_UPDATE_LOCATION_REQUEST: 04010809710000000156f0280102
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_INIT}: state_chg to UPD_HLR_VLR_S_WAIT_FOR_DATA
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
gsup_tx_confirmed == 1
lu_result_sent == 0
- HLR sends _INSERT_DATA_REQUEST, VLR responds with _INSERT_DATA_RESULT
<-- GSUP rx OSMO_GSUP_MSGT_INSERT_DATA_REQUEST: 10010809710000000156f00804032443f2
DVLR GSUP rx 17: 10010809710000000156f00804032443f2
DREF VLR subscr IMSI:901700000010650 usage increases to: 2
DVLR IMSI:901700000010650 has MSISDN:42342
VLR: log subscriber update * move log helpers to generic header * log subscriber update It's handy for troubleshooting issues with subscriber update via GSUP from HLR. Change-Id: I1958aeeb3ea99831c7e2c5ee9a6b59834baf4520
2017-12-27 12:23:44 +00:00
DVLR SUBSCR(MSISDN:42342) VLR: update for IMSI=901700000010650 (MSISDN=42342, used=2)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DVLR GSUP tx: 12010809710000000156f0
GSUP --> HLR: OSMO_GSUP_MSGT_INSERT_DATA_RESULT: 12010809710000000156f0
DREF VLR subscr MSISDN:42342 usage decreases to: 1
<-- GSUP rx OSMO_GSUP_MSGT_INSERT_DATA_REQUEST: vlr_gsupc_read_cb() returns 0
lu_result_sent == 0
- HLR also sends GSUP _UPDATE_LOCATION_RESULT
<-- GSUP rx OSMO_GSUP_MSGT_UPDATE_LOCATION_RESULT: 06010809710000000156f0
DVLR GSUP rx 11: 06010809710000000156f0
DREF VLR subscr MSISDN:42342 usage increases to: 2
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_HLR_UPD}: Received Event VLR_ULA_E_HLR_LU_RES
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_WAIT_FOR_DATA}: Received Event UPD_HLR_VLR_E_UPD_LOC_ACK
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_WAIT_FOR_DATA}: state_chg to UPD_HLR_VLR_S_DONE
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_DONE}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_DONE}: Removing from parent vlr_lu_fsm(LU:901700000010650)
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_DONE}: Freeing instance
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_DONE}: Deallocated
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_HLR_UPD}: Received Event VLR_ULA_E_UPD_HLR_COMPL
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_HLR_UPD}: state_chg to VLR_ULA_S_WAIT_LU_COMPL
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_INIT}: Allocated
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_INIT}: is child of vlr_lu_fsm(LU:901700000010650)
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_INIT}: Received Event LU_COMPL_VLR_E_START
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_INIT}: state_chg to LU_COMPL_VLR_S_WAIT_SUB_PRES
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_WAIT_SUB_PRES}: Received Event LU_COMPL_VLR_E_SUB_PRES_COMPL
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_WAIT_SUB_PRES}: lu_compl_vlr_new_tmsi()
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_WAIT_SUB_PRES}: state_chg to LU_COMPL_VLR_S_WAIT_TMSI_CNF
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
- sending LU Accept for MSISDN:42342, with TMSI 0x03020100
DREF VLR subscr MSISDN:42342 usage decreases to: 1
<-- GSUP rx OSMO_GSUP_MSGT_UPDATE_LOCATION_RESULT: vlr_gsupc_read_cb() returns 0
lu_result_sent == 1
- a LU Accept with a new TMSI was sent, waiting for TMSI Realloc Compl
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
llist_count(&net->ran_conns) == 1
ran_conn_is_accepted() == false
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
- MS sends TMSI Realloc Complete
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_MM_TMSI_REALL_COMPL
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + dtap == 1 (0x2: dtap)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DRLL Dispatching 04.08 message GSM48_MT_MM_TMSI_REALL_COMPL (0x5:0x1b)
DMM TMSI Reallocation Completed. Subscriber: MSISDN:42342
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_LU_COMPL}: Received Event VLR_ULA_E_NEW_TMSI_ACK
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_WAIT_TMSI_CNF}: Received Event LU_COMPL_VLR_E_NEW_TMSI_ACK
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DREF VLR subscr MSISDN:42342 usage increases to: 2
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_WAIT_TMSI_CNF}: state_chg to LU_COMPL_VLR_S_DONE
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_LU_COMPL}: Received Event VLR_ULA_E_LU_COMPL_SUCCESS
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_DONE}: Terminating (cause = OSMO_FSM_TERM_PARENT)
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_DONE}: Removing from parent vlr_lu_fsm(LU:901700000010650)
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_DONE}: Freeing instance
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_DONE}: Deallocated
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_LU_COMPL}: state_chg to VLR_ULA_S_DONE
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_ACCEPTED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_AUTH_CIPH}: state_chg to RAN_CONN_S_ACCEPTED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_ACCEPTED}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_ACCEPTED}: state_chg to RAN_CONN_S_RELEASING
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + release == 2 (0x102: dtap,release)
refactor subscr_conn and subscr_conn_fsm de-/alloc Refactor: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. 2. Add separate AUTH_CIPH state to the FSM. 3. Use conn->use_count to trigger conn release. 4. Add separate RELEASING state to the FSM. 5. Add rate counters for each of the three Complete Layer 3 types. Details: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. Historically, a gsm_subscriber_connection was allocated in libbsc land, and only upon Complete Layer 3 did libmsc add the fsm instance. After splitting openbsc.git into a separate osmo-msc, this is no longer necessary, hence: Closely tie gsm_subscriber_connection allocation to the subscr_conn_fsm instance: talloc the conn as a child of the FSM instance, and discard the conn as soon as the FSM terminates. 2. Add separate AUTH_CIPH state to the FSM. Decoding the Complete Layer 3 message is distinctly separate from waiting for the VLR FSMs to conclude. Use the NEW state as "we don't know if this is a valid message yet", and the AUTH_CIPH state as "evaluating, don't release". A profound effect of this: should we for any odd reason fail to leave the FSM's NEW state, the conn will be released right at the end of msc_compl_l3(), without needing to trigger release in each code path. 3. Use conn->use_count to trigger conn release. Before, the FSM itself would hold a use count on the conn, and hence we would need to ask it whether it is ready to release the conn yet by dispatching events, to achieve a use_count decrement. Instead, unite the FSM instance and conn, and do not hold a use count by the FSM. Hence, trigger an FSM "UNUSED" event only when the use_count reaches zero. As long as use counts are done correctly, the FSM will terminate correctly. These exceptions: - The new AUTH_CIPH state explicitly ignores UNUSED events, since we expect the use count to reach zero while evaluating Authentication and Ciphering. (I experimented with holding a use count by AUTH_CIPH onenter() and releasing by onleave(), but the use count and thus the conn are released before the next state can initiate transactions that would increment the use count again. Same thing for the VLR FSMs holding a use count, they should be done before we advance to the next state. The easiest is to simply expect zero use count during the AUTH_CIPH state.) - A CM Service Request means that even though the MSC would be through with all it wants to do, we shall still wait for a request to follow from the MS. Hence the FSM holds a use count on itself while a CM Service is pending. - While waiting for a Release/Clear Complete, the FSM holds a use count on itself. 4. Add separate RELEASING state to the FSM. If we decide to release for other reasons than a use count reaching zero, we still need to be able to wait for the msc_dtap() use count on the conn to release. (An upcoming patch will further use the RELEASING state to properly wait for Clear Complete / Release Complete messages.) 5. Add rate counters for each of the three Complete Layer 3 types. Besides LU, also count CM Service Request and Paging Response acceptance/rejections. Without these counters, only very few of the auth+ciph outcomes actually show in the counters. Related: OS#3122 Change-Id: I55feb379e176a96a831e105b86202b17a0ffe889
2018-03-30 22:02:14 +00:00
DREF VLR subscr MSISDN:42342 usage increases to: 3
DREF VLR subscr MSISDN:42342 usage decreases to: 2
- Iu Release --RAN_UTRAN_IU--> MS
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - dtap == 1 (0x100: release)
properly receive BSSMAP Clear Complete and Iu Release Complete When sending a BSSMAP Clear or Iu Release, do not immediately discard the conn, but wait until a BSSMAP Clear Complete / Iu Release Complete has been received. Hence we will no longer show in the log that an incoming Release/Clear Complete belongs to an unknown subscriber, but will still be around to properly log the release. Related: OS#3122 Change-Id: Ie4c6aaba3866d6e5b98004e8870a215e8cf8ffc1
2018-04-01 18:55:54 +00:00
iu_release_sent == 1
- LU was successful, and the conn has already been closed
- RNC sends Iu Release Complete
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - release == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_RELEASING}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_RELEASING}: state_chg to RAN_CONN_S_RELEASED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_RELEASED}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_DONE}: Terminating (cause = OSMO_FSM_TERM_PARENT)
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_DONE}: Removing from parent RAN_conn(LU:901700000010650)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_DONE}: fsm_lu_cleanup called with cause OSMO_FSM_TERM_PARENT
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_DONE}: Freeing instance
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_DONE}: Deallocated
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DRLL MSISDN:42342: Freeing RAN connection
refactor subscr_conn and subscr_conn_fsm de-/alloc Refactor: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. 2. Add separate AUTH_CIPH state to the FSM. 3. Use conn->use_count to trigger conn release. 4. Add separate RELEASING state to the FSM. 5. Add rate counters for each of the three Complete Layer 3 types. Details: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. Historically, a gsm_subscriber_connection was allocated in libbsc land, and only upon Complete Layer 3 did libmsc add the fsm instance. After splitting openbsc.git into a separate osmo-msc, this is no longer necessary, hence: Closely tie gsm_subscriber_connection allocation to the subscr_conn_fsm instance: talloc the conn as a child of the FSM instance, and discard the conn as soon as the FSM terminates. 2. Add separate AUTH_CIPH state to the FSM. Decoding the Complete Layer 3 message is distinctly separate from waiting for the VLR FSMs to conclude. Use the NEW state as "we don't know if this is a valid message yet", and the AUTH_CIPH state as "evaluating, don't release". A profound effect of this: should we for any odd reason fail to leave the FSM's NEW state, the conn will be released right at the end of msc_compl_l3(), without needing to trigger release in each code path. 3. Use conn->use_count to trigger conn release. Before, the FSM itself would hold a use count on the conn, and hence we would need to ask it whether it is ready to release the conn yet by dispatching events, to achieve a use_count decrement. Instead, unite the FSM instance and conn, and do not hold a use count by the FSM. Hence, trigger an FSM "UNUSED" event only when the use_count reaches zero. As long as use counts are done correctly, the FSM will terminate correctly. These exceptions: - The new AUTH_CIPH state explicitly ignores UNUSED events, since we expect the use count to reach zero while evaluating Authentication and Ciphering. (I experimented with holding a use count by AUTH_CIPH onenter() and releasing by onleave(), but the use count and thus the conn are released before the next state can initiate transactions that would increment the use count again. Same thing for the VLR FSMs holding a use count, they should be done before we advance to the next state. The easiest is to simply expect zero use count during the AUTH_CIPH state.) - A CM Service Request means that even though the MSC would be through with all it wants to do, we shall still wait for a request to follow from the MS. Hence the FSM holds a use count on itself while a CM Service is pending. - While waiting for a Release/Clear Complete, the FSM holds a use count on itself. 4. Add separate RELEASING state to the FSM. If we decide to release for other reasons than a use count reaching zero, we still need to be able to wait for the msc_dtap() use count on the conn to release. (An upcoming patch will further use the RELEASING state to properly wait for Clear Complete / Release Complete messages.) 5. Add rate counters for each of the three Complete Layer 3 types. Besides LU, also count CM Service Request and Paging Response acceptance/rejections. Without these counters, only very few of the auth+ciph outcomes actually show in the counters. Related: OS#3122 Change-Id: I55feb379e176a96a831e105b86202b17a0ffe889
2018-03-30 22:02:14 +00:00
DREF VLR subscr MSISDN:42342 usage decreases to: 1
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_RELEASED}: Freeing instance
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_RELEASED}: Deallocated
llist_count(&net->ran_conns) == 0
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DREF VLR subscr MSISDN:42342 usage increases to: 2
vsub != NULL == 1
strcmp(vsub->imsi, IMSI) == 0
vsub->lac == 23
DREF VLR subscr MSISDN:42342 usage decreases to: 1
---
- after a while, MNCC asks us to setup a call, causing Paging
DMNCC receive message MNCC_SETUP_REQ
DREF VLR subscr MSISDN:42342 usage increases to: 2
cosmetic: msc_vlr_tests: enable CC logging in debug There are a number of bad failures in CC teardown handling we're solving. It helps to see CC logging in the msc_vlr_tests. Change-Id: I56ac269d46b48b6b85efad81c4d2343bfc41ea90
2018-03-12 22:59:37 +00:00
DCC (ti ff sub MSISDN:42342 callref 423) New transaction
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DREF VLR subscr MSISDN:42342 usage increases to: 3
DMM Subscriber MSISDN:42342 not paged yet, start paging.
RAN_UTRAN_IU sends out paging request to IMSI 901700000010650, TMSI 0x03020100, LAC 23
strcmp(paging_expecting_imsi, imsi) == 0
DREF VLR subscr MSISDN:42342 usage increases to: 4
DREF VLR subscr MSISDN:42342 usage decreases to: 3
paging_sent == 1
paging_stopped == 0
- MS replies with Paging Response, and VLR sends Auth Request
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_RR_PAG_RESP
new conn
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn{RAN_CONN_S_NEW}: Allocated
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF unknown: MSC conn use + compl_l3 == 1 (0x1: compl_l3)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DRLL Dispatching 04.08 message GSM48_MT_RR_PAG_RESP (0x6:0x27)
DRR PAGING RESPONSE: MI(IMSI)=901700000010650
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_NEW}: Updated ID
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_INIT}: Allocated
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_INIT}: is child of RAN_conn(PAGING_RESP:901700000010650)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_INIT}: rev=R99 net=UTRAN Auth+Ciph
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_INIT}: Received Event PR_ARQ_E_START
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DREF VLR subscr MSISDN:42342 usage increases to: 4
DREF VLR subscr MSISDN:42342 usage increases to: 5
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_INIT}: proc_arq_vlr_fn_post_imsi()
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_INIT}: state_chg to PR_ARQ_S_WAIT_AUTH
DVLR VLR_Authenticate(PAGING_RESP:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: Allocated
DVLR VLR_Authenticate(PAGING_RESP:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: is child of Process_Access_Request_VLR(PAGING_RESP:901700000010650)
DVLR VLR_Authenticate(PAGING_RESP:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: Received Event VLR_AUTH_E_START
DVLR VLR_Authenticate(PAGING_RESP:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: state_chg to VLR_SUB_AS_WAIT_RESP
DVLR VLR_Authenticate(PAGING_RESP:901700000010650){VLR_SUB_AS_WAIT_RESP}: got auth tuple: use_count=1 key_seq=1 -- will use UMTS AKA (is_r99=yes, at->vec.auth_types=0x3)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
- sending UMTS Auth Request for MSISDN:42342: tuple use_count=1 key_seq=1 auth_types=0x3 and...
- ...rand=c187a53a5e6b9d573cac7c74451fd46d
- ...autn=1843a645b98d00005b2d666af46c45d9
- ...expecting res=7db47cf7f81e4dc7
DREF VLR subscr MSISDN:42342 usage decreases to: 4
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_NEW}: Received Event RAN_CONN_E_COMPLETE_LAYER_3
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_NEW}: state_chg to RAN_CONN_S_AUTH_CIPH
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - compl_l3 == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_AUTH_CIPH}: Awaiting results for Auth+Ciph, overruling event RAN_CONN_E_UNUSED
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
auth_request_sent == 1
- MS sends Authen Response, VLR accepts and sends SecurityModeControl
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_MM_AUTH_RESP
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + dtap == 1 (0x2: dtap)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DRLL Dispatching 04.08 message GSM48_MT_MM_AUTH_RESP (0x5:0x14)
cosmetic: gsm48_rx_mm_auth_resp(): log 'UMTS AUTH', not 'R99 AUTH' Change-Id: Iba43c685cbe238d96175267e9cc954b2f2f3e7fc
2018-03-10 03:51:39 +00:00
DMM MSISDN:42342: MM UMTS AUTHENTICATION RESPONSE (res = 7db47cf7f81e4dc7)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(PAGING_RESP:901700000010650){VLR_SUB_AS_WAIT_RESP}: Received Event VLR_AUTH_E_MS_AUTH_RESP
cosmetic: vlr_auth_fsm: log RAN and size along with SRES/RES Change-Id: Ib0f9f573ffac2302fbd3ee28f48ccd8fce5fe286
2018-03-10 03:22:50 +00:00
DVLR SUBSCR(MSISDN:42342) AUTH on UTRAN received RES: 7db47cf7f81e4dc7 (8 bytes)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DVLR SUBSCR(MSISDN:42342) AUTH established UMTS security context
refactor VLR FSM result handling Instead of keeping separate enums for FSM results and translating between those and the actual 04.08 reject causes that will ultimately reach the MS, just pass enum gsm48_reject_value cause codes around everywhere. Collapse some VLR *_timeout() and *_cancel() api to just *_cancel() with a gsm48 cause arg. (Hopefully) improve a few reject causes, but otherwise just aim for more transparent decisions on which cause value is used, for future fixes of returned causes. Depends: I6661f139e68a498fb1bef10c266c2f064b72774a (libosmocore) Change-Id: I27bf8d68737ff1f8dc6d11fb1eac3d391aab0cb1
2018-04-06 00:57:51 +00:00
DVLR VLR_Authenticate(PAGING_RESP:901700000010650){VLR_SUB_AS_WAIT_RESP}: Authentication terminating with result PASSED
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(PAGING_RESP:901700000010650){VLR_SUB_AS_WAIT_RESP}: state_chg to VLR_SUB_AS_AUTHENTICATED
DVLR VLR_Authenticate(PAGING_RESP:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
DVLR VLR_Authenticate(PAGING_RESP:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Removing from parent Process_Access_Request_VLR(PAGING_RESP:901700000010650)
DVLR VLR_Authenticate(PAGING_RESP:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Freeing instance
DVLR VLR_Authenticate(PAGING_RESP:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Deallocated
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_WAIT_AUTH}: Received Event PR_ARQ_E_AUTH_RES
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_WAIT_AUTH}: _proc_arq_vlr_node2()
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_WAIT_AUTH}: Set Ciphering Mode
msc_vlr_tests: improve cipher mode coverage Actually call msc_vlr_set_ciph_mode() and wrap away a_iface_tx_cipher_mode() and ranap_iu_tx_sec_mode_cmd(). Hence we'll see decisions and errors in msc_vlr_set_ciph_mode() as well. Change-Id: Id23bc245d4b5707edcd27c44db272fbb211bf9bd
2018-03-02 00:50:09 +00:00
DMM -> SECURITY MODE CONTROL MSISDN:42342
- sending SecurityModeControl for UE ctx 42 send_ck=0 new_key=1
- ...ik=1159ec926a50e98c034a6b7d7c9f418d
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_WAIT_AUTH}: state_chg to PR_ARQ_S_WAIT_CIPH
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - dtap == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_AUTH_CIPH}: Awaiting results for Auth+Ciph, overruling event RAN_CONN_E_UNUSED
msc_vlr_tests: clearly separate Ciph Mode from Security Mode checking Clearly distinguish between Ciphering Mode Command on GERAN and Security Mode Control on UTRAN. Cosmetic: explicitly verify the key strings in the testing code (not only in the expected output). Change-Id: Ica93ed06c4c63dc6768736d25231de8068001114
2018-03-10 01:06:47 +00:00
security_mode_ctrl_sent == 1
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
- MS sends SecurityModeControl acceptance, VLR accepts, sends CC Setup
DMM <- SECURITY MODE COMPLETE MSISDN:42342
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_WAIT_CIPH}: Received Event PR_ARQ_E_CIPH_RES
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_WAIT_CIPH}: _proc_arq_vlr_node2_post_ciph()
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DIUCS MSISDN:42342: tx CommonID 901700000010650
- Iu Common ID --RAN_UTRAN_IU--> MS (IMSI=901700000010650)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_WAIT_CIPH}: _proc_arq_vlr_node2_post_vlr()
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_WAIT_CIPH}: _proc_arq_vlr_post_pres()
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_WAIT_CIPH}: _proc_arq_vlr_post_trace()
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_WAIT_CIPH}: _proc_arq_vlr_post_imei()
refactor VLR FSM result handling Instead of keeping separate enums for FSM results and translating between those and the actual 04.08 reject causes that will ultimately reach the MS, just pass enum gsm48_reject_value cause codes around everywhere. Collapse some VLR *_timeout() and *_cancel() api to just *_cancel() with a gsm48 cause arg. (Hopefully) improve a few reject causes, but otherwise just aim for more transparent decisions on which cause value is used, for future fixes of returned causes. Depends: I6661f139e68a498fb1bef10c266c2f064b72774a (libosmocore) Change-Id: I27bf8d68737ff1f8dc6d11fb1eac3d391aab0cb1
2018-04-06 00:57:51 +00:00
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_WAIT_CIPH}: proc_arq_fsm_done(PASSED)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_WAIT_CIPH}: state_chg to PR_ARQ_S_DONE
refactor VLR FSM result handling Instead of keeping separate enums for FSM results and translating between those and the actual 04.08 reject causes that will ultimately reach the MS, just pass enum gsm48_reject_value cause codes around everywhere. Collapse some VLR *_timeout() and *_cancel() api to just *_cancel() with a gsm48 cause arg. (Hopefully) improve a few reject causes, but otherwise just aim for more transparent decisions on which cause value is used, for future fixes of returned causes. Depends: I6661f139e68a498fb1bef10c266c2f064b72774a (libosmocore) Change-Id: I27bf8d68737ff1f8dc6d11fb1eac3d391aab0cb1
2018-04-06 00:57:51 +00:00
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_DONE}: Process Access Request result: PASSED
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_ACCEPTED
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DPAG Paging success for MSISDN:42342 (event=0)
DPAG Calling paging cbfn.
cosmetic: msc_vlr_tests: enable CC logging in debug There are a number of bad failures in CC teardown handling we're solving. It helps to see CC logging in the msc_vlr_tests. Change-Id: I56ac269d46b48b6b85efad81c4d2343bfc41ea90
2018-03-12 22:59:37 +00:00
DCC Paging subscr 42342 succeeded!
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + trans_cc == 1 (0x10: trans_cc)
cosmetic: msc_vlr_tests: enable CC logging in debug There are a number of bad failures in CC teardown handling we're solving. It helps to see CC logging in the msc_vlr_tests. Change-Id: I56ac269d46b48b6b85efad81c4d2343bfc41ea90
2018-03-12 22:59:37 +00:00
DCC starting timer T303 with 30 seconds
DCC (ti 00 sub MSISDN:42342) new state NULL -> CALL_PRESENT
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DMSC msc_tx 2 bytes to MSISDN:42342 via RAN_UTRAN_IU
- DTAP --RAN_UTRAN_IU--> MS: GSM48_MT_CC_SETUP: 0305
- DTAP matches expected message
DREF VLR subscr MSISDN:42342 usage decreases to: 3
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_AUTH_CIPH}: state_chg to RAN_CONN_S_ACCEPTED
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_ACCEPTED}: ran_conn_fsm_has_active_transactions: connection still has active transaction: CC
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
paging_stopped == 1
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_CC_CALL_CONF
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + dtap == 2 (0x12: dtap,trans_cc)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DRLL Dispatching 04.08 message GSM48_MT_CC_CALL_CONF (0x3:0x8)
cosmetic: msc_vlr_tests: enable CC logging in debug There are a number of bad failures in CC teardown handling we're solving. It helps to see CC logging in the msc_vlr_tests. Change-Id: I56ac269d46b48b6b85efad81c4d2343bfc41ea90
2018-03-12 22:59:37 +00:00
DCC stopping pending timer T303
DCC starting timer T310 with 30 seconds
DCC (ti 00 sub MSISDN:42342) new state CALL_PRESENT -> MO_TERM_CALL_CONF
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
MS <--Call Assignment-- MSC: subscr=MSISDN:42342 callref=0x423
DMNCC transmit message MNCC_CALL_CONF_IND
cosmetic: msc_vlr_tests: enable CC logging in debug There are a number of bad failures in CC teardown handling we're solving. It helps to see CC logging in the msc_vlr_tests. Change-Id: I56ac269d46b48b6b85efad81c4d2343bfc41ea90
2018-03-12 22:59:37 +00:00
DCC Sending 'MNCC_CALL_CONF_IND' to MNCC.
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
MSC --> MNCC: callref 0x423: MNCC_CALL_CONF_IND
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_ACCEPTED}: Received Event RAN_CONN_E_COMMUNICATING
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_ACCEPTED}: state_chg to RAN_CONN_S_COMMUNICATING
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - dtap == 1 (0x10: trans_cc)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
- Total time passed: 1.000023 s
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_CC_ALERTING
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + dtap == 2 (0x12: dtap,trans_cc)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DRLL Dispatching 04.08 message GSM48_MT_CC_ALERTING (0x3:0x1)
cosmetic: msc_vlr_tests: enable CC logging in debug There are a number of bad failures in CC teardown handling we're solving. It helps to see CC logging in the msc_vlr_tests. Change-Id: I56ac269d46b48b6b85efad81c4d2343bfc41ea90
2018-03-12 22:59:37 +00:00
DCC stopping pending timer T310
DCC starting timer T301 with 180 seconds
DCC (ti 00 sub MSISDN:42342) new state MO_TERM_CALL_CONF -> CALL_RECEIVED
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DMNCC transmit message MNCC_ALERT_IND
cosmetic: msc_vlr_tests: enable CC logging in debug There are a number of bad failures in CC teardown handling we're solving. It helps to see CC logging in the msc_vlr_tests. Change-Id: I56ac269d46b48b6b85efad81c4d2343bfc41ea90
2018-03-12 22:59:37 +00:00
DCC Sending 'MNCC_ALERT_IND' to MNCC.
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
MSC --> MNCC: callref 0x423: MNCC_ALERT_IND
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_COMMUNICATING}: Received Event RAN_CONN_E_COMMUNICATING
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - dtap == 1 (0x10: trans_cc)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
- Total time passed: 2.000046 s
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_CC_CONNECT
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + dtap == 2 (0x12: dtap,trans_cc)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DRLL Dispatching 04.08 message GSM48_MT_CC_CONNECT (0x3:0x7)
cosmetic: msc_vlr_tests: enable CC logging in debug There are a number of bad failures in CC teardown handling we're solving. It helps to see CC logging in the msc_vlr_tests. Change-Id: I56ac269d46b48b6b85efad81c4d2343bfc41ea90
2018-03-12 22:59:37 +00:00
DCC stopping pending timer T301
DCC (ti 00 sub MSISDN:42342) new state CALL_RECEIVED -> CONNECT_REQUEST
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DMNCC transmit message MNCC_SETUP_CNF
cosmetic: msc_vlr_tests: enable CC logging in debug There are a number of bad failures in CC teardown handling we're solving. It helps to see CC logging in the msc_vlr_tests. Change-Id: I56ac269d46b48b6b85efad81c4d2343bfc41ea90
2018-03-12 22:59:37 +00:00
DCC Sending 'MNCC_SETUP_CNF' to MNCC.
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
MSC --> MNCC: callref 0x423: MNCC_SETUP_CNF
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_COMMUNICATING}: Received Event RAN_CONN_E_COMMUNICATING
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - dtap == 1 (0x10: trans_cc)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DMNCC receive message MNCC_SETUP_COMPL_REQ
gsm_04_08_cc: Add global guard timer for MNCC The external MNCC handler may hang indefinitely in cases where the remote end of the MNCC ceases to work properly. Add a global guard timer to make sure the call reaches ACTIVE state. Change-Id: I7375d1e17cd746aac4eadfe1e587e82cf1630d3d Related: OS#3599
2018-10-10 15:00:49 +00:00
DCC (sub 42342) starting guard timer with 180 seconds
cosmetic: msc_vlr_tests: enable CC logging in debug There are a number of bad failures in CC teardown handling we're solving. It helps to see CC logging in the msc_vlr_tests. Change-Id: I56ac269d46b48b6b85efad81c4d2343bfc41ea90
2018-03-12 22:59:37 +00:00
DCC (ti 00 sub 42342) Received 'MNCC_SETUP_COMPL_REQ' from MNCC in state 8 (CONNECT_REQUEST)
DCC (ti 00 sub MSISDN:42342) new state CONNECT_REQUEST -> ACTIVE
gsm_04_08_cc: Add global guard timer for MNCC The external MNCC handler may hang indefinitely in cases where the remote end of the MNCC ceases to work properly. Add a global guard timer to make sure the call reaches ACTIVE state. Change-Id: I7375d1e17cd746aac4eadfe1e587e82cf1630d3d Related: OS#3599
2018-10-10 15:00:49 +00:00
DCC (sub 42342) stopping pending guard timer
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DMSC msc_tx 2 bytes to MSISDN:42342 via RAN_UTRAN_IU
- DTAP --RAN_UTRAN_IU--> MS: GSM48_MT_CC_CONNECT_ACK: 030f
- DTAP matches expected message
---
- RTP stream goes ahead, not shown here.
- Total time passed: 125.000091 s
---
- Call ends
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_CC_DISCONNECT
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + dtap == 2 (0x12: dtap,trans_cc)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DRLL Dispatching 04.08 message GSM48_MT_CC_DISCONNECT (0x3:0x25)
cosmetic: msc_vlr_tests: enable CC logging in debug There are a number of bad failures in CC teardown handling we're solving. It helps to see CC logging in the msc_vlr_tests. Change-Id: I56ac269d46b48b6b85efad81c4d2343bfc41ea90
2018-03-12 22:59:37 +00:00
DCC (ti 00 sub MSISDN:42342) new state ACTIVE -> DISCONNECT_IND
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DMNCC transmit message MNCC_DISC_IND
cosmetic: msc_vlr_tests: enable CC logging in debug There are a number of bad failures in CC teardown handling we're solving. It helps to see CC logging in the msc_vlr_tests. Change-Id: I56ac269d46b48b6b85efad81c4d2343bfc41ea90
2018-03-12 22:59:37 +00:00
DCC Sending 'MNCC_DISC_IND' to MNCC.
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
MSC --> MNCC: callref 0x423: MNCC_DISC_IND
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_COMMUNICATING}: Received Event RAN_CONN_E_COMMUNICATING
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - dtap == 1 (0x10: trans_cc)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DMNCC receive message MNCC_REL_REQ
gsm_04_08_cc: Add global guard timer for MNCC The external MNCC handler may hang indefinitely in cases where the remote end of the MNCC ceases to work properly. Add a global guard timer to make sure the call reaches ACTIVE state. Change-Id: I7375d1e17cd746aac4eadfe1e587e82cf1630d3d Related: OS#3599
2018-10-10 15:00:49 +00:00
DCC (sub 42342) starting guard timer with 180 seconds
cosmetic: msc_vlr_tests: enable CC logging in debug There are a number of bad failures in CC teardown handling we're solving. It helps to see CC logging in the msc_vlr_tests. Change-Id: I56ac269d46b48b6b85efad81c4d2343bfc41ea90
2018-03-12 22:59:37 +00:00
DCC (ti 00 sub 42342) Received 'MNCC_REL_REQ' from MNCC in state 12 (DISCONNECT_IND)
DCC starting timer T308 with 10 seconds
DCC (ti 00 sub MSISDN:42342) new state DISCONNECT_IND -> RELEASE_REQ
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DMSC msc_tx 2 bytes to MSISDN:42342 via RAN_UTRAN_IU
- DTAP --RAN_UTRAN_IU--> MS: GSM48_MT_CC_RELEASE: 032d
- DTAP matches expected message
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_CC_RELEASE_COMPL
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + dtap == 2 (0x12: dtap,trans_cc)
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DRLL Dispatching 04.08 message GSM48_MT_CC_RELEASE_COMPL (0x3:0x2a)
cosmetic: msc_vlr_tests: enable CC logging in debug There are a number of bad failures in CC teardown handling we're solving. It helps to see CC logging in the msc_vlr_tests. Change-Id: I56ac269d46b48b6b85efad81c4d2343bfc41ea90
2018-03-12 22:59:37 +00:00
DCC stopping pending timer T308
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DMNCC transmit message MNCC_REL_CNF
cosmetic: msc_vlr_tests: enable CC logging in debug There are a number of bad failures in CC teardown handling we're solving. It helps to see CC logging in the msc_vlr_tests. Change-Id: I56ac269d46b48b6b85efad81c4d2343bfc41ea90
2018-03-12 22:59:37 +00:00
DCC Sending 'MNCC_REL_CNF' to MNCC.
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
MSC --> MNCC: callref 0x423: MNCC_REL_CNF
MS <--Call Release-- MSC: subscr=MSISDN:42342 callref=0x0
cosmetic: msc_vlr_tests: enable CC logging in debug There are a number of bad failures in CC teardown handling we're solving. It helps to see CC logging in the msc_vlr_tests. Change-Id: I56ac269d46b48b6b85efad81c4d2343bfc41ea90
2018-03-12 22:59:37 +00:00
DCC (ti 00 sub MSISDN:42342) new state RELEASE_REQ -> NULL
gsm_04_08_cc: Add global guard timer for MNCC The external MNCC handler may hang indefinitely in cases where the remote end of the MNCC ceases to work properly. Add a global guard timer to make sure the call reaches ACTIVE state. Change-Id: I7375d1e17cd746aac4eadfe1e587e82cf1630d3d Related: OS#3599
2018-10-10 15:00:49 +00:00
DCC (sub 42342) stopping pending guard timer
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DREF VLR subscr MSISDN:42342 usage decreases to: 2
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - trans_cc == 1 (0x2: dtap)
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_COMMUNICATING}: Received Event RAN_CONN_E_COMMUNICATING
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - dtap == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_COMMUNICATING}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_COMMUNICATING}: state_chg to RAN_CONN_S_RELEASING
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + release == 1 (0x100: release)
refactor subscr_conn and subscr_conn_fsm de-/alloc Refactor: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. 2. Add separate AUTH_CIPH state to the FSM. 3. Use conn->use_count to trigger conn release. 4. Add separate RELEASING state to the FSM. 5. Add rate counters for each of the three Complete Layer 3 types. Details: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. Historically, a gsm_subscriber_connection was allocated in libbsc land, and only upon Complete Layer 3 did libmsc add the fsm instance. After splitting openbsc.git into a separate osmo-msc, this is no longer necessary, hence: Closely tie gsm_subscriber_connection allocation to the subscr_conn_fsm instance: talloc the conn as a child of the FSM instance, and discard the conn as soon as the FSM terminates. 2. Add separate AUTH_CIPH state to the FSM. Decoding the Complete Layer 3 message is distinctly separate from waiting for the VLR FSMs to conclude. Use the NEW state as "we don't know if this is a valid message yet", and the AUTH_CIPH state as "evaluating, don't release". A profound effect of this: should we for any odd reason fail to leave the FSM's NEW state, the conn will be released right at the end of msc_compl_l3(), without needing to trigger release in each code path. 3. Use conn->use_count to trigger conn release. Before, the FSM itself would hold a use count on the conn, and hence we would need to ask it whether it is ready to release the conn yet by dispatching events, to achieve a use_count decrement. Instead, unite the FSM instance and conn, and do not hold a use count by the FSM. Hence, trigger an FSM "UNUSED" event only when the use_count reaches zero. As long as use counts are done correctly, the FSM will terminate correctly. These exceptions: - The new AUTH_CIPH state explicitly ignores UNUSED events, since we expect the use count to reach zero while evaluating Authentication and Ciphering. (I experimented with holding a use count by AUTH_CIPH onenter() and releasing by onleave(), but the use count and thus the conn are released before the next state can initiate transactions that would increment the use count again. Same thing for the VLR FSMs holding a use count, they should be done before we advance to the next state. The easiest is to simply expect zero use count during the AUTH_CIPH state.) - A CM Service Request means that even though the MSC would be through with all it wants to do, we shall still wait for a request to follow from the MS. Hence the FSM holds a use count on itself while a CM Service is pending. - While waiting for a Release/Clear Complete, the FSM holds a use count on itself. 4. Add separate RELEASING state to the FSM. If we decide to release for other reasons than a use count reaching zero, we still need to be able to wait for the msc_dtap() use count on the conn to release. (An upcoming patch will further use the RELEASING state to properly wait for Clear Complete / Release Complete messages.) 5. Add rate counters for each of the three Complete Layer 3 types. Besides LU, also count CM Service Request and Paging Response acceptance/rejections. Without these counters, only very few of the auth+ciph outcomes actually show in the counters. Related: OS#3122 Change-Id: I55feb379e176a96a831e105b86202b17a0ffe889
2018-03-30 22:02:14 +00:00
DREF VLR subscr MSISDN:42342 usage increases to: 3
DREF VLR subscr MSISDN:42342 usage decreases to: 2
- Iu Release --RAN_UTRAN_IU--> MS
properly receive BSSMAP Clear Complete and Iu Release Complete When sending a BSSMAP Clear or Iu Release, do not immediately discard the conn, but wait until a BSSMAP Clear Complete / Iu Release Complete has been received. Hence we will no longer show in the log that an incoming Release/Clear Complete belongs to an unknown subscriber, but will still be around to properly log the release. Related: OS#3122 Change-Id: Ie4c6aaba3866d6e5b98004e8870a215e8cf8ffc1
2018-04-01 18:55:54 +00:00
- RNC sends Iu Release Complete
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - release == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_RELEASING}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_RELEASING}: state_chg to RAN_CONN_S_RELEASED
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_RELEASED}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_DONE}: Terminating (cause = OSMO_FSM_TERM_PARENT)
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_DONE}: Removing from parent RAN_conn(PAGING_RESP:901700000010650)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_DONE}: Freeing instance
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_DONE}: Deallocated
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DRLL MSISDN:42342: Freeing RAN connection
refactor subscr_conn and subscr_conn_fsm de-/alloc Refactor: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. 2. Add separate AUTH_CIPH state to the FSM. 3. Use conn->use_count to trigger conn release. 4. Add separate RELEASING state to the FSM. 5. Add rate counters for each of the three Complete Layer 3 types. Details: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. Historically, a gsm_subscriber_connection was allocated in libbsc land, and only upon Complete Layer 3 did libmsc add the fsm instance. After splitting openbsc.git into a separate osmo-msc, this is no longer necessary, hence: Closely tie gsm_subscriber_connection allocation to the subscr_conn_fsm instance: talloc the conn as a child of the FSM instance, and discard the conn as soon as the FSM terminates. 2. Add separate AUTH_CIPH state to the FSM. Decoding the Complete Layer 3 message is distinctly separate from waiting for the VLR FSMs to conclude. Use the NEW state as "we don't know if this is a valid message yet", and the AUTH_CIPH state as "evaluating, don't release". A profound effect of this: should we for any odd reason fail to leave the FSM's NEW state, the conn will be released right at the end of msc_compl_l3(), without needing to trigger release in each code path. 3. Use conn->use_count to trigger conn release. Before, the FSM itself would hold a use count on the conn, and hence we would need to ask it whether it is ready to release the conn yet by dispatching events, to achieve a use_count decrement. Instead, unite the FSM instance and conn, and do not hold a use count by the FSM. Hence, trigger an FSM "UNUSED" event only when the use_count reaches zero. As long as use counts are done correctly, the FSM will terminate correctly. These exceptions: - The new AUTH_CIPH state explicitly ignores UNUSED events, since we expect the use count to reach zero while evaluating Authentication and Ciphering. (I experimented with holding a use count by AUTH_CIPH onenter() and releasing by onleave(), but the use count and thus the conn are released before the next state can initiate transactions that would increment the use count again. Same thing for the VLR FSMs holding a use count, they should be done before we advance to the next state. The easiest is to simply expect zero use count during the AUTH_CIPH state.) - A CM Service Request means that even though the MSC would be through with all it wants to do, we shall still wait for a request to follow from the MS. Hence the FSM holds a use count on itself while a CM Service is pending. - While waiting for a Release/Clear Complete, the FSM holds a use count on itself. 4. Add separate RELEASING state to the FSM. If we decide to release for other reasons than a use count reaching zero, we still need to be able to wait for the msc_dtap() use count on the conn to release. (An upcoming patch will further use the RELEASING state to properly wait for Clear Complete / Release Complete messages.) 5. Add rate counters for each of the three Complete Layer 3 types. Besides LU, also count CM Service Request and Paging Response acceptance/rejections. Without these counters, only very few of the auth+ciph outcomes actually show in the counters. Related: OS#3122 Change-Id: I55feb379e176a96a831e105b86202b17a0ffe889
2018-03-30 22:02:14 +00:00
DREF VLR subscr MSISDN:42342 usage decreases to: 1
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_RELEASED}: Freeing instance
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_RELEASED}: Deallocated
llist_count(&net->ran_conns) == 0
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
DREF freeing VLR subscr MSISDN:42342
msc_vlr_tests: revert IMSI parameter and test nr output Three recently merged commits take the msc_vlr_tests in a wrong direction. The IMSI is usually encoded in the hex streams. The rationale behind hex streams is that it is a) easily copied from a wireshark trace and b) exactly the bytes as sent by an actual phone. It is hard to parameterize the IMSI because we would have to employ our encoding functions, which I intentionally want to keep out of the loop here. The test number should not appear in the normal test output, so that adding a test or changing their order does not affect expected output for following tests. The nr is simply for manual invocation, only seen when invoked with -v. Revert - "VLR tests: always print test parameters" b0a4314911140b1599cccfc8171fcdab4cd9bfab. - "Expand VLR tests" d5feadeee8dd24f991df2892d6bcf0be8b0cf707. - "Move IMSI into test parameters" 093300d141c300651954473d73138b72de04d931. Change-Id: Ie1b49237746751021da88f6f07bbb9f780d077c9
2018-03-01 23:40:58 +00:00
===== test_call_mt: SUCCESS
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
full talloc report on 'msgb' (total 0 bytes in 1 blocks)
make gsup ipa name configurable in osmo-msc.cfg Add a 'ipa-name' VTY command which overrides the default IPA name used by the MSC. This is a prerequisite for inter-MSC handover. Related: OS#3355 Change-Id: I317d6c59f77e92fbb2b875a83dc0ec2fa5cb6006
2018-12-06 11:06:59 +00:00
talloc_total_blocks(tall_bsc_ctx) == 13
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
===== test_call_mt2
- Total time passed: 0.000000 s
- Location Update request causes a GSUP Send Auth Info request to HLR
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_MM_LOC_UPD_REQUEST
new conn
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn{RAN_CONN_S_NEW}: Allocated
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF unknown: MSC conn use + compl_l3 == 1 (0x1: compl_l3)
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
DRLL Dispatching 04.08 message GSM48_MT_MM_LOC_UPD_REQUEST (0x5:0x8)
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_NEW}: Updated ID
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
DMM LOCATION UPDATING REQUEST: MI(IMSI)=901700000010650 type=NORMAL
DMM LU/new-LAC: 23/23
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_IDLE}: Allocated
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_IDLE}: is child of RAN_conn(LU:901700000010650)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_IDLE}: rev=R99 net=UTRAN Auth+Ciph
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_IDLE}: Received Event VLR_ULA_E_UPDATE_LA
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
DREF VLR subscr unknown usage increases to: 1
DVLR set IMSI on subscriber; IMSI=901700000010650 id=901700000010650
DVLR New subscr, IMSI: 901700000010650
DREF VLR subscr IMSI:901700000010650 usage increases to: 2
DREF VLR subscr IMSI:901700000010650 usage decreases to: 1
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_IDLE}: vlr_loc_upd_node1()
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_IDLE}: state_chg to VLR_ULA_S_WAIT_AUTH
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: Allocated
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: is child of vlr_lu_fsm(LU:901700000010650)
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: Received Event VLR_AUTH_E_START
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
DVLR GSUP tx: 08010809710000000156f0
GSUP --> HLR: OSMO_GSUP_MSGT_SEND_AUTH_INFO_REQUEST: 08010809710000000156f0
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: state_chg to VLR_SUB_AS_NEEDS_AUTH_WAIT_AI
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_NEW}: Received Event RAN_CONN_E_COMPLETE_LAYER_3
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_NEW}: state_chg to RAN_CONN_S_AUTH_CIPH
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF IMSI:901700000010650: MSC conn use - compl_l3 == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_AUTH_CIPH}: Awaiting results for Auth+Ciph, overruling event RAN_CONN_E_UNUSED
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
lu_result_sent == 0
- from HLR, rx _SEND_AUTH_INFO_RESULT; VLR sends Auth Req to MS
<-- GSUP rx OSMO_GSUP_MSGT_SEND_AUTH_INFO_RESULT: 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
DVLR GSUP rx 511: 0a010809710000000156f00362201039fa2f4e3d523d8619a73b4f65c3e14d21049b36efdf2208059a4f668f6fbe39231027497388b6cb044648f396aa155b95ef2410f64735036e5871319c679f4742a75ea125108704f5ba55f30000d2ee44b22c8ea9192708e229c19e791f2e4103622010c187a53a5e6b9d573cac7c74451fd46d210485aa31302208d3d50a000bf04f6e23101159ec926a50e98c034a6b7d7c9f418d2410df3a03d9ca5335641efc8e36d76cd20b25101843a645b98d00005b2d666af46c45d927087db47cf7f81e4dc703622010efa9c29a9742148d5c9070348716e1bb210469d5f9fb22083df176f0c29f1a3d2310eb50e770ddcc3060101d2f43b6c2b884241076542abce5ff9345b0e8947f4c6e019c2510f9375e6d41e1000096e7fe4ff1c27e392708706f996719ba609c03622010f023d5a3b24726e0631b64b3840f82532104d570c03f2208ec011be8919883d62310c4e58af4ba43f3bcd904e16984f086d724100593f65e752e5cb7f473862bda05aa0a2510541ff1f077270000c5ea00d658bc7e9a27083fd26072eaa2a04d036220102f8f90c780d6a9c0c53da7ac57b6707e2104b072446f220823f39f9f425ad6e6231065af0527fda95b0dc5ae4aa515cdf32f2410537c3b35a3b13b08d08eeb28098f45cc25104bf4e564f75300009bc796706bc6574427080edb0eadbea94ac2
DREF VLR subscr IMSI:901700000010650 usage increases to: 2
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_NEEDS_AUTH_WAIT_AI}: Received Event VLR_AUTH_E_HLR_SAI_ACK
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
DVLR SUBSCR(IMSI:901700000010650) Received 5 auth tuples
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_NEEDS_AUTH_WAIT_AI}: state_chg to VLR_SUB_AS_WAIT_RESP
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_WAIT_RESP}: got auth tuple: use_count=1 key_seq=0 -- will use UMTS AKA (is_r99=yes, at->vec.auth_types=0x3)
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
- sending UMTS Auth Request for IMSI:901700000010650: tuple use_count=1 key_seq=0 auth_types=0x3 and...
- ...rand=39fa2f4e3d523d8619a73b4f65c3e14d
- ...autn=8704f5ba55f30000d2ee44b22c8ea919
- ...expecting res=e229c19e791f2e41
DREF VLR subscr IMSI:901700000010650 usage decreases to: 1
<-- GSUP rx OSMO_GSUP_MSGT_SEND_AUTH_INFO_RESULT: vlr_gsupc_read_cb() returns 0
auth_request_sent == 1
lu_result_sent == 0
- MS sends Authen Response, VLR accepts and sends SecurityModeControl
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_MM_AUTH_RESP
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF IMSI:901700000010650: MSC conn use + dtap == 1 (0x2: dtap)
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
DRLL Dispatching 04.08 message GSM48_MT_MM_AUTH_RESP (0x5:0x14)
DMM IMSI:901700000010650: MM UMTS AUTHENTICATION RESPONSE (res = e229c19e791f2e41)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_WAIT_RESP}: Received Event VLR_AUTH_E_MS_AUTH_RESP
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
DVLR SUBSCR(IMSI:901700000010650) AUTH on UTRAN received RES: e229c19e791f2e41 (8 bytes)
DVLR SUBSCR(IMSI:901700000010650) AUTH established UMTS security context
refactor VLR FSM result handling Instead of keeping separate enums for FSM results and translating between those and the actual 04.08 reject causes that will ultimately reach the MS, just pass enum gsm48_reject_value cause codes around everywhere. Collapse some VLR *_timeout() and *_cancel() api to just *_cancel() with a gsm48 cause arg. (Hopefully) improve a few reject causes, but otherwise just aim for more transparent decisions on which cause value is used, for future fixes of returned causes. Depends: I6661f139e68a498fb1bef10c266c2f064b72774a (libosmocore) Change-Id: I27bf8d68737ff1f8dc6d11fb1eac3d391aab0cb1
2018-04-06 00:57:51 +00:00
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_WAIT_RESP}: Authentication terminating with result PASSED
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_WAIT_RESP}: state_chg to VLR_SUB_AS_AUTHENTICATED
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Removing from parent vlr_lu_fsm(LU:901700000010650)
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Freeing instance
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Deallocated
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_AUTH}: Received Event VLR_ULA_E_AUTH_RES
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_AUTH}: vlr_loc_upd_post_auth()
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_AUTH}: Set Ciphering Mode
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
DMM -> SECURITY MODE CONTROL IMSI:901700000010650
- sending SecurityModeControl for UE ctx 42 send_ck=0 new_key=1
- ...ik=27497388b6cb044648f396aa155b95ef
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_AUTH}: state_chg to VLR_ULA_S_WAIT_CIPH
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF IMSI:901700000010650: MSC conn use - dtap == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_AUTH_CIPH}: Awaiting results for Auth+Ciph, overruling event RAN_CONN_E_UNUSED
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
security_mode_ctrl_sent == 1
lu_result_sent == 0
- MS sends SecurityModeControl acceptance, VLR accepts and sends GSUP LU Req to HLR
DMM <- SECURITY MODE COMPLETE IMSI:901700000010650
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_CIPH}: Received Event VLR_ULA_E_CIPH_RES
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_CIPH}: vlr_loc_upd_post_ciph()
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
DIUCS IMSI:901700000010650: tx CommonID 901700000010650
- Iu Common ID --RAN_UTRAN_IU--> MS (IMSI=901700000010650)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_CIPH}: vlr_loc_upd_node_4()
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_CIPH}: state_chg to VLR_ULA_S_WAIT_HLR_UPD
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_INIT}: Allocated
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_INIT}: is child of vlr_lu_fsm(LU:901700000010650)
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_INIT}: Received Event UPD_HLR_VLR_E_START
GSUP client: send CN domain IE on LU request Give the HLR a chance to send us updated subscriber data by indicating the CN domain to be Circuit Switched, only during a LU Request GSUP message. Adjust msc_vlr_tests to expect the added GSUP CN domain IE to indicate CS, i.e. append '280102'. Related: OS#3601 Change-Id: I0c2d33fbfdb4728e480679120d06b7f3a2ccfd76
2018-09-28 00:41:39 +00:00
DVLR GSUP tx: 04010809710000000156f0280102
GSUP --> HLR: OSMO_GSUP_MSGT_UPDATE_LOCATION_REQUEST: 04010809710000000156f0280102
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_INIT}: state_chg to UPD_HLR_VLR_S_WAIT_FOR_DATA
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
gsup_tx_confirmed == 1
lu_result_sent == 0
- HLR sends _INSERT_DATA_REQUEST, VLR responds with _INSERT_DATA_RESULT
<-- GSUP rx OSMO_GSUP_MSGT_INSERT_DATA_REQUEST: 10010809710000000156f00804032443f2
DVLR GSUP rx 17: 10010809710000000156f00804032443f2
DREF VLR subscr IMSI:901700000010650 usage increases to: 2
DVLR IMSI:901700000010650 has MSISDN:42342
DVLR SUBSCR(MSISDN:42342) VLR: update for IMSI=901700000010650 (MSISDN=42342, used=2)
DVLR GSUP tx: 12010809710000000156f0
GSUP --> HLR: OSMO_GSUP_MSGT_INSERT_DATA_RESULT: 12010809710000000156f0
DREF VLR subscr MSISDN:42342 usage decreases to: 1
<-- GSUP rx OSMO_GSUP_MSGT_INSERT_DATA_REQUEST: vlr_gsupc_read_cb() returns 0
lu_result_sent == 0
- HLR also sends GSUP _UPDATE_LOCATION_RESULT
<-- GSUP rx OSMO_GSUP_MSGT_UPDATE_LOCATION_RESULT: 06010809710000000156f0
DVLR GSUP rx 11: 06010809710000000156f0
DREF VLR subscr MSISDN:42342 usage increases to: 2
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_HLR_UPD}: Received Event VLR_ULA_E_HLR_LU_RES
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_WAIT_FOR_DATA}: Received Event UPD_HLR_VLR_E_UPD_LOC_ACK
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_WAIT_FOR_DATA}: state_chg to UPD_HLR_VLR_S_DONE
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_DONE}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_DONE}: Removing from parent vlr_lu_fsm(LU:901700000010650)
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_DONE}: Freeing instance
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_DONE}: Deallocated
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_HLR_UPD}: Received Event VLR_ULA_E_UPD_HLR_COMPL
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_HLR_UPD}: state_chg to VLR_ULA_S_WAIT_LU_COMPL
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_INIT}: Allocated
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_INIT}: is child of vlr_lu_fsm(LU:901700000010650)
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_INIT}: Received Event LU_COMPL_VLR_E_START
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_INIT}: state_chg to LU_COMPL_VLR_S_WAIT_SUB_PRES
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_WAIT_SUB_PRES}: Received Event LU_COMPL_VLR_E_SUB_PRES_COMPL
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_WAIT_SUB_PRES}: lu_compl_vlr_new_tmsi()
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_WAIT_SUB_PRES}: state_chg to LU_COMPL_VLR_S_WAIT_TMSI_CNF
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
- sending LU Accept for MSISDN:42342, with TMSI 0x03020100
DREF VLR subscr MSISDN:42342 usage decreases to: 1
<-- GSUP rx OSMO_GSUP_MSGT_UPDATE_LOCATION_RESULT: vlr_gsupc_read_cb() returns 0
lu_result_sent == 1
- a LU Accept with a new TMSI was sent, waiting for TMSI Realloc Compl
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
llist_count(&net->ran_conns) == 1
ran_conn_is_accepted() == false
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
- MS sends TMSI Realloc Complete
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_MM_TMSI_REALL_COMPL
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + dtap == 1 (0x2: dtap)
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
DRLL Dispatching 04.08 message GSM48_MT_MM_TMSI_REALL_COMPL (0x5:0x1b)
DMM TMSI Reallocation Completed. Subscriber: MSISDN:42342
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_LU_COMPL}: Received Event VLR_ULA_E_NEW_TMSI_ACK
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_WAIT_TMSI_CNF}: Received Event LU_COMPL_VLR_E_NEW_TMSI_ACK
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
DREF VLR subscr MSISDN:42342 usage increases to: 2
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_WAIT_TMSI_CNF}: state_chg to LU_COMPL_VLR_S_DONE
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_LU_COMPL}: Received Event VLR_ULA_E_LU_COMPL_SUCCESS
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_DONE}: Terminating (cause = OSMO_FSM_TERM_PARENT)
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_DONE}: Removing from parent vlr_lu_fsm(LU:901700000010650)
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_DONE}: Freeing instance
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_DONE}: Deallocated
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_LU_COMPL}: state_chg to VLR_ULA_S_DONE
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_ACCEPTED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_AUTH_CIPH}: state_chg to RAN_CONN_S_ACCEPTED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_ACCEPTED}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_ACCEPTED}: state_chg to RAN_CONN_S_RELEASING
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + release == 2 (0x102: dtap,release)
refactor subscr_conn and subscr_conn_fsm de-/alloc Refactor: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. 2. Add separate AUTH_CIPH state to the FSM. 3. Use conn->use_count to trigger conn release. 4. Add separate RELEASING state to the FSM. 5. Add rate counters for each of the three Complete Layer 3 types. Details: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. Historically, a gsm_subscriber_connection was allocated in libbsc land, and only upon Complete Layer 3 did libmsc add the fsm instance. After splitting openbsc.git into a separate osmo-msc, this is no longer necessary, hence: Closely tie gsm_subscriber_connection allocation to the subscr_conn_fsm instance: talloc the conn as a child of the FSM instance, and discard the conn as soon as the FSM terminates. 2. Add separate AUTH_CIPH state to the FSM. Decoding the Complete Layer 3 message is distinctly separate from waiting for the VLR FSMs to conclude. Use the NEW state as "we don't know if this is a valid message yet", and the AUTH_CIPH state as "evaluating, don't release". A profound effect of this: should we for any odd reason fail to leave the FSM's NEW state, the conn will be released right at the end of msc_compl_l3(), without needing to trigger release in each code path. 3. Use conn->use_count to trigger conn release. Before, the FSM itself would hold a use count on the conn, and hence we would need to ask it whether it is ready to release the conn yet by dispatching events, to achieve a use_count decrement. Instead, unite the FSM instance and conn, and do not hold a use count by the FSM. Hence, trigger an FSM "UNUSED" event only when the use_count reaches zero. As long as use counts are done correctly, the FSM will terminate correctly. These exceptions: - The new AUTH_CIPH state explicitly ignores UNUSED events, since we expect the use count to reach zero while evaluating Authentication and Ciphering. (I experimented with holding a use count by AUTH_CIPH onenter() and releasing by onleave(), but the use count and thus the conn are released before the next state can initiate transactions that would increment the use count again. Same thing for the VLR FSMs holding a use count, they should be done before we advance to the next state. The easiest is to simply expect zero use count during the AUTH_CIPH state.) - A CM Service Request means that even though the MSC would be through with all it wants to do, we shall still wait for a request to follow from the MS. Hence the FSM holds a use count on itself while a CM Service is pending. - While waiting for a Release/Clear Complete, the FSM holds a use count on itself. 4. Add separate RELEASING state to the FSM. If we decide to release for other reasons than a use count reaching zero, we still need to be able to wait for the msc_dtap() use count on the conn to release. (An upcoming patch will further use the RELEASING state to properly wait for Clear Complete / Release Complete messages.) 5. Add rate counters for each of the three Complete Layer 3 types. Besides LU, also count CM Service Request and Paging Response acceptance/rejections. Without these counters, only very few of the auth+ciph outcomes actually show in the counters. Related: OS#3122 Change-Id: I55feb379e176a96a831e105b86202b17a0ffe889
2018-03-30 22:02:14 +00:00
DREF VLR subscr MSISDN:42342 usage increases to: 3
DREF VLR subscr MSISDN:42342 usage decreases to: 2
- Iu Release --RAN_UTRAN_IU--> MS
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - dtap == 1 (0x100: release)
properly receive BSSMAP Clear Complete and Iu Release Complete When sending a BSSMAP Clear or Iu Release, do not immediately discard the conn, but wait until a BSSMAP Clear Complete / Iu Release Complete has been received. Hence we will no longer show in the log that an incoming Release/Clear Complete belongs to an unknown subscriber, but will still be around to properly log the release. Related: OS#3122 Change-Id: Ie4c6aaba3866d6e5b98004e8870a215e8cf8ffc1
2018-04-01 18:55:54 +00:00
iu_release_sent == 1
- LU was successful, and the conn has already been closed
- RNC sends Iu Release Complete
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - release == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_RELEASING}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_RELEASING}: state_chg to RAN_CONN_S_RELEASED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_RELEASED}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_DONE}: Terminating (cause = OSMO_FSM_TERM_PARENT)
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_DONE}: Removing from parent RAN_conn(LU:901700000010650)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_DONE}: fsm_lu_cleanup called with cause OSMO_FSM_TERM_PARENT
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_DONE}: Freeing instance
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_DONE}: Deallocated
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DRLL MSISDN:42342: Freeing RAN connection
refactor subscr_conn and subscr_conn_fsm de-/alloc Refactor: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. 2. Add separate AUTH_CIPH state to the FSM. 3. Use conn->use_count to trigger conn release. 4. Add separate RELEASING state to the FSM. 5. Add rate counters for each of the three Complete Layer 3 types. Details: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. Historically, a gsm_subscriber_connection was allocated in libbsc land, and only upon Complete Layer 3 did libmsc add the fsm instance. After splitting openbsc.git into a separate osmo-msc, this is no longer necessary, hence: Closely tie gsm_subscriber_connection allocation to the subscr_conn_fsm instance: talloc the conn as a child of the FSM instance, and discard the conn as soon as the FSM terminates. 2. Add separate AUTH_CIPH state to the FSM. Decoding the Complete Layer 3 message is distinctly separate from waiting for the VLR FSMs to conclude. Use the NEW state as "we don't know if this is a valid message yet", and the AUTH_CIPH state as "evaluating, don't release". A profound effect of this: should we for any odd reason fail to leave the FSM's NEW state, the conn will be released right at the end of msc_compl_l3(), without needing to trigger release in each code path. 3. Use conn->use_count to trigger conn release. Before, the FSM itself would hold a use count on the conn, and hence we would need to ask it whether it is ready to release the conn yet by dispatching events, to achieve a use_count decrement. Instead, unite the FSM instance and conn, and do not hold a use count by the FSM. Hence, trigger an FSM "UNUSED" event only when the use_count reaches zero. As long as use counts are done correctly, the FSM will terminate correctly. These exceptions: - The new AUTH_CIPH state explicitly ignores UNUSED events, since we expect the use count to reach zero while evaluating Authentication and Ciphering. (I experimented with holding a use count by AUTH_CIPH onenter() and releasing by onleave(), but the use count and thus the conn are released before the next state can initiate transactions that would increment the use count again. Same thing for the VLR FSMs holding a use count, they should be done before we advance to the next state. The easiest is to simply expect zero use count during the AUTH_CIPH state.) - A CM Service Request means that even though the MSC would be through with all it wants to do, we shall still wait for a request to follow from the MS. Hence the FSM holds a use count on itself while a CM Service is pending. - While waiting for a Release/Clear Complete, the FSM holds a use count on itself. 4. Add separate RELEASING state to the FSM. If we decide to release for other reasons than a use count reaching zero, we still need to be able to wait for the msc_dtap() use count on the conn to release. (An upcoming patch will further use the RELEASING state to properly wait for Clear Complete / Release Complete messages.) 5. Add rate counters for each of the three Complete Layer 3 types. Besides LU, also count CM Service Request and Paging Response acceptance/rejections. Without these counters, only very few of the auth+ciph outcomes actually show in the counters. Related: OS#3122 Change-Id: I55feb379e176a96a831e105b86202b17a0ffe889
2018-03-30 22:02:14 +00:00
DREF VLR subscr MSISDN:42342 usage decreases to: 1
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_RELEASED}: Freeing instance
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_RELEASED}: Deallocated
llist_count(&net->ran_conns) == 0
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
DREF VLR subscr MSISDN:42342 usage increases to: 2
vsub != NULL == 1
strcmp(vsub->imsi, IMSI) == 0
vsub->lac == 23
DREF VLR subscr MSISDN:42342 usage decreases to: 1
---
- after a while, MNCC asks us to setup a call, causing Paging
DMNCC receive message MNCC_SETUP_REQ
DREF VLR subscr MSISDN:42342 usage increases to: 2
DCC (ti ff sub MSISDN:42342 callref 423) New transaction
DREF VLR subscr MSISDN:42342 usage increases to: 3
DMM Subscriber MSISDN:42342 not paged yet, start paging.
RAN_UTRAN_IU sends out paging request to IMSI 901700000010650, TMSI 0x03020100, LAC 23
strcmp(paging_expecting_imsi, imsi) == 0
DREF VLR subscr MSISDN:42342 usage increases to: 4
DREF VLR subscr MSISDN:42342 usage decreases to: 3
paging_sent == 1
paging_stopped == 0
- MS replies with Paging Response, and VLR sends Auth Request
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_RR_PAG_RESP
new conn
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn{RAN_CONN_S_NEW}: Allocated
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF unknown: MSC conn use + compl_l3 == 1 (0x1: compl_l3)
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
DRLL Dispatching 04.08 message GSM48_MT_RR_PAG_RESP (0x6:0x27)
DRR PAGING RESPONSE: MI(IMSI)=901700000010650
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_NEW}: Updated ID
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_INIT}: Allocated
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_INIT}: is child of RAN_conn(PAGING_RESP:901700000010650)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_INIT}: rev=R99 net=UTRAN Auth+Ciph
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_INIT}: Received Event PR_ARQ_E_START
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
DREF VLR subscr MSISDN:42342 usage increases to: 4
DREF VLR subscr MSISDN:42342 usage increases to: 5
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_INIT}: proc_arq_vlr_fn_post_imsi()
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_INIT}: state_chg to PR_ARQ_S_WAIT_AUTH
DVLR VLR_Authenticate(PAGING_RESP:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: Allocated
DVLR VLR_Authenticate(PAGING_RESP:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: is child of Process_Access_Request_VLR(PAGING_RESP:901700000010650)
DVLR VLR_Authenticate(PAGING_RESP:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: Received Event VLR_AUTH_E_START
DVLR VLR_Authenticate(PAGING_RESP:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: state_chg to VLR_SUB_AS_WAIT_RESP
DVLR VLR_Authenticate(PAGING_RESP:901700000010650){VLR_SUB_AS_WAIT_RESP}: got auth tuple: use_count=1 key_seq=1 -- will use UMTS AKA (is_r99=yes, at->vec.auth_types=0x3)
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
- sending UMTS Auth Request for MSISDN:42342: tuple use_count=1 key_seq=1 auth_types=0x3 and...
- ...rand=c187a53a5e6b9d573cac7c74451fd46d
- ...autn=1843a645b98d00005b2d666af46c45d9
- ...expecting res=7db47cf7f81e4dc7
DREF VLR subscr MSISDN:42342 usage decreases to: 4
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_NEW}: Received Event RAN_CONN_E_COMPLETE_LAYER_3
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_NEW}: state_chg to RAN_CONN_S_AUTH_CIPH
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - compl_l3 == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_AUTH_CIPH}: Awaiting results for Auth+Ciph, overruling event RAN_CONN_E_UNUSED
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
auth_request_sent == 1
- MS sends Authen Response, VLR accepts and sends SecurityModeControl
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_MM_AUTH_RESP
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + dtap == 1 (0x2: dtap)
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
DRLL Dispatching 04.08 message GSM48_MT_MM_AUTH_RESP (0x5:0x14)
DMM MSISDN:42342: MM UMTS AUTHENTICATION RESPONSE (res = 7db47cf7f81e4dc7)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(PAGING_RESP:901700000010650){VLR_SUB_AS_WAIT_RESP}: Received Event VLR_AUTH_E_MS_AUTH_RESP
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
DVLR SUBSCR(MSISDN:42342) AUTH on UTRAN received RES: 7db47cf7f81e4dc7 (8 bytes)
DVLR SUBSCR(MSISDN:42342) AUTH established UMTS security context
refactor VLR FSM result handling Instead of keeping separate enums for FSM results and translating between those and the actual 04.08 reject causes that will ultimately reach the MS, just pass enum gsm48_reject_value cause codes around everywhere. Collapse some VLR *_timeout() and *_cancel() api to just *_cancel() with a gsm48 cause arg. (Hopefully) improve a few reject causes, but otherwise just aim for more transparent decisions on which cause value is used, for future fixes of returned causes. Depends: I6661f139e68a498fb1bef10c266c2f064b72774a (libosmocore) Change-Id: I27bf8d68737ff1f8dc6d11fb1eac3d391aab0cb1
2018-04-06 00:57:51 +00:00
DVLR VLR_Authenticate(PAGING_RESP:901700000010650){VLR_SUB_AS_WAIT_RESP}: Authentication terminating with result PASSED
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(PAGING_RESP:901700000010650){VLR_SUB_AS_WAIT_RESP}: state_chg to VLR_SUB_AS_AUTHENTICATED
DVLR VLR_Authenticate(PAGING_RESP:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
DVLR VLR_Authenticate(PAGING_RESP:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Removing from parent Process_Access_Request_VLR(PAGING_RESP:901700000010650)
DVLR VLR_Authenticate(PAGING_RESP:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Freeing instance
DVLR VLR_Authenticate(PAGING_RESP:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Deallocated
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_WAIT_AUTH}: Received Event PR_ARQ_E_AUTH_RES
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_WAIT_AUTH}: _proc_arq_vlr_node2()
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_WAIT_AUTH}: Set Ciphering Mode
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
DMM -> SECURITY MODE CONTROL MSISDN:42342
- sending SecurityModeControl for UE ctx 42 send_ck=0 new_key=1
- ...ik=1159ec926a50e98c034a6b7d7c9f418d
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_WAIT_AUTH}: state_chg to PR_ARQ_S_WAIT_CIPH
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - dtap == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_AUTH_CIPH}: Awaiting results for Auth+Ciph, overruling event RAN_CONN_E_UNUSED
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
security_mode_ctrl_sent == 1
- MS sends SecurityModeControl acceptance, VLR accepts, sends CC Setup
DMM <- SECURITY MODE COMPLETE MSISDN:42342
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_WAIT_CIPH}: Received Event PR_ARQ_E_CIPH_RES
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_WAIT_CIPH}: _proc_arq_vlr_node2_post_ciph()
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
DIUCS MSISDN:42342: tx CommonID 901700000010650
- Iu Common ID --RAN_UTRAN_IU--> MS (IMSI=901700000010650)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_WAIT_CIPH}: _proc_arq_vlr_node2_post_vlr()
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_WAIT_CIPH}: _proc_arq_vlr_post_pres()
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_WAIT_CIPH}: _proc_arq_vlr_post_trace()
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_WAIT_CIPH}: _proc_arq_vlr_post_imei()
refactor VLR FSM result handling Instead of keeping separate enums for FSM results and translating between those and the actual 04.08 reject causes that will ultimately reach the MS, just pass enum gsm48_reject_value cause codes around everywhere. Collapse some VLR *_timeout() and *_cancel() api to just *_cancel() with a gsm48 cause arg. (Hopefully) improve a few reject causes, but otherwise just aim for more transparent decisions on which cause value is used, for future fixes of returned causes. Depends: I6661f139e68a498fb1bef10c266c2f064b72774a (libosmocore) Change-Id: I27bf8d68737ff1f8dc6d11fb1eac3d391aab0cb1
2018-04-06 00:57:51 +00:00
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_WAIT_CIPH}: proc_arq_fsm_done(PASSED)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_WAIT_CIPH}: state_chg to PR_ARQ_S_DONE
refactor VLR FSM result handling Instead of keeping separate enums for FSM results and translating between those and the actual 04.08 reject causes that will ultimately reach the MS, just pass enum gsm48_reject_value cause codes around everywhere. Collapse some VLR *_timeout() and *_cancel() api to just *_cancel() with a gsm48 cause arg. (Hopefully) improve a few reject causes, but otherwise just aim for more transparent decisions on which cause value is used, for future fixes of returned causes. Depends: I6661f139e68a498fb1bef10c266c2f064b72774a (libosmocore) Change-Id: I27bf8d68737ff1f8dc6d11fb1eac3d391aab0cb1
2018-04-06 00:57:51 +00:00
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_DONE}: Process Access Request result: PASSED
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_ACCEPTED
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
DPAG Paging success for MSISDN:42342 (event=0)
DPAG Calling paging cbfn.
DCC Paging subscr 42342 succeeded!
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + trans_cc == 1 (0x10: trans_cc)
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
DCC starting timer T303 with 30 seconds
DCC (ti 00 sub MSISDN:42342) new state NULL -> CALL_PRESENT
DMSC msc_tx 2 bytes to MSISDN:42342 via RAN_UTRAN_IU
- DTAP --RAN_UTRAN_IU--> MS: GSM48_MT_CC_SETUP: 0305
- DTAP matches expected message
DREF VLR subscr MSISDN:42342 usage decreases to: 3
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_AUTH_CIPH}: state_chg to RAN_CONN_S_ACCEPTED
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_ACCEPTED}: ran_conn_fsm_has_active_transactions: connection still has active transaction: CC
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
paging_stopped == 1
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_CC_CALL_CONF
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + dtap == 2 (0x12: dtap,trans_cc)
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
DRLL Dispatching 04.08 message GSM48_MT_CC_CALL_CONF (0x3:0x8)
DCC stopping pending timer T303
DCC starting timer T310 with 30 seconds
DCC (ti 00 sub MSISDN:42342) new state CALL_PRESENT -> MO_TERM_CALL_CONF
MS <--Call Assignment-- MSC: subscr=MSISDN:42342 callref=0x423
DMNCC transmit message MNCC_CALL_CONF_IND
DCC Sending 'MNCC_CALL_CONF_IND' to MNCC.
MSC --> MNCC: callref 0x423: MNCC_CALL_CONF_IND
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_ACCEPTED}: Received Event RAN_CONN_E_COMMUNICATING
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_ACCEPTED}: state_chg to RAN_CONN_S_COMMUNICATING
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - dtap == 1 (0x10: trans_cc)
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
- Total time passed: 1.000023 s
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_CC_ALERTING
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + dtap == 2 (0x12: dtap,trans_cc)
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
DRLL Dispatching 04.08 message GSM48_MT_CC_ALERTING (0x3:0x1)
DCC stopping pending timer T310
DCC starting timer T301 with 180 seconds
DCC (ti 00 sub MSISDN:42342) new state MO_TERM_CALL_CONF -> CALL_RECEIVED
DMNCC transmit message MNCC_ALERT_IND
DCC Sending 'MNCC_ALERT_IND' to MNCC.
MSC --> MNCC: callref 0x423: MNCC_ALERT_IND
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_COMMUNICATING}: Received Event RAN_CONN_E_COMMUNICATING
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - dtap == 1 (0x10: trans_cc)
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
- Total time passed: 16.000046 s
- The call failed, the BSC sends a BSSMAP Clear Request
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_COMMUNICATING}: Received Event RAN_CONN_E_CN_CLOSE
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_COMMUNICATING}: state_chg to RAN_CONN_S_RELEASING
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + release == 2 (0x110: trans_cc,release)
refactor subscr_conn and subscr_conn_fsm de-/alloc Refactor: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. 2. Add separate AUTH_CIPH state to the FSM. 3. Use conn->use_count to trigger conn release. 4. Add separate RELEASING state to the FSM. 5. Add rate counters for each of the three Complete Layer 3 types. Details: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. Historically, a gsm_subscriber_connection was allocated in libbsc land, and only upon Complete Layer 3 did libmsc add the fsm instance. After splitting openbsc.git into a separate osmo-msc, this is no longer necessary, hence: Closely tie gsm_subscriber_connection allocation to the subscr_conn_fsm instance: talloc the conn as a child of the FSM instance, and discard the conn as soon as the FSM terminates. 2. Add separate AUTH_CIPH state to the FSM. Decoding the Complete Layer 3 message is distinctly separate from waiting for the VLR FSMs to conclude. Use the NEW state as "we don't know if this is a valid message yet", and the AUTH_CIPH state as "evaluating, don't release". A profound effect of this: should we for any odd reason fail to leave the FSM's NEW state, the conn will be released right at the end of msc_compl_l3(), without needing to trigger release in each code path. 3. Use conn->use_count to trigger conn release. Before, the FSM itself would hold a use count on the conn, and hence we would need to ask it whether it is ready to release the conn yet by dispatching events, to achieve a use_count decrement. Instead, unite the FSM instance and conn, and do not hold a use count by the FSM. Hence, trigger an FSM "UNUSED" event only when the use_count reaches zero. As long as use counts are done correctly, the FSM will terminate correctly. These exceptions: - The new AUTH_CIPH state explicitly ignores UNUSED events, since we expect the use count to reach zero while evaluating Authentication and Ciphering. (I experimented with holding a use count by AUTH_CIPH onenter() and releasing by onleave(), but the use count and thus the conn are released before the next state can initiate transactions that would increment the use count again. Same thing for the VLR FSMs holding a use count, they should be done before we advance to the next state. The easiest is to simply expect zero use count during the AUTH_CIPH state.) - A CM Service Request means that even though the MSC would be through with all it wants to do, we shall still wait for a request to follow from the MS. Hence the FSM holds a use count on itself while a CM Service is pending. - While waiting for a Release/Clear Complete, the FSM holds a use count on itself. 4. Add separate RELEASING state to the FSM. If we decide to release for other reasons than a use count reaching zero, we still need to be able to wait for the msc_dtap() use count on the conn to release. (An upcoming patch will further use the RELEASING state to properly wait for Clear Complete / Release Complete messages.) 5. Add rate counters for each of the three Complete Layer 3 types. Besides LU, also count CM Service Request and Paging Response acceptance/rejections. Without these counters, only very few of the auth+ciph outcomes actually show in the counters. Related: OS#3122 Change-Id: I55feb379e176a96a831e105b86202b17a0ffe889
2018-03-30 22:02:14 +00:00
DREF VLR subscr MSISDN:42342 usage increases to: 4
DREF VLR subscr MSISDN:42342 usage decreases to: 3
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
DCC stopping pending timer T301
MS <--Call Release-- MSC: subscr=MSISDN:42342 callref=0x423
DMNCC receive message MNCC_REL_REQ
gsm_04_08_cc: Add global guard timer for MNCC The external MNCC handler may hang indefinitely in cases where the remote end of the MNCC ceases to work properly. Add a global guard timer to make sure the call reaches ACTIVE state. Change-Id: I7375d1e17cd746aac4eadfe1e587e82cf1630d3d Related: OS#3599
2018-10-10 15:00:49 +00:00
DCC (sub 42342) starting guard timer with 180 seconds
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
DCC (ti 00 sub 42342) Received 'MNCC_REL_REQ' from MNCC in state 7 (CALL_RECEIVED)
DCC starting timer T308 with 10 seconds
DCC (ti 00 sub MSISDN:42342) new state CALL_RECEIVED -> RELEASE_REQ
DMSC msc_tx 2 bytes to MSISDN:42342 via RAN_UTRAN_IU
- DTAP --RAN_UTRAN_IU--> MS: GSM48_MT_CC_RELEASE: 032d
- DTAP matches expected message
DMNCC transmit message MNCC_REL_CNF
DCC Sending 'MNCC_REL_CNF' to MNCC.
MSC --> MNCC: callref 0x423: MNCC_REL_CNF
CC: intentionally release T308 on BSSMAP Clear Request from BSC So far we hit a running T308 during CC release when caused by a BSSMAP Clear Request, and we loudly log that as error. However, now I understand that T308 is a direct cause of the dispatch of a REL IND towards MNCC, which is used to indicate teardown to MNCC. So during _gsm48_cc_trans_free(), we first clear all timers, then invoke mncc_release_ind() which starts another timer (useful for graceful CC Release, but in this code path the intention is immediate release). Simply immediately cancel the timer again and release the conn. A separate question is whether a BSSMAP Clear Request should be less aggressive in releasing the connections; i.e. instead of calling trans_free() all around, to rather ask each transaction to "please stop soon", somehow. Related: OS#3062 Change-Id: I231fdb574a086a206321148474cbdc7ca9cf39f0
2018-04-09 21:29:14 +00:00
DCC stopping pending timer T308
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
DCC (ti 00 sub MSISDN:42342) new state RELEASE_REQ -> NULL
gsm_04_08_cc: Add global guard timer for MNCC The external MNCC handler may hang indefinitely in cases where the remote end of the MNCC ceases to work properly. Add a global guard timer to make sure the call reaches ACTIVE state. Change-Id: I7375d1e17cd746aac4eadfe1e587e82cf1630d3d Related: OS#3599
2018-10-10 15:00:49 +00:00
DCC (sub 42342) stopping pending guard timer
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
DREF VLR subscr MSISDN:42342 usage decreases to: 2
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - trans_cc == 1 (0x100: release)
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
- Iu Release --RAN_UTRAN_IU--> MS
properly receive BSSMAP Clear Complete and Iu Release Complete When sending a BSSMAP Clear or Iu Release, do not immediately discard the conn, but wait until a BSSMAP Clear Complete / Iu Release Complete has been received. Hence we will no longer show in the log that an incoming Release/Clear Complete belongs to an unknown subscriber, but will still be around to properly log the release. Related: OS#3122 Change-Id: Ie4c6aaba3866d6e5b98004e8870a215e8cf8ffc1
2018-04-01 18:55:54 +00:00
- RNC sends Iu Release Complete
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - release == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_RELEASING}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_RELEASING}: state_chg to RAN_CONN_S_RELEASED
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_RELEASED}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_DONE}: Terminating (cause = OSMO_FSM_TERM_PARENT)
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_DONE}: Removing from parent RAN_conn(PAGING_RESP:901700000010650)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_DONE}: Freeing instance
DVLR Process_Access_Request_VLR(PAGING_RESP:901700000010650){PR_ARQ_S_DONE}: Deallocated
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DRLL MSISDN:42342: Freeing RAN connection
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
DREF VLR subscr MSISDN:42342 usage decreases to: 1
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_RELEASED}: Freeing instance
DMM RAN_conn(PAGING_RESP:901700000010650){RAN_CONN_S_RELEASED}: Deallocated
llist_count(&net->ran_conns) == 0
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
- Total time passed: 31.000069 s
DREF freeing VLR subscr MSISDN:42342
===== test_call_mt2: SUCCESS
full talloc report on 'msgb' (total 0 bytes in 1 blocks)
make gsup ipa name configurable in osmo-msc.cfg Add a 'ipa-name' VTY command which overrides the default IPA name used by the MSC. This is a prerequisite for inter-MSC handover. Related: OS#3355 Change-Id: I317d6c59f77e92fbb2b875a83dc0ec2fa5cb6006
2018-12-06 11:06:59 +00:00
talloc_total_blocks(tall_bsc_ctx) == 13
msc_vlr_test_call: reproduce OS#3062 A related ttcn3 test is added in Ic80646e1fba37bb6163ca3a7eead7980b4ad7a51 Related: OS#3062 Change-Id: Ice7197b48d4e163a3c4d97b559fdcd7e88c4107e
2018-03-13 16:52:07 +00:00
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
===== test_call_mo_to_unknown
- Total time passed: 0.000000 s
- Location Update request causes a GSUP Send Auth Info request to HLR
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_MM_LOC_UPD_REQUEST
new conn
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn{RAN_CONN_S_NEW}: Allocated
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF unknown: MSC conn use + compl_l3 == 1 (0x1: compl_l3)
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DRLL Dispatching 04.08 message GSM48_MT_MM_LOC_UPD_REQUEST (0x5:0x8)
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_NEW}: Updated ID
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DMM LOCATION UPDATING REQUEST: MI(IMSI)=901700000010650 type=NORMAL
DMM LU/new-LAC: 23/23
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_IDLE}: Allocated
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_IDLE}: is child of RAN_conn(LU:901700000010650)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_IDLE}: rev=R99 net=UTRAN Auth+Ciph
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_IDLE}: Received Event VLR_ULA_E_UPDATE_LA
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DREF VLR subscr unknown usage increases to: 1
DVLR set IMSI on subscriber; IMSI=901700000010650 id=901700000010650
DVLR New subscr, IMSI: 901700000010650
DREF VLR subscr IMSI:901700000010650 usage increases to: 2
DREF VLR subscr IMSI:901700000010650 usage decreases to: 1
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_IDLE}: vlr_loc_upd_node1()
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_IDLE}: state_chg to VLR_ULA_S_WAIT_AUTH
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: Allocated
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: is child of vlr_lu_fsm(LU:901700000010650)
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: Received Event VLR_AUTH_E_START
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DVLR GSUP tx: 08010809710000000156f0
GSUP --> HLR: OSMO_GSUP_MSGT_SEND_AUTH_INFO_REQUEST: 08010809710000000156f0
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: state_chg to VLR_SUB_AS_NEEDS_AUTH_WAIT_AI
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_NEW}: Received Event RAN_CONN_E_COMPLETE_LAYER_3
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_NEW}: state_chg to RAN_CONN_S_AUTH_CIPH
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF IMSI:901700000010650: MSC conn use - compl_l3 == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_AUTH_CIPH}: Awaiting results for Auth+Ciph, overruling event RAN_CONN_E_UNUSED
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
lu_result_sent == 0
- from HLR, rx _SEND_AUTH_INFO_RESULT; VLR sends Auth Req to MS
<-- GSUP rx OSMO_GSUP_MSGT_SEND_AUTH_INFO_RESULT: 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
DVLR GSUP rx 511: 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
DREF VLR subscr IMSI:901700000010650 usage increases to: 2
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_NEEDS_AUTH_WAIT_AI}: Received Event VLR_AUTH_E_HLR_SAI_ACK
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DVLR SUBSCR(IMSI:901700000010650) Received 5 auth tuples
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_NEEDS_AUTH_WAIT_AI}: state_chg to VLR_SUB_AS_WAIT_RESP
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_WAIT_RESP}: got auth tuple: use_count=1 key_seq=0 -- will use UMTS AKA (is_r99=yes, at->vec.auth_types=0x3)
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
- sending UMTS Auth Request for IMSI:901700000010650: tuple use_count=1 key_seq=0 auth_types=0x3 and...
- ...rand=39fa2f4e3d523d8619a73b4f65c3e14d
- ...autn=8704f5ba55f30000d2ee44b22c8ea919
- ...expecting res=e229c19e791f2e41
DREF VLR subscr IMSI:901700000010650 usage decreases to: 1
<-- GSUP rx OSMO_GSUP_MSGT_SEND_AUTH_INFO_RESULT: vlr_gsupc_read_cb() returns 0
auth_request_sent == 1
lu_result_sent == 0
- MS sends Authen Response, VLR accepts and sends SecurityModeControl
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_MM_AUTH_RESP
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF IMSI:901700000010650: MSC conn use + dtap == 1 (0x2: dtap)
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DRLL Dispatching 04.08 message GSM48_MT_MM_AUTH_RESP (0x5:0x14)
DMM IMSI:901700000010650: MM UMTS AUTHENTICATION RESPONSE (res = e229c19e791f2e41)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_WAIT_RESP}: Received Event VLR_AUTH_E_MS_AUTH_RESP
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DVLR SUBSCR(IMSI:901700000010650) AUTH on UTRAN received RES: e229c19e791f2e41 (8 bytes)
DVLR SUBSCR(IMSI:901700000010650) AUTH established UMTS security context
refactor VLR FSM result handling Instead of keeping separate enums for FSM results and translating between those and the actual 04.08 reject causes that will ultimately reach the MS, just pass enum gsm48_reject_value cause codes around everywhere. Collapse some VLR *_timeout() and *_cancel() api to just *_cancel() with a gsm48 cause arg. (Hopefully) improve a few reject causes, but otherwise just aim for more transparent decisions on which cause value is used, for future fixes of returned causes. Depends: I6661f139e68a498fb1bef10c266c2f064b72774a (libosmocore) Change-Id: I27bf8d68737ff1f8dc6d11fb1eac3d391aab0cb1
2018-04-06 00:57:51 +00:00
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_WAIT_RESP}: Authentication terminating with result PASSED
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_WAIT_RESP}: state_chg to VLR_SUB_AS_AUTHENTICATED
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Removing from parent vlr_lu_fsm(LU:901700000010650)
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Freeing instance
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Deallocated
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_AUTH}: Received Event VLR_ULA_E_AUTH_RES
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_AUTH}: vlr_loc_upd_post_auth()
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_AUTH}: Set Ciphering Mode
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DMM -> SECURITY MODE CONTROL IMSI:901700000010650
- sending SecurityModeControl for UE ctx 42 send_ck=0 new_key=1
- ...ik=27497388b6cb044648f396aa155b95ef
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_AUTH}: state_chg to VLR_ULA_S_WAIT_CIPH
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF IMSI:901700000010650: MSC conn use - dtap == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_AUTH_CIPH}: Awaiting results for Auth+Ciph, overruling event RAN_CONN_E_UNUSED
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
security_mode_ctrl_sent == 1
lu_result_sent == 0
- MS sends SecurityModeControl acceptance, VLR accepts and sends GSUP LU Req to HLR
DMM <- SECURITY MODE COMPLETE IMSI:901700000010650
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_CIPH}: Received Event VLR_ULA_E_CIPH_RES
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_CIPH}: vlr_loc_upd_post_ciph()
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DIUCS IMSI:901700000010650: tx CommonID 901700000010650
- Iu Common ID --RAN_UTRAN_IU--> MS (IMSI=901700000010650)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_CIPH}: vlr_loc_upd_node_4()
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_CIPH}: state_chg to VLR_ULA_S_WAIT_HLR_UPD
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_INIT}: Allocated
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_INIT}: is child of vlr_lu_fsm(LU:901700000010650)
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_INIT}: Received Event UPD_HLR_VLR_E_START
GSUP client: send CN domain IE on LU request Give the HLR a chance to send us updated subscriber data by indicating the CN domain to be Circuit Switched, only during a LU Request GSUP message. Adjust msc_vlr_tests to expect the added GSUP CN domain IE to indicate CS, i.e. append '280102'. Related: OS#3601 Change-Id: I0c2d33fbfdb4728e480679120d06b7f3a2ccfd76
2018-09-28 00:41:39 +00:00
DVLR GSUP tx: 04010809710000000156f0280102
GSUP --> HLR: OSMO_GSUP_MSGT_UPDATE_LOCATION_REQUEST: 04010809710000000156f0280102
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_INIT}: state_chg to UPD_HLR_VLR_S_WAIT_FOR_DATA
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
gsup_tx_confirmed == 1
lu_result_sent == 0
- HLR sends _INSERT_DATA_REQUEST, VLR responds with _INSERT_DATA_RESULT
<-- GSUP rx OSMO_GSUP_MSGT_INSERT_DATA_REQUEST: 10010809710000000156f00804032443f2
DVLR GSUP rx 17: 10010809710000000156f00804032443f2
DREF VLR subscr IMSI:901700000010650 usage increases to: 2
DVLR IMSI:901700000010650 has MSISDN:42342
DVLR SUBSCR(MSISDN:42342) VLR: update for IMSI=901700000010650 (MSISDN=42342, used=2)
DVLR GSUP tx: 12010809710000000156f0
GSUP --> HLR: OSMO_GSUP_MSGT_INSERT_DATA_RESULT: 12010809710000000156f0
DREF VLR subscr MSISDN:42342 usage decreases to: 1
<-- GSUP rx OSMO_GSUP_MSGT_INSERT_DATA_REQUEST: vlr_gsupc_read_cb() returns 0
lu_result_sent == 0
- HLR also sends GSUP _UPDATE_LOCATION_RESULT
<-- GSUP rx OSMO_GSUP_MSGT_UPDATE_LOCATION_RESULT: 06010809710000000156f0
DVLR GSUP rx 11: 06010809710000000156f0
DREF VLR subscr MSISDN:42342 usage increases to: 2
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_HLR_UPD}: Received Event VLR_ULA_E_HLR_LU_RES
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_WAIT_FOR_DATA}: Received Event UPD_HLR_VLR_E_UPD_LOC_ACK
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_WAIT_FOR_DATA}: state_chg to UPD_HLR_VLR_S_DONE
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_DONE}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_DONE}: Removing from parent vlr_lu_fsm(LU:901700000010650)
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_DONE}: Freeing instance
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_DONE}: Deallocated
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_HLR_UPD}: Received Event VLR_ULA_E_UPD_HLR_COMPL
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_HLR_UPD}: state_chg to VLR_ULA_S_WAIT_LU_COMPL
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_INIT}: Allocated
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_INIT}: is child of vlr_lu_fsm(LU:901700000010650)
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_INIT}: Received Event LU_COMPL_VLR_E_START
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_INIT}: state_chg to LU_COMPL_VLR_S_WAIT_SUB_PRES
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_WAIT_SUB_PRES}: Received Event LU_COMPL_VLR_E_SUB_PRES_COMPL
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_WAIT_SUB_PRES}: lu_compl_vlr_new_tmsi()
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_WAIT_SUB_PRES}: state_chg to LU_COMPL_VLR_S_WAIT_TMSI_CNF
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
- sending LU Accept for MSISDN:42342, with TMSI 0x03020100
DREF VLR subscr MSISDN:42342 usage decreases to: 1
<-- GSUP rx OSMO_GSUP_MSGT_UPDATE_LOCATION_RESULT: vlr_gsupc_read_cb() returns 0
lu_result_sent == 1
- a LU Accept with a new TMSI was sent, waiting for TMSI Realloc Compl
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
llist_count(&net->ran_conns) == 1
ran_conn_is_accepted() == false
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
- MS sends TMSI Realloc Complete
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_MM_TMSI_REALL_COMPL
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + dtap == 1 (0x2: dtap)
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DRLL Dispatching 04.08 message GSM48_MT_MM_TMSI_REALL_COMPL (0x5:0x1b)
DMM TMSI Reallocation Completed. Subscriber: MSISDN:42342
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_LU_COMPL}: Received Event VLR_ULA_E_NEW_TMSI_ACK
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_WAIT_TMSI_CNF}: Received Event LU_COMPL_VLR_E_NEW_TMSI_ACK
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DREF VLR subscr MSISDN:42342 usage increases to: 2
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_WAIT_TMSI_CNF}: state_chg to LU_COMPL_VLR_S_DONE
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_LU_COMPL}: Received Event VLR_ULA_E_LU_COMPL_SUCCESS
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_DONE}: Terminating (cause = OSMO_FSM_TERM_PARENT)
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_DONE}: Removing from parent vlr_lu_fsm(LU:901700000010650)
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_DONE}: Freeing instance
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_DONE}: Deallocated
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_LU_COMPL}: state_chg to VLR_ULA_S_DONE
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_ACCEPTED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_AUTH_CIPH}: state_chg to RAN_CONN_S_ACCEPTED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_ACCEPTED}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_ACCEPTED}: state_chg to RAN_CONN_S_RELEASING
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + release == 2 (0x102: dtap,release)
refactor subscr_conn and subscr_conn_fsm de-/alloc Refactor: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. 2. Add separate AUTH_CIPH state to the FSM. 3. Use conn->use_count to trigger conn release. 4. Add separate RELEASING state to the FSM. 5. Add rate counters for each of the three Complete Layer 3 types. Details: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. Historically, a gsm_subscriber_connection was allocated in libbsc land, and only upon Complete Layer 3 did libmsc add the fsm instance. After splitting openbsc.git into a separate osmo-msc, this is no longer necessary, hence: Closely tie gsm_subscriber_connection allocation to the subscr_conn_fsm instance: talloc the conn as a child of the FSM instance, and discard the conn as soon as the FSM terminates. 2. Add separate AUTH_CIPH state to the FSM. Decoding the Complete Layer 3 message is distinctly separate from waiting for the VLR FSMs to conclude. Use the NEW state as "we don't know if this is a valid message yet", and the AUTH_CIPH state as "evaluating, don't release". A profound effect of this: should we for any odd reason fail to leave the FSM's NEW state, the conn will be released right at the end of msc_compl_l3(), without needing to trigger release in each code path. 3. Use conn->use_count to trigger conn release. Before, the FSM itself would hold a use count on the conn, and hence we would need to ask it whether it is ready to release the conn yet by dispatching events, to achieve a use_count decrement. Instead, unite the FSM instance and conn, and do not hold a use count by the FSM. Hence, trigger an FSM "UNUSED" event only when the use_count reaches zero. As long as use counts are done correctly, the FSM will terminate correctly. These exceptions: - The new AUTH_CIPH state explicitly ignores UNUSED events, since we expect the use count to reach zero while evaluating Authentication and Ciphering. (I experimented with holding a use count by AUTH_CIPH onenter() and releasing by onleave(), but the use count and thus the conn are released before the next state can initiate transactions that would increment the use count again. Same thing for the VLR FSMs holding a use count, they should be done before we advance to the next state. The easiest is to simply expect zero use count during the AUTH_CIPH state.) - A CM Service Request means that even though the MSC would be through with all it wants to do, we shall still wait for a request to follow from the MS. Hence the FSM holds a use count on itself while a CM Service is pending. - While waiting for a Release/Clear Complete, the FSM holds a use count on itself. 4. Add separate RELEASING state to the FSM. If we decide to release for other reasons than a use count reaching zero, we still need to be able to wait for the msc_dtap() use count on the conn to release. (An upcoming patch will further use the RELEASING state to properly wait for Clear Complete / Release Complete messages.) 5. Add rate counters for each of the three Complete Layer 3 types. Besides LU, also count CM Service Request and Paging Response acceptance/rejections. Without these counters, only very few of the auth+ciph outcomes actually show in the counters. Related: OS#3122 Change-Id: I55feb379e176a96a831e105b86202b17a0ffe889
2018-03-30 22:02:14 +00:00
DREF VLR subscr MSISDN:42342 usage increases to: 3
DREF VLR subscr MSISDN:42342 usage decreases to: 2
- Iu Release --RAN_UTRAN_IU--> MS
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - dtap == 1 (0x100: release)
properly receive BSSMAP Clear Complete and Iu Release Complete When sending a BSSMAP Clear or Iu Release, do not immediately discard the conn, but wait until a BSSMAP Clear Complete / Iu Release Complete has been received. Hence we will no longer show in the log that an incoming Release/Clear Complete belongs to an unknown subscriber, but will still be around to properly log the release. Related: OS#3122 Change-Id: Ie4c6aaba3866d6e5b98004e8870a215e8cf8ffc1
2018-04-01 18:55:54 +00:00
iu_release_sent == 1
- LU was successful, and the conn has already been closed
- RNC sends Iu Release Complete
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - release == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_RELEASING}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_RELEASING}: state_chg to RAN_CONN_S_RELEASED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_RELEASED}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_DONE}: Terminating (cause = OSMO_FSM_TERM_PARENT)
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_DONE}: Removing from parent RAN_conn(LU:901700000010650)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_DONE}: fsm_lu_cleanup called with cause OSMO_FSM_TERM_PARENT
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_DONE}: Freeing instance
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_DONE}: Deallocated
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DRLL MSISDN:42342: Freeing RAN connection
refactor subscr_conn and subscr_conn_fsm de-/alloc Refactor: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. 2. Add separate AUTH_CIPH state to the FSM. 3. Use conn->use_count to trigger conn release. 4. Add separate RELEASING state to the FSM. 5. Add rate counters for each of the three Complete Layer 3 types. Details: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. Historically, a gsm_subscriber_connection was allocated in libbsc land, and only upon Complete Layer 3 did libmsc add the fsm instance. After splitting openbsc.git into a separate osmo-msc, this is no longer necessary, hence: Closely tie gsm_subscriber_connection allocation to the subscr_conn_fsm instance: talloc the conn as a child of the FSM instance, and discard the conn as soon as the FSM terminates. 2. Add separate AUTH_CIPH state to the FSM. Decoding the Complete Layer 3 message is distinctly separate from waiting for the VLR FSMs to conclude. Use the NEW state as "we don't know if this is a valid message yet", and the AUTH_CIPH state as "evaluating, don't release". A profound effect of this: should we for any odd reason fail to leave the FSM's NEW state, the conn will be released right at the end of msc_compl_l3(), without needing to trigger release in each code path. 3. Use conn->use_count to trigger conn release. Before, the FSM itself would hold a use count on the conn, and hence we would need to ask it whether it is ready to release the conn yet by dispatching events, to achieve a use_count decrement. Instead, unite the FSM instance and conn, and do not hold a use count by the FSM. Hence, trigger an FSM "UNUSED" event only when the use_count reaches zero. As long as use counts are done correctly, the FSM will terminate correctly. These exceptions: - The new AUTH_CIPH state explicitly ignores UNUSED events, since we expect the use count to reach zero while evaluating Authentication and Ciphering. (I experimented with holding a use count by AUTH_CIPH onenter() and releasing by onleave(), but the use count and thus the conn are released before the next state can initiate transactions that would increment the use count again. Same thing for the VLR FSMs holding a use count, they should be done before we advance to the next state. The easiest is to simply expect zero use count during the AUTH_CIPH state.) - A CM Service Request means that even though the MSC would be through with all it wants to do, we shall still wait for a request to follow from the MS. Hence the FSM holds a use count on itself while a CM Service is pending. - While waiting for a Release/Clear Complete, the FSM holds a use count on itself. 4. Add separate RELEASING state to the FSM. If we decide to release for other reasons than a use count reaching zero, we still need to be able to wait for the msc_dtap() use count on the conn to release. (An upcoming patch will further use the RELEASING state to properly wait for Clear Complete / Release Complete messages.) 5. Add rate counters for each of the three Complete Layer 3 types. Besides LU, also count CM Service Request and Paging Response acceptance/rejections. Without these counters, only very few of the auth+ciph outcomes actually show in the counters. Related: OS#3122 Change-Id: I55feb379e176a96a831e105b86202b17a0ffe889
2018-03-30 22:02:14 +00:00
DREF VLR subscr MSISDN:42342 usage decreases to: 1
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_RELEASED}: Freeing instance
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_RELEASED}: Deallocated
llist_count(&net->ran_conns) == 0
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DREF VLR subscr MSISDN:42342 usage increases to: 2
vsub != NULL == 1
strcmp(vsub->imsi, IMSI) == 0
vsub->lac == 23
DREF VLR subscr MSISDN:42342 usage decreases to: 1
---
- after a while, a new conn sends a CM Service Request. VLR responds with Auth Req, 2nd auth vector
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_MM_CM_SERV_REQ
new conn
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn{RAN_CONN_S_NEW}: Allocated
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF unknown: MSC conn use + compl_l3 == 1 (0x1: compl_l3)
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DRLL Dispatching 04.08 message GSM48_MT_MM_CM_SERV_REQ (0x5:0x24)
DMM <- CM SERVICE REQUEST serv_type=0x08 MI(IMSI)=901700000010650
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_NEW}: Updated ID
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_INIT}: Allocated
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_INIT}: is child of RAN_conn(CM_SERVICE_REQ:901700000010650)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_INIT}: rev=R99 net=UTRAN Auth+Ciph
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_INIT}: Received Event PR_ARQ_E_START
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DREF VLR subscr MSISDN:42342 usage increases to: 2
DREF VLR subscr MSISDN:42342 usage increases to: 3
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_INIT}: proc_arq_vlr_fn_post_imsi()
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_INIT}: state_chg to PR_ARQ_S_WAIT_AUTH
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: Allocated
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: is child of Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650)
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: Received Event VLR_AUTH_E_START
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: state_chg to VLR_SUB_AS_WAIT_RESP
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_WAIT_RESP}: got auth tuple: use_count=1 key_seq=1 -- will use UMTS AKA (is_r99=yes, at->vec.auth_types=0x3)
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
- sending UMTS Auth Request for MSISDN:42342: tuple use_count=1 key_seq=1 auth_types=0x3 and...
- ...rand=c187a53a5e6b9d573cac7c74451fd46d
- ...autn=1843a645b98d00005b2d666af46c45d9
- ...expecting res=7db47cf7f81e4dc7
DREF VLR subscr MSISDN:42342 usage decreases to: 2
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_NEW}: Received Event RAN_CONN_E_COMPLETE_LAYER_3
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_NEW}: state_chg to RAN_CONN_S_AUTH_CIPH
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - compl_l3 == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_AUTH_CIPH}: Awaiting results for Auth+Ciph, overruling event RAN_CONN_E_UNUSED
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
cm_service_result_sent == 0
auth_request_sent == 1
- needs auth, not yet accepted
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
ran_conn_is_accepted() == false
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
- MS sends Authen Response, VLR accepts and sends SecurityModeControl
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_MM_AUTH_RESP
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + dtap == 1 (0x2: dtap)
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DRLL Dispatching 04.08 message GSM48_MT_MM_AUTH_RESP (0x5:0x14)
DMM MSISDN:42342: MM UMTS AUTHENTICATION RESPONSE (res = 7db47cf7f81e4dc7)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_WAIT_RESP}: Received Event VLR_AUTH_E_MS_AUTH_RESP
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DVLR SUBSCR(MSISDN:42342) AUTH on UTRAN received RES: 7db47cf7f81e4dc7 (8 bytes)
DVLR SUBSCR(MSISDN:42342) AUTH established UMTS security context
refactor VLR FSM result handling Instead of keeping separate enums for FSM results and translating between those and the actual 04.08 reject causes that will ultimately reach the MS, just pass enum gsm48_reject_value cause codes around everywhere. Collapse some VLR *_timeout() and *_cancel() api to just *_cancel() with a gsm48 cause arg. (Hopefully) improve a few reject causes, but otherwise just aim for more transparent decisions on which cause value is used, for future fixes of returned causes. Depends: I6661f139e68a498fb1bef10c266c2f064b72774a (libosmocore) Change-Id: I27bf8d68737ff1f8dc6d11fb1eac3d391aab0cb1
2018-04-06 00:57:51 +00:00
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_WAIT_RESP}: Authentication terminating with result PASSED
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_WAIT_RESP}: state_chg to VLR_SUB_AS_AUTHENTICATED
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Removing from parent Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650)
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Freeing instance
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Deallocated
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_AUTH}: Received Event PR_ARQ_E_AUTH_RES
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_AUTH}: _proc_arq_vlr_node2()
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_AUTH}: Set Ciphering Mode
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DMM -> SECURITY MODE CONTROL MSISDN:42342
- sending SecurityModeControl for UE ctx 42 send_ck=0 new_key=1
- ...ik=1159ec926a50e98c034a6b7d7c9f418d
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_AUTH}: state_chg to PR_ARQ_S_WAIT_CIPH
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - dtap == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_AUTH_CIPH}: Awaiting results for Auth+Ciph, overruling event RAN_CONN_E_UNUSED
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
security_mode_ctrl_sent == 1
cm_service_result_sent == 0
- MS sends SecurityModeControl acceptance, VLR accepts; above Ciphering is an implicit CM Service Accept
DMM <- SECURITY MODE COMPLETE MSISDN:42342
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_CIPH}: Received Event PR_ARQ_E_CIPH_RES
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_CIPH}: _proc_arq_vlr_node2_post_ciph()
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DIUCS MSISDN:42342: tx CommonID 901700000010650
- Iu Common ID --RAN_UTRAN_IU--> MS (IMSI=901700000010650)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_CIPH}: _proc_arq_vlr_node2_post_vlr()
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_CIPH}: _proc_arq_vlr_post_pres()
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_CIPH}: _proc_arq_vlr_post_trace()
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_CIPH}: _proc_arq_vlr_post_imei()
refactor VLR FSM result handling Instead of keeping separate enums for FSM results and translating between those and the actual 04.08 reject causes that will ultimately reach the MS, just pass enum gsm48_reject_value cause codes around everywhere. Collapse some VLR *_timeout() and *_cancel() api to just *_cancel() with a gsm48 cause arg. (Hopefully) improve a few reject causes, but otherwise just aim for more transparent decisions on which cause value is used, for future fixes of returned causes. Depends: I6661f139e68a498fb1bef10c266c2f064b72774a (libosmocore) Change-Id: I27bf8d68737ff1f8dc6d11fb1eac3d391aab0cb1
2018-04-06 00:57:51 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_CIPH}: proc_arq_fsm_done(PASSED)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_CIPH}: state_chg to PR_ARQ_S_DONE
refactor VLR FSM result handling Instead of keeping separate enums for FSM results and translating between those and the actual 04.08 reject causes that will ultimately reach the MS, just pass enum gsm48_reject_value cause codes around everywhere. Collapse some VLR *_timeout() and *_cancel() api to just *_cancel() with a gsm48 cause arg. (Hopefully) improve a few reject causes, but otherwise just aim for more transparent decisions on which cause value is used, for future fixes of returned causes. Depends: I6661f139e68a498fb1bef10c266c2f064b72774a (libosmocore) Change-Id: I27bf8d68737ff1f8dc6d11fb1eac3d391aab0cb1
2018-04-06 00:57:51 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_DONE}: Process Access Request result: PASSED
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_ACCEPTED
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + cm_service == 1 (0x8: cm_service)
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_AUTH_CIPH}: state_chg to RAN_CONN_S_ACCEPTED
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_ACCEPTED}: ran_conn_fsm_has_active_transactions: still awaiting first request after a CM Service Request
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
cm_service_result_sent == 0
---
- a call is initiated
- SETUP gets forwarded to MNCC
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_CC_SETUP
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + dtap == 2 (0xa: dtap,cm_service)
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DRLL Dispatching 04.08 message GSM48_MT_CC_SETUP (0x3:0x5)
DCC Unknown transaction ID 8, creating new trans.
DCC (ti 08 sub MSISDN:42342 callref 80000002) New transaction
DREF VLR subscr MSISDN:42342 usage increases to: 3
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + trans_cc == 3 (0x1a: dtap,cm_service,trans_cc)
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DMM MSISDN:42342: rx msg GSM48_MT_CC_SETUP: received_cm_service_request changes to false
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - cm_service == 2 (0x12: dtap,trans_cc)
gsm_04_08_cc: Add global guard timer for MNCC The external MNCC handler may hang indefinitely in cases where the remote end of the MNCC ceases to work properly. Add a global guard timer to make sure the call reaches ACTIVE state. Change-Id: I7375d1e17cd746aac4eadfe1e587e82cf1630d3d Related: OS#3599
2018-10-10 15:00:49 +00:00
DCC (sub 42342) starting guard timer with 180 seconds
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DCC (ti 08 sub MSISDN:42342) new state NULL -> INITIATED
DCC Subscriber MSISDN:42342 (42342) sends SETUP to 123
DMNCC transmit message MNCC_SETUP_IND
DCC Sending 'MNCC_SETUP_IND' to MNCC.
MSC --> MNCC: callref 0x80000002: MNCC_SETUP_IND
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_ACCEPTED}: Received Event RAN_CONN_E_COMMUNICATING
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_ACCEPTED}: state_chg to RAN_CONN_S_COMMUNICATING
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - dtap == 1 (0x10: trans_cc)
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
- MNCC says that's fine
DMNCC receive message MNCC_CALL_PROC_REQ
gsm_04_08_cc: Add global guard timer for MNCC The external MNCC handler may hang indefinitely in cases where the remote end of the MNCC ceases to work properly. Add a global guard timer to make sure the call reaches ACTIVE state. Change-Id: I7375d1e17cd746aac4eadfe1e587e82cf1630d3d Related: OS#3599
2018-10-10 15:00:49 +00:00
DCC (sub 42342) stopping pending guard timer
DCC (sub 42342) starting guard timer with 180 seconds
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DCC (ti 08 sub 42342) Received 'MNCC_CALL_PROC_REQ' from MNCC in state 1 (INITIATED)
DCC (ti 08 sub MSISDN:42342) new state INITIATED -> MO_CALL_PROC
DMSC msc_tx 2 bytes to MSISDN:42342 via RAN_UTRAN_IU
- DTAP --RAN_UTRAN_IU--> MS: GSM48_MT_CC_CALL_PROC: 8302
- DTAP matches expected message
MS <--Call Assignment-- MSC: subscr=MSISDN:42342 callref=0x80000002
- But the other side's MSISDN could not be resolved, MNCC tells us to cancel
DMNCC receive message MNCC_REL_REQ
gsm_04_08_cc: Add global guard timer for MNCC The external MNCC handler may hang indefinitely in cases where the remote end of the MNCC ceases to work properly. Add a global guard timer to make sure the call reaches ACTIVE state. Change-Id: I7375d1e17cd746aac4eadfe1e587e82cf1630d3d Related: OS#3599
2018-10-10 15:00:49 +00:00
DCC (sub 42342) stopping pending guard timer
DCC (sub 42342) starting guard timer with 180 seconds
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DCC (ti 08 sub 42342) Received 'MNCC_REL_REQ' from MNCC in state 3 (MO_CALL_PROC)
DCC starting timer T308 with 10 seconds
DCC (ti 08 sub MSISDN:42342) new state MO_CALL_PROC -> RELEASE_REQ
DMSC msc_tx 2 bytes to MSISDN:42342 via RAN_UTRAN_IU
- DTAP --RAN_UTRAN_IU--> MS: GSM48_MT_CC_RELEASE: 832d
- DTAP matches expected message
- Total time passed: 10.000023 s
DCC starting timer T308 with 10 seconds
DMSC msc_tx 2 bytes to MSISDN:42342 via RAN_UTRAN_IU
- DTAP --RAN_UTRAN_IU--> MS: GSM48_MT_CC_RELEASE: 832d
- DTAP matches expected message
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_CC_RELEASE_COMPL
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + dtap == 2 (0x12: dtap,trans_cc)
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DRLL Dispatching 04.08 message GSM48_MT_CC_RELEASE_COMPL (0x3:0x2a)
DCC stopping pending timer T308
DMNCC transmit message MNCC_REL_CNF
DCC Sending 'MNCC_REL_CNF' to MNCC.
MSC --> MNCC: callref 0x80000002: MNCC_REL_CNF
MS <--Call Release-- MSC: subscr=MSISDN:42342 callref=0x0
DCC (ti 08 sub MSISDN:42342) new state RELEASE_REQ -> NULL
gsm_04_08_cc: Add global guard timer for MNCC The external MNCC handler may hang indefinitely in cases where the remote end of the MNCC ceases to work properly. Add a global guard timer to make sure the call reaches ACTIVE state. Change-Id: I7375d1e17cd746aac4eadfe1e587e82cf1630d3d Related: OS#3599
2018-10-10 15:00:49 +00:00
DCC (sub 42342) stopping pending guard timer
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DREF VLR subscr MSISDN:42342 usage decreases to: 2
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - trans_cc == 1 (0x2: dtap)
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_COMMUNICATING}: Received Event RAN_CONN_E_COMMUNICATING
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - dtap == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_COMMUNICATING}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_COMMUNICATING}: state_chg to RAN_CONN_S_RELEASING
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + release == 1 (0x100: release)
refactor subscr_conn and subscr_conn_fsm de-/alloc Refactor: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. 2. Add separate AUTH_CIPH state to the FSM. 3. Use conn->use_count to trigger conn release. 4. Add separate RELEASING state to the FSM. 5. Add rate counters for each of the three Complete Layer 3 types. Details: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. Historically, a gsm_subscriber_connection was allocated in libbsc land, and only upon Complete Layer 3 did libmsc add the fsm instance. After splitting openbsc.git into a separate osmo-msc, this is no longer necessary, hence: Closely tie gsm_subscriber_connection allocation to the subscr_conn_fsm instance: talloc the conn as a child of the FSM instance, and discard the conn as soon as the FSM terminates. 2. Add separate AUTH_CIPH state to the FSM. Decoding the Complete Layer 3 message is distinctly separate from waiting for the VLR FSMs to conclude. Use the NEW state as "we don't know if this is a valid message yet", and the AUTH_CIPH state as "evaluating, don't release". A profound effect of this: should we for any odd reason fail to leave the FSM's NEW state, the conn will be released right at the end of msc_compl_l3(), without needing to trigger release in each code path. 3. Use conn->use_count to trigger conn release. Before, the FSM itself would hold a use count on the conn, and hence we would need to ask it whether it is ready to release the conn yet by dispatching events, to achieve a use_count decrement. Instead, unite the FSM instance and conn, and do not hold a use count by the FSM. Hence, trigger an FSM "UNUSED" event only when the use_count reaches zero. As long as use counts are done correctly, the FSM will terminate correctly. These exceptions: - The new AUTH_CIPH state explicitly ignores UNUSED events, since we expect the use count to reach zero while evaluating Authentication and Ciphering. (I experimented with holding a use count by AUTH_CIPH onenter() and releasing by onleave(), but the use count and thus the conn are released before the next state can initiate transactions that would increment the use count again. Same thing for the VLR FSMs holding a use count, they should be done before we advance to the next state. The easiest is to simply expect zero use count during the AUTH_CIPH state.) - A CM Service Request means that even though the MSC would be through with all it wants to do, we shall still wait for a request to follow from the MS. Hence the FSM holds a use count on itself while a CM Service is pending. - While waiting for a Release/Clear Complete, the FSM holds a use count on itself. 4. Add separate RELEASING state to the FSM. If we decide to release for other reasons than a use count reaching zero, we still need to be able to wait for the msc_dtap() use count on the conn to release. (An upcoming patch will further use the RELEASING state to properly wait for Clear Complete / Release Complete messages.) 5. Add rate counters for each of the three Complete Layer 3 types. Besides LU, also count CM Service Request and Paging Response acceptance/rejections. Without these counters, only very few of the auth+ciph outcomes actually show in the counters. Related: OS#3122 Change-Id: I55feb379e176a96a831e105b86202b17a0ffe889
2018-03-30 22:02:14 +00:00
DREF VLR subscr MSISDN:42342 usage increases to: 3
DREF VLR subscr MSISDN:42342 usage decreases to: 2
- Iu Release --RAN_UTRAN_IU--> MS
properly receive BSSMAP Clear Complete and Iu Release Complete When sending a BSSMAP Clear or Iu Release, do not immediately discard the conn, but wait until a BSSMAP Clear Complete / Iu Release Complete has been received. Hence we will no longer show in the log that an incoming Release/Clear Complete belongs to an unknown subscriber, but will still be around to properly log the release. Related: OS#3122 Change-Id: Ie4c6aaba3866d6e5b98004e8870a215e8cf8ffc1
2018-04-01 18:55:54 +00:00
- RNC sends Iu Release Complete
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - release == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_RELEASING}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_RELEASING}: state_chg to RAN_CONN_S_RELEASED
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_RELEASED}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_DONE}: Terminating (cause = OSMO_FSM_TERM_PARENT)
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_DONE}: Removing from parent RAN_conn(CM_SERVICE_REQ:901700000010650)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_DONE}: Freeing instance
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_DONE}: Deallocated
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DRLL MSISDN:42342: Freeing RAN connection
refactor subscr_conn and subscr_conn_fsm de-/alloc Refactor: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. 2. Add separate AUTH_CIPH state to the FSM. 3. Use conn->use_count to trigger conn release. 4. Add separate RELEASING state to the FSM. 5. Add rate counters for each of the three Complete Layer 3 types. Details: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. Historically, a gsm_subscriber_connection was allocated in libbsc land, and only upon Complete Layer 3 did libmsc add the fsm instance. After splitting openbsc.git into a separate osmo-msc, this is no longer necessary, hence: Closely tie gsm_subscriber_connection allocation to the subscr_conn_fsm instance: talloc the conn as a child of the FSM instance, and discard the conn as soon as the FSM terminates. 2. Add separate AUTH_CIPH state to the FSM. Decoding the Complete Layer 3 message is distinctly separate from waiting for the VLR FSMs to conclude. Use the NEW state as "we don't know if this is a valid message yet", and the AUTH_CIPH state as "evaluating, don't release". A profound effect of this: should we for any odd reason fail to leave the FSM's NEW state, the conn will be released right at the end of msc_compl_l3(), without needing to trigger release in each code path. 3. Use conn->use_count to trigger conn release. Before, the FSM itself would hold a use count on the conn, and hence we would need to ask it whether it is ready to release the conn yet by dispatching events, to achieve a use_count decrement. Instead, unite the FSM instance and conn, and do not hold a use count by the FSM. Hence, trigger an FSM "UNUSED" event only when the use_count reaches zero. As long as use counts are done correctly, the FSM will terminate correctly. These exceptions: - The new AUTH_CIPH state explicitly ignores UNUSED events, since we expect the use count to reach zero while evaluating Authentication and Ciphering. (I experimented with holding a use count by AUTH_CIPH onenter() and releasing by onleave(), but the use count and thus the conn are released before the next state can initiate transactions that would increment the use count again. Same thing for the VLR FSMs holding a use count, they should be done before we advance to the next state. The easiest is to simply expect zero use count during the AUTH_CIPH state.) - A CM Service Request means that even though the MSC would be through with all it wants to do, we shall still wait for a request to follow from the MS. Hence the FSM holds a use count on itself while a CM Service is pending. - While waiting for a Release/Clear Complete, the FSM holds a use count on itself. 4. Add separate RELEASING state to the FSM. If we decide to release for other reasons than a use count reaching zero, we still need to be able to wait for the msc_dtap() use count on the conn to release. (An upcoming patch will further use the RELEASING state to properly wait for Clear Complete / Release Complete messages.) 5. Add rate counters for each of the three Complete Layer 3 types. Besides LU, also count CM Service Request and Paging Response acceptance/rejections. Without these counters, only very few of the auth+ciph outcomes actually show in the counters. Related: OS#3122 Change-Id: I55feb379e176a96a831e105b86202b17a0ffe889
2018-03-30 22:02:14 +00:00
DREF VLR subscr MSISDN:42342 usage decreases to: 1
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_RELEASED}: Freeing instance
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_RELEASED}: Deallocated
llist_count(&net->ran_conns) == 0
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DREF freeing VLR subscr MSISDN:42342
===== test_call_mo_to_unknown: SUCCESS
full talloc report on 'msgb' (total 0 bytes in 1 blocks)
make gsup ipa name configurable in osmo-msc.cfg Add a 'ipa-name' VTY command which overrides the default IPA name used by the MSC. This is a prerequisite for inter-MSC handover. Related: OS#3355 Change-Id: I317d6c59f77e92fbb2b875a83dc0ec2fa5cb6006
2018-12-06 11:06:59 +00:00
talloc_total_blocks(tall_bsc_ctx) == 13
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
===== test_call_mo_to_unknown_timeout
- Total time passed: 0.000000 s
- Location Update request causes a GSUP Send Auth Info request to HLR
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_MM_LOC_UPD_REQUEST
new conn
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn{RAN_CONN_S_NEW}: Allocated
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF unknown: MSC conn use + compl_l3 == 1 (0x1: compl_l3)
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DRLL Dispatching 04.08 message GSM48_MT_MM_LOC_UPD_REQUEST (0x5:0x8)
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_NEW}: Updated ID
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DMM LOCATION UPDATING REQUEST: MI(IMSI)=901700000010650 type=NORMAL
DMM LU/new-LAC: 23/23
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_IDLE}: Allocated
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_IDLE}: is child of RAN_conn(LU:901700000010650)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_IDLE}: rev=R99 net=UTRAN Auth+Ciph
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_IDLE}: Received Event VLR_ULA_E_UPDATE_LA
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DREF VLR subscr unknown usage increases to: 1
DVLR set IMSI on subscriber; IMSI=901700000010650 id=901700000010650
DVLR New subscr, IMSI: 901700000010650
DREF VLR subscr IMSI:901700000010650 usage increases to: 2
DREF VLR subscr IMSI:901700000010650 usage decreases to: 1
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_IDLE}: vlr_loc_upd_node1()
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_IDLE}: state_chg to VLR_ULA_S_WAIT_AUTH
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: Allocated
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: is child of vlr_lu_fsm(LU:901700000010650)
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: Received Event VLR_AUTH_E_START
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DVLR GSUP tx: 08010809710000000156f0
GSUP --> HLR: OSMO_GSUP_MSGT_SEND_AUTH_INFO_REQUEST: 08010809710000000156f0
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: state_chg to VLR_SUB_AS_NEEDS_AUTH_WAIT_AI
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_NEW}: Received Event RAN_CONN_E_COMPLETE_LAYER_3
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_NEW}: state_chg to RAN_CONN_S_AUTH_CIPH
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF IMSI:901700000010650: MSC conn use - compl_l3 == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_AUTH_CIPH}: Awaiting results for Auth+Ciph, overruling event RAN_CONN_E_UNUSED
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
lu_result_sent == 0
- from HLR, rx _SEND_AUTH_INFO_RESULT; VLR sends Auth Req to MS
<-- GSUP rx OSMO_GSUP_MSGT_SEND_AUTH_INFO_RESULT: 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
DVLR GSUP rx 511: 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
DREF VLR subscr IMSI:901700000010650 usage increases to: 2
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_NEEDS_AUTH_WAIT_AI}: Received Event VLR_AUTH_E_HLR_SAI_ACK
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DVLR SUBSCR(IMSI:901700000010650) Received 5 auth tuples
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_NEEDS_AUTH_WAIT_AI}: state_chg to VLR_SUB_AS_WAIT_RESP
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_WAIT_RESP}: got auth tuple: use_count=1 key_seq=0 -- will use UMTS AKA (is_r99=yes, at->vec.auth_types=0x3)
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
- sending UMTS Auth Request for IMSI:901700000010650: tuple use_count=1 key_seq=0 auth_types=0x3 and...
- ...rand=39fa2f4e3d523d8619a73b4f65c3e14d
- ...autn=8704f5ba55f30000d2ee44b22c8ea919
- ...expecting res=e229c19e791f2e41
DREF VLR subscr IMSI:901700000010650 usage decreases to: 1
<-- GSUP rx OSMO_GSUP_MSGT_SEND_AUTH_INFO_RESULT: vlr_gsupc_read_cb() returns 0
auth_request_sent == 1
lu_result_sent == 0
- MS sends Authen Response, VLR accepts and sends SecurityModeControl
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_MM_AUTH_RESP
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF IMSI:901700000010650: MSC conn use + dtap == 1 (0x2: dtap)
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DRLL Dispatching 04.08 message GSM48_MT_MM_AUTH_RESP (0x5:0x14)
DMM IMSI:901700000010650: MM UMTS AUTHENTICATION RESPONSE (res = e229c19e791f2e41)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_WAIT_RESP}: Received Event VLR_AUTH_E_MS_AUTH_RESP
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DVLR SUBSCR(IMSI:901700000010650) AUTH on UTRAN received RES: e229c19e791f2e41 (8 bytes)
DVLR SUBSCR(IMSI:901700000010650) AUTH established UMTS security context
refactor VLR FSM result handling Instead of keeping separate enums for FSM results and translating between those and the actual 04.08 reject causes that will ultimately reach the MS, just pass enum gsm48_reject_value cause codes around everywhere. Collapse some VLR *_timeout() and *_cancel() api to just *_cancel() with a gsm48 cause arg. (Hopefully) improve a few reject causes, but otherwise just aim for more transparent decisions on which cause value is used, for future fixes of returned causes. Depends: I6661f139e68a498fb1bef10c266c2f064b72774a (libosmocore) Change-Id: I27bf8d68737ff1f8dc6d11fb1eac3d391aab0cb1
2018-04-06 00:57:51 +00:00
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_WAIT_RESP}: Authentication terminating with result PASSED
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_WAIT_RESP}: state_chg to VLR_SUB_AS_AUTHENTICATED
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Removing from parent vlr_lu_fsm(LU:901700000010650)
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Freeing instance
DVLR VLR_Authenticate(LU:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Deallocated
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_AUTH}: Received Event VLR_ULA_E_AUTH_RES
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_AUTH}: vlr_loc_upd_post_auth()
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_AUTH}: Set Ciphering Mode
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DMM -> SECURITY MODE CONTROL IMSI:901700000010650
- sending SecurityModeControl for UE ctx 42 send_ck=0 new_key=1
- ...ik=27497388b6cb044648f396aa155b95ef
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_AUTH}: state_chg to VLR_ULA_S_WAIT_CIPH
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF IMSI:901700000010650: MSC conn use - dtap == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_AUTH_CIPH}: Awaiting results for Auth+Ciph, overruling event RAN_CONN_E_UNUSED
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
security_mode_ctrl_sent == 1
lu_result_sent == 0
- MS sends SecurityModeControl acceptance, VLR accepts and sends GSUP LU Req to HLR
DMM <- SECURITY MODE COMPLETE IMSI:901700000010650
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_CIPH}: Received Event VLR_ULA_E_CIPH_RES
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_CIPH}: vlr_loc_upd_post_ciph()
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DIUCS IMSI:901700000010650: tx CommonID 901700000010650
- Iu Common ID --RAN_UTRAN_IU--> MS (IMSI=901700000010650)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_CIPH}: vlr_loc_upd_node_4()
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_CIPH}: state_chg to VLR_ULA_S_WAIT_HLR_UPD
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_INIT}: Allocated
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_INIT}: is child of vlr_lu_fsm(LU:901700000010650)
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_INIT}: Received Event UPD_HLR_VLR_E_START
GSUP client: send CN domain IE on LU request Give the HLR a chance to send us updated subscriber data by indicating the CN domain to be Circuit Switched, only during a LU Request GSUP message. Adjust msc_vlr_tests to expect the added GSUP CN domain IE to indicate CS, i.e. append '280102'. Related: OS#3601 Change-Id: I0c2d33fbfdb4728e480679120d06b7f3a2ccfd76
2018-09-28 00:41:39 +00:00
DVLR GSUP tx: 04010809710000000156f0280102
GSUP --> HLR: OSMO_GSUP_MSGT_UPDATE_LOCATION_REQUEST: 04010809710000000156f0280102
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_INIT}: state_chg to UPD_HLR_VLR_S_WAIT_FOR_DATA
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
gsup_tx_confirmed == 1
lu_result_sent == 0
- HLR sends _INSERT_DATA_REQUEST, VLR responds with _INSERT_DATA_RESULT
<-- GSUP rx OSMO_GSUP_MSGT_INSERT_DATA_REQUEST: 10010809710000000156f00804032443f2
DVLR GSUP rx 17: 10010809710000000156f00804032443f2
DREF VLR subscr IMSI:901700000010650 usage increases to: 2
DVLR IMSI:901700000010650 has MSISDN:42342
DVLR SUBSCR(MSISDN:42342) VLR: update for IMSI=901700000010650 (MSISDN=42342, used=2)
DVLR GSUP tx: 12010809710000000156f0
GSUP --> HLR: OSMO_GSUP_MSGT_INSERT_DATA_RESULT: 12010809710000000156f0
DREF VLR subscr MSISDN:42342 usage decreases to: 1
<-- GSUP rx OSMO_GSUP_MSGT_INSERT_DATA_REQUEST: vlr_gsupc_read_cb() returns 0
lu_result_sent == 0
- HLR also sends GSUP _UPDATE_LOCATION_RESULT
<-- GSUP rx OSMO_GSUP_MSGT_UPDATE_LOCATION_RESULT: 06010809710000000156f0
DVLR GSUP rx 11: 06010809710000000156f0
DREF VLR subscr MSISDN:42342 usage increases to: 2
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_HLR_UPD}: Received Event VLR_ULA_E_HLR_LU_RES
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_WAIT_FOR_DATA}: Received Event UPD_HLR_VLR_E_UPD_LOC_ACK
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_WAIT_FOR_DATA}: state_chg to UPD_HLR_VLR_S_DONE
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_DONE}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_DONE}: Removing from parent vlr_lu_fsm(LU:901700000010650)
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_DONE}: Freeing instance
DVLR upd_hlr_vlr_fsm(LU:901700000010650){UPD_HLR_VLR_S_DONE}: Deallocated
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_HLR_UPD}: Received Event VLR_ULA_E_UPD_HLR_COMPL
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_HLR_UPD}: state_chg to VLR_ULA_S_WAIT_LU_COMPL
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_INIT}: Allocated
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_INIT}: is child of vlr_lu_fsm(LU:901700000010650)
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_INIT}: Received Event LU_COMPL_VLR_E_START
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_INIT}: state_chg to LU_COMPL_VLR_S_WAIT_SUB_PRES
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_WAIT_SUB_PRES}: Received Event LU_COMPL_VLR_E_SUB_PRES_COMPL
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_WAIT_SUB_PRES}: lu_compl_vlr_new_tmsi()
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_WAIT_SUB_PRES}: state_chg to LU_COMPL_VLR_S_WAIT_TMSI_CNF
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
- sending LU Accept for MSISDN:42342, with TMSI 0x03020100
DREF VLR subscr MSISDN:42342 usage decreases to: 1
<-- GSUP rx OSMO_GSUP_MSGT_UPDATE_LOCATION_RESULT: vlr_gsupc_read_cb() returns 0
lu_result_sent == 1
- a LU Accept with a new TMSI was sent, waiting for TMSI Realloc Compl
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
llist_count(&net->ran_conns) == 1
ran_conn_is_accepted() == false
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
- MS sends TMSI Realloc Complete
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_MM_TMSI_REALL_COMPL
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + dtap == 1 (0x2: dtap)
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DRLL Dispatching 04.08 message GSM48_MT_MM_TMSI_REALL_COMPL (0x5:0x1b)
DMM TMSI Reallocation Completed. Subscriber: MSISDN:42342
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_LU_COMPL}: Received Event VLR_ULA_E_NEW_TMSI_ACK
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_WAIT_TMSI_CNF}: Received Event LU_COMPL_VLR_E_NEW_TMSI_ACK
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DREF VLR subscr MSISDN:42342 usage increases to: 2
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_WAIT_TMSI_CNF}: state_chg to LU_COMPL_VLR_S_DONE
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_LU_COMPL}: Received Event VLR_ULA_E_LU_COMPL_SUCCESS
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_DONE}: Terminating (cause = OSMO_FSM_TERM_PARENT)
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_DONE}: Removing from parent vlr_lu_fsm(LU:901700000010650)
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_DONE}: Freeing instance
DVLR lu_compl_vlr_fsm(LU:901700000010650){LU_COMPL_VLR_S_DONE}: Deallocated
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_WAIT_LU_COMPL}: state_chg to VLR_ULA_S_DONE
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_ACCEPTED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_AUTH_CIPH}: state_chg to RAN_CONN_S_ACCEPTED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_ACCEPTED}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_ACCEPTED}: state_chg to RAN_CONN_S_RELEASING
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + release == 2 (0x102: dtap,release)
refactor subscr_conn and subscr_conn_fsm de-/alloc Refactor: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. 2. Add separate AUTH_CIPH state to the FSM. 3. Use conn->use_count to trigger conn release. 4. Add separate RELEASING state to the FSM. 5. Add rate counters for each of the three Complete Layer 3 types. Details: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. Historically, a gsm_subscriber_connection was allocated in libbsc land, and only upon Complete Layer 3 did libmsc add the fsm instance. After splitting openbsc.git into a separate osmo-msc, this is no longer necessary, hence: Closely tie gsm_subscriber_connection allocation to the subscr_conn_fsm instance: talloc the conn as a child of the FSM instance, and discard the conn as soon as the FSM terminates. 2. Add separate AUTH_CIPH state to the FSM. Decoding the Complete Layer 3 message is distinctly separate from waiting for the VLR FSMs to conclude. Use the NEW state as "we don't know if this is a valid message yet", and the AUTH_CIPH state as "evaluating, don't release". A profound effect of this: should we for any odd reason fail to leave the FSM's NEW state, the conn will be released right at the end of msc_compl_l3(), without needing to trigger release in each code path. 3. Use conn->use_count to trigger conn release. Before, the FSM itself would hold a use count on the conn, and hence we would need to ask it whether it is ready to release the conn yet by dispatching events, to achieve a use_count decrement. Instead, unite the FSM instance and conn, and do not hold a use count by the FSM. Hence, trigger an FSM "UNUSED" event only when the use_count reaches zero. As long as use counts are done correctly, the FSM will terminate correctly. These exceptions: - The new AUTH_CIPH state explicitly ignores UNUSED events, since we expect the use count to reach zero while evaluating Authentication and Ciphering. (I experimented with holding a use count by AUTH_CIPH onenter() and releasing by onleave(), but the use count and thus the conn are released before the next state can initiate transactions that would increment the use count again. Same thing for the VLR FSMs holding a use count, they should be done before we advance to the next state. The easiest is to simply expect zero use count during the AUTH_CIPH state.) - A CM Service Request means that even though the MSC would be through with all it wants to do, we shall still wait for a request to follow from the MS. Hence the FSM holds a use count on itself while a CM Service is pending. - While waiting for a Release/Clear Complete, the FSM holds a use count on itself. 4. Add separate RELEASING state to the FSM. If we decide to release for other reasons than a use count reaching zero, we still need to be able to wait for the msc_dtap() use count on the conn to release. (An upcoming patch will further use the RELEASING state to properly wait for Clear Complete / Release Complete messages.) 5. Add rate counters for each of the three Complete Layer 3 types. Besides LU, also count CM Service Request and Paging Response acceptance/rejections. Without these counters, only very few of the auth+ciph outcomes actually show in the counters. Related: OS#3122 Change-Id: I55feb379e176a96a831e105b86202b17a0ffe889
2018-03-30 22:02:14 +00:00
DREF VLR subscr MSISDN:42342 usage increases to: 3
DREF VLR subscr MSISDN:42342 usage decreases to: 2
- Iu Release --RAN_UTRAN_IU--> MS
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - dtap == 1 (0x100: release)
properly receive BSSMAP Clear Complete and Iu Release Complete When sending a BSSMAP Clear or Iu Release, do not immediately discard the conn, but wait until a BSSMAP Clear Complete / Iu Release Complete has been received. Hence we will no longer show in the log that an incoming Release/Clear Complete belongs to an unknown subscriber, but will still be around to properly log the release. Related: OS#3122 Change-Id: Ie4c6aaba3866d6e5b98004e8870a215e8cf8ffc1
2018-04-01 18:55:54 +00:00
iu_release_sent == 1
- LU was successful, and the conn has already been closed
- RNC sends Iu Release Complete
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - release == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_RELEASING}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_RELEASING}: state_chg to RAN_CONN_S_RELEASED
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_RELEASED}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_DONE}: Terminating (cause = OSMO_FSM_TERM_PARENT)
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_DONE}: Removing from parent RAN_conn(LU:901700000010650)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_DONE}: fsm_lu_cleanup called with cause OSMO_FSM_TERM_PARENT
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_DONE}: Freeing instance
DVLR vlr_lu_fsm(LU:901700000010650){VLR_ULA_S_DONE}: Deallocated
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DRLL MSISDN:42342: Freeing RAN connection
refactor subscr_conn and subscr_conn_fsm de-/alloc Refactor: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. 2. Add separate AUTH_CIPH state to the FSM. 3. Use conn->use_count to trigger conn release. 4. Add separate RELEASING state to the FSM. 5. Add rate counters for each of the three Complete Layer 3 types. Details: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. Historically, a gsm_subscriber_connection was allocated in libbsc land, and only upon Complete Layer 3 did libmsc add the fsm instance. After splitting openbsc.git into a separate osmo-msc, this is no longer necessary, hence: Closely tie gsm_subscriber_connection allocation to the subscr_conn_fsm instance: talloc the conn as a child of the FSM instance, and discard the conn as soon as the FSM terminates. 2. Add separate AUTH_CIPH state to the FSM. Decoding the Complete Layer 3 message is distinctly separate from waiting for the VLR FSMs to conclude. Use the NEW state as "we don't know if this is a valid message yet", and the AUTH_CIPH state as "evaluating, don't release". A profound effect of this: should we for any odd reason fail to leave the FSM's NEW state, the conn will be released right at the end of msc_compl_l3(), without needing to trigger release in each code path. 3. Use conn->use_count to trigger conn release. Before, the FSM itself would hold a use count on the conn, and hence we would need to ask it whether it is ready to release the conn yet by dispatching events, to achieve a use_count decrement. Instead, unite the FSM instance and conn, and do not hold a use count by the FSM. Hence, trigger an FSM "UNUSED" event only when the use_count reaches zero. As long as use counts are done correctly, the FSM will terminate correctly. These exceptions: - The new AUTH_CIPH state explicitly ignores UNUSED events, since we expect the use count to reach zero while evaluating Authentication and Ciphering. (I experimented with holding a use count by AUTH_CIPH onenter() and releasing by onleave(), but the use count and thus the conn are released before the next state can initiate transactions that would increment the use count again. Same thing for the VLR FSMs holding a use count, they should be done before we advance to the next state. The easiest is to simply expect zero use count during the AUTH_CIPH state.) - A CM Service Request means that even though the MSC would be through with all it wants to do, we shall still wait for a request to follow from the MS. Hence the FSM holds a use count on itself while a CM Service is pending. - While waiting for a Release/Clear Complete, the FSM holds a use count on itself. 4. Add separate RELEASING state to the FSM. If we decide to release for other reasons than a use count reaching zero, we still need to be able to wait for the msc_dtap() use count on the conn to release. (An upcoming patch will further use the RELEASING state to properly wait for Clear Complete / Release Complete messages.) 5. Add rate counters for each of the three Complete Layer 3 types. Besides LU, also count CM Service Request and Paging Response acceptance/rejections. Without these counters, only very few of the auth+ciph outcomes actually show in the counters. Related: OS#3122 Change-Id: I55feb379e176a96a831e105b86202b17a0ffe889
2018-03-30 22:02:14 +00:00
DREF VLR subscr MSISDN:42342 usage decreases to: 1
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_RELEASED}: Freeing instance
DMM RAN_conn(LU:901700000010650){RAN_CONN_S_RELEASED}: Deallocated
llist_count(&net->ran_conns) == 0
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DREF VLR subscr MSISDN:42342 usage increases to: 2
vsub != NULL == 1
strcmp(vsub->imsi, IMSI) == 0
vsub->lac == 23
DREF VLR subscr MSISDN:42342 usage decreases to: 1
---
- after a while, a new conn sends a CM Service Request. VLR responds with Auth Req, 2nd auth vector
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_MM_CM_SERV_REQ
new conn
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn{RAN_CONN_S_NEW}: Allocated
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF unknown: MSC conn use + compl_l3 == 1 (0x1: compl_l3)
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DRLL Dispatching 04.08 message GSM48_MT_MM_CM_SERV_REQ (0x5:0x24)
DMM <- CM SERVICE REQUEST serv_type=0x08 MI(IMSI)=901700000010650
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_NEW}: Updated ID
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_INIT}: Allocated
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_INIT}: is child of RAN_conn(CM_SERVICE_REQ:901700000010650)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_INIT}: rev=R99 net=UTRAN Auth+Ciph
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_INIT}: Received Event PR_ARQ_E_START
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DREF VLR subscr MSISDN:42342 usage increases to: 2
DREF VLR subscr MSISDN:42342 usage increases to: 3
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_INIT}: proc_arq_vlr_fn_post_imsi()
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_INIT}: state_chg to PR_ARQ_S_WAIT_AUTH
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: Allocated
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: is child of Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650)
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: Received Event VLR_AUTH_E_START
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_NEEDS_AUTH}: state_chg to VLR_SUB_AS_WAIT_RESP
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_WAIT_RESP}: got auth tuple: use_count=1 key_seq=1 -- will use UMTS AKA (is_r99=yes, at->vec.auth_types=0x3)
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
- sending UMTS Auth Request for MSISDN:42342: tuple use_count=1 key_seq=1 auth_types=0x3 and...
- ...rand=c187a53a5e6b9d573cac7c74451fd46d
- ...autn=1843a645b98d00005b2d666af46c45d9
- ...expecting res=7db47cf7f81e4dc7
DREF VLR subscr MSISDN:42342 usage decreases to: 2
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_NEW}: Received Event RAN_CONN_E_COMPLETE_LAYER_3
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_NEW}: state_chg to RAN_CONN_S_AUTH_CIPH
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - compl_l3 == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_AUTH_CIPH}: Awaiting results for Auth+Ciph, overruling event RAN_CONN_E_UNUSED
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
cm_service_result_sent == 0
auth_request_sent == 1
- needs auth, not yet accepted
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
ran_conn_is_accepted() == false
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
- MS sends Authen Response, VLR accepts and sends SecurityModeControl
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_MM_AUTH_RESP
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + dtap == 1 (0x2: dtap)
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DRLL Dispatching 04.08 message GSM48_MT_MM_AUTH_RESP (0x5:0x14)
DMM MSISDN:42342: MM UMTS AUTHENTICATION RESPONSE (res = 7db47cf7f81e4dc7)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_WAIT_RESP}: Received Event VLR_AUTH_E_MS_AUTH_RESP
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DVLR SUBSCR(MSISDN:42342) AUTH on UTRAN received RES: 7db47cf7f81e4dc7 (8 bytes)
DVLR SUBSCR(MSISDN:42342) AUTH established UMTS security context
refactor VLR FSM result handling Instead of keeping separate enums for FSM results and translating between those and the actual 04.08 reject causes that will ultimately reach the MS, just pass enum gsm48_reject_value cause codes around everywhere. Collapse some VLR *_timeout() and *_cancel() api to just *_cancel() with a gsm48 cause arg. (Hopefully) improve a few reject causes, but otherwise just aim for more transparent decisions on which cause value is used, for future fixes of returned causes. Depends: I6661f139e68a498fb1bef10c266c2f064b72774a (libosmocore) Change-Id: I27bf8d68737ff1f8dc6d11fb1eac3d391aab0cb1
2018-04-06 00:57:51 +00:00
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_WAIT_RESP}: Authentication terminating with result PASSED
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_WAIT_RESP}: state_chg to VLR_SUB_AS_AUTHENTICATED
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Removing from parent Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650)
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Freeing instance
DVLR VLR_Authenticate(CM_SERVICE_REQ:901700000010650){VLR_SUB_AS_AUTHENTICATED}: Deallocated
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_AUTH}: Received Event PR_ARQ_E_AUTH_RES
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_AUTH}: _proc_arq_vlr_node2()
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_AUTH}: Set Ciphering Mode
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DMM -> SECURITY MODE CONTROL MSISDN:42342
- sending SecurityModeControl for UE ctx 42 send_ck=0 new_key=1
- ...ik=1159ec926a50e98c034a6b7d7c9f418d
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_AUTH}: state_chg to PR_ARQ_S_WAIT_CIPH
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - dtap == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_AUTH_CIPH}: Awaiting results for Auth+Ciph, overruling event RAN_CONN_E_UNUSED
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
security_mode_ctrl_sent == 1
cm_service_result_sent == 0
- MS sends SecurityModeControl acceptance, VLR accepts; above Ciphering is an implicit CM Service Accept
DMM <- SECURITY MODE COMPLETE MSISDN:42342
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_CIPH}: Received Event PR_ARQ_E_CIPH_RES
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_CIPH}: _proc_arq_vlr_node2_post_ciph()
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DIUCS MSISDN:42342: tx CommonID 901700000010650
- Iu Common ID --RAN_UTRAN_IU--> MS (IMSI=901700000010650)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_CIPH}: _proc_arq_vlr_node2_post_vlr()
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_CIPH}: _proc_arq_vlr_post_pres()
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_CIPH}: _proc_arq_vlr_post_trace()
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_CIPH}: _proc_arq_vlr_post_imei()
refactor VLR FSM result handling Instead of keeping separate enums for FSM results and translating between those and the actual 04.08 reject causes that will ultimately reach the MS, just pass enum gsm48_reject_value cause codes around everywhere. Collapse some VLR *_timeout() and *_cancel() api to just *_cancel() with a gsm48 cause arg. (Hopefully) improve a few reject causes, but otherwise just aim for more transparent decisions on which cause value is used, for future fixes of returned causes. Depends: I6661f139e68a498fb1bef10c266c2f064b72774a (libosmocore) Change-Id: I27bf8d68737ff1f8dc6d11fb1eac3d391aab0cb1
2018-04-06 00:57:51 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_CIPH}: proc_arq_fsm_done(PASSED)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_WAIT_CIPH}: state_chg to PR_ARQ_S_DONE
refactor VLR FSM result handling Instead of keeping separate enums for FSM results and translating between those and the actual 04.08 reject causes that will ultimately reach the MS, just pass enum gsm48_reject_value cause codes around everywhere. Collapse some VLR *_timeout() and *_cancel() api to just *_cancel() with a gsm48 cause arg. (Hopefully) improve a few reject causes, but otherwise just aim for more transparent decisions on which cause value is used, for future fixes of returned causes. Depends: I6661f139e68a498fb1bef10c266c2f064b72774a (libosmocore) Change-Id: I27bf8d68737ff1f8dc6d11fb1eac3d391aab0cb1
2018-04-06 00:57:51 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_DONE}: Process Access Request result: PASSED
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_ACCEPTED
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + cm_service == 1 (0x8: cm_service)
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_AUTH_CIPH}: state_chg to RAN_CONN_S_ACCEPTED
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_ACCEPTED}: ran_conn_fsm_has_active_transactions: still awaiting first request after a CM Service Request
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
cm_service_result_sent == 0
---
- a call is initiated
- SETUP gets forwarded to MNCC
MSC <--RAN_UTRAN_IU-- MS: GSM48_MT_CC_SETUP
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + dtap == 2 (0xa: dtap,cm_service)
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DRLL Dispatching 04.08 message GSM48_MT_CC_SETUP (0x3:0x5)
DCC Unknown transaction ID 8, creating new trans.
DCC (ti 08 sub MSISDN:42342 callref 80000003) New transaction
DREF VLR subscr MSISDN:42342 usage increases to: 3
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + trans_cc == 3 (0x1a: dtap,cm_service,trans_cc)
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DMM MSISDN:42342: rx msg GSM48_MT_CC_SETUP: received_cm_service_request changes to false
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - cm_service == 2 (0x12: dtap,trans_cc)
gsm_04_08_cc: Add global guard timer for MNCC The external MNCC handler may hang indefinitely in cases where the remote end of the MNCC ceases to work properly. Add a global guard timer to make sure the call reaches ACTIVE state. Change-Id: I7375d1e17cd746aac4eadfe1e587e82cf1630d3d Related: OS#3599
2018-10-10 15:00:49 +00:00
DCC (sub 42342) starting guard timer with 180 seconds
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DCC (ti 08 sub MSISDN:42342) new state NULL -> INITIATED
DCC Subscriber MSISDN:42342 (42342) sends SETUP to 123
DMNCC transmit message MNCC_SETUP_IND
DCC Sending 'MNCC_SETUP_IND' to MNCC.
MSC --> MNCC: callref 0x80000003: MNCC_SETUP_IND
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_ACCEPTED}: Received Event RAN_CONN_E_COMMUNICATING
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_ACCEPTED}: state_chg to RAN_CONN_S_COMMUNICATING
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - dtap == 1 (0x10: trans_cc)
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
- MNCC says that's fine
DMNCC receive message MNCC_CALL_PROC_REQ
gsm_04_08_cc: Add global guard timer for MNCC The external MNCC handler may hang indefinitely in cases where the remote end of the MNCC ceases to work properly. Add a global guard timer to make sure the call reaches ACTIVE state. Change-Id: I7375d1e17cd746aac4eadfe1e587e82cf1630d3d Related: OS#3599
2018-10-10 15:00:49 +00:00
DCC (sub 42342) stopping pending guard timer
DCC (sub 42342) starting guard timer with 180 seconds
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DCC (ti 08 sub 42342) Received 'MNCC_CALL_PROC_REQ' from MNCC in state 1 (INITIATED)
DCC (ti 08 sub MSISDN:42342) new state INITIATED -> MO_CALL_PROC
DMSC msc_tx 2 bytes to MSISDN:42342 via RAN_UTRAN_IU
- DTAP --RAN_UTRAN_IU--> MS: GSM48_MT_CC_CALL_PROC: 8302
- DTAP matches expected message
MS <--Call Assignment-- MSC: subscr=MSISDN:42342 callref=0x80000003
- But the other side's MSISDN could not be resolved, MNCC tells us to cancel
DMNCC receive message MNCC_REL_REQ
gsm_04_08_cc: Add global guard timer for MNCC The external MNCC handler may hang indefinitely in cases where the remote end of the MNCC ceases to work properly. Add a global guard timer to make sure the call reaches ACTIVE state. Change-Id: I7375d1e17cd746aac4eadfe1e587e82cf1630d3d Related: OS#3599
2018-10-10 15:00:49 +00:00
DCC (sub 42342) stopping pending guard timer
DCC (sub 42342) starting guard timer with 180 seconds
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DCC (ti 08 sub 42342) Received 'MNCC_REL_REQ' from MNCC in state 3 (MO_CALL_PROC)
DCC starting timer T308 with 10 seconds
DCC (ti 08 sub MSISDN:42342) new state MO_CALL_PROC -> RELEASE_REQ
DMSC msc_tx 2 bytes to MSISDN:42342 via RAN_UTRAN_IU
- DTAP --RAN_UTRAN_IU--> MS: GSM48_MT_CC_RELEASE: 832d
- DTAP matches expected message
- Despite our repeated CC Release Requests, the MS does not respond anymore
- Total time passed: 10.000023 s
DCC starting timer T308 with 10 seconds
DMSC msc_tx 2 bytes to MSISDN:42342 via RAN_UTRAN_IU
- DTAP --RAN_UTRAN_IU--> MS: GSM48_MT_CC_RELEASE: 832d
- DTAP matches expected message
- The CC Release times out and we still properly clear the conn
- Total time passed: 20.000046 s
MS <--Call Release-- MSC: subscr=MSISDN:42342 callref=0x80000003
DMNCC transmit message MNCC_REL_CNF
DCC Sending 'MNCC_REL_CNF' to MNCC.
MSC --> MNCC: callref 0x80000003: MNCC_REL_CNF
DCC (ti 08 sub MSISDN:42342) new state RELEASE_REQ -> NULL
gsm_04_08_cc: Add global guard timer for MNCC The external MNCC handler may hang indefinitely in cases where the remote end of the MNCC ceases to work properly. Add a global guard timer to make sure the call reaches ACTIVE state. Change-Id: I7375d1e17cd746aac4eadfe1e587e82cf1630d3d Related: OS#3599
2018-10-10 15:00:49 +00:00
DCC (sub 42342) stopping pending guard timer
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DREF VLR subscr MSISDN:42342 usage decreases to: 2
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - trans_cc == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_COMMUNICATING}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_COMMUNICATING}: state_chg to RAN_CONN_S_RELEASING
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use + release == 1 (0x100: release)
refactor subscr_conn and subscr_conn_fsm de-/alloc Refactor: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. 2. Add separate AUTH_CIPH state to the FSM. 3. Use conn->use_count to trigger conn release. 4. Add separate RELEASING state to the FSM. 5. Add rate counters for each of the three Complete Layer 3 types. Details: 1. Glue the gsm_subscriber_connection alloc to the subscr_conn_fsm. Historically, a gsm_subscriber_connection was allocated in libbsc land, and only upon Complete Layer 3 did libmsc add the fsm instance. After splitting openbsc.git into a separate osmo-msc, this is no longer necessary, hence: Closely tie gsm_subscriber_connection allocation to the subscr_conn_fsm instance: talloc the conn as a child of the FSM instance, and discard the conn as soon as the FSM terminates. 2. Add separate AUTH_CIPH state to the FSM. Decoding the Complete Layer 3 message is distinctly separate from waiting for the VLR FSMs to conclude. Use the NEW state as "we don't know if this is a valid message yet", and the AUTH_CIPH state as "evaluating, don't release". A profound effect of this: should we for any odd reason fail to leave the FSM's NEW state, the conn will be released right at the end of msc_compl_l3(), without needing to trigger release in each code path. 3. Use conn->use_count to trigger conn release. Before, the FSM itself would hold a use count on the conn, and hence we would need to ask it whether it is ready to release the conn yet by dispatching events, to achieve a use_count decrement. Instead, unite the FSM instance and conn, and do not hold a use count by the FSM. Hence, trigger an FSM "UNUSED" event only when the use_count reaches zero. As long as use counts are done correctly, the FSM will terminate correctly. These exceptions: - The new AUTH_CIPH state explicitly ignores UNUSED events, since we expect the use count to reach zero while evaluating Authentication and Ciphering. (I experimented with holding a use count by AUTH_CIPH onenter() and releasing by onleave(), but the use count and thus the conn are released before the next state can initiate transactions that would increment the use count again. Same thing for the VLR FSMs holding a use count, they should be done before we advance to the next state. The easiest is to simply expect zero use count during the AUTH_CIPH state.) - A CM Service Request means that even though the MSC would be through with all it wants to do, we shall still wait for a request to follow from the MS. Hence the FSM holds a use count on itself while a CM Service is pending. - While waiting for a Release/Clear Complete, the FSM holds a use count on itself. 4. Add separate RELEASING state to the FSM. If we decide to release for other reasons than a use count reaching zero, we still need to be able to wait for the msc_dtap() use count on the conn to release. (An upcoming patch will further use the RELEASING state to properly wait for Clear Complete / Release Complete messages.) 5. Add rate counters for each of the three Complete Layer 3 types. Besides LU, also count CM Service Request and Paging Response acceptance/rejections. Without these counters, only very few of the auth+ciph outcomes actually show in the counters. Related: OS#3122 Change-Id: I55feb379e176a96a831e105b86202b17a0ffe889
2018-03-30 22:02:14 +00:00
DREF VLR subscr MSISDN:42342 usage increases to: 3
DREF VLR subscr MSISDN:42342 usage decreases to: 2
- Iu Release --RAN_UTRAN_IU--> MS
properly receive BSSMAP Clear Complete and Iu Release Complete When sending a BSSMAP Clear or Iu Release, do not immediately discard the conn, but wait until a BSSMAP Clear Complete / Iu Release Complete has been received. Hence we will no longer show in the log that an incoming Release/Clear Complete belongs to an unknown subscriber, but will still be around to properly log the release. Related: OS#3122 Change-Id: Ie4c6aaba3866d6e5b98004e8870a215e8cf8ffc1
2018-04-01 18:55:54 +00:00
- RNC sends Iu Release Complete
msc conn ref counts: log human readable list of conn owners Change-Id: I2a09efafbdbdde0399238f7d79feea8612605201
2018-04-09 18:44:56 +00:00
DREF MSISDN:42342: MSC conn use - release == 0 (0x0: )
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_RELEASING}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_RELEASING}: state_chg to RAN_CONN_S_RELEASED
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_RELEASED}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_DONE}: Terminating (cause = OSMO_FSM_TERM_PARENT)
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_DONE}: Removing from parent RAN_conn(CM_SERVICE_REQ:901700000010650)
cosmetic: embed compl_l3_type in FSM id In the subscr_conn_fsm instance's ID, include the Complete Layer 3 type, so that we can see on the first glance whether a state transition belongs to MO or MT. The huge patch is due to the cosmetic change affecting nearly every single log line in the msc_vlr_tests, by nature of changing the FSM's ID. Related: OS#3122 Change-Id: I2a7e27e0f16df1872dcda64cb928c3b8528ea3f7
2018-04-02 21:17:50 +00:00
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_DONE}: Freeing instance
DVLR Process_Access_Request_VLR(CM_SERVICE_REQ:901700000010650){PR_ARQ_S_DONE}: Deallocated
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DRLL MSISDN:42342: Freeing RAN connection
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DREF VLR subscr MSISDN:42342 usage decreases to: 1
rename gsm_subscriber_connection to ran_conn In preparation for inter-BSC and inter-MSC handover, we need to separate the subscriber management logic from the actual RAN connections. What better time to finally rename gsm_subscriber_connection. * Name choice: In 2G, this is a connection to the BSS, but even though 3GPP TS commonly talk of "BSS-A" and "BSS-B" when explaining handover, it's not good to call it "bss_conn": in 3G a BSS is called RNS, IIUC. The overall term for 2G (GERAN) and 3G (UTRAN) is RAN: Radio Access Network. * Rationale: A subscriber in the MSC so far has only one RAN connection, but e.g. for inter-BSC handover, a second one needs to be created to handover to. Most of the items in the former gsm_subscriber_connection are actually related to the RAN, with only a few MM and RTP related items. So, as a first step, just rename it to ran_conn, to cosmetically prepare for moving the not strictly RAN related items away later. Also: - Rename some functions from msc_subscr_conn_* to ran_conn_* - Rename "Subscr_Conn" FSM instance name to "RAN_conn" - Rename SUBSCR_CONN_* to RAN_CONN_* Change-Id: Ic595f7a558d3553c067f77dc67543ab59659707a
2018-11-29 21:37:51 +00:00
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_RELEASED}: Freeing instance
DMM RAN_conn(CM_SERVICE_REQ:901700000010650){RAN_CONN_S_RELEASED}: Deallocated
llist_count(&net->ran_conns) == 0
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
DREF freeing VLR subscr MSISDN:42342
===== test_call_mo_to_unknown_timeout: SUCCESS
full talloc report on 'msgb' (total 0 bytes in 1 blocks)
make gsup ipa name configurable in osmo-msc.cfg Add a 'ipa-name' VTY command which overrides the default IPA name used by the MSC. This is a prerequisite for inter-MSC handover. Related: OS#3355 Change-Id: I317d6c59f77e92fbb2b875a83dc0ec2fa5cb6006
2018-12-06 11:06:59 +00:00
talloc_total_blocks(tall_bsc_ctx) == 13
msc_vlr_tests: add CC Release test and test to catch OS#2779 These tests helped to debug issue OS#2779. Now that they're here we might as well keep them. The test test_call_mo_to_unknown shows that an MS answering to the Release Request works as it should: the conn is torn down. The test test_call_mo_to_unknown_timeout currently expects the error: the conn remains active if the CC Release times out. This bug and the test expectations will be fixed in I46ff2e9b09b67e4e0d79cccf8c04936f17281fcb. Change-Id: Ic3c84520bff8c3fc82512d03ff6ab97d21b8fb7a
2018-03-15 12:42:10 +00:00
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00
full talloc report on 'msgb' (total 0 bytes in 1 blocks)
make gsup ipa name configurable in osmo-msc.cfg Add a 'ipa-name' VTY command which overrides the default IPA name used by the MSC. This is a prerequisite for inter-MSC handover. Related: OS#3355 Change-Id: I317d6c59f77e92fbb2b875a83dc0ec2fa5cb6006
2018-12-06 11:06:59 +00:00
talloc_total_blocks(tall_bsc_ctx) == 13
add msc_vlr_test_call to reproduce a sanitizer error On MT call, there is a bug in CC conn use which leads to an early free and use-after-free. Add msc_vlr_test_call to show both MO and MT call legs separately and reproduce the failure. It is visible in a sanitizer build (on debian 9). A subsequent patch will fix the bug: If0659a878deb383ed0300217e2c41c8c79b2b6a5 Related: OS#2672 Change-Id: I6c3ca0c660388b1e2c82df17ec540c846201b0c7
2017-11-21 16:13:23 +00:00