2018-03-30 21:04:04 +00:00
full talloc report on 'msgb' (total 0 bytes in 1 blocks)
2018-12-06 11:06:59 +00:00
talloc_total_blocks(tall_bsc_ctx) == 13
2018-03-30 21:04:04 +00:00
2018-03-01 23:40:58 +00:00
===== test_ms_timeout_lu_auth_resp
2017-01-25 14:04:16 +00:00
- Total time passed: 0.000000 s
- Location Update request causes a GSUP Send Auth Info request to HLR
2018-12-25 23:40:18 +00:00
MSC <--GERAN-A-- MS: GSM48_MT_MM_LOC_UPD_REQUEST
2017-01-25 14:04:16 +00:00
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
2018-04-09 18:44:56 +00:00
DREF unknown: MSC conn use + compl_l3 == 1 (0x1: compl_l3)
2017-03-10 01:15:20 +00:00
DRLL Dispatching 04.08 message GSM48_MT_MM_LOC_UPD_REQUEST (0x5:0x8)
2019-01-03 01:32:14 +00:00
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_NEW}: Updated ID
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_NEW}: LOCATION UPDATING REQUEST: MI=IMSI-901700000004620 LU-type=IMSI-ATTACH
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_NEW}: LU/new-LAC: 1/23
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_IDLE}: Allocated
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_IDLE}: is child of RAN_conn(IMSI-901700000004620:GERAN-A-0:LU)
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_IDLE}: rev=GSM net=GERAN Auth (no Ciph)
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_IDLE}: Received Event VLR_ULA_E_UPDATE_LA
2017-01-25 14:04:16 +00:00
DREF VLR subscr unknown usage increases to: 1
DVLR set IMSI on subscriber; IMSI=901700000004620 id=901700000004620
DVLR New subscr, IMSI: 901700000004620
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620 usage increases to: 2
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_NEW}: Updated ID
DREF VLR subscr IMSI-901700000004620 usage decreases to: 1
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_IDLE}: vlr_loc_upd_node1()
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_IDLE}: state_chg to VLR_ULA_S_WAIT_AUTH
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_NEEDS_AUTH}: Allocated
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_NEEDS_AUTH}: is child of vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU)
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_NEEDS_AUTH}: Received Event VLR_AUTH_E_START
2017-01-25 14:04:16 +00:00
DVLR GSUP tx: 08010809710000004026f0
GSUP --> HLR: OSMO_GSUP_MSGT_SEND_AUTH_INFO_REQUEST: 08010809710000004026f0
2019-01-03 01:32:14 +00:00
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_NEEDS_AUTH}: state_chg to VLR_SUB_AS_NEEDS_AUTH_WAIT_AI
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_NEW}: Received Event RAN_CONN_E_COMPLETE_LAYER_3
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_NEW}: state_chg to RAN_CONN_S_AUTH_CIPH
DREF IMSI-901700000004620: MSC conn use - compl_l3 == 0 (0x0: )
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_AUTH_CIPH}: Awaiting results for Auth+Ciph, overruling event RAN_CONN_E_UNUSED
2017-01-25 14:04:16 +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: 0a010809710000004026f003222010585df1ae287f6e273dce07090d61320b21042d8b2c3e220861855fb81fc2a800
DVLR GSUP rx 47: 0a010809710000004026f003222010585df1ae287f6e273dce07090d61320b21042d8b2c3e220861855fb81fc2a800
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620 usage increases to: 2
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_NEEDS_AUTH_WAIT_AI}: Received Event VLR_AUTH_E_HLR_SAI_ACK
DVLR SUBSCR(IMSI-901700000004620) Received 1 auth tuples
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_NEEDS_AUTH_WAIT_AI}: state_chg to VLR_SUB_AS_WAIT_RESP
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_WAIT_RESP}: got auth tuple: use_count=1 key_seq=0 -- will use GSM AKA (is_r99=no, at->vec.auth_types=0x1)
- sending GSM Auth Request for IMSI-901700000004620: tuple use_count=1 key_seq=0 auth_types=0x1 and...
2017-01-25 14:04:16 +00:00
- ...rand=585df1ae287f6e273dce07090d61320b
- ...expecting sres=2d8b2c3e
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620 usage decreases to: 1
2017-01-25 14:04:16 +00:00
<-- GSUP rx OSMO_GSUP_MSGT_SEND_AUTH_INFO_RESULT: vlr_gsupc_read_cb() returns 0
auth_request_sent == 1
lu_result_sent == 0
---
- MS fails to send an Authentication Response
- At first, we're still waiting
- Total time passed: 0.000423 s
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
2017-01-25 14:04:16 +00:00
lu_result_sent == 0
- Total time passed: 1.000658 s
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
2017-01-25 14:04:16 +00:00
lu_result_sent == 0
- Total time passed: 2.000893 s
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
2017-01-25 14:04:16 +00:00
lu_result_sent == 0
- Total time passed: 3.001128 s
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
2017-01-25 14:04:16 +00:00
lu_result_sent == 0
- Total time passed: 4.001363 s
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
2017-01-25 14:04:16 +00:00
lu_result_sent == 0
- Total time passed: 5.001598 s
2019-01-03 01:32:14 +00:00
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_AUTH_CIPH}: Timeout of T0
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_CN_CLOSE
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_AUTH_CIPH}: Close event, cause: CONGESTION
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_AUTH_CIPH}: state_chg to RAN_CONN_S_RELEASING
DREF IMSI-901700000004620: MSC conn use + release == 1 (0x100: release)
DREF VLR subscr IMSI-901700000004620 usage increases to: 2
- sending LU Reject for IMSI-901700000004620, cause 22
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_WAIT_AUTH}: state_chg to VLR_ULA_S_DONE
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_RELEASING}: Received Event RAN_CONN_E_CN_CLOSE
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_RELEASING}: Event RAN_CONN_E_CN_CLOSE not permitted
DREF VLR subscr IMSI-901700000004620 usage decreases to: 1
2018-12-25 23:40:18 +00:00
- BSSAP Clear --GERAN-A--> MS
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_TIMEOUT has passed, conn is gone.
2018-04-01 18:55:54 +00:00
bssap_clear_sent == 1
- BSS sends BSSMAP Clear Complete
2019-01-03 01:32:14 +00:00
DREF IMSI-901700000004620: MSC conn use - release == 0 (0x0: )
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_RELEASING}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_RELEASING}: state_chg to RAN_CONN_S_RELEASED
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_RELEASED}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_DONE}: Terminating (cause = OSMO_FSM_TERM_PARENT)
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_WAIT_RESP}: Terminating (cause = OSMO_FSM_TERM_PARENT)
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_WAIT_RESP}: Removing from parent vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU)
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_WAIT_RESP}: Freeing instance
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_WAIT_RESP}: Deallocated
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_DONE}: Removing from parent RAN_conn(IMSI-901700000004620:GERAN-A-0:LU)
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_DONE}: fsm_lu_cleanup called with cause OSMO_FSM_TERM_PARENT
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_DONE}: Freeing instance
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_DONE}: Deallocated
DRLL IMSI-901700000004620: Freeing RAN connection
DREF VLR subscr IMSI-901700000004620 usage decreases to: 0
DREF freeing VLR subscr IMSI-901700000004620
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_RELEASED}: Freeing instance
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_RELEASED}: 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
llist_count(&net->ran_conns) == 0
2017-01-25 14:04:16 +00:00
lu_result_sent == 2
2018-03-01 23:40:58 +00:00
===== test_ms_timeout_lu_auth_resp: SUCCESS
2017-01-25 14:04:16 +00:00
full talloc report on 'msgb' (total 0 bytes in 1 blocks)
2018-12-06 11:06:59 +00:00
talloc_total_blocks(tall_bsc_ctx) == 13
2017-01-25 14:04:16 +00:00
2018-03-01 23:40:58 +00:00
===== test_ms_timeout_cm_auth_resp
2017-01-25 14:04:16 +00:00
- Total time passed: 0.000000 s
- Location Update request causes a GSUP Send Auth Info request to HLR
2018-12-25 23:40:18 +00:00
MSC <--GERAN-A-- MS: GSM48_MT_MM_LOC_UPD_REQUEST
2017-01-25 14:04:16 +00:00
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
2018-04-09 18:44:56 +00:00
DREF unknown: MSC conn use + compl_l3 == 1 (0x1: compl_l3)
2017-03-10 01:15:20 +00:00
DRLL Dispatching 04.08 message GSM48_MT_MM_LOC_UPD_REQUEST (0x5:0x8)
2019-01-03 01:32:14 +00:00
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_NEW}: Updated ID
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_NEW}: LOCATION UPDATING REQUEST: MI=IMSI-901700000004620 LU-type=IMSI-ATTACH
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_NEW}: LU/new-LAC: 1/23
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_IDLE}: Allocated
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_IDLE}: is child of RAN_conn(IMSI-901700000004620:GERAN-A-0:LU)
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_IDLE}: rev=GSM net=GERAN Auth (no Ciph)
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_IDLE}: Received Event VLR_ULA_E_UPDATE_LA
2017-01-25 14:04:16 +00:00
DREF VLR subscr unknown usage increases to: 1
DVLR set IMSI on subscriber; IMSI=901700000004620 id=901700000004620
DVLR New subscr, IMSI: 901700000004620
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620 usage increases to: 2
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_NEW}: Updated ID
DREF VLR subscr IMSI-901700000004620 usage decreases to: 1
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_IDLE}: vlr_loc_upd_node1()
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_IDLE}: state_chg to VLR_ULA_S_WAIT_AUTH
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_NEEDS_AUTH}: Allocated
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_NEEDS_AUTH}: is child of vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU)
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_NEEDS_AUTH}: Received Event VLR_AUTH_E_START
2017-01-25 14:04:16 +00:00
DVLR GSUP tx: 08010809710000004026f0
GSUP --> HLR: OSMO_GSUP_MSGT_SEND_AUTH_INFO_REQUEST: 08010809710000004026f0
2019-01-03 01:32:14 +00:00
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_NEEDS_AUTH}: state_chg to VLR_SUB_AS_NEEDS_AUTH_WAIT_AI
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_NEW}: Received Event RAN_CONN_E_COMPLETE_LAYER_3
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_NEW}: state_chg to RAN_CONN_S_AUTH_CIPH
DREF IMSI-901700000004620: MSC conn use - compl_l3 == 0 (0x0: )
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_AUTH_CIPH}: Awaiting results for Auth+Ciph, overruling event RAN_CONN_E_UNUSED
2017-01-25 14:04:16 +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: 0a010809710000004026f003222010585df1ae287f6e273dce07090d61320b21042d8b2c3e220861855fb81fc2a8000322201012aca96fb4ffdea5c985cbafa9b6e18b210420bde240220807fa7502e07e1c0003222010e7c03ba7cf0e2fde82b2dc4d63077d422104a29514ae2208e2b234f80788640003222010fa8f20b781b5881329d4fea26b1a3c5121045afc8d7222082392f14f709ae000032220100fd4cc8dbe8715d1f439e304edfd68dc2104bc8d1c5b2208da7cdd6bfe2d7000
DVLR GSUP rx 191: 0a010809710000004026f003222010585df1ae287f6e273dce07090d61320b21042d8b2c3e220861855fb81fc2a8000322201012aca96fb4ffdea5c985cbafa9b6e18b210420bde240220807fa7502e07e1c0003222010e7c03ba7cf0e2fde82b2dc4d63077d422104a29514ae2208e2b234f80788640003222010fa8f20b781b5881329d4fea26b1a3c5121045afc8d7222082392f14f709ae000032220100fd4cc8dbe8715d1f439e304edfd68dc2104bc8d1c5b2208da7cdd6bfe2d7000
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620 usage increases to: 2
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_NEEDS_AUTH_WAIT_AI}: Received Event VLR_AUTH_E_HLR_SAI_ACK
DVLR SUBSCR(IMSI-901700000004620) Received 5 auth tuples
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_NEEDS_AUTH_WAIT_AI}: state_chg to VLR_SUB_AS_WAIT_RESP
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_WAIT_RESP}: got auth tuple: use_count=1 key_seq=0 -- will use GSM AKA (is_r99=no, at->vec.auth_types=0x1)
- sending GSM Auth Request for IMSI-901700000004620: tuple use_count=1 key_seq=0 auth_types=0x1 and...
2017-01-25 14:04:16 +00:00
- ...rand=585df1ae287f6e273dce07090d61320b
- ...expecting sres=2d8b2c3e
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620 usage decreases to: 1
2017-01-25 14:04:16 +00:00
<-- 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 GSUP LU Req to HLR
2018-12-25 23:40:18 +00:00
MSC <--GERAN-A-- MS: GSM48_MT_MM_AUTH_RESP
2019-01-03 01:32:14 +00:00
DREF IMSI-901700000004620: MSC conn use + dtap == 1 (0x2: dtap)
2017-03-10 01:15:20 +00:00
DRLL Dispatching 04.08 message GSM48_MT_MM_AUTH_RESP (0x5:0x14)
2019-01-03 01:32:14 +00:00
DMM IMSI-901700000004620: MM GSM AUTHENTICATION RESPONSE (sres = 2d8b2c3e)
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_WAIT_RESP}: Received Event VLR_AUTH_E_MS_AUTH_RESP
DVLR SUBSCR(IMSI-901700000004620) AUTH on GERAN received SRES/RES: 2d8b2c3e (4 bytes)
DVLR SUBSCR(IMSI-901700000004620) AUTH established GSM security context
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_WAIT_RESP}: Authentication terminating with result PASSED
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_WAIT_RESP}: state_chg to VLR_SUB_AS_AUTHENTICATED
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_AUTHENTICATED}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_AUTHENTICATED}: Removing from parent vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU)
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_AUTHENTICATED}: Freeing instance
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_AUTHENTICATED}: Deallocated
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_WAIT_AUTH}: Received Event VLR_ULA_E_AUTH_RES
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_WAIT_AUTH}: vlr_loc_upd_post_auth()
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_WAIT_AUTH}: vlr_loc_upd_post_ciph()
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_WAIT_AUTH}: vlr_loc_upd_node_4()
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_WAIT_AUTH}: state_chg to VLR_ULA_S_WAIT_HLR_UPD
DVLR upd_hlr_vlr_fsm(IMSI-901700000004620:GERAN-A-0:LU){UPD_HLR_VLR_S_INIT}: Allocated
DVLR upd_hlr_vlr_fsm(IMSI-901700000004620:GERAN-A-0:LU){UPD_HLR_VLR_S_INIT}: is child of vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU)
DVLR upd_hlr_vlr_fsm(IMSI-901700000004620:GERAN-A-0:LU){UPD_HLR_VLR_S_INIT}: Received Event UPD_HLR_VLR_E_START
2018-09-28 00:41:39 +00:00
DVLR GSUP tx: 04010809710000004026f0280102
GSUP --> HLR: OSMO_GSUP_MSGT_UPDATE_LOCATION_REQUEST: 04010809710000004026f0280102
2019-01-03 01:32:14 +00:00
DVLR upd_hlr_vlr_fsm(IMSI-901700000004620:GERAN-A-0:LU){UPD_HLR_VLR_S_INIT}: state_chg to UPD_HLR_VLR_S_WAIT_FOR_DATA
DREF IMSI-901700000004620: MSC conn use - dtap == 0 (0x0: )
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_AUTH_CIPH}: Awaiting results for Auth+Ciph, overruling event RAN_CONN_E_UNUSED
2017-01-25 14:04:16 +00:00
lu_result_sent == 0
- HLR sends _INSERT_DATA_REQUEST, VLR responds with _INSERT_DATA_RESULT
<-- GSUP rx OSMO_GSUP_MSGT_INSERT_DATA_REQUEST: 10010809710000004026f00804036470f1
DVLR GSUP rx 17: 10010809710000004026f00804036470f1
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620 usage increases to: 2
2017-01-25 14:04:16 +00:00
DVLR IMSI:901700000004620 has MSISDN:46071
2019-01-03 01:32:14 +00:00
DVLR SUBSCR(IMSI-901700000004620:MSISDN-46071) VLR: update for IMSI=901700000004620 (MSISDN=46071, used=2)
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){RAN_CONN_S_AUTH_CIPH}: Updated ID
2017-01-25 14:04:16 +00:00
DVLR GSUP tx: 12010809710000004026f0
GSUP --> HLR: OSMO_GSUP_MSGT_INSERT_DATA_RESULT: 12010809710000004026f0
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage decreases to: 1
2017-01-25 14:04:16 +00:00
<-- 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: 06010809710000004026f0
DVLR GSUP rx 11: 06010809710000004026f0
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage increases to: 2
DVLR vlr_lu_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){VLR_ULA_S_WAIT_HLR_UPD}: Received Event VLR_ULA_E_HLR_LU_RES
DVLR upd_hlr_vlr_fsm(IMSI-901700000004620:GERAN-A-0:LU){UPD_HLR_VLR_S_WAIT_FOR_DATA}: Received Event UPD_HLR_VLR_E_UPD_LOC_ACK
DVLR upd_hlr_vlr_fsm(IMSI-901700000004620:GERAN-A-0:LU){UPD_HLR_VLR_S_WAIT_FOR_DATA}: state_chg to UPD_HLR_VLR_S_DONE
DVLR upd_hlr_vlr_fsm(IMSI-901700000004620:GERAN-A-0:LU){UPD_HLR_VLR_S_DONE}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
DVLR upd_hlr_vlr_fsm(IMSI-901700000004620:GERAN-A-0:LU){UPD_HLR_VLR_S_DONE}: Removing from parent vlr_lu_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU)
DVLR upd_hlr_vlr_fsm(IMSI-901700000004620:GERAN-A-0:LU){UPD_HLR_VLR_S_DONE}: Freeing instance
DVLR upd_hlr_vlr_fsm(IMSI-901700000004620:GERAN-A-0:LU){UPD_HLR_VLR_S_DONE}: Deallocated
DVLR vlr_lu_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){VLR_ULA_S_WAIT_HLR_UPD}: Received Event VLR_ULA_E_UPD_HLR_COMPL
DVLR vlr_lu_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){VLR_ULA_S_WAIT_HLR_UPD}: state_chg to VLR_ULA_S_WAIT_LU_COMPL
DVLR lu_compl_vlr_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){LU_COMPL_VLR_S_INIT}: Allocated
DVLR lu_compl_vlr_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){LU_COMPL_VLR_S_INIT}: is child of vlr_lu_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU)
DVLR lu_compl_vlr_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){LU_COMPL_VLR_S_INIT}: Received Event LU_COMPL_VLR_E_START
DVLR lu_compl_vlr_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){LU_COMPL_VLR_S_INIT}: state_chg to LU_COMPL_VLR_S_WAIT_SUB_PRES
DVLR lu_compl_vlr_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){LU_COMPL_VLR_S_WAIT_SUB_PRES}: Received Event LU_COMPL_VLR_E_SUB_PRES_COMPL
- sending LU Accept for IMSI-901700000004620:MSISDN-46071
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage increases to: 3
DVLR lu_compl_vlr_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){LU_COMPL_VLR_S_WAIT_SUB_PRES}: state_chg to LU_COMPL_VLR_S_DONE
DVLR vlr_lu_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){VLR_ULA_S_WAIT_LU_COMPL}: Received Event VLR_ULA_E_LU_COMPL_SUCCESS
DVLR lu_compl_vlr_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){LU_COMPL_VLR_S_DONE}: Terminating (cause = OSMO_FSM_TERM_PARENT)
DVLR lu_compl_vlr_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){LU_COMPL_VLR_S_DONE}: Removing from parent vlr_lu_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU)
DVLR lu_compl_vlr_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){LU_COMPL_VLR_S_DONE}: Freeing instance
DVLR lu_compl_vlr_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){LU_COMPL_VLR_S_DONE}: Deallocated
DVLR vlr_lu_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){VLR_ULA_S_WAIT_LU_COMPL}: state_chg to VLR_ULA_S_DONE
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_ACCEPTED
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){RAN_CONN_S_AUTH_CIPH}: state_chg to RAN_CONN_S_ACCEPTED
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){RAN_CONN_S_ACCEPTED}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){RAN_CONN_S_ACCEPTED}: state_chg to RAN_CONN_S_RELEASING
DREF IMSI-901700000004620:MSISDN-46071: MSC conn use + release == 1 (0x100: release)
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage increases to: 4
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage decreases to: 3
2018-12-25 23:40:18 +00:00
- BSSAP Clear --GERAN-A--> MS
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage decreases to: 2
2018-04-01 18:55:54 +00:00
<-- GSUP rx OSMO_GSUP_MSGT_UPDATE_LOCATION_RESULT: vlr_gsupc_read_cb() returns 0
bssap_clear_sent == 1
- LU was successful, and the conn has already been closed
lu_result_sent == 1
- BSS sends BSSMAP Clear Complete
2019-01-03 01:32:14 +00:00
DREF IMSI-901700000004620:MSISDN-46071: MSC conn use - release == 0 (0x0: )
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){RAN_CONN_S_RELEASING}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){RAN_CONN_S_RELEASING}: state_chg to RAN_CONN_S_RELEASED
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){RAN_CONN_S_RELEASED}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
DVLR vlr_lu_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){VLR_ULA_S_DONE}: Terminating (cause = OSMO_FSM_TERM_PARENT)
DVLR vlr_lu_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){VLR_ULA_S_DONE}: Removing from parent RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU)
DVLR vlr_lu_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){VLR_ULA_S_DONE}: fsm_lu_cleanup called with cause OSMO_FSM_TERM_PARENT
DVLR vlr_lu_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){VLR_ULA_S_DONE}: Freeing instance
DVLR vlr_lu_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){VLR_ULA_S_DONE}: Deallocated
DRLL IMSI-901700000004620:MSISDN-46071: Freeing RAN connection
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage decreases to: 1
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){RAN_CONN_S_RELEASED}: Freeing instance
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){RAN_CONN_S_RELEASED}: 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
llist_count(&net->ran_conns) == 0
2017-01-25 14:04:16 +00:00
---
- after a while, a new conn sends a CM Service Request. VLR responds with Auth Req, 2nd auth vector
2018-12-25 23:40:18 +00:00
MSC <--GERAN-A-- MS: GSM48_MT_MM_CM_SERV_REQ
2017-01-25 14:04:16 +00:00
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
2018-04-09 18:44:56 +00:00
DREF unknown: MSC conn use + compl_l3 == 1 (0x1: compl_l3)
2017-03-10 01:15:20 +00:00
DRLL Dispatching 04.08 message GSM48_MT_MM_CM_SERV_REQ (0x5:0x24)
2019-01-03 01:32:14 +00:00
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:CM_SERVICE_REQ){RAN_CONN_S_NEW}: Updated ID
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:CM_SERVICE_REQ){RAN_CONN_S_NEW}: Rx CM SERVICE REQUEST cm_service_type=0x08
DVLR Process_Access_Request_VLR(IMSI-901700000004620:GERAN-A-0:CM_SERVICE_REQ){PR_ARQ_S_INIT}: Allocated
DVLR Process_Access_Request_VLR(IMSI-901700000004620:GERAN-A-0:CM_SERVICE_REQ){PR_ARQ_S_INIT}: is child of RAN_conn(IMSI-901700000004620:GERAN-A-0:CM_SERVICE_REQ)
DVLR Process_Access_Request_VLR(IMSI-901700000004620:GERAN-A-0:CM_SERVICE_REQ){PR_ARQ_S_INIT}: rev=GSM net=GERAN Auth (no Ciph)
DVLR Process_Access_Request_VLR(IMSI-901700000004620:GERAN-A-0:CM_SERVICE_REQ){PR_ARQ_S_INIT}: Received Event PR_ARQ_E_START
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage increases to: 2
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage increases to: 3
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){RAN_CONN_S_NEW}: Updated ID
DVLR Process_Access_Request_VLR(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){PR_ARQ_S_INIT}: proc_arq_vlr_fn_post_imsi()
DVLR Process_Access_Request_VLR(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){PR_ARQ_S_INIT}: state_chg to PR_ARQ_S_WAIT_AUTH
DVLR VLR_Authenticate(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){VLR_SUB_AS_NEEDS_AUTH}: Allocated
DVLR VLR_Authenticate(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){VLR_SUB_AS_NEEDS_AUTH}: is child of Process_Access_Request_VLR(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ)
DVLR VLR_Authenticate(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){VLR_SUB_AS_NEEDS_AUTH}: Received Event VLR_AUTH_E_START
DVLR VLR_Authenticate(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){VLR_SUB_AS_NEEDS_AUTH}: state_chg to VLR_SUB_AS_WAIT_RESP
DVLR VLR_Authenticate(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){VLR_SUB_AS_WAIT_RESP}: got auth tuple: use_count=1 key_seq=1 -- will use GSM AKA (is_r99=no, at->vec.auth_types=0x1)
- sending GSM Auth Request for IMSI-901700000004620:MSISDN-46071: tuple use_count=1 key_seq=1 auth_types=0x1 and...
2017-01-25 14:04:16 +00:00
- ...rand=12aca96fb4ffdea5c985cbafa9b6e18b
- ...expecting sres=20bde240
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage decreases to: 2
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){RAN_CONN_S_NEW}: Received Event RAN_CONN_E_COMPLETE_LAYER_3
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){RAN_CONN_S_NEW}: state_chg to RAN_CONN_S_AUTH_CIPH
DREF IMSI-901700000004620:MSISDN-46071: MSC conn use - compl_l3 == 0 (0x0: )
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){RAN_CONN_S_AUTH_CIPH}: Awaiting results for Auth+Ciph, overruling event RAN_CONN_E_UNUSED
2017-01-25 14:04:16 +00:00
cm_service_result_sent == 0
auth_request_sent == 1
---
- MS fails to send an Authentication Response
- At first, we're still waiting
- Total time passed: 0.000423 s
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
2017-01-25 14:04:16 +00:00
cm_service_result_sent == 0
- Total time passed: 1.000658 s
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
2017-01-25 14:04:16 +00:00
cm_service_result_sent == 0
- Total time passed: 2.000893 s
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
2017-01-25 14:04:16 +00:00
cm_service_result_sent == 0
- Total time passed: 3.001128 s
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
2017-01-25 14:04:16 +00:00
cm_service_result_sent == 0
- Total time passed: 4.001363 s
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
2017-01-25 14:04:16 +00:00
cm_service_result_sent == 0
- Total time passed: 5.001598 s
2019-01-03 01:32:14 +00:00
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){RAN_CONN_S_AUTH_CIPH}: Timeout of T0
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_CN_CLOSE
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){RAN_CONN_S_AUTH_CIPH}: Close event, cause: CONGESTION
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){RAN_CONN_S_AUTH_CIPH}: state_chg to RAN_CONN_S_RELEASING
DREF IMSI-901700000004620:MSISDN-46071: MSC conn use + release == 1 (0x100: release)
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage increases to: 3
DVLR Process_Access_Request_VLR(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){PR_ARQ_S_WAIT_AUTH}: Cancel: OSMO_FSM_TERM_ERROR
DVLR Process_Access_Request_VLR(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){PR_ARQ_S_WAIT_AUTH}: proc_arq_fsm_done(CONGESTION)
DVLR Process_Access_Request_VLR(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){PR_ARQ_S_WAIT_AUTH}: state_chg to PR_ARQ_S_DONE
DVLR Process_Access_Request_VLR(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){PR_ARQ_S_DONE}: Process Access Request result: CONGESTION
- sending CM Service Reject for IMSI-901700000004620:MSISDN-46071, cause: CONGESTION
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){RAN_CONN_S_RELEASING}: Received Event RAN_CONN_E_CN_CLOSE
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){RAN_CONN_S_RELEASING}: Event RAN_CONN_E_CN_CLOSE not permitted
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage decreases to: 2
2018-12-25 23:40:18 +00:00
- BSSAP Clear --GERAN-A--> MS
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_TIMEOUT has passed, conn is gone.
2018-04-01 18:55:54 +00:00
bssap_clear_sent == 1
- BSS sends BSSMAP Clear Complete
2019-01-03 01:32:14 +00:00
DREF IMSI-901700000004620:MSISDN-46071: MSC conn use - release == 0 (0x0: )
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){RAN_CONN_S_RELEASING}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){RAN_CONN_S_RELEASING}: state_chg to RAN_CONN_S_RELEASED
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){RAN_CONN_S_RELEASED}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
DVLR Process_Access_Request_VLR(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){PR_ARQ_S_DONE}: Terminating (cause = OSMO_FSM_TERM_PARENT)
DVLR VLR_Authenticate(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){VLR_SUB_AS_WAIT_RESP}: Terminating (cause = OSMO_FSM_TERM_PARENT)
DVLR VLR_Authenticate(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){VLR_SUB_AS_WAIT_RESP}: Removing from parent Process_Access_Request_VLR(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ)
DVLR VLR_Authenticate(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){VLR_SUB_AS_WAIT_RESP}: Freeing instance
DVLR VLR_Authenticate(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){VLR_SUB_AS_WAIT_RESP}: Deallocated
DVLR Process_Access_Request_VLR(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){PR_ARQ_S_DONE}: Removing from parent RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ)
DVLR Process_Access_Request_VLR(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){PR_ARQ_S_DONE}: Freeing instance
DVLR Process_Access_Request_VLR(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){PR_ARQ_S_DONE}: Deallocated
DRLL IMSI-901700000004620:MSISDN-46071: Freeing RAN connection
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage decreases to: 1
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){RAN_CONN_S_RELEASED}: Freeing instance
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:CM_SERVICE_REQ){RAN_CONN_S_RELEASED}: 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
llist_count(&net->ran_conns) == 0
2017-01-25 14:04:16 +00:00
cm_service_result_sent == 2
2019-01-03 01:32:14 +00:00
DREF freeing VLR subscr IMSI-901700000004620:MSISDN-46071
2018-03-01 23:40:58 +00:00
===== test_ms_timeout_cm_auth_resp: SUCCESS
2017-01-25 14:04:16 +00:00
full talloc report on 'msgb' (total 0 bytes in 1 blocks)
2018-12-06 11:06:59 +00:00
talloc_total_blocks(tall_bsc_ctx) == 13
2017-01-25 14:04:16 +00:00
2018-03-01 23:40:58 +00:00
===== test_ms_timeout_paging
fix paging: add timeout to discard unsuccessful paging
Currently, if there is no reply from the BSS / RNC, a subscriber will remain as
"already paged" forever, and is never going to be paged again. Even on IMSI
Detach, the pending request will keep a ref count on the vlr_subscr.
Add a paging timeout, as gsm_network->paging_timeout and in the VTY on the
'msc' node as 'paging timeout (default|<1-65535>'. (There is a 'network' /
'T3113' in OsmoBSC, but to not confuse the two, give this a different name.)
Add test_ms_timeout_paging() test to verify the timeout works.
I hit this while testing Paging across multiple hNodeB, when a UE lost
connection to the hNodeB. I noticed that no matter how long I wait, no Paging
is sent out anymore, and found this embarrassing issue. Good grief...
The choice of 10 seconds is taken from https://osmocom.org/issues/2756
Change-Id: I2db6f1e2ad341cf9c2cc7a21ec2fca0bae5b2db5
2017-12-15 02:02:27 +00:00
- Total time passed: 0.000000 s
- Location Update request causes a GSUP LU request to HLR
2018-12-25 23:40:18 +00:00
MSC <--GERAN-A-- MS: GSM48_MT_MM_LOC_UPD_REQUEST
fix paging: add timeout to discard unsuccessful paging
Currently, if there is no reply from the BSS / RNC, a subscriber will remain as
"already paged" forever, and is never going to be paged again. Even on IMSI
Detach, the pending request will keep a ref count on the vlr_subscr.
Add a paging timeout, as gsm_network->paging_timeout and in the VTY on the
'msc' node as 'paging timeout (default|<1-65535>'. (There is a 'network' /
'T3113' in OsmoBSC, but to not confuse the two, give this a different name.)
Add test_ms_timeout_paging() test to verify the timeout works.
I hit this while testing Paging across multiple hNodeB, when a UE lost
connection to the hNodeB. I noticed that no matter how long I wait, no Paging
is sent out anymore, and found this embarrassing issue. Good grief...
The choice of 10 seconds is taken from https://osmocom.org/issues/2756
Change-Id: I2db6f1e2ad341cf9c2cc7a21ec2fca0bae5b2db5
2017-12-15 02:02:27 +00:00
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
2018-04-09 18:44:56 +00:00
DREF unknown: MSC conn use + compl_l3 == 1 (0x1: compl_l3)
fix paging: add timeout to discard unsuccessful paging
Currently, if there is no reply from the BSS / RNC, a subscriber will remain as
"already paged" forever, and is never going to be paged again. Even on IMSI
Detach, the pending request will keep a ref count on the vlr_subscr.
Add a paging timeout, as gsm_network->paging_timeout and in the VTY on the
'msc' node as 'paging timeout (default|<1-65535>'. (There is a 'network' /
'T3113' in OsmoBSC, but to not confuse the two, give this a different name.)
Add test_ms_timeout_paging() test to verify the timeout works.
I hit this while testing Paging across multiple hNodeB, when a UE lost
connection to the hNodeB. I noticed that no matter how long I wait, no Paging
is sent out anymore, and found this embarrassing issue. Good grief...
The choice of 10 seconds is taken from https://osmocom.org/issues/2756
Change-Id: I2db6f1e2ad341cf9c2cc7a21ec2fca0bae5b2db5
2017-12-15 02:02:27 +00:00
DRLL Dispatching 04.08 message GSM48_MT_MM_LOC_UPD_REQUEST (0x5:0x8)
2019-01-03 01:32:14 +00:00
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_NEW}: Updated ID
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_NEW}: LOCATION UPDATING REQUEST: MI=IMSI-901700000004620 LU-type=IMSI-ATTACH
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_NEW}: LU/new-LAC: 1/23
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_IDLE}: Allocated
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_IDLE}: is child of RAN_conn(IMSI-901700000004620:GERAN-A-0:LU)
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_IDLE}: rev=GSM net=GERAN (no Auth)
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_IDLE}: Received Event VLR_ULA_E_UPDATE_LA
fix paging: add timeout to discard unsuccessful paging
Currently, if there is no reply from the BSS / RNC, a subscriber will remain as
"already paged" forever, and is never going to be paged again. Even on IMSI
Detach, the pending request will keep a ref count on the vlr_subscr.
Add a paging timeout, as gsm_network->paging_timeout and in the VTY on the
'msc' node as 'paging timeout (default|<1-65535>'. (There is a 'network' /
'T3113' in OsmoBSC, but to not confuse the two, give this a different name.)
Add test_ms_timeout_paging() test to verify the timeout works.
I hit this while testing Paging across multiple hNodeB, when a UE lost
connection to the hNodeB. I noticed that no matter how long I wait, no Paging
is sent out anymore, and found this embarrassing issue. Good grief...
The choice of 10 seconds is taken from https://osmocom.org/issues/2756
Change-Id: I2db6f1e2ad341cf9c2cc7a21ec2fca0bae5b2db5
2017-12-15 02:02:27 +00:00
DREF VLR subscr unknown usage increases to: 1
DVLR set IMSI on subscriber; IMSI=901700000004620 id=901700000004620
DVLR New subscr, IMSI: 901700000004620
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620 usage increases to: 2
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_NEW}: Updated ID
DREF VLR subscr IMSI-901700000004620 usage decreases to: 1
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_IDLE}: vlr_loc_upd_node1()
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_IDLE}: vlr_loc_upd_post_auth()
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_IDLE}: vlr_loc_upd_post_ciph()
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_IDLE}: vlr_loc_upd_node_4()
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_IDLE}: state_chg to VLR_ULA_S_WAIT_HLR_UPD
DVLR upd_hlr_vlr_fsm(IMSI-901700000004620:GERAN-A-0:LU){UPD_HLR_VLR_S_INIT}: Allocated
DVLR upd_hlr_vlr_fsm(IMSI-901700000004620:GERAN-A-0:LU){UPD_HLR_VLR_S_INIT}: is child of vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU)
DVLR upd_hlr_vlr_fsm(IMSI-901700000004620:GERAN-A-0:LU){UPD_HLR_VLR_S_INIT}: Received Event UPD_HLR_VLR_E_START
2018-09-28 00:41:39 +00:00
DVLR GSUP tx: 04010809710000004026f0280102
GSUP --> HLR: OSMO_GSUP_MSGT_UPDATE_LOCATION_REQUEST: 04010809710000004026f0280102
2019-01-03 01:32:14 +00:00
DVLR upd_hlr_vlr_fsm(IMSI-901700000004620:GERAN-A-0:LU){UPD_HLR_VLR_S_INIT}: state_chg to UPD_HLR_VLR_S_WAIT_FOR_DATA
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_NEW}: Received Event RAN_CONN_E_COMPLETE_LAYER_3
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_NEW}: state_chg to RAN_CONN_S_AUTH_CIPH
DREF IMSI-901700000004620: MSC conn use - compl_l3 == 0 (0x0: )
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_AUTH_CIPH}: Awaiting results for Auth+Ciph, overruling event RAN_CONN_E_UNUSED
fix paging: add timeout to discard unsuccessful paging
Currently, if there is no reply from the BSS / RNC, a subscriber will remain as
"already paged" forever, and is never going to be paged again. Even on IMSI
Detach, the pending request will keep a ref count on the vlr_subscr.
Add a paging timeout, as gsm_network->paging_timeout and in the VTY on the
'msc' node as 'paging timeout (default|<1-65535>'. (There is a 'network' /
'T3113' in OsmoBSC, but to not confuse the two, give this a different name.)
Add test_ms_timeout_paging() test to verify the timeout works.
I hit this while testing Paging across multiple hNodeB, when a UE lost
connection to the hNodeB. I noticed that no matter how long I wait, no Paging
is sent out anymore, and found this embarrassing issue. Good grief...
The choice of 10 seconds is taken from https://osmocom.org/issues/2756
Change-Id: I2db6f1e2ad341cf9c2cc7a21ec2fca0bae5b2db5
2017-12-15 02:02:27 +00:00
lu_result_sent == 0
- HLR sends _INSERT_DATA_REQUEST, VLR responds with _INSERT_DATA_RESULT
<-- GSUP rx OSMO_GSUP_MSGT_INSERT_DATA_REQUEST: 10010809710000004026f00804036470f1
DVLR GSUP rx 17: 10010809710000004026f00804036470f1
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620 usage increases to: 2
fix paging: add timeout to discard unsuccessful paging
Currently, if there is no reply from the BSS / RNC, a subscriber will remain as
"already paged" forever, and is never going to be paged again. Even on IMSI
Detach, the pending request will keep a ref count on the vlr_subscr.
Add a paging timeout, as gsm_network->paging_timeout and in the VTY on the
'msc' node as 'paging timeout (default|<1-65535>'. (There is a 'network' /
'T3113' in OsmoBSC, but to not confuse the two, give this a different name.)
Add test_ms_timeout_paging() test to verify the timeout works.
I hit this while testing Paging across multiple hNodeB, when a UE lost
connection to the hNodeB. I noticed that no matter how long I wait, no Paging
is sent out anymore, and found this embarrassing issue. Good grief...
The choice of 10 seconds is taken from https://osmocom.org/issues/2756
Change-Id: I2db6f1e2ad341cf9c2cc7a21ec2fca0bae5b2db5
2017-12-15 02:02:27 +00:00
DVLR IMSI:901700000004620 has MSISDN:46071
2019-01-03 01:32:14 +00:00
DVLR SUBSCR(IMSI-901700000004620:MSISDN-46071) VLR: update for IMSI=901700000004620 (MSISDN=46071, used=2)
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){RAN_CONN_S_AUTH_CIPH}: Updated ID
fix paging: add timeout to discard unsuccessful paging
Currently, if there is no reply from the BSS / RNC, a subscriber will remain as
"already paged" forever, and is never going to be paged again. Even on IMSI
Detach, the pending request will keep a ref count on the vlr_subscr.
Add a paging timeout, as gsm_network->paging_timeout and in the VTY on the
'msc' node as 'paging timeout (default|<1-65535>'. (There is a 'network' /
'T3113' in OsmoBSC, but to not confuse the two, give this a different name.)
Add test_ms_timeout_paging() test to verify the timeout works.
I hit this while testing Paging across multiple hNodeB, when a UE lost
connection to the hNodeB. I noticed that no matter how long I wait, no Paging
is sent out anymore, and found this embarrassing issue. Good grief...
The choice of 10 seconds is taken from https://osmocom.org/issues/2756
Change-Id: I2db6f1e2ad341cf9c2cc7a21ec2fca0bae5b2db5
2017-12-15 02:02:27 +00:00
DVLR GSUP tx: 12010809710000004026f0
GSUP --> HLR: OSMO_GSUP_MSGT_INSERT_DATA_RESULT: 12010809710000004026f0
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage decreases to: 1
fix paging: add timeout to discard unsuccessful paging
Currently, if there is no reply from the BSS / RNC, a subscriber will remain as
"already paged" forever, and is never going to be paged again. Even on IMSI
Detach, the pending request will keep a ref count on the vlr_subscr.
Add a paging timeout, as gsm_network->paging_timeout and in the VTY on the
'msc' node as 'paging timeout (default|<1-65535>'. (There is a 'network' /
'T3113' in OsmoBSC, but to not confuse the two, give this a different name.)
Add test_ms_timeout_paging() test to verify the timeout works.
I hit this while testing Paging across multiple hNodeB, when a UE lost
connection to the hNodeB. I noticed that no matter how long I wait, no Paging
is sent out anymore, and found this embarrassing issue. Good grief...
The choice of 10 seconds is taken from https://osmocom.org/issues/2756
Change-Id: I2db6f1e2ad341cf9c2cc7a21ec2fca0bae5b2db5
2017-12-15 02:02:27 +00:00
<-- 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: 06010809710000004026f0
DVLR GSUP rx 11: 06010809710000004026f0
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage increases to: 2
DVLR vlr_lu_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){VLR_ULA_S_WAIT_HLR_UPD}: Received Event VLR_ULA_E_HLR_LU_RES
DVLR upd_hlr_vlr_fsm(IMSI-901700000004620:GERAN-A-0:LU){UPD_HLR_VLR_S_WAIT_FOR_DATA}: Received Event UPD_HLR_VLR_E_UPD_LOC_ACK
DVLR upd_hlr_vlr_fsm(IMSI-901700000004620:GERAN-A-0:LU){UPD_HLR_VLR_S_WAIT_FOR_DATA}: state_chg to UPD_HLR_VLR_S_DONE
DVLR upd_hlr_vlr_fsm(IMSI-901700000004620:GERAN-A-0:LU){UPD_HLR_VLR_S_DONE}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
DVLR upd_hlr_vlr_fsm(IMSI-901700000004620:GERAN-A-0:LU){UPD_HLR_VLR_S_DONE}: Removing from parent vlr_lu_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU)
DVLR upd_hlr_vlr_fsm(IMSI-901700000004620:GERAN-A-0:LU){UPD_HLR_VLR_S_DONE}: Freeing instance
DVLR upd_hlr_vlr_fsm(IMSI-901700000004620:GERAN-A-0:LU){UPD_HLR_VLR_S_DONE}: Deallocated
DVLR vlr_lu_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){VLR_ULA_S_WAIT_HLR_UPD}: Received Event VLR_ULA_E_UPD_HLR_COMPL
DVLR vlr_lu_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){VLR_ULA_S_WAIT_HLR_UPD}: state_chg to VLR_ULA_S_WAIT_LU_COMPL
DVLR lu_compl_vlr_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){LU_COMPL_VLR_S_INIT}: Allocated
DVLR lu_compl_vlr_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){LU_COMPL_VLR_S_INIT}: is child of vlr_lu_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU)
DVLR lu_compl_vlr_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){LU_COMPL_VLR_S_INIT}: Received Event LU_COMPL_VLR_E_START
DVLR lu_compl_vlr_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){LU_COMPL_VLR_S_INIT}: state_chg to LU_COMPL_VLR_S_WAIT_SUB_PRES
DVLR lu_compl_vlr_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){LU_COMPL_VLR_S_WAIT_SUB_PRES}: Received Event LU_COMPL_VLR_E_SUB_PRES_COMPL
- sending LU Accept for IMSI-901700000004620:MSISDN-46071
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage increases to: 3
DVLR lu_compl_vlr_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){LU_COMPL_VLR_S_WAIT_SUB_PRES}: state_chg to LU_COMPL_VLR_S_DONE
DVLR vlr_lu_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){VLR_ULA_S_WAIT_LU_COMPL}: Received Event VLR_ULA_E_LU_COMPL_SUCCESS
DVLR lu_compl_vlr_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){LU_COMPL_VLR_S_DONE}: Terminating (cause = OSMO_FSM_TERM_PARENT)
DVLR lu_compl_vlr_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){LU_COMPL_VLR_S_DONE}: Removing from parent vlr_lu_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU)
DVLR lu_compl_vlr_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){LU_COMPL_VLR_S_DONE}: Freeing instance
DVLR lu_compl_vlr_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){LU_COMPL_VLR_S_DONE}: Deallocated
DVLR vlr_lu_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){VLR_ULA_S_WAIT_LU_COMPL}: state_chg to VLR_ULA_S_DONE
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_ACCEPTED
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){RAN_CONN_S_AUTH_CIPH}: state_chg to RAN_CONN_S_ACCEPTED
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){RAN_CONN_S_ACCEPTED}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){RAN_CONN_S_ACCEPTED}: state_chg to RAN_CONN_S_RELEASING
DREF IMSI-901700000004620:MSISDN-46071: MSC conn use + release == 1 (0x100: release)
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage increases to: 4
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage decreases to: 3
2018-12-25 23:40:18 +00:00
- BSSAP Clear --GERAN-A--> MS
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage decreases to: 2
2018-04-01 18:55:54 +00:00
<-- GSUP rx OSMO_GSUP_MSGT_UPDATE_LOCATION_RESULT: vlr_gsupc_read_cb() returns 0
- LU was successful, and the conn has already been closed
lu_result_sent == 1
bssap_clear_sent == 1
- BSS sends BSSMAP Clear Complete
2019-01-03 01:32:14 +00:00
DREF IMSI-901700000004620:MSISDN-46071: MSC conn use - release == 0 (0x0: )
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){RAN_CONN_S_RELEASING}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){RAN_CONN_S_RELEASING}: state_chg to RAN_CONN_S_RELEASED
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){RAN_CONN_S_RELEASED}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
DVLR vlr_lu_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){VLR_ULA_S_DONE}: Terminating (cause = OSMO_FSM_TERM_PARENT)
DVLR vlr_lu_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){VLR_ULA_S_DONE}: Removing from parent RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU)
DVLR vlr_lu_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){VLR_ULA_S_DONE}: fsm_lu_cleanup called with cause OSMO_FSM_TERM_PARENT
DVLR vlr_lu_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){VLR_ULA_S_DONE}: Freeing instance
DVLR vlr_lu_fsm(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){VLR_ULA_S_DONE}: Deallocated
DRLL IMSI-901700000004620:MSISDN-46071: Freeing RAN connection
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage decreases to: 1
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){RAN_CONN_S_RELEASED}: Freeing instance
DMM RAN_conn(IMSI-901700000004620:MSISDN-46071:GERAN-A-0:LU){RAN_CONN_S_RELEASED}: 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
llist_count(&net->ran_conns) == 0
fix paging: add timeout to discard unsuccessful paging
Currently, if there is no reply from the BSS / RNC, a subscriber will remain as
"already paged" forever, and is never going to be paged again. Even on IMSI
Detach, the pending request will keep a ref count on the vlr_subscr.
Add a paging timeout, as gsm_network->paging_timeout and in the VTY on the
'msc' node as 'paging timeout (default|<1-65535>'. (There is a 'network' /
'T3113' in OsmoBSC, but to not confuse the two, give this a different name.)
Add test_ms_timeout_paging() test to verify the timeout works.
I hit this while testing Paging across multiple hNodeB, when a UE lost
connection to the hNodeB. I noticed that no matter how long I wait, no Paging
is sent out anymore, and found this embarrassing issue. Good grief...
The choice of 10 seconds is taken from https://osmocom.org/issues/2756
Change-Id: I2db6f1e2ad341cf9c2cc7a21ec2fca0bae5b2db5
2017-12-15 02:02:27 +00:00
---
- an SMS is sent, MS is paged
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage increases to: 2
fix paging: add timeout to discard unsuccessful paging
Currently, if there is no reply from the BSS / RNC, a subscriber will remain as
"already paged" forever, and is never going to be paged again. Even on IMSI
Detach, the pending request will keep a ref count on the vlr_subscr.
Add a paging timeout, as gsm_network->paging_timeout and in the VTY on the
'msc' node as 'paging timeout (default|<1-65535>'. (There is a 'network' /
'T3113' in OsmoBSC, but to not confuse the two, give this a different name.)
Add test_ms_timeout_paging() test to verify the timeout works.
I hit this while testing Paging across multiple hNodeB, when a UE lost
connection to the hNodeB. I noticed that no matter how long I wait, no Paging
is sent out anymore, and found this embarrassing issue. Good grief...
The choice of 10 seconds is taken from https://osmocom.org/issues/2756
Change-Id: I2db6f1e2ad341cf9c2cc7a21ec2fca0bae5b2db5
2017-12-15 02:02:27 +00:00
llist_count(&vsub->cs.requests) == 0
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage increases to: 3
2019-01-16 11:47:39 +00:00
DLSMS Going to send a MT SMS
2019-01-03 01:32:14 +00:00
DCC (ti 00 sub IMSI-901700000004620:MSISDN-46071 callref 40000001) New transaction
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage increases to: 4
2019-01-16 11:47:39 +00:00
DLSMS SMC(0) instance created for network
DLSMS SMR(0) instance created for network.
DLSMS SMR(0) message SM-RL-DATA_REQ received in state IDLE
DLSMS SMR(0) TX SMS RP-DATA
DLSMS SMR(0) new RP state IDLE -> WAIT_FOR_RP_ACK
DLSMS SMC(0) message MNSMS-EST-REQ received in state IDLE
DLSMS SMC(0) new CP state IDLE -> MM_CONN_PENDING
DLSMS Initiating Paging procedure for IMSI-901700000004620:MSISDN-46071 due to MMSMS_EST_REQ
2019-01-03 01:32:14 +00:00
DMM Subscriber IMSI-901700000004620:MSISDN-46071 not paged yet, start paging.
2018-12-25 23:40:18 +00:00
GERAN-A sends out paging request to IMSI 901700000004620, TMSI 0xffffffff, LAC 23
fix paging: add timeout to discard unsuccessful paging
Currently, if there is no reply from the BSS / RNC, a subscriber will remain as
"already paged" forever, and is never going to be paged again. Even on IMSI
Detach, the pending request will keep a ref count on the vlr_subscr.
Add a paging timeout, as gsm_network->paging_timeout and in the VTY on the
'msc' node as 'paging timeout (default|<1-65535>'. (There is a 'network' /
'T3113' in OsmoBSC, but to not confuse the two, give this a different name.)
Add test_ms_timeout_paging() test to verify the timeout works.
I hit this while testing Paging across multiple hNodeB, when a UE lost
connection to the hNodeB. I noticed that no matter how long I wait, no Paging
is sent out anymore, and found this embarrassing issue. Good grief...
The choice of 10 seconds is taken from https://osmocom.org/issues/2756
Change-Id: I2db6f1e2ad341cf9c2cc7a21ec2fca0bae5b2db5
2017-12-15 02:02:27 +00:00
strcmp(paging_expecting_imsi, imsi) == 0
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage increases to: 5
fix paging: add timeout to discard unsuccessful paging
Currently, if there is no reply from the BSS / RNC, a subscriber will remain as
"already paged" forever, and is never going to be paged again. Even on IMSI
Detach, the pending request will keep a ref count on the vlr_subscr.
Add a paging timeout, as gsm_network->paging_timeout and in the VTY on the
'msc' node as 'paging timeout (default|<1-65535>'. (There is a 'network' /
'T3113' in OsmoBSC, but to not confuse the two, give this a different name.)
Add test_ms_timeout_paging() test to verify the timeout works.
I hit this while testing Paging across multiple hNodeB, when a UE lost
connection to the hNodeB. I noticed that no matter how long I wait, no Paging
is sent out anymore, and found this embarrassing issue. Good grief...
The choice of 10 seconds is taken from https://osmocom.org/issues/2756
Change-Id: I2db6f1e2ad341cf9c2cc7a21ec2fca0bae5b2db5
2017-12-15 02:02:27 +00:00
llist_count(&vsub->cs.requests) == 1
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage decreases to: 4
fix paging: add timeout to discard unsuccessful paging
Currently, if there is no reply from the BSS / RNC, a subscriber will remain as
"already paged" forever, and is never going to be paged again. Even on IMSI
Detach, the pending request will keep a ref count on the vlr_subscr.
Add a paging timeout, as gsm_network->paging_timeout and in the VTY on the
'msc' node as 'paging timeout (default|<1-65535>'. (There is a 'network' /
'T3113' in OsmoBSC, but to not confuse the two, give this a different name.)
Add test_ms_timeout_paging() test to verify the timeout works.
I hit this while testing Paging across multiple hNodeB, when a UE lost
connection to the hNodeB. I noticed that no matter how long I wait, no Paging
is sent out anymore, and found this embarrassing issue. Good grief...
The choice of 10 seconds is taken from https://osmocom.org/issues/2756
Change-Id: I2db6f1e2ad341cf9c2cc7a21ec2fca0bae5b2db5
2017-12-15 02:02:27 +00:00
paging_sent == 1
paging_stopped == 0
- time passes and no paging result is received
- Total time passed: 9.000000 s
- the paging timeout has not yet expired
paging_stopped == 0
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage increases to: 5
fix paging: add timeout to discard unsuccessful paging
Currently, if there is no reply from the BSS / RNC, a subscriber will remain as
"already paged" forever, and is never going to be paged again. Even on IMSI
Detach, the pending request will keep a ref count on the vlr_subscr.
Add a paging timeout, as gsm_network->paging_timeout and in the VTY on the
'msc' node as 'paging timeout (default|<1-65535>'. (There is a 'network' /
'T3113' in OsmoBSC, but to not confuse the two, give this a different name.)
Add test_ms_timeout_paging() test to verify the timeout works.
I hit this while testing Paging across multiple hNodeB, when a UE lost
connection to the hNodeB. I noticed that no matter how long I wait, no Paging
is sent out anymore, and found this embarrassing issue. Good grief...
The choice of 10 seconds is taken from https://osmocom.org/issues/2756
Change-Id: I2db6f1e2ad341cf9c2cc7a21ec2fca0bae5b2db5
2017-12-15 02:02:27 +00:00
vsub->cs.is_paging == 1
- another request is added to the list but does not cause another paging
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage increases to: 6
2019-01-16 11:47:39 +00:00
DLSMS Going to send a MT SMS
2019-01-03 01:32:14 +00:00
DCC (ti 01 sub IMSI-901700000004620:MSISDN-46071 callref 40000002) New transaction
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage increases to: 7
2019-01-16 11:47:39 +00:00
DLSMS SMC(0) instance created for network
DLSMS SMR(0) instance created for network.
DLSMS SMR(0) message SM-RL-DATA_REQ received in state IDLE
DLSMS SMR(0) TX SMS RP-DATA
DLSMS SMR(0) new RP state IDLE -> WAIT_FOR_RP_ACK
DLSMS SMC(0) message MNSMS-EST-REQ received in state IDLE
DLSMS SMC(0) new CP state IDLE -> MM_CONN_PENDING
DLSMS Initiating Paging procedure for IMSI-901700000004620:MSISDN-46071 due to MMSMS_EST_REQ
2019-01-03 01:32:14 +00:00
DMM Subscriber IMSI-901700000004620:MSISDN-46071 already paged.
fix paging: add timeout to discard unsuccessful paging
Currently, if there is no reply from the BSS / RNC, a subscriber will remain as
"already paged" forever, and is never going to be paged again. Even on IMSI
Detach, the pending request will keep a ref count on the vlr_subscr.
Add a paging timeout, as gsm_network->paging_timeout and in the VTY on the
'msc' node as 'paging timeout (default|<1-65535>'. (There is a 'network' /
'T3113' in OsmoBSC, but to not confuse the two, give this a different name.)
Add test_ms_timeout_paging() test to verify the timeout works.
I hit this while testing Paging across multiple hNodeB, when a UE lost
connection to the hNodeB. I noticed that no matter how long I wait, no Paging
is sent out anymore, and found this embarrassing issue. Good grief...
The choice of 10 seconds is taken from https://osmocom.org/issues/2756
Change-Id: I2db6f1e2ad341cf9c2cc7a21ec2fca0bae5b2db5
2017-12-15 02:02:27 +00:00
llist_count(&vsub->cs.requests) == 2
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage decreases to: 6
fix paging: add timeout to discard unsuccessful paging
Currently, if there is no reply from the BSS / RNC, a subscriber will remain as
"already paged" forever, and is never going to be paged again. Even on IMSI
Detach, the pending request will keep a ref count on the vlr_subscr.
Add a paging timeout, as gsm_network->paging_timeout and in the VTY on the
'msc' node as 'paging timeout (default|<1-65535>'. (There is a 'network' /
'T3113' in OsmoBSC, but to not confuse the two, give this a different name.)
Add test_ms_timeout_paging() test to verify the timeout works.
I hit this while testing Paging across multiple hNodeB, when a UE lost
connection to the hNodeB. I noticed that no matter how long I wait, no Paging
is sent out anymore, and found this embarrassing issue. Good grief...
The choice of 10 seconds is taken from https://osmocom.org/issues/2756
Change-Id: I2db6f1e2ad341cf9c2cc7a21ec2fca0bae5b2db5
2017-12-15 02:02:27 +00:00
paging_sent == 0
- the paging timeout expires, the paging as well as the requests are canceled
- Total time passed: 11.000000 s
2019-01-03 01:32:14 +00:00
DPAG Paging failure for IMSI-901700000004620:MSISDN-46071 (event=1)
fix paging: add timeout to discard unsuccessful paging
Currently, if there is no reply from the BSS / RNC, a subscriber will remain as
"already paged" forever, and is never going to be paged again. Even on IMSI
Detach, the pending request will keep a ref count on the vlr_subscr.
Add a paging timeout, as gsm_network->paging_timeout and in the VTY on the
'msc' node as 'paging timeout (default|<1-65535>'. (There is a 'network' /
'T3113' in OsmoBSC, but to not confuse the two, give this a different name.)
Add test_ms_timeout_paging() test to verify the timeout works.
I hit this while testing Paging across multiple hNodeB, when a UE lost
connection to the hNodeB. I noticed that no matter how long I wait, no Paging
is sent out anymore, and found this embarrassing issue. Good grief...
The choice of 10 seconds is taken from https://osmocom.org/issues/2756
Change-Id: I2db6f1e2ad341cf9c2cc7a21ec2fca0bae5b2db5
2017-12-15 02:02:27 +00:00
DPAG Calling paging cbfn.
2019-01-16 11:47:39 +00:00
DLSMS paging_cb_mmsms_est_req(hooknum=1, event=1)
DLSMS SMC(0) message MMSMS-REL-IND received in state MM_CONN_PENDING
DLSMS SMC(0) MM layer is released
DLSMS SMC(0) new CP state MM_CONN_PENDING -> IDLE
DLSMS MNSMS-ERROR-IND, no cause
DLSMS SMR(0) message MNSMS-ERROR-IND received in state WAIT_FOR_RP_ACK
DLSMS SMR(0) TX SMS MNSMS-ERROR-IND
DLSMS SMR(0) new RP state WAIT_FOR_RP_ACK -> IDLE
DLSMS SMR(0) TX: MNSMS-REL-REQ
DLSMS SMC(0) message MNSMS-REL-REQ received in state IDLE
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage decreases to: 5
2019-01-16 11:47:39 +00:00
DLSMS SMR(0) clearing SMR instance
DLSMS SMC(0) clearing instance
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage decreases to: 4
fix paging: add timeout to discard unsuccessful paging
Currently, if there is no reply from the BSS / RNC, a subscriber will remain as
"already paged" forever, and is never going to be paged again. Even on IMSI
Detach, the pending request will keep a ref count on the vlr_subscr.
Add a paging timeout, as gsm_network->paging_timeout and in the VTY on the
'msc' node as 'paging timeout (default|<1-65535>'. (There is a 'network' /
'T3113' in OsmoBSC, but to not confuse the two, give this a different name.)
Add test_ms_timeout_paging() test to verify the timeout works.
I hit this while testing Paging across multiple hNodeB, when a UE lost
connection to the hNodeB. I noticed that no matter how long I wait, no Paging
is sent out anymore, and found this embarrassing issue. Good grief...
The choice of 10 seconds is taken from https://osmocom.org/issues/2756
Change-Id: I2db6f1e2ad341cf9c2cc7a21ec2fca0bae5b2db5
2017-12-15 02:02:27 +00:00
DPAG Calling paging cbfn.
2019-01-16 11:47:39 +00:00
DLSMS paging_cb_mmsms_est_req(hooknum=1, event=1)
DLSMS SMC(0) message MMSMS-REL-IND received in state MM_CONN_PENDING
DLSMS SMC(0) MM layer is released
DLSMS SMC(0) new CP state MM_CONN_PENDING -> IDLE
DLSMS MNSMS-ERROR-IND, no cause
DLSMS SMR(0) message MNSMS-ERROR-IND received in state WAIT_FOR_RP_ACK
DLSMS SMR(0) TX SMS MNSMS-ERROR-IND
DLSMS SMR(0) new RP state WAIT_FOR_RP_ACK -> IDLE
DLSMS SMR(0) TX: MNSMS-REL-REQ
DLSMS SMC(0) message MNSMS-REL-REQ received in state IDLE
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage decreases to: 3
2019-01-16 11:47:39 +00:00
DLSMS SMR(0) clearing SMR instance
DLSMS SMC(0) clearing instance
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage decreases to: 2
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage decreases to: 1
fix paging: add timeout to discard unsuccessful paging
Currently, if there is no reply from the BSS / RNC, a subscriber will remain as
"already paged" forever, and is never going to be paged again. Even on IMSI
Detach, the pending request will keep a ref count on the vlr_subscr.
Add a paging timeout, as gsm_network->paging_timeout and in the VTY on the
'msc' node as 'paging timeout (default|<1-65535>'. (There is a 'network' /
'T3113' in OsmoBSC, but to not confuse the two, give this a different name.)
Add test_ms_timeout_paging() test to verify the timeout works.
I hit this while testing Paging across multiple hNodeB, when a UE lost
connection to the hNodeB. I noticed that no matter how long I wait, no Paging
is sent out anymore, and found this embarrassing issue. Good grief...
The choice of 10 seconds is taken from https://osmocom.org/issues/2756
Change-Id: I2db6f1e2ad341cf9c2cc7a21ec2fca0bae5b2db5
2017-12-15 02:02:27 +00:00
paging_stopped == 1
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage increases to: 2
fix paging: add timeout to discard unsuccessful paging
Currently, if there is no reply from the BSS / RNC, a subscriber will remain as
"already paged" forever, and is never going to be paged again. Even on IMSI
Detach, the pending request will keep a ref count on the vlr_subscr.
Add a paging timeout, as gsm_network->paging_timeout and in the VTY on the
'msc' node as 'paging timeout (default|<1-65535>'. (There is a 'network' /
'T3113' in OsmoBSC, but to not confuse the two, give this a different name.)
Add test_ms_timeout_paging() test to verify the timeout works.
I hit this while testing Paging across multiple hNodeB, when a UE lost
connection to the hNodeB. I noticed that no matter how long I wait, no Paging
is sent out anymore, and found this embarrassing issue. Good grief...
The choice of 10 seconds is taken from https://osmocom.org/issues/2756
Change-Id: I2db6f1e2ad341cf9c2cc7a21ec2fca0bae5b2db5
2017-12-15 02:02:27 +00:00
vsub->cs.is_paging == 0
llist_count(&vsub->cs.requests) == 0
---
2017-12-15 02:48:48 +00:00
- Now that the timeout has expired, another Paging is sent on request
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage increases to: 3
2019-01-16 11:47:39 +00:00
DLSMS Going to send a MT SMS
2019-01-03 01:32:14 +00:00
DCC (ti 00 sub IMSI-901700000004620:MSISDN-46071 callref 40000003) New transaction
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage increases to: 4
2019-01-16 11:47:39 +00:00
DLSMS SMC(0) instance created for network
DLSMS SMR(0) instance created for network.
DLSMS SMR(0) message SM-RL-DATA_REQ received in state IDLE
DLSMS SMR(0) TX SMS RP-DATA
DLSMS SMR(0) new RP state IDLE -> WAIT_FOR_RP_ACK
DLSMS SMC(0) message MNSMS-EST-REQ received in state IDLE
DLSMS SMC(0) new CP state IDLE -> MM_CONN_PENDING
DLSMS Initiating Paging procedure for IMSI-901700000004620:MSISDN-46071 due to MMSMS_EST_REQ
2019-01-03 01:32:14 +00:00
DMM Subscriber IMSI-901700000004620:MSISDN-46071 not paged yet, start paging.
2018-12-25 23:40:18 +00:00
GERAN-A sends out paging request to IMSI 901700000004620, TMSI 0xffffffff, LAC 23
2017-12-15 02:48:48 +00:00
strcmp(paging_expecting_imsi, imsi) == 0
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage increases to: 5
2017-12-15 02:48:48 +00:00
llist_count(&vsub->cs.requests) == 1
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage decreases to: 4
2017-12-15 02:48:48 +00:00
paging_sent == 1
paging_stopped == 0
---
- subscriber detaches, pagings are canceled
2018-12-25 23:40:18 +00:00
MSC <--GERAN-A-- MS: GSM48_MT_MM_IMSI_DETACH_IND
fix paging: add timeout to discard unsuccessful paging
Currently, if there is no reply from the BSS / RNC, a subscriber will remain as
"already paged" forever, and is never going to be paged again. Even on IMSI
Detach, the pending request will keep a ref count on the vlr_subscr.
Add a paging timeout, as gsm_network->paging_timeout and in the VTY on the
'msc' node as 'paging timeout (default|<1-65535>'. (There is a 'network' /
'T3113' in OsmoBSC, but to not confuse the two, give this a different name.)
Add test_ms_timeout_paging() test to verify the timeout works.
I hit this while testing Paging across multiple hNodeB, when a UE lost
connection to the hNodeB. I noticed that no matter how long I wait, no Paging
is sent out anymore, and found this embarrassing issue. Good grief...
The choice of 10 seconds is taken from https://osmocom.org/issues/2756
Change-Id: I2db6f1e2ad341cf9c2cc7a21ec2fca0bae5b2db5
2017-12-15 02:02:27 +00:00
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
2018-04-09 18:44:56 +00:00
DREF unknown: MSC conn use + compl_l3 == 1 (0x1: compl_l3)
fix paging: add timeout to discard unsuccessful paging
Currently, if there is no reply from the BSS / RNC, a subscriber will remain as
"already paged" forever, and is never going to be paged again. Even on IMSI
Detach, the pending request will keep a ref count on the vlr_subscr.
Add a paging timeout, as gsm_network->paging_timeout and in the VTY on the
'msc' node as 'paging timeout (default|<1-65535>'. (There is a 'network' /
'T3113' in OsmoBSC, but to not confuse the two, give this a different name.)
Add test_ms_timeout_paging() test to verify the timeout works.
I hit this while testing Paging across multiple hNodeB, when a UE lost
connection to the hNodeB. I noticed that no matter how long I wait, no Paging
is sent out anymore, and found this embarrassing issue. Good grief...
The choice of 10 seconds is taken from https://osmocom.org/issues/2756
Change-Id: I2db6f1e2ad341cf9c2cc7a21ec2fca0bae5b2db5
2017-12-15 02:02:27 +00:00
DRLL Dispatching 04.08 message GSM48_MT_MM_IMSI_DETACH_IND (0x5:0x1)
DMM IMSI DETACH INDICATION: MI(IMSI)=901700000004620
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage increases to: 5
DMM IMSI DETACH for IMSI-901700000004620:MSISDN-46071
DPAG Paging failure for IMSI-901700000004620:MSISDN-46071 (event=1)
2017-12-15 02:48:48 +00:00
DPAG Calling paging cbfn.
2019-01-16 11:47:39 +00:00
DLSMS paging_cb_mmsms_est_req(hooknum=1, event=1)
DLSMS SMC(0) message MMSMS-REL-IND received in state MM_CONN_PENDING
DLSMS SMC(0) MM layer is released
DLSMS SMC(0) new CP state MM_CONN_PENDING -> IDLE
DLSMS MNSMS-ERROR-IND, no cause
DLSMS SMR(0) message MNSMS-ERROR-IND received in state WAIT_FOR_RP_ACK
DLSMS SMR(0) TX SMS MNSMS-ERROR-IND
DLSMS SMR(0) new RP state WAIT_FOR_RP_ACK -> IDLE
DLSMS SMR(0) TX: MNSMS-REL-REQ
DLSMS SMC(0) message MNSMS-REL-REQ received in state IDLE
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage decreases to: 4
2019-01-16 11:47:39 +00:00
DLSMS SMR(0) clearing SMR instance
DLSMS SMC(0) clearing instance
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage decreases to: 3
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage decreases to: 2
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage increases to: 3
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage decreases to: 2
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage decreases to: 1
DREF VLR subscr IMSI-901700000004620:MSISDN-46071 usage decreases to: 0
DREF freeing VLR subscr IMSI-901700000004620:MSISDN-46071
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}: Received Event RAN_CONN_E_CN_CLOSE
DMM RAN_conn{RAN_CONN_S_NEW}: state_chg to RAN_CONN_S_RELEASING
2018-04-09 18:44:56 +00:00
DREF unknown: MSC conn use + release == 2 (0x101: compl_l3,release)
2018-12-25 23:40:18 +00:00
- BSSAP Clear --GERAN-A--> MS
2018-04-09 18:44:56 +00:00
DREF unknown: MSC conn use - compl_l3 == 1 (0x100: release)
2018-04-01 18:55:54 +00:00
bssap_clear_sent == 1
paging_stopped == 1
- BSS sends BSSMAP Clear Complete
2018-04-09 18:44:56 +00:00
DREF unknown: 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{RAN_CONN_S_RELEASING}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn{RAN_CONN_S_RELEASING}: state_chg to RAN_CONN_S_RELEASED
DMM RAN_conn{RAN_CONN_S_RELEASED}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
DRLL Freeing RAN connection with NULL subscriber
DMM RAN_conn{RAN_CONN_S_RELEASED}: Freeing instance
DMM RAN_conn{RAN_CONN_S_RELEASED}: Deallocated
llist_count(&net->ran_conns) == 0
2018-03-01 23:40:58 +00:00
===== test_ms_timeout_paging: SUCCESS
fix paging: add timeout to discard unsuccessful paging
Currently, if there is no reply from the BSS / RNC, a subscriber will remain as
"already paged" forever, and is never going to be paged again. Even on IMSI
Detach, the pending request will keep a ref count on the vlr_subscr.
Add a paging timeout, as gsm_network->paging_timeout and in the VTY on the
'msc' node as 'paging timeout (default|<1-65535>'. (There is a 'network' /
'T3113' in OsmoBSC, but to not confuse the two, give this a different name.)
Add test_ms_timeout_paging() test to verify the timeout works.
I hit this while testing Paging across multiple hNodeB, when a UE lost
connection to the hNodeB. I noticed that no matter how long I wait, no Paging
is sent out anymore, and found this embarrassing issue. Good grief...
The choice of 10 seconds is taken from https://osmocom.org/issues/2756
Change-Id: I2db6f1e2ad341cf9c2cc7a21ec2fca0bae5b2db5
2017-12-15 02:02:27 +00:00
full talloc report on 'msgb' (total 0 bytes in 1 blocks)
2018-12-06 11:06:59 +00:00
talloc_total_blocks(tall_bsc_ctx) == 13
fix paging: add timeout to discard unsuccessful paging
Currently, if there is no reply from the BSS / RNC, a subscriber will remain as
"already paged" forever, and is never going to be paged again. Even on IMSI
Detach, the pending request will keep a ref count on the vlr_subscr.
Add a paging timeout, as gsm_network->paging_timeout and in the VTY on the
'msc' node as 'paging timeout (default|<1-65535>'. (There is a 'network' /
'T3113' in OsmoBSC, but to not confuse the two, give this a different name.)
Add test_ms_timeout_paging() test to verify the timeout works.
I hit this while testing Paging across multiple hNodeB, when a UE lost
connection to the hNodeB. I noticed that no matter how long I wait, no Paging
is sent out anymore, and found this embarrassing issue. Good grief...
The choice of 10 seconds is taken from https://osmocom.org/issues/2756
Change-Id: I2db6f1e2ad341cf9c2cc7a21ec2fca0bae5b2db5
2017-12-15 02:02:27 +00:00
A5/n Ciph: request Classmark Update if missing
When the VLR requests a Ciphering Mode with vlr_ops.set_ciph_mode(), and if we
need a ciph algo flag from a Classmark information that is not yet known
(usually CM 2 during LU), send a BSSMAP Classmark Request to get it.
To manage the intermission of the Classmark Request, add
- msc_classmark_request_then_cipher_mode_cmd(),
- state SUBSCR_CONN_S_WAIT_CLASSMARK_UPDATE,
- event SUBSCR_CONN_E_CLASSMARK_UPDATE.
From state AUTH_CIPH, switch to state WAIT_CLASSMARK_UPDATE. Once the BSSMAP
Classmark Response, is received, switch back to SUBSCR_CONN_S_AUTH_CIPH and
re-initiate Ciphering Mode.
To be able to re-enter the Ciphering Mode algo decision, factor it out into
msc_geran_set_cipher_mode().
Rationale:
In the following commit, essentially we stopped supporting A5/3 ciphering:
commit 71330720b6efdda2fcfd3e9c0cb45f89e32e5670
"MSC: Intersect configured A5 algorithms with MS-supported ones"
Change-Id: Id124923ee52a357cb7d3e04d33f585214774f3a3
A5/3 was no longer supported because from that commit on, we strictly checked
the MS-supported ciphers, but we did not have Classmark 2 available during
Location Updating.
This patch changes that: when Classmark 2 is missing, actively request it by a
BSSMAP Classmark Request; continue Ciphering only after the Response. Always
request missing Classmark, even if a lesser cipher were configured available.
If the Classmark Update response fails to come in, cause an attach failure.
Instead, we could attempt to use a lesser cipher that is also enabled. That is
left as a future feature, should that become relevant. I think it's unlikely.
Technically, we could now end up requesting a Classmark Updating both during LU
(vlr_lu_fsm) and CM Service/Paging Response (proc_arq_fsm), but in practice the
only time we lack a Classmark is: during Location Updating with A5/3 enabled.
A5/1 support is indicated in CM1 which is always available, and A5/3 support is
indicated in CM2, which is always available during CM Service Request as well
as Paging Response. So this patch has practical relevance only for Location
Updating. For networks that permit only A5/3, this patch fixes Location
Updating. For networks that support A5/3 and A5/1, so far we always used A5/1
during LU, and after this patch we request CM2 and likely use A5/3 instead.
In msc_vlr_test_gsm_ciph, verify that requesting Classmark 2 for A5/3 works
during LU. Also verify that the lack of a Classmark Response results in attach
failure.
In msc_vlr_test_gsm_ciph, a hacky unit test fakes a situation where a CM2 is
missing during proc_arq_fsm and proves that that code path works, even though
the practical relevance is currently zero. It would only become interesting if
ciphering algorithms A5/4 and higher became relevant, because support of those
would be indicated in Classmark 3, which would always require a Classmark
Request.
Related: OS#3043
Depends: I4a2e1d3923e33912579c4180aa1ff8e8f5abb7e7 (libosmocore)
Change-Id: I73c7cb6a86624695bd9c0f59abb72e2fdc655131
2018-09-13 01:23:07 +00:00
===== test_classmark_update_timeout
- Total time passed: 0.000000 s
- Location Update request causes a GSUP Send Auth Info request to HLR
2018-12-25 23:40:18 +00:00
MSC <--GERAN-A-- MS: GSM48_MT_MM_LOC_UPD_REQUEST
A5/n Ciph: request Classmark Update if missing
When the VLR requests a Ciphering Mode with vlr_ops.set_ciph_mode(), and if we
need a ciph algo flag from a Classmark information that is not yet known
(usually CM 2 during LU), send a BSSMAP Classmark Request to get it.
To manage the intermission of the Classmark Request, add
- msc_classmark_request_then_cipher_mode_cmd(),
- state SUBSCR_CONN_S_WAIT_CLASSMARK_UPDATE,
- event SUBSCR_CONN_E_CLASSMARK_UPDATE.
From state AUTH_CIPH, switch to state WAIT_CLASSMARK_UPDATE. Once the BSSMAP
Classmark Response, is received, switch back to SUBSCR_CONN_S_AUTH_CIPH and
re-initiate Ciphering Mode.
To be able to re-enter the Ciphering Mode algo decision, factor it out into
msc_geran_set_cipher_mode().
Rationale:
In the following commit, essentially we stopped supporting A5/3 ciphering:
commit 71330720b6efdda2fcfd3e9c0cb45f89e32e5670
"MSC: Intersect configured A5 algorithms with MS-supported ones"
Change-Id: Id124923ee52a357cb7d3e04d33f585214774f3a3
A5/3 was no longer supported because from that commit on, we strictly checked
the MS-supported ciphers, but we did not have Classmark 2 available during
Location Updating.
This patch changes that: when Classmark 2 is missing, actively request it by a
BSSMAP Classmark Request; continue Ciphering only after the Response. Always
request missing Classmark, even if a lesser cipher were configured available.
If the Classmark Update response fails to come in, cause an attach failure.
Instead, we could attempt to use a lesser cipher that is also enabled. That is
left as a future feature, should that become relevant. I think it's unlikely.
Technically, we could now end up requesting a Classmark Updating both during LU
(vlr_lu_fsm) and CM Service/Paging Response (proc_arq_fsm), but in practice the
only time we lack a Classmark is: during Location Updating with A5/3 enabled.
A5/1 support is indicated in CM1 which is always available, and A5/3 support is
indicated in CM2, which is always available during CM Service Request as well
as Paging Response. So this patch has practical relevance only for Location
Updating. For networks that permit only A5/3, this patch fixes Location
Updating. For networks that support A5/3 and A5/1, so far we always used A5/1
during LU, and after this patch we request CM2 and likely use A5/3 instead.
In msc_vlr_test_gsm_ciph, verify that requesting Classmark 2 for A5/3 works
during LU. Also verify that the lack of a Classmark Response results in attach
failure.
In msc_vlr_test_gsm_ciph, a hacky unit test fakes a situation where a CM2 is
missing during proc_arq_fsm and proves that that code path works, even though
the practical relevance is currently zero. It would only become interesting if
ciphering algorithms A5/4 and higher became relevant, because support of those
would be indicated in Classmark 3, which would always require a Classmark
Request.
Related: OS#3043
Depends: I4a2e1d3923e33912579c4180aa1ff8e8f5abb7e7 (libosmocore)
Change-Id: I73c7cb6a86624695bd9c0f59abb72e2fdc655131
2018-09-13 01:23:07 +00:00
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
A5/n Ciph: request Classmark Update if missing
When the VLR requests a Ciphering Mode with vlr_ops.set_ciph_mode(), and if we
need a ciph algo flag from a Classmark information that is not yet known
(usually CM 2 during LU), send a BSSMAP Classmark Request to get it.
To manage the intermission of the Classmark Request, add
- msc_classmark_request_then_cipher_mode_cmd(),
- state SUBSCR_CONN_S_WAIT_CLASSMARK_UPDATE,
- event SUBSCR_CONN_E_CLASSMARK_UPDATE.
From state AUTH_CIPH, switch to state WAIT_CLASSMARK_UPDATE. Once the BSSMAP
Classmark Response, is received, switch back to SUBSCR_CONN_S_AUTH_CIPH and
re-initiate Ciphering Mode.
To be able to re-enter the Ciphering Mode algo decision, factor it out into
msc_geran_set_cipher_mode().
Rationale:
In the following commit, essentially we stopped supporting A5/3 ciphering:
commit 71330720b6efdda2fcfd3e9c0cb45f89e32e5670
"MSC: Intersect configured A5 algorithms with MS-supported ones"
Change-Id: Id124923ee52a357cb7d3e04d33f585214774f3a3
A5/3 was no longer supported because from that commit on, we strictly checked
the MS-supported ciphers, but we did not have Classmark 2 available during
Location Updating.
This patch changes that: when Classmark 2 is missing, actively request it by a
BSSMAP Classmark Request; continue Ciphering only after the Response. Always
request missing Classmark, even if a lesser cipher were configured available.
If the Classmark Update response fails to come in, cause an attach failure.
Instead, we could attempt to use a lesser cipher that is also enabled. That is
left as a future feature, should that become relevant. I think it's unlikely.
Technically, we could now end up requesting a Classmark Updating both during LU
(vlr_lu_fsm) and CM Service/Paging Response (proc_arq_fsm), but in practice the
only time we lack a Classmark is: during Location Updating with A5/3 enabled.
A5/1 support is indicated in CM1 which is always available, and A5/3 support is
indicated in CM2, which is always available during CM Service Request as well
as Paging Response. So this patch has practical relevance only for Location
Updating. For networks that permit only A5/3, this patch fixes Location
Updating. For networks that support A5/3 and A5/1, so far we always used A5/1
during LU, and after this patch we request CM2 and likely use A5/3 instead.
In msc_vlr_test_gsm_ciph, verify that requesting Classmark 2 for A5/3 works
during LU. Also verify that the lack of a Classmark Response results in attach
failure.
In msc_vlr_test_gsm_ciph, a hacky unit test fakes a situation where a CM2 is
missing during proc_arq_fsm and proves that that code path works, even though
the practical relevance is currently zero. It would only become interesting if
ciphering algorithms A5/4 and higher became relevant, because support of those
would be indicated in Classmark 3, which would always require a Classmark
Request.
Related: OS#3043
Depends: I4a2e1d3923e33912579c4180aa1ff8e8f5abb7e7 (libosmocore)
Change-Id: I73c7cb6a86624695bd9c0f59abb72e2fdc655131
2018-09-13 01:23:07 +00:00
DREF unknown: MSC conn use + compl_l3 == 1 (0x1: compl_l3)
DRLL Dispatching 04.08 message GSM48_MT_MM_LOC_UPD_REQUEST (0x5:0x8)
2019-01-03 01:32:14 +00:00
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_NEW}: Updated ID
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_NEW}: LOCATION UPDATING REQUEST: MI=IMSI-901700000004620 LU-type=IMSI-ATTACH
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_NEW}: LU/new-LAC: 1/23
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_IDLE}: Allocated
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_IDLE}: is child of RAN_conn(IMSI-901700000004620:GERAN-A-0:LU)
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_IDLE}: rev=GSM net=GERAN Auth+Ciph
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_IDLE}: Received Event VLR_ULA_E_UPDATE_LA
A5/n Ciph: request Classmark Update if missing
When the VLR requests a Ciphering Mode with vlr_ops.set_ciph_mode(), and if we
need a ciph algo flag from a Classmark information that is not yet known
(usually CM 2 during LU), send a BSSMAP Classmark Request to get it.
To manage the intermission of the Classmark Request, add
- msc_classmark_request_then_cipher_mode_cmd(),
- state SUBSCR_CONN_S_WAIT_CLASSMARK_UPDATE,
- event SUBSCR_CONN_E_CLASSMARK_UPDATE.
From state AUTH_CIPH, switch to state WAIT_CLASSMARK_UPDATE. Once the BSSMAP
Classmark Response, is received, switch back to SUBSCR_CONN_S_AUTH_CIPH and
re-initiate Ciphering Mode.
To be able to re-enter the Ciphering Mode algo decision, factor it out into
msc_geran_set_cipher_mode().
Rationale:
In the following commit, essentially we stopped supporting A5/3 ciphering:
commit 71330720b6efdda2fcfd3e9c0cb45f89e32e5670
"MSC: Intersect configured A5 algorithms with MS-supported ones"
Change-Id: Id124923ee52a357cb7d3e04d33f585214774f3a3
A5/3 was no longer supported because from that commit on, we strictly checked
the MS-supported ciphers, but we did not have Classmark 2 available during
Location Updating.
This patch changes that: when Classmark 2 is missing, actively request it by a
BSSMAP Classmark Request; continue Ciphering only after the Response. Always
request missing Classmark, even if a lesser cipher were configured available.
If the Classmark Update response fails to come in, cause an attach failure.
Instead, we could attempt to use a lesser cipher that is also enabled. That is
left as a future feature, should that become relevant. I think it's unlikely.
Technically, we could now end up requesting a Classmark Updating both during LU
(vlr_lu_fsm) and CM Service/Paging Response (proc_arq_fsm), but in practice the
only time we lack a Classmark is: during Location Updating with A5/3 enabled.
A5/1 support is indicated in CM1 which is always available, and A5/3 support is
indicated in CM2, which is always available during CM Service Request as well
as Paging Response. So this patch has practical relevance only for Location
Updating. For networks that permit only A5/3, this patch fixes Location
Updating. For networks that support A5/3 and A5/1, so far we always used A5/1
during LU, and after this patch we request CM2 and likely use A5/3 instead.
In msc_vlr_test_gsm_ciph, verify that requesting Classmark 2 for A5/3 works
during LU. Also verify that the lack of a Classmark Response results in attach
failure.
In msc_vlr_test_gsm_ciph, a hacky unit test fakes a situation where a CM2 is
missing during proc_arq_fsm and proves that that code path works, even though
the practical relevance is currently zero. It would only become interesting if
ciphering algorithms A5/4 and higher became relevant, because support of those
would be indicated in Classmark 3, which would always require a Classmark
Request.
Related: OS#3043
Depends: I4a2e1d3923e33912579c4180aa1ff8e8f5abb7e7 (libosmocore)
Change-Id: I73c7cb6a86624695bd9c0f59abb72e2fdc655131
2018-09-13 01:23:07 +00:00
DREF VLR subscr unknown usage increases to: 1
DVLR set IMSI on subscriber; IMSI=901700000004620 id=901700000004620
DVLR New subscr, IMSI: 901700000004620
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620 usage increases to: 2
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_NEW}: Updated ID
DREF VLR subscr IMSI-901700000004620 usage decreases to: 1
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_IDLE}: vlr_loc_upd_node1()
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_IDLE}: state_chg to VLR_ULA_S_WAIT_AUTH
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_NEEDS_AUTH}: Allocated
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_NEEDS_AUTH}: is child of vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU)
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_NEEDS_AUTH}: Received Event VLR_AUTH_E_START
A5/n Ciph: request Classmark Update if missing
When the VLR requests a Ciphering Mode with vlr_ops.set_ciph_mode(), and if we
need a ciph algo flag from a Classmark information that is not yet known
(usually CM 2 during LU), send a BSSMAP Classmark Request to get it.
To manage the intermission of the Classmark Request, add
- msc_classmark_request_then_cipher_mode_cmd(),
- state SUBSCR_CONN_S_WAIT_CLASSMARK_UPDATE,
- event SUBSCR_CONN_E_CLASSMARK_UPDATE.
From state AUTH_CIPH, switch to state WAIT_CLASSMARK_UPDATE. Once the BSSMAP
Classmark Response, is received, switch back to SUBSCR_CONN_S_AUTH_CIPH and
re-initiate Ciphering Mode.
To be able to re-enter the Ciphering Mode algo decision, factor it out into
msc_geran_set_cipher_mode().
Rationale:
In the following commit, essentially we stopped supporting A5/3 ciphering:
commit 71330720b6efdda2fcfd3e9c0cb45f89e32e5670
"MSC: Intersect configured A5 algorithms with MS-supported ones"
Change-Id: Id124923ee52a357cb7d3e04d33f585214774f3a3
A5/3 was no longer supported because from that commit on, we strictly checked
the MS-supported ciphers, but we did not have Classmark 2 available during
Location Updating.
This patch changes that: when Classmark 2 is missing, actively request it by a
BSSMAP Classmark Request; continue Ciphering only after the Response. Always
request missing Classmark, even if a lesser cipher were configured available.
If the Classmark Update response fails to come in, cause an attach failure.
Instead, we could attempt to use a lesser cipher that is also enabled. That is
left as a future feature, should that become relevant. I think it's unlikely.
Technically, we could now end up requesting a Classmark Updating both during LU
(vlr_lu_fsm) and CM Service/Paging Response (proc_arq_fsm), but in practice the
only time we lack a Classmark is: during Location Updating with A5/3 enabled.
A5/1 support is indicated in CM1 which is always available, and A5/3 support is
indicated in CM2, which is always available during CM Service Request as well
as Paging Response. So this patch has practical relevance only for Location
Updating. For networks that permit only A5/3, this patch fixes Location
Updating. For networks that support A5/3 and A5/1, so far we always used A5/1
during LU, and after this patch we request CM2 and likely use A5/3 instead.
In msc_vlr_test_gsm_ciph, verify that requesting Classmark 2 for A5/3 works
during LU. Also verify that the lack of a Classmark Response results in attach
failure.
In msc_vlr_test_gsm_ciph, a hacky unit test fakes a situation where a CM2 is
missing during proc_arq_fsm and proves that that code path works, even though
the practical relevance is currently zero. It would only become interesting if
ciphering algorithms A5/4 and higher became relevant, because support of those
would be indicated in Classmark 3, which would always require a Classmark
Request.
Related: OS#3043
Depends: I4a2e1d3923e33912579c4180aa1ff8e8f5abb7e7 (libosmocore)
Change-Id: I73c7cb6a86624695bd9c0f59abb72e2fdc655131
2018-09-13 01:23:07 +00:00
DVLR GSUP tx: 08010809710000004026f0
GSUP --> HLR: OSMO_GSUP_MSGT_SEND_AUTH_INFO_REQUEST: 08010809710000004026f0
2019-01-03 01:32:14 +00:00
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_NEEDS_AUTH}: state_chg to VLR_SUB_AS_NEEDS_AUTH_WAIT_AI
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_NEW}: Received Event RAN_CONN_E_COMPLETE_LAYER_3
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_NEW}: state_chg to RAN_CONN_S_AUTH_CIPH
DREF IMSI-901700000004620: MSC conn use - compl_l3 == 0 (0x0: )
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_AUTH_CIPH}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_AUTH_CIPH}: Awaiting results for Auth+Ciph, overruling event RAN_CONN_E_UNUSED
A5/n Ciph: request Classmark Update if missing
When the VLR requests a Ciphering Mode with vlr_ops.set_ciph_mode(), and if we
need a ciph algo flag from a Classmark information that is not yet known
(usually CM 2 during LU), send a BSSMAP Classmark Request to get it.
To manage the intermission of the Classmark Request, add
- msc_classmark_request_then_cipher_mode_cmd(),
- state SUBSCR_CONN_S_WAIT_CLASSMARK_UPDATE,
- event SUBSCR_CONN_E_CLASSMARK_UPDATE.
From state AUTH_CIPH, switch to state WAIT_CLASSMARK_UPDATE. Once the BSSMAP
Classmark Response, is received, switch back to SUBSCR_CONN_S_AUTH_CIPH and
re-initiate Ciphering Mode.
To be able to re-enter the Ciphering Mode algo decision, factor it out into
msc_geran_set_cipher_mode().
Rationale:
In the following commit, essentially we stopped supporting A5/3 ciphering:
commit 71330720b6efdda2fcfd3e9c0cb45f89e32e5670
"MSC: Intersect configured A5 algorithms with MS-supported ones"
Change-Id: Id124923ee52a357cb7d3e04d33f585214774f3a3
A5/3 was no longer supported because from that commit on, we strictly checked
the MS-supported ciphers, but we did not have Classmark 2 available during
Location Updating.
This patch changes that: when Classmark 2 is missing, actively request it by a
BSSMAP Classmark Request; continue Ciphering only after the Response. Always
request missing Classmark, even if a lesser cipher were configured available.
If the Classmark Update response fails to come in, cause an attach failure.
Instead, we could attempt to use a lesser cipher that is also enabled. That is
left as a future feature, should that become relevant. I think it's unlikely.
Technically, we could now end up requesting a Classmark Updating both during LU
(vlr_lu_fsm) and CM Service/Paging Response (proc_arq_fsm), but in practice the
only time we lack a Classmark is: during Location Updating with A5/3 enabled.
A5/1 support is indicated in CM1 which is always available, and A5/3 support is
indicated in CM2, which is always available during CM Service Request as well
as Paging Response. So this patch has practical relevance only for Location
Updating. For networks that permit only A5/3, this patch fixes Location
Updating. For networks that support A5/3 and A5/1, so far we always used A5/1
during LU, and after this patch we request CM2 and likely use A5/3 instead.
In msc_vlr_test_gsm_ciph, verify that requesting Classmark 2 for A5/3 works
during LU. Also verify that the lack of a Classmark Response results in attach
failure.
In msc_vlr_test_gsm_ciph, a hacky unit test fakes a situation where a CM2 is
missing during proc_arq_fsm and proves that that code path works, even though
the practical relevance is currently zero. It would only become interesting if
ciphering algorithms A5/4 and higher became relevant, because support of those
would be indicated in Classmark 3, which would always require a Classmark
Request.
Related: OS#3043
Depends: I4a2e1d3923e33912579c4180aa1ff8e8f5abb7e7 (libosmocore)
Change-Id: I73c7cb6a86624695bd9c0f59abb72e2fdc655131
2018-09-13 01:23:07 +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: 0a010809710000004026f003222010585df1ae287f6e273dce07090d61320b21042d8b2c3e220861855fb81fc2a800
DVLR GSUP rx 47: 0a010809710000004026f003222010585df1ae287f6e273dce07090d61320b21042d8b2c3e220861855fb81fc2a800
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620 usage increases to: 2
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_NEEDS_AUTH_WAIT_AI}: Received Event VLR_AUTH_E_HLR_SAI_ACK
DVLR SUBSCR(IMSI-901700000004620) Received 1 auth tuples
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_NEEDS_AUTH_WAIT_AI}: state_chg to VLR_SUB_AS_WAIT_RESP
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_WAIT_RESP}: got auth tuple: use_count=1 key_seq=0 -- will use GSM AKA (is_r99=no, at->vec.auth_types=0x1)
- sending GSM Auth Request for IMSI-901700000004620: tuple use_count=1 key_seq=0 auth_types=0x1 and...
A5/n Ciph: request Classmark Update if missing
When the VLR requests a Ciphering Mode with vlr_ops.set_ciph_mode(), and if we
need a ciph algo flag from a Classmark information that is not yet known
(usually CM 2 during LU), send a BSSMAP Classmark Request to get it.
To manage the intermission of the Classmark Request, add
- msc_classmark_request_then_cipher_mode_cmd(),
- state SUBSCR_CONN_S_WAIT_CLASSMARK_UPDATE,
- event SUBSCR_CONN_E_CLASSMARK_UPDATE.
From state AUTH_CIPH, switch to state WAIT_CLASSMARK_UPDATE. Once the BSSMAP
Classmark Response, is received, switch back to SUBSCR_CONN_S_AUTH_CIPH and
re-initiate Ciphering Mode.
To be able to re-enter the Ciphering Mode algo decision, factor it out into
msc_geran_set_cipher_mode().
Rationale:
In the following commit, essentially we stopped supporting A5/3 ciphering:
commit 71330720b6efdda2fcfd3e9c0cb45f89e32e5670
"MSC: Intersect configured A5 algorithms with MS-supported ones"
Change-Id: Id124923ee52a357cb7d3e04d33f585214774f3a3
A5/3 was no longer supported because from that commit on, we strictly checked
the MS-supported ciphers, but we did not have Classmark 2 available during
Location Updating.
This patch changes that: when Classmark 2 is missing, actively request it by a
BSSMAP Classmark Request; continue Ciphering only after the Response. Always
request missing Classmark, even if a lesser cipher were configured available.
If the Classmark Update response fails to come in, cause an attach failure.
Instead, we could attempt to use a lesser cipher that is also enabled. That is
left as a future feature, should that become relevant. I think it's unlikely.
Technically, we could now end up requesting a Classmark Updating both during LU
(vlr_lu_fsm) and CM Service/Paging Response (proc_arq_fsm), but in practice the
only time we lack a Classmark is: during Location Updating with A5/3 enabled.
A5/1 support is indicated in CM1 which is always available, and A5/3 support is
indicated in CM2, which is always available during CM Service Request as well
as Paging Response. So this patch has practical relevance only for Location
Updating. For networks that permit only A5/3, this patch fixes Location
Updating. For networks that support A5/3 and A5/1, so far we always used A5/1
during LU, and after this patch we request CM2 and likely use A5/3 instead.
In msc_vlr_test_gsm_ciph, verify that requesting Classmark 2 for A5/3 works
during LU. Also verify that the lack of a Classmark Response results in attach
failure.
In msc_vlr_test_gsm_ciph, a hacky unit test fakes a situation where a CM2 is
missing during proc_arq_fsm and proves that that code path works, even though
the practical relevance is currently zero. It would only become interesting if
ciphering algorithms A5/4 and higher became relevant, because support of those
would be indicated in Classmark 3, which would always require a Classmark
Request.
Related: OS#3043
Depends: I4a2e1d3923e33912579c4180aa1ff8e8f5abb7e7 (libosmocore)
Change-Id: I73c7cb6a86624695bd9c0f59abb72e2fdc655131
2018-09-13 01:23:07 +00:00
- ...rand=585df1ae287f6e273dce07090d61320b
- ...expecting sres=2d8b2c3e
2019-01-03 01:32:14 +00:00
DREF VLR subscr IMSI-901700000004620 usage decreases to: 1
A5/n Ciph: request Classmark Update if missing
When the VLR requests a Ciphering Mode with vlr_ops.set_ciph_mode(), and if we
need a ciph algo flag from a Classmark information that is not yet known
(usually CM 2 during LU), send a BSSMAP Classmark Request to get it.
To manage the intermission of the Classmark Request, add
- msc_classmark_request_then_cipher_mode_cmd(),
- state SUBSCR_CONN_S_WAIT_CLASSMARK_UPDATE,
- event SUBSCR_CONN_E_CLASSMARK_UPDATE.
From state AUTH_CIPH, switch to state WAIT_CLASSMARK_UPDATE. Once the BSSMAP
Classmark Response, is received, switch back to SUBSCR_CONN_S_AUTH_CIPH and
re-initiate Ciphering Mode.
To be able to re-enter the Ciphering Mode algo decision, factor it out into
msc_geran_set_cipher_mode().
Rationale:
In the following commit, essentially we stopped supporting A5/3 ciphering:
commit 71330720b6efdda2fcfd3e9c0cb45f89e32e5670
"MSC: Intersect configured A5 algorithms with MS-supported ones"
Change-Id: Id124923ee52a357cb7d3e04d33f585214774f3a3
A5/3 was no longer supported because from that commit on, we strictly checked
the MS-supported ciphers, but we did not have Classmark 2 available during
Location Updating.
This patch changes that: when Classmark 2 is missing, actively request it by a
BSSMAP Classmark Request; continue Ciphering only after the Response. Always
request missing Classmark, even if a lesser cipher were configured available.
If the Classmark Update response fails to come in, cause an attach failure.
Instead, we could attempt to use a lesser cipher that is also enabled. That is
left as a future feature, should that become relevant. I think it's unlikely.
Technically, we could now end up requesting a Classmark Updating both during LU
(vlr_lu_fsm) and CM Service/Paging Response (proc_arq_fsm), but in practice the
only time we lack a Classmark is: during Location Updating with A5/3 enabled.
A5/1 support is indicated in CM1 which is always available, and A5/3 support is
indicated in CM2, which is always available during CM Service Request as well
as Paging Response. So this patch has practical relevance only for Location
Updating. For networks that permit only A5/3, this patch fixes Location
Updating. For networks that support A5/3 and A5/1, so far we always used A5/1
during LU, and after this patch we request CM2 and likely use A5/3 instead.
In msc_vlr_test_gsm_ciph, verify that requesting Classmark 2 for A5/3 works
during LU. Also verify that the lack of a Classmark Response results in attach
failure.
In msc_vlr_test_gsm_ciph, a hacky unit test fakes a situation where a CM2 is
missing during proc_arq_fsm and proves that that code path works, even though
the practical relevance is currently zero. It would only become interesting if
ciphering algorithms A5/4 and higher became relevant, because support of those
would be indicated in Classmark 3, which would always require a Classmark
Request.
Related: OS#3043
Depends: I4a2e1d3923e33912579c4180aa1ff8e8f5abb7e7 (libosmocore)
Change-Id: I73c7cb6a86624695bd9c0f59abb72e2fdc655131
2018-09-13 01:23:07 +00:00
<-- GSUP rx OSMO_GSUP_MSGT_SEND_AUTH_INFO_RESULT: vlr_gsupc_read_cb() returns 0
lu_result_sent == 0
auth_request_sent == 1
---
- MS sends Authen Response, VLR accepts and wants to send Ciphering Mode Command to MS -- but needs Classmark 2 to determine whether A5/3 is supported
2018-12-25 23:40:18 +00:00
MSC <--GERAN-A-- MS: GSM48_MT_MM_AUTH_RESP
2019-01-03 01:32:14 +00:00
DREF IMSI-901700000004620: MSC conn use + dtap == 1 (0x2: dtap)
A5/n Ciph: request Classmark Update if missing
When the VLR requests a Ciphering Mode with vlr_ops.set_ciph_mode(), and if we
need a ciph algo flag from a Classmark information that is not yet known
(usually CM 2 during LU), send a BSSMAP Classmark Request to get it.
To manage the intermission of the Classmark Request, add
- msc_classmark_request_then_cipher_mode_cmd(),
- state SUBSCR_CONN_S_WAIT_CLASSMARK_UPDATE,
- event SUBSCR_CONN_E_CLASSMARK_UPDATE.
From state AUTH_CIPH, switch to state WAIT_CLASSMARK_UPDATE. Once the BSSMAP
Classmark Response, is received, switch back to SUBSCR_CONN_S_AUTH_CIPH and
re-initiate Ciphering Mode.
To be able to re-enter the Ciphering Mode algo decision, factor it out into
msc_geran_set_cipher_mode().
Rationale:
In the following commit, essentially we stopped supporting A5/3 ciphering:
commit 71330720b6efdda2fcfd3e9c0cb45f89e32e5670
"MSC: Intersect configured A5 algorithms with MS-supported ones"
Change-Id: Id124923ee52a357cb7d3e04d33f585214774f3a3
A5/3 was no longer supported because from that commit on, we strictly checked
the MS-supported ciphers, but we did not have Classmark 2 available during
Location Updating.
This patch changes that: when Classmark 2 is missing, actively request it by a
BSSMAP Classmark Request; continue Ciphering only after the Response. Always
request missing Classmark, even if a lesser cipher were configured available.
If the Classmark Update response fails to come in, cause an attach failure.
Instead, we could attempt to use a lesser cipher that is also enabled. That is
left as a future feature, should that become relevant. I think it's unlikely.
Technically, we could now end up requesting a Classmark Updating both during LU
(vlr_lu_fsm) and CM Service/Paging Response (proc_arq_fsm), but in practice the
only time we lack a Classmark is: during Location Updating with A5/3 enabled.
A5/1 support is indicated in CM1 which is always available, and A5/3 support is
indicated in CM2, which is always available during CM Service Request as well
as Paging Response. So this patch has practical relevance only for Location
Updating. For networks that permit only A5/3, this patch fixes Location
Updating. For networks that support A5/3 and A5/1, so far we always used A5/1
during LU, and after this patch we request CM2 and likely use A5/3 instead.
In msc_vlr_test_gsm_ciph, verify that requesting Classmark 2 for A5/3 works
during LU. Also verify that the lack of a Classmark Response results in attach
failure.
In msc_vlr_test_gsm_ciph, a hacky unit test fakes a situation where a CM2 is
missing during proc_arq_fsm and proves that that code path works, even though
the practical relevance is currently zero. It would only become interesting if
ciphering algorithms A5/4 and higher became relevant, because support of those
would be indicated in Classmark 3, which would always require a Classmark
Request.
Related: OS#3043
Depends: I4a2e1d3923e33912579c4180aa1ff8e8f5abb7e7 (libosmocore)
Change-Id: I73c7cb6a86624695bd9c0f59abb72e2fdc655131
2018-09-13 01:23:07 +00:00
DRLL Dispatching 04.08 message GSM48_MT_MM_AUTH_RESP (0x5:0x14)
2019-01-03 01:32:14 +00:00
DMM IMSI-901700000004620: MM GSM AUTHENTICATION RESPONSE (sres = 2d8b2c3e)
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_WAIT_RESP}: Received Event VLR_AUTH_E_MS_AUTH_RESP
DVLR SUBSCR(IMSI-901700000004620) AUTH on GERAN received SRES/RES: 2d8b2c3e (4 bytes)
DVLR SUBSCR(IMSI-901700000004620) AUTH established GSM security context
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_WAIT_RESP}: Authentication terminating with result PASSED
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_WAIT_RESP}: state_chg to VLR_SUB_AS_AUTHENTICATED
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_AUTHENTICATED}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_AUTHENTICATED}: Removing from parent vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU)
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_AUTHENTICATED}: Freeing instance
DVLR VLR_Authenticate(IMSI-901700000004620:GERAN-A-0:LU){VLR_SUB_AS_AUTHENTICATED}: Deallocated
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_WAIT_AUTH}: Received Event VLR_ULA_E_AUTH_RES
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_WAIT_AUTH}: vlr_loc_upd_post_auth()
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_WAIT_AUTH}: Set Ciphering Mode
DMM IMSI-901700000004620: to determine whether A5/3 is supported, first ask for a Classmark Update to obtain Classmark 2
DBSSAP RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_AUTH_CIPH}: Tx BSSMAP CLASSMARK REQUEST to BSC
A5/n Ciph: request Classmark Update if missing
When the VLR requests a Ciphering Mode with vlr_ops.set_ciph_mode(), and if we
need a ciph algo flag from a Classmark information that is not yet known
(usually CM 2 during LU), send a BSSMAP Classmark Request to get it.
To manage the intermission of the Classmark Request, add
- msc_classmark_request_then_cipher_mode_cmd(),
- state SUBSCR_CONN_S_WAIT_CLASSMARK_UPDATE,
- event SUBSCR_CONN_E_CLASSMARK_UPDATE.
From state AUTH_CIPH, switch to state WAIT_CLASSMARK_UPDATE. Once the BSSMAP
Classmark Response, is received, switch back to SUBSCR_CONN_S_AUTH_CIPH and
re-initiate Ciphering Mode.
To be able to re-enter the Ciphering Mode algo decision, factor it out into
msc_geran_set_cipher_mode().
Rationale:
In the following commit, essentially we stopped supporting A5/3 ciphering:
commit 71330720b6efdda2fcfd3e9c0cb45f89e32e5670
"MSC: Intersect configured A5 algorithms with MS-supported ones"
Change-Id: Id124923ee52a357cb7d3e04d33f585214774f3a3
A5/3 was no longer supported because from that commit on, we strictly checked
the MS-supported ciphers, but we did not have Classmark 2 available during
Location Updating.
This patch changes that: when Classmark 2 is missing, actively request it by a
BSSMAP Classmark Request; continue Ciphering only after the Response. Always
request missing Classmark, even if a lesser cipher were configured available.
If the Classmark Update response fails to come in, cause an attach failure.
Instead, we could attempt to use a lesser cipher that is also enabled. That is
left as a future feature, should that become relevant. I think it's unlikely.
Technically, we could now end up requesting a Classmark Updating both during LU
(vlr_lu_fsm) and CM Service/Paging Response (proc_arq_fsm), but in practice the
only time we lack a Classmark is: during Location Updating with A5/3 enabled.
A5/1 support is indicated in CM1 which is always available, and A5/3 support is
indicated in CM2, which is always available during CM Service Request as well
as Paging Response. So this patch has practical relevance only for Location
Updating. For networks that permit only A5/3, this patch fixes Location
Updating. For networks that support A5/3 and A5/1, so far we always used A5/1
during LU, and after this patch we request CM2 and likely use A5/3 instead.
In msc_vlr_test_gsm_ciph, verify that requesting Classmark 2 for A5/3 works
during LU. Also verify that the lack of a Classmark Response results in attach
failure.
In msc_vlr_test_gsm_ciph, a hacky unit test fakes a situation where a CM2 is
missing during proc_arq_fsm and proves that that code path works, even though
the practical relevance is currently zero. It would only become interesting if
ciphering algorithms A5/4 and higher became relevant, because support of those
would be indicated in Classmark 3, which would always require a Classmark
Request.
Related: OS#3043
Depends: I4a2e1d3923e33912579c4180aa1ff8e8f5abb7e7 (libosmocore)
Change-Id: I73c7cb6a86624695bd9c0f59abb72e2fdc655131
2018-09-13 01:23:07 +00:00
BSC <--BSSAP-BSS-MANAGEMENT-- MSC: CLASSMARK REQ [L3]> 00 01 58
2019-01-03 01:32:14 +00:00
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_AUTH_CIPH}: state_chg to RAN_CONN_S_WAIT_CLASSMARK_UPDATE
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_WAIT_AUTH}: state_chg to VLR_ULA_S_WAIT_CIPH
DREF IMSI-901700000004620: MSC conn use - dtap == 0 (0x0: )
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_WAIT_CLASSMARK_UPDATE}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_WAIT_CLASSMARK_UPDATE}: Awaiting results for Auth+Ciph, overruling event RAN_CONN_E_UNUSED
A5/n Ciph: request Classmark Update if missing
When the VLR requests a Ciphering Mode with vlr_ops.set_ciph_mode(), and if we
need a ciph algo flag from a Classmark information that is not yet known
(usually CM 2 during LU), send a BSSMAP Classmark Request to get it.
To manage the intermission of the Classmark Request, add
- msc_classmark_request_then_cipher_mode_cmd(),
- state SUBSCR_CONN_S_WAIT_CLASSMARK_UPDATE,
- event SUBSCR_CONN_E_CLASSMARK_UPDATE.
From state AUTH_CIPH, switch to state WAIT_CLASSMARK_UPDATE. Once the BSSMAP
Classmark Response, is received, switch back to SUBSCR_CONN_S_AUTH_CIPH and
re-initiate Ciphering Mode.
To be able to re-enter the Ciphering Mode algo decision, factor it out into
msc_geran_set_cipher_mode().
Rationale:
In the following commit, essentially we stopped supporting A5/3 ciphering:
commit 71330720b6efdda2fcfd3e9c0cb45f89e32e5670
"MSC: Intersect configured A5 algorithms with MS-supported ones"
Change-Id: Id124923ee52a357cb7d3e04d33f585214774f3a3
A5/3 was no longer supported because from that commit on, we strictly checked
the MS-supported ciphers, but we did not have Classmark 2 available during
Location Updating.
This patch changes that: when Classmark 2 is missing, actively request it by a
BSSMAP Classmark Request; continue Ciphering only after the Response. Always
request missing Classmark, even if a lesser cipher were configured available.
If the Classmark Update response fails to come in, cause an attach failure.
Instead, we could attempt to use a lesser cipher that is also enabled. That is
left as a future feature, should that become relevant. I think it's unlikely.
Technically, we could now end up requesting a Classmark Updating both during LU
(vlr_lu_fsm) and CM Service/Paging Response (proc_arq_fsm), but in practice the
only time we lack a Classmark is: during Location Updating with A5/3 enabled.
A5/1 support is indicated in CM1 which is always available, and A5/3 support is
indicated in CM2, which is always available during CM Service Request as well
as Paging Response. So this patch has practical relevance only for Location
Updating. For networks that permit only A5/3, this patch fixes Location
Updating. For networks that support A5/3 and A5/1, so far we always used A5/1
during LU, and after this patch we request CM2 and likely use A5/3 instead.
In msc_vlr_test_gsm_ciph, verify that requesting Classmark 2 for A5/3 works
during LU. Also verify that the lack of a Classmark Response results in attach
failure.
In msc_vlr_test_gsm_ciph, a hacky unit test fakes a situation where a CM2 is
missing during proc_arq_fsm and proves that that code path works, even though
the practical relevance is currently zero. It would only become interesting if
ciphering algorithms A5/4 and higher became relevant, because support of those
would be indicated in Classmark 3, which would always require a Classmark
Request.
Related: OS#3043
Depends: I4a2e1d3923e33912579c4180aa1ff8e8f5abb7e7 (libosmocore)
Change-Id: I73c7cb6a86624695bd9c0f59abb72e2fdc655131
2018-09-13 01:23:07 +00:00
lu_result_sent == 0
---
- But the BSSMAP Classmark Update never arrives
- At first, we're still waiting
- Total time passed: 0.000423 s
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
A5/n Ciph: request Classmark Update if missing
When the VLR requests a Ciphering Mode with vlr_ops.set_ciph_mode(), and if we
need a ciph algo flag from a Classmark information that is not yet known
(usually CM 2 during LU), send a BSSMAP Classmark Request to get it.
To manage the intermission of the Classmark Request, add
- msc_classmark_request_then_cipher_mode_cmd(),
- state SUBSCR_CONN_S_WAIT_CLASSMARK_UPDATE,
- event SUBSCR_CONN_E_CLASSMARK_UPDATE.
From state AUTH_CIPH, switch to state WAIT_CLASSMARK_UPDATE. Once the BSSMAP
Classmark Response, is received, switch back to SUBSCR_CONN_S_AUTH_CIPH and
re-initiate Ciphering Mode.
To be able to re-enter the Ciphering Mode algo decision, factor it out into
msc_geran_set_cipher_mode().
Rationale:
In the following commit, essentially we stopped supporting A5/3 ciphering:
commit 71330720b6efdda2fcfd3e9c0cb45f89e32e5670
"MSC: Intersect configured A5 algorithms with MS-supported ones"
Change-Id: Id124923ee52a357cb7d3e04d33f585214774f3a3
A5/3 was no longer supported because from that commit on, we strictly checked
the MS-supported ciphers, but we did not have Classmark 2 available during
Location Updating.
This patch changes that: when Classmark 2 is missing, actively request it by a
BSSMAP Classmark Request; continue Ciphering only after the Response. Always
request missing Classmark, even if a lesser cipher were configured available.
If the Classmark Update response fails to come in, cause an attach failure.
Instead, we could attempt to use a lesser cipher that is also enabled. That is
left as a future feature, should that become relevant. I think it's unlikely.
Technically, we could now end up requesting a Classmark Updating both during LU
(vlr_lu_fsm) and CM Service/Paging Response (proc_arq_fsm), but in practice the
only time we lack a Classmark is: during Location Updating with A5/3 enabled.
A5/1 support is indicated in CM1 which is always available, and A5/3 support is
indicated in CM2, which is always available during CM Service Request as well
as Paging Response. So this patch has practical relevance only for Location
Updating. For networks that permit only A5/3, this patch fixes Location
Updating. For networks that support A5/3 and A5/1, so far we always used A5/1
during LU, and after this patch we request CM2 and likely use A5/3 instead.
In msc_vlr_test_gsm_ciph, verify that requesting Classmark 2 for A5/3 works
during LU. Also verify that the lack of a Classmark Response results in attach
failure.
In msc_vlr_test_gsm_ciph, a hacky unit test fakes a situation where a CM2 is
missing during proc_arq_fsm and proves that that code path works, even though
the practical relevance is currently zero. It would only become interesting if
ciphering algorithms A5/4 and higher became relevant, because support of those
would be indicated in Classmark 3, which would always require a Classmark
Request.
Related: OS#3043
Depends: I4a2e1d3923e33912579c4180aa1ff8e8f5abb7e7 (libosmocore)
Change-Id: I73c7cb6a86624695bd9c0f59abb72e2fdc655131
2018-09-13 01:23:07 +00:00
lu_result_sent == 0
- Total time passed: 1.000658 s
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
A5/n Ciph: request Classmark Update if missing
When the VLR requests a Ciphering Mode with vlr_ops.set_ciph_mode(), and if we
need a ciph algo flag from a Classmark information that is not yet known
(usually CM 2 during LU), send a BSSMAP Classmark Request to get it.
To manage the intermission of the Classmark Request, add
- msc_classmark_request_then_cipher_mode_cmd(),
- state SUBSCR_CONN_S_WAIT_CLASSMARK_UPDATE,
- event SUBSCR_CONN_E_CLASSMARK_UPDATE.
From state AUTH_CIPH, switch to state WAIT_CLASSMARK_UPDATE. Once the BSSMAP
Classmark Response, is received, switch back to SUBSCR_CONN_S_AUTH_CIPH and
re-initiate Ciphering Mode.
To be able to re-enter the Ciphering Mode algo decision, factor it out into
msc_geran_set_cipher_mode().
Rationale:
In the following commit, essentially we stopped supporting A5/3 ciphering:
commit 71330720b6efdda2fcfd3e9c0cb45f89e32e5670
"MSC: Intersect configured A5 algorithms with MS-supported ones"
Change-Id: Id124923ee52a357cb7d3e04d33f585214774f3a3
A5/3 was no longer supported because from that commit on, we strictly checked
the MS-supported ciphers, but we did not have Classmark 2 available during
Location Updating.
This patch changes that: when Classmark 2 is missing, actively request it by a
BSSMAP Classmark Request; continue Ciphering only after the Response. Always
request missing Classmark, even if a lesser cipher were configured available.
If the Classmark Update response fails to come in, cause an attach failure.
Instead, we could attempt to use a lesser cipher that is also enabled. That is
left as a future feature, should that become relevant. I think it's unlikely.
Technically, we could now end up requesting a Classmark Updating both during LU
(vlr_lu_fsm) and CM Service/Paging Response (proc_arq_fsm), but in practice the
only time we lack a Classmark is: during Location Updating with A5/3 enabled.
A5/1 support is indicated in CM1 which is always available, and A5/3 support is
indicated in CM2, which is always available during CM Service Request as well
as Paging Response. So this patch has practical relevance only for Location
Updating. For networks that permit only A5/3, this patch fixes Location
Updating. For networks that support A5/3 and A5/1, so far we always used A5/1
during LU, and after this patch we request CM2 and likely use A5/3 instead.
In msc_vlr_test_gsm_ciph, verify that requesting Classmark 2 for A5/3 works
during LU. Also verify that the lack of a Classmark Response results in attach
failure.
In msc_vlr_test_gsm_ciph, a hacky unit test fakes a situation where a CM2 is
missing during proc_arq_fsm and proves that that code path works, even though
the practical relevance is currently zero. It would only become interesting if
ciphering algorithms A5/4 and higher became relevant, because support of those
would be indicated in Classmark 3, which would always require a Classmark
Request.
Related: OS#3043
Depends: I4a2e1d3923e33912579c4180aa1ff8e8f5abb7e7 (libosmocore)
Change-Id: I73c7cb6a86624695bd9c0f59abb72e2fdc655131
2018-09-13 01:23:07 +00:00
lu_result_sent == 0
- Total time passed: 2.000893 s
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
A5/n Ciph: request Classmark Update if missing
When the VLR requests a Ciphering Mode with vlr_ops.set_ciph_mode(), and if we
need a ciph algo flag from a Classmark information that is not yet known
(usually CM 2 during LU), send a BSSMAP Classmark Request to get it.
To manage the intermission of the Classmark Request, add
- msc_classmark_request_then_cipher_mode_cmd(),
- state SUBSCR_CONN_S_WAIT_CLASSMARK_UPDATE,
- event SUBSCR_CONN_E_CLASSMARK_UPDATE.
From state AUTH_CIPH, switch to state WAIT_CLASSMARK_UPDATE. Once the BSSMAP
Classmark Response, is received, switch back to SUBSCR_CONN_S_AUTH_CIPH and
re-initiate Ciphering Mode.
To be able to re-enter the Ciphering Mode algo decision, factor it out into
msc_geran_set_cipher_mode().
Rationale:
In the following commit, essentially we stopped supporting A5/3 ciphering:
commit 71330720b6efdda2fcfd3e9c0cb45f89e32e5670
"MSC: Intersect configured A5 algorithms with MS-supported ones"
Change-Id: Id124923ee52a357cb7d3e04d33f585214774f3a3
A5/3 was no longer supported because from that commit on, we strictly checked
the MS-supported ciphers, but we did not have Classmark 2 available during
Location Updating.
This patch changes that: when Classmark 2 is missing, actively request it by a
BSSMAP Classmark Request; continue Ciphering only after the Response. Always
request missing Classmark, even if a lesser cipher were configured available.
If the Classmark Update response fails to come in, cause an attach failure.
Instead, we could attempt to use a lesser cipher that is also enabled. That is
left as a future feature, should that become relevant. I think it's unlikely.
Technically, we could now end up requesting a Classmark Updating both during LU
(vlr_lu_fsm) and CM Service/Paging Response (proc_arq_fsm), but in practice the
only time we lack a Classmark is: during Location Updating with A5/3 enabled.
A5/1 support is indicated in CM1 which is always available, and A5/3 support is
indicated in CM2, which is always available during CM Service Request as well
as Paging Response. So this patch has practical relevance only for Location
Updating. For networks that permit only A5/3, this patch fixes Location
Updating. For networks that support A5/3 and A5/1, so far we always used A5/1
during LU, and after this patch we request CM2 and likely use A5/3 instead.
In msc_vlr_test_gsm_ciph, verify that requesting Classmark 2 for A5/3 works
during LU. Also verify that the lack of a Classmark Response results in attach
failure.
In msc_vlr_test_gsm_ciph, a hacky unit test fakes a situation where a CM2 is
missing during proc_arq_fsm and proves that that code path works, even though
the practical relevance is currently zero. It would only become interesting if
ciphering algorithms A5/4 and higher became relevant, because support of those
would be indicated in Classmark 3, which would always require a Classmark
Request.
Related: OS#3043
Depends: I4a2e1d3923e33912579c4180aa1ff8e8f5abb7e7 (libosmocore)
Change-Id: I73c7cb6a86624695bd9c0f59abb72e2fdc655131
2018-09-13 01:23:07 +00:00
lu_result_sent == 0
- Total time passed: 3.001128 s
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
A5/n Ciph: request Classmark Update if missing
When the VLR requests a Ciphering Mode with vlr_ops.set_ciph_mode(), and if we
need a ciph algo flag from a Classmark information that is not yet known
(usually CM 2 during LU), send a BSSMAP Classmark Request to get it.
To manage the intermission of the Classmark Request, add
- msc_classmark_request_then_cipher_mode_cmd(),
- state SUBSCR_CONN_S_WAIT_CLASSMARK_UPDATE,
- event SUBSCR_CONN_E_CLASSMARK_UPDATE.
From state AUTH_CIPH, switch to state WAIT_CLASSMARK_UPDATE. Once the BSSMAP
Classmark Response, is received, switch back to SUBSCR_CONN_S_AUTH_CIPH and
re-initiate Ciphering Mode.
To be able to re-enter the Ciphering Mode algo decision, factor it out into
msc_geran_set_cipher_mode().
Rationale:
In the following commit, essentially we stopped supporting A5/3 ciphering:
commit 71330720b6efdda2fcfd3e9c0cb45f89e32e5670
"MSC: Intersect configured A5 algorithms with MS-supported ones"
Change-Id: Id124923ee52a357cb7d3e04d33f585214774f3a3
A5/3 was no longer supported because from that commit on, we strictly checked
the MS-supported ciphers, but we did not have Classmark 2 available during
Location Updating.
This patch changes that: when Classmark 2 is missing, actively request it by a
BSSMAP Classmark Request; continue Ciphering only after the Response. Always
request missing Classmark, even if a lesser cipher were configured available.
If the Classmark Update response fails to come in, cause an attach failure.
Instead, we could attempt to use a lesser cipher that is also enabled. That is
left as a future feature, should that become relevant. I think it's unlikely.
Technically, we could now end up requesting a Classmark Updating both during LU
(vlr_lu_fsm) and CM Service/Paging Response (proc_arq_fsm), but in practice the
only time we lack a Classmark is: during Location Updating with A5/3 enabled.
A5/1 support is indicated in CM1 which is always available, and A5/3 support is
indicated in CM2, which is always available during CM Service Request as well
as Paging Response. So this patch has practical relevance only for Location
Updating. For networks that permit only A5/3, this patch fixes Location
Updating. For networks that support A5/3 and A5/1, so far we always used A5/1
during LU, and after this patch we request CM2 and likely use A5/3 instead.
In msc_vlr_test_gsm_ciph, verify that requesting Classmark 2 for A5/3 works
during LU. Also verify that the lack of a Classmark Response results in attach
failure.
In msc_vlr_test_gsm_ciph, a hacky unit test fakes a situation where a CM2 is
missing during proc_arq_fsm and proves that that code path works, even though
the practical relevance is currently zero. It would only become interesting if
ciphering algorithms A5/4 and higher became relevant, because support of those
would be indicated in Classmark 3, which would always require a Classmark
Request.
Related: OS#3043
Depends: I4a2e1d3923e33912579c4180aa1ff8e8f5abb7e7 (libosmocore)
Change-Id: I73c7cb6a86624695bd9c0f59abb72e2fdc655131
2018-09-13 01:23:07 +00:00
lu_result_sent == 0
- Total time passed: 4.001363 s
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
A5/n Ciph: request Classmark Update if missing
When the VLR requests a Ciphering Mode with vlr_ops.set_ciph_mode(), and if we
need a ciph algo flag from a Classmark information that is not yet known
(usually CM 2 during LU), send a BSSMAP Classmark Request to get it.
To manage the intermission of the Classmark Request, add
- msc_classmark_request_then_cipher_mode_cmd(),
- state SUBSCR_CONN_S_WAIT_CLASSMARK_UPDATE,
- event SUBSCR_CONN_E_CLASSMARK_UPDATE.
From state AUTH_CIPH, switch to state WAIT_CLASSMARK_UPDATE. Once the BSSMAP
Classmark Response, is received, switch back to SUBSCR_CONN_S_AUTH_CIPH and
re-initiate Ciphering Mode.
To be able to re-enter the Ciphering Mode algo decision, factor it out into
msc_geran_set_cipher_mode().
Rationale:
In the following commit, essentially we stopped supporting A5/3 ciphering:
commit 71330720b6efdda2fcfd3e9c0cb45f89e32e5670
"MSC: Intersect configured A5 algorithms with MS-supported ones"
Change-Id: Id124923ee52a357cb7d3e04d33f585214774f3a3
A5/3 was no longer supported because from that commit on, we strictly checked
the MS-supported ciphers, but we did not have Classmark 2 available during
Location Updating.
This patch changes that: when Classmark 2 is missing, actively request it by a
BSSMAP Classmark Request; continue Ciphering only after the Response. Always
request missing Classmark, even if a lesser cipher were configured available.
If the Classmark Update response fails to come in, cause an attach failure.
Instead, we could attempt to use a lesser cipher that is also enabled. That is
left as a future feature, should that become relevant. I think it's unlikely.
Technically, we could now end up requesting a Classmark Updating both during LU
(vlr_lu_fsm) and CM Service/Paging Response (proc_arq_fsm), but in practice the
only time we lack a Classmark is: during Location Updating with A5/3 enabled.
A5/1 support is indicated in CM1 which is always available, and A5/3 support is
indicated in CM2, which is always available during CM Service Request as well
as Paging Response. So this patch has practical relevance only for Location
Updating. For networks that permit only A5/3, this patch fixes Location
Updating. For networks that support A5/3 and A5/1, so far we always used A5/1
during LU, and after this patch we request CM2 and likely use A5/3 instead.
In msc_vlr_test_gsm_ciph, verify that requesting Classmark 2 for A5/3 works
during LU. Also verify that the lack of a Classmark Response results in attach
failure.
In msc_vlr_test_gsm_ciph, a hacky unit test fakes a situation where a CM2 is
missing during proc_arq_fsm and proves that that code path works, even though
the practical relevance is currently zero. It would only become interesting if
ciphering algorithms A5/4 and higher became relevant, because support of those
would be indicated in Classmark 3, which would always require a Classmark
Request.
Related: OS#3043
Depends: I4a2e1d3923e33912579c4180aa1ff8e8f5abb7e7 (libosmocore)
Change-Id: I73c7cb6a86624695bd9c0f59abb72e2fdc655131
2018-09-13 01:23:07 +00:00
lu_result_sent == 0
- Total time passed: 5.001598 s
2019-01-03 01:32:14 +00:00
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_WAIT_CLASSMARK_UPDATE}: Timeout of T0
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_WAIT_CLASSMARK_UPDATE}: Received Event RAN_CONN_E_CN_CLOSE
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_WAIT_CLASSMARK_UPDATE}: Close event, cause: CONGESTION
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_WAIT_CLASSMARK_UPDATE}: state_chg to RAN_CONN_S_RELEASING
DREF IMSI-901700000004620: MSC conn use + release == 1 (0x100: release)
DREF VLR subscr IMSI-901700000004620 usage increases to: 2
- sending LU Reject for IMSI-901700000004620, cause 22
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_WAIT_CIPH}: state_chg to VLR_ULA_S_DONE
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_RELEASING}: Received Event RAN_CONN_E_CN_CLOSE
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_RELEASING}: Event RAN_CONN_E_CN_CLOSE not permitted
DREF VLR subscr IMSI-901700000004620 usage decreases to: 1
2018-12-25 23:40:18 +00:00
- BSSAP Clear --GERAN-A--> MS
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_TIMEOUT has passed, conn is gone.
A5/n Ciph: request Classmark Update if missing
When the VLR requests a Ciphering Mode with vlr_ops.set_ciph_mode(), and if we
need a ciph algo flag from a Classmark information that is not yet known
(usually CM 2 during LU), send a BSSMAP Classmark Request to get it.
To manage the intermission of the Classmark Request, add
- msc_classmark_request_then_cipher_mode_cmd(),
- state SUBSCR_CONN_S_WAIT_CLASSMARK_UPDATE,
- event SUBSCR_CONN_E_CLASSMARK_UPDATE.
From state AUTH_CIPH, switch to state WAIT_CLASSMARK_UPDATE. Once the BSSMAP
Classmark Response, is received, switch back to SUBSCR_CONN_S_AUTH_CIPH and
re-initiate Ciphering Mode.
To be able to re-enter the Ciphering Mode algo decision, factor it out into
msc_geran_set_cipher_mode().
Rationale:
In the following commit, essentially we stopped supporting A5/3 ciphering:
commit 71330720b6efdda2fcfd3e9c0cb45f89e32e5670
"MSC: Intersect configured A5 algorithms with MS-supported ones"
Change-Id: Id124923ee52a357cb7d3e04d33f585214774f3a3
A5/3 was no longer supported because from that commit on, we strictly checked
the MS-supported ciphers, but we did not have Classmark 2 available during
Location Updating.
This patch changes that: when Classmark 2 is missing, actively request it by a
BSSMAP Classmark Request; continue Ciphering only after the Response. Always
request missing Classmark, even if a lesser cipher were configured available.
If the Classmark Update response fails to come in, cause an attach failure.
Instead, we could attempt to use a lesser cipher that is also enabled. That is
left as a future feature, should that become relevant. I think it's unlikely.
Technically, we could now end up requesting a Classmark Updating both during LU
(vlr_lu_fsm) and CM Service/Paging Response (proc_arq_fsm), but in practice the
only time we lack a Classmark is: during Location Updating with A5/3 enabled.
A5/1 support is indicated in CM1 which is always available, and A5/3 support is
indicated in CM2, which is always available during CM Service Request as well
as Paging Response. So this patch has practical relevance only for Location
Updating. For networks that permit only A5/3, this patch fixes Location
Updating. For networks that support A5/3 and A5/1, so far we always used A5/1
during LU, and after this patch we request CM2 and likely use A5/3 instead.
In msc_vlr_test_gsm_ciph, verify that requesting Classmark 2 for A5/3 works
during LU. Also verify that the lack of a Classmark Response results in attach
failure.
In msc_vlr_test_gsm_ciph, a hacky unit test fakes a situation where a CM2 is
missing during proc_arq_fsm and proves that that code path works, even though
the practical relevance is currently zero. It would only become interesting if
ciphering algorithms A5/4 and higher became relevant, because support of those
would be indicated in Classmark 3, which would always require a Classmark
Request.
Related: OS#3043
Depends: I4a2e1d3923e33912579c4180aa1ff8e8f5abb7e7 (libosmocore)
Change-Id: I73c7cb6a86624695bd9c0f59abb72e2fdc655131
2018-09-13 01:23:07 +00:00
bssap_clear_sent == 1
- BSS sends BSSMAP Clear Complete
2019-01-03 01:32:14 +00:00
DREF IMSI-901700000004620: MSC conn use - release == 0 (0x0: )
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_RELEASING}: Received Event RAN_CONN_E_UNUSED
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_RELEASING}: state_chg to RAN_CONN_S_RELEASED
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_RELEASED}: Terminating (cause = OSMO_FSM_TERM_REGULAR)
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_DONE}: Terminating (cause = OSMO_FSM_TERM_PARENT)
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_DONE}: Removing from parent RAN_conn(IMSI-901700000004620:GERAN-A-0:LU)
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_DONE}: fsm_lu_cleanup called with cause OSMO_FSM_TERM_PARENT
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_DONE}: Freeing instance
DVLR vlr_lu_fsm(IMSI-901700000004620:GERAN-A-0:LU){VLR_ULA_S_DONE}: Deallocated
DRLL IMSI-901700000004620: Freeing RAN connection
DREF VLR subscr IMSI-901700000004620 usage decreases to: 0
DREF freeing VLR subscr IMSI-901700000004620
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_RELEASED}: Freeing instance
DMM RAN_conn(IMSI-901700000004620:GERAN-A-0:LU){RAN_CONN_S_RELEASED}: 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
llist_count(&net->ran_conns) == 0
A5/n Ciph: request Classmark Update if missing
When the VLR requests a Ciphering Mode with vlr_ops.set_ciph_mode(), and if we
need a ciph algo flag from a Classmark information that is not yet known
(usually CM 2 during LU), send a BSSMAP Classmark Request to get it.
To manage the intermission of the Classmark Request, add
- msc_classmark_request_then_cipher_mode_cmd(),
- state SUBSCR_CONN_S_WAIT_CLASSMARK_UPDATE,
- event SUBSCR_CONN_E_CLASSMARK_UPDATE.
From state AUTH_CIPH, switch to state WAIT_CLASSMARK_UPDATE. Once the BSSMAP
Classmark Response, is received, switch back to SUBSCR_CONN_S_AUTH_CIPH and
re-initiate Ciphering Mode.
To be able to re-enter the Ciphering Mode algo decision, factor it out into
msc_geran_set_cipher_mode().
Rationale:
In the following commit, essentially we stopped supporting A5/3 ciphering:
commit 71330720b6efdda2fcfd3e9c0cb45f89e32e5670
"MSC: Intersect configured A5 algorithms with MS-supported ones"
Change-Id: Id124923ee52a357cb7d3e04d33f585214774f3a3
A5/3 was no longer supported because from that commit on, we strictly checked
the MS-supported ciphers, but we did not have Classmark 2 available during
Location Updating.
This patch changes that: when Classmark 2 is missing, actively request it by a
BSSMAP Classmark Request; continue Ciphering only after the Response. Always
request missing Classmark, even if a lesser cipher were configured available.
If the Classmark Update response fails to come in, cause an attach failure.
Instead, we could attempt to use a lesser cipher that is also enabled. That is
left as a future feature, should that become relevant. I think it's unlikely.
Technically, we could now end up requesting a Classmark Updating both during LU
(vlr_lu_fsm) and CM Service/Paging Response (proc_arq_fsm), but in practice the
only time we lack a Classmark is: during Location Updating with A5/3 enabled.
A5/1 support is indicated in CM1 which is always available, and A5/3 support is
indicated in CM2, which is always available during CM Service Request as well
as Paging Response. So this patch has practical relevance only for Location
Updating. For networks that permit only A5/3, this patch fixes Location
Updating. For networks that support A5/3 and A5/1, so far we always used A5/1
during LU, and after this patch we request CM2 and likely use A5/3 instead.
In msc_vlr_test_gsm_ciph, verify that requesting Classmark 2 for A5/3 works
during LU. Also verify that the lack of a Classmark Response results in attach
failure.
In msc_vlr_test_gsm_ciph, a hacky unit test fakes a situation where a CM2 is
missing during proc_arq_fsm and proves that that code path works, even though
the practical relevance is currently zero. It would only become interesting if
ciphering algorithms A5/4 and higher became relevant, because support of those
would be indicated in Classmark 3, which would always require a Classmark
Request.
Related: OS#3043
Depends: I4a2e1d3923e33912579c4180aa1ff8e8f5abb7e7 (libosmocore)
Change-Id: I73c7cb6a86624695bd9c0f59abb72e2fdc655131
2018-09-13 01:23:07 +00:00
lu_result_sent == 2
===== test_classmark_update_timeout: SUCCESS
full talloc report on 'msgb' (total 0 bytes in 1 blocks)
2018-12-06 11:06:59 +00:00
talloc_total_blocks(tall_bsc_ctx) == 13
A5/n Ciph: request Classmark Update if missing
When the VLR requests a Ciphering Mode with vlr_ops.set_ciph_mode(), and if we
need a ciph algo flag from a Classmark information that is not yet known
(usually CM 2 during LU), send a BSSMAP Classmark Request to get it.
To manage the intermission of the Classmark Request, add
- msc_classmark_request_then_cipher_mode_cmd(),
- state SUBSCR_CONN_S_WAIT_CLASSMARK_UPDATE,
- event SUBSCR_CONN_E_CLASSMARK_UPDATE.
From state AUTH_CIPH, switch to state WAIT_CLASSMARK_UPDATE. Once the BSSMAP
Classmark Response, is received, switch back to SUBSCR_CONN_S_AUTH_CIPH and
re-initiate Ciphering Mode.
To be able to re-enter the Ciphering Mode algo decision, factor it out into
msc_geran_set_cipher_mode().
Rationale:
In the following commit, essentially we stopped supporting A5/3 ciphering:
commit 71330720b6efdda2fcfd3e9c0cb45f89e32e5670
"MSC: Intersect configured A5 algorithms with MS-supported ones"
Change-Id: Id124923ee52a357cb7d3e04d33f585214774f3a3
A5/3 was no longer supported because from that commit on, we strictly checked
the MS-supported ciphers, but we did not have Classmark 2 available during
Location Updating.
This patch changes that: when Classmark 2 is missing, actively request it by a
BSSMAP Classmark Request; continue Ciphering only after the Response. Always
request missing Classmark, even if a lesser cipher were configured available.
If the Classmark Update response fails to come in, cause an attach failure.
Instead, we could attempt to use a lesser cipher that is also enabled. That is
left as a future feature, should that become relevant. I think it's unlikely.
Technically, we could now end up requesting a Classmark Updating both during LU
(vlr_lu_fsm) and CM Service/Paging Response (proc_arq_fsm), but in practice the
only time we lack a Classmark is: during Location Updating with A5/3 enabled.
A5/1 support is indicated in CM1 which is always available, and A5/3 support is
indicated in CM2, which is always available during CM Service Request as well
as Paging Response. So this patch has practical relevance only for Location
Updating. For networks that permit only A5/3, this patch fixes Location
Updating. For networks that support A5/3 and A5/1, so far we always used A5/1
during LU, and after this patch we request CM2 and likely use A5/3 instead.
In msc_vlr_test_gsm_ciph, verify that requesting Classmark 2 for A5/3 works
during LU. Also verify that the lack of a Classmark Response results in attach
failure.
In msc_vlr_test_gsm_ciph, a hacky unit test fakes a situation where a CM2 is
missing during proc_arq_fsm and proves that that code path works, even though
the practical relevance is currently zero. It would only become interesting if
ciphering algorithms A5/4 and higher became relevant, because support of those
would be indicated in Classmark 3, which would always require a Classmark
Request.
Related: OS#3043
Depends: I4a2e1d3923e33912579c4180aa1ff8e8f5abb7e7 (libosmocore)
Change-Id: I73c7cb6a86624695bd9c0f59abb72e2fdc655131
2018-09-13 01:23:07 +00:00
2017-01-25 14:04:16 +00:00
full talloc report on 'msgb' (total 0 bytes in 1 blocks)
2018-12-06 11:06:59 +00:00
talloc_total_blocks(tall_bsc_ctx) == 13
2017-01-25 14:04:16 +00:00
