Commit Graph

26 Commits

Author SHA1 Message Date
Jacob Erlbeck 98a95ac17f sgsn: Add a subscriber based authentication phase
This implements the MAP way of subscriber validation when the MS
tries to perform an Attach Request:

  1. perform authentication (optionally invoke the sendAuthInfo
     procedure), starts the Auth & Ciph procedure
  2. perform update location
  3. insert subscriber data
  4. finish the update location
  5. Attach Accept / Attach Reject

The authentication triplets are used and eventually updated if all of
them have been used.

This is currently accessible via the VTY interface by the following
commands:

  - update-subscriber imsi IMSI update-auth-info
  - update-subscriber imsi IMSI update-location-result (ok|ERR-CAUSE)

Sponsored-by: On-Waves ehf
2014-12-24 16:50:30 +01:00
Jacob Erlbeck 7921ab1593 sgsn: Add support for authentication triplets
This commit add data structures, functions, initialization, and VTY
commands for per subscriber authentication triplets.

The following VTY command is added:

  - update-subscriber imsi IMSI \
    insert auth-triplet <1-5> sres SRES rand RAND kc KC

Note that the triplets are not really used by the SGSN yet.

Sponsored-by: On-Waves ehf
2014-12-24 16:50:16 +01:00
Jacob Erlbeck 2e5e94c328 sgsn: Support subscriber based authentication
This commit mainly extends sgsn_auth.c to use and support the
auth_state SGSN_AUTH_AUTHENTICATE. It will be activated when IMSI and
IMEI are available, authentication is required
(subscr->sgsn_data->authenticate is set), but the MM context is not
marked as authenticated. If the state has been set to
SGSN_AUTH_AUTHENTICATE and sgsn_auth_update() is called, the GMM
layer will be informed by invoking gsm0408_gprs_authenticate().

Sponsored-by: On-Waves ehf
2014-12-10 12:45:22 +01:00
Jacob Erlbeck 98647ca0ef sgsn: Add gsm0408_gprs_access_cancelled
This function is called to delete an established MM context
silently without invoking a detach procedure.

It is called when a subscriber is cancelled by the HLR. This
generally happens, when an MS has moved to another routing area and
has to use another SGSN.

Sponsored-by: On-Waves ehf
2014-12-09 09:48:25 +01:00
Jacob Erlbeck be2c8d9358 sgsn: Integrate subscriber handling into the SGSN
This commit adds a new authorization policy 'remote' and uses
the subscriber cache for authorization when this policy is being used.

Note that there is no remote backend implemented yet. After the
IMSI/IMEI have been acquired, a request would be sent to the remote
peer. The attach/auth-ciph procedure continues when authorization
info has been received from the peer. This means, that
gprs_subscr_update() must be called then to tell the GMM layer
that it can proceed. A later commit will add VTY commands to do this
manually.

Sponsored-by: On-Waves ehf
2014-12-09 09:27:20 +01:00
Jacob Erlbeck 33b6dadc88 sgsn: Add gprs_subscriber.c
This patch adds GPRS specific functions for gsm_subscriber objects
(allocation, retrieval, deletion) and subscriber data
requests/updates. The sgsn_update_subscriber_data callback is used to
notify the sgsn about updates and is extended by a parameter that
passes a reference to a gsm_subscriber.

Sponsored-by: On-Waves ehf
2014-12-09 09:23:11 +01:00
Jacob Erlbeck de4bbc7146 sgsn/test: Add test case for unexpected Detach Accepts
The commit "sgsn: Don't send XID reset after Detach Accept" fixed the
SGSN to not respond to a Detach Accept message when there is no MM
context.

This patch adds a test case to verify, that there is really no
message sent in that case.

The test fails when the commit mentioned aboved is reverted.

Sponsored-by: On-Waves ehf
2014-12-02 09:18:45 +01:00
Jacob Erlbeck aec03a1f13 sgsn/test: Don't rely on the actual PRNG sequence
Currently the expected P-TMSI generated by the SGSN is hard-coded
into the test. This adds a dependency on the implementation of rand()
and thus the libc used. This breaks the test e.g. on FreeBSD.

This patch modifies test_gmm_attach() to srand(1) first, generate the
P-TMSI, and finally srand(1) again before starting the test.

Sponsored-by: On-Waves ehf
2014-12-02 09:18:19 +01:00
Jacob Erlbeck a0b6efb368 sgsn: Remove explicit sgsn_instance parameters
Currently the function in sgsn_auth.c either have an sgsn_config or
an sgsn_instance parameter. Since then global sgsn variable is
already being used in that file and since other parts of the SGSN
related code also rely on a global sgsn singleton, these parameters
pretend to provide a flexibility that is not really supported.

Therefore this patch removes these parameters except for the ACL
related functions, which do not call code that uses the sgsn
variable.

Sponsored-by: On-Waves ehf
2014-11-14 10:26:06 +01:00
Jacob Erlbeck f6e7d99d54 sgsn: Don't assign a new P-TMSI if one is pending
Currently every time an RA Update Req or an Attach Req is processed, a
new P-TMSI is allocated. When an MS issues another of these messages
before it has completed the first procedure, old_ptmsi is replaced by
ptmsi (and thus lost) and ptmsi is replaced by the newly allocated
P-TMSI. This can confuse the gbproxy, which can loose track of the
logical link then. At least a Blackberry emits a double set of RA Upd
Req messages from time to time which may be just 20ms apart.

This patch adds a check whether mm->ptmsi or mm->old_ptmsi are set.
If both are set, the P-TMSI is not re-allocated. This is only the
case, when the Complete message has not been received yet, since that
message will reset old_ptmsi.

Sponsored-by: On-Waves ehf
2014-11-14 10:19:29 +01:00
Jacob Erlbeck 80d07e30c7 sgsn: Cleanup after RA Update Reject / Attach Reject
Currently, the LLME is not cleaned up after sending an RA Update
Reject. This happens after entering a routing area from outside,
since in that case the SGSN sends an RA Update Reject (implicitly
detached) which causes the MS to restart the attach procedure.
The LLME is also not updated if an Attach Request with message errors
(encoding, invalid MI type) is received or if an MM context cannot be
allocated.

This patch changes gsm48_rx_gmm_ra_upd_req and gsm48_rx_gmm_att_req
to unassign the LLME or free the MM context (if available) after a
Reject message has been sent.

Ticket: OW#1324
Sponsored-by: On-Waves ehf
2014-11-14 10:07:42 +01:00
Jacob Erlbeck 106f547733 sgsn: Add 'acl-only' authentication policy
Currently the VTY 'auth-policy' command results in setting or clearing
the acl_enabled flag. This also enables the matching of the MCC/MNC
prefix of the IMSI.

This patch adds an additional policy 'acl-only' which disables the
MCC/MNC matching and relies on the ACL only.

Sponsored-by: On-Waves ehf
2014-11-14 10:07:28 +01:00
Jacob Erlbeck 0c06f98ced sgsn: Move IMSI authorization to gsm48_gmm_authorize
Currently the IMSI is only checked immediately when an Attach Request
is received that contains an IMSI IE. If it contains a P-TMSI
instead, access is always granted.

This commit moves the IMSI check to gsm48_gmm_authorize where it is
applied when IMSI and IMEI have been acquired. This fixes the
authorization when the Attach Accept doesn't contain an IMSI.

Sponsored-by: On-Waves ehf
2014-11-14 09:27:23 +01:00
Jacob Erlbeck 0074a77424 sgsn: Cleanup GMM state transitions
Currently the GMM state is set to GMM-REGISTERED when an Attach
Accept or a RA Update Accept message is sent, even if a new P-TMSI is
included. In this case 04.08 requires (see 4.7.3.1.3 and 4.7.5.1.3),
that the state is set to GMM-COMMON-PROCEDURE-INITIATED when the
Accept is sent. When the Complete is received, the SGSN shall set
the state to GMM-REGISTERED.

This patch modifies the state updates accordingly.

Sponsored-by: On-Waves ehf
2014-11-14 09:12:10 +01:00
Jacob Erlbeck 27cb4d57e0 sgsn/test: Add test_gmm_attach
This test checks the attach procedure until the Attach Complete is
received.

Note that authorization and GMM state updates are not working
properly yet.

Sponsored-by: On-Waves ehf
2014-11-11 22:47:59 +01:00
Jacob Erlbeck 94ef1c0da9 sgsn/test: Move message sending to send_0408_message
This replaces serveral occurences of duplicated code for message
creation and sending (passing to gsm0408_gprs_rcvmsg) into a single
function. In addition, the sgsn_tx_counter is always reset within
send_0408_message to simplify the code that checks for the number of
messages sent.

Sponsored-by: On-Waves ehf
2014-11-11 22:47:40 +01:00
Jacob Erlbeck 14ae582064 sgsn: Unassign the LLME after GMM Status without mmctx
Currently the LLME is not deleted when a GMM Status message is
received for which a mmctx cannot be found. This can fill the LLME
list with unneeded entries.

This patch adds code to unassign the LLME in that case.

Ticket: OW#1324
Sponsored-by: On-Waves ehf
2014-11-05 00:24:02 +01:00
Jacob Erlbeck 5a38f6470e sgsn: Handle Detach Requests even when there is no mmctx
Currently, when a Detach Request is received with an unknown TLLI,
it is answered by another Detach Request (!), even when a power_off
Type is used.

This patch uses gsm48_rx_gmm_det_req to handle the message instead.
So this function is changed to cope with a NULL mmctx. In that case
it doesn't unassign the llme, so this must be done manually
afterwards.

Sponsored-by: On-Waves ehf
2014-10-27 15:34:14 +01:00
Jacob Erlbeck b9ab0d4f39 sgsn: Only send Detach Accept (MO) if power_off isn't set
Currently, every time the SGSN received a Detach Request from the MS
via an established logical link, it is answered by a Detach Accept.
This violates the specification (GSM 04.08, 4.7.4.1.2 and .3), which
states, that it should only be sent, if "the detach type IE value
indicates that the detach request has not been sent due to switching
off".

This patch adds a conditional to limit the sending of Detach Accept
accordingly.

Sponsored-by: On-Waves ehf
2014-10-27 15:21:31 +01:00
Jacob Erlbeck 189999d654 sgsn/test: Add test case for Detach Request (MO, power-off = 1)
Currently only a Detach Request (MO) message with power_off = 0 is
checked.

This commit adds a new test case with power_off set to 1. It also
adds checks for the number of messages generated by the SGSN to
verify that these messages are handled differently.

Note that the handling of power_off isn't implemented yet. Therefore
the corresponding assertion is being disabled yet.

Sponsored-by: On-Waves ehf
2014-10-27 15:21:11 +01:00
Jacob Erlbeck abc16a55b2 sgsn/test: Move MM context allocation into separate function
This refactores serveral code lines needed for the allocation of MM
contexts into the new function alloc_mm_ctx.

Sponsored-by: On-Waves ehf
2014-10-27 15:20:07 +01:00
Jacob Erlbeck 3b5d407203 sgsn: Moved IMSI ACL management to sgsn_auth.c
Currently the ACL code is located in sgsn_vty.c.

This commit moves this to a new file sgsn_auth.c as a first step to
make authorization more flexible in order to implement remote
acquisition on subsciber data.

Sponsored-by: On-Waves ehf
2014-10-27 13:06:55 +01:00
Jacob Erlbeck 258ce3ded5 sgsn: Free MM context after receiving a Detach Request
On a Detach/Re-attach cycle the Address Sanitizer detected a
use-after-free kind of problem. That is because we tried to
destroy the LLME twice. The first time it is destroyed as part
of the Detach handling ans the second time it is destroyed as
part of destroying the old MM context.

In case the GPRS GMM detach message is lost the SGSN needs
to reply besides not having a MM entry.

The alternative would have been to add NULL checks for all
usages of ctx->llme which would not have helped with the
readability.

Sponsored-by: On-Waves ehf
2014-10-09 17:22:34 +02:00
Holger Hans Peter Freyther fe92133132 sgsn: Add test that exposes a dangling pointer to the LLME
On detach the LLME get's unassigned (and hence destroyed) but the
GMM context will still point to that dead structure.
2014-10-09 17:22:34 +02:00
Holger Hans Peter Freyther 4299c0560f sgsn: Create testcase that verifies that llmes get deleted
On an "unassignment" this code verifies that the LLME will vanish
from the list of LLMEs. We assume that this doesn't create a
memory leak.
2014-10-09 17:22:34 +02:00
Holger Hans Peter Freyther 68c6f887c2 sgsn: Add boilerplate code for a SGSN test 2014-10-09 17:22:34 +02:00