openbsc.git branch jerlbeck/wip/gprs-patching updated. 0.13.0-878-g64d0007

gitosis at gitosis at
Wed Jan 14 10:28:30 UTC 2015

This is an automated email from the git hooks/post-receive script. It was
generated because a ref change was pushed to the repository containing
the project "The OpenBSC GSM Base Station Controller (+MSC/HLR/SGSN)".

The branch, jerlbeck/wip/gprs-patching has been updated
  discards  89b5af1f8d450e039602030d06ea6b0bd62925a2 (commit)
  discards  488cc66679e20973fc85c4bc9a39eacf46548f80 (commit)
  discards  51d8be51f7a4c099a72a5ebc025f3daf8710918d (commit)
  discards  cfe222efff4a215043b6828d246b433e2faa0710 (commit)
  discards  71e61686c3a469f057636eec6ff2003affdbe1c1 (commit)
  discards  c20303e00196a96b08d5be4e716774983ea92fce (commit)
  discards  328c5c1d40aecfe2c72e352d0d6beda4b7c86f86 (commit)
  discards  5355202011c9ac4a751608ecdb7d825c7f335d1e (commit)
  discards  7ed031b2c876d8c2ab1471d27569131c683011ee (commit)
  discards  77057d0820690d6f67716670ff94b240b370c118 (commit)
  discards  157392a168916b6828676d53d943b099e0d0708d (commit)
  discards  0df73b1093c7d720cc64fa967a2268a13a929c72 (commit)
  discards  aee0cdc4db6d34855597ebe7e09abc647d8e692e (commit)
  discards  43e71d624c7be3cce2cec7cb419f8b4129977a9e (commit)
  discards  f1393090b4a164bb480f45cce9f8ab54b1bc6089 (commit)
  discards  1636e5f607325fb2cf0fa39c7c51eea0df49a9a3 (commit)
  discards  9c420d46836a7bee7dbb74b75763c92a7ee1d6ad (commit)
  discards  a5a1af953fefa425b7a1611dab0ab67f8fe9285d (commit)
  discards  9d3c9d60d842bf053a1481703c9615bac0312295 (commit)
  discards  4fc87b30c4dda8c6897b148bb0d9e400c8e22c4f (commit)
  discards  1fcb2dca60a43cffedb9e40fca13e86a1832b7b4 (commit)
  discards  866eb9fbc5711a66538fb5da2ac2238c9b035116 (commit)
  discards  287cf4ee746ff1eb49eea52858de7304b1735b87 (commit)
  discards  df61af7043081b1a20646dec088677b09d67fa78 (commit)
  discards  a77c6904227149fdea29b8ed27ac3dddf2e9cf47 (commit)
  discards  b13a67d78b35170fed891413569b203c8b63c8a5 (commit)
  discards  48abb3007302f2dab329e301bbabcb75782b3a4a (commit)
  discards  af5d51b0f6494f4b222880abef35cc696e52a013 (commit)
  discards  1fdcbf613508ededda08689db16c2785de614d96 (commit)
  discards  7c83c0e1c4670ba2c70b3093b4135f4a8ba31ed1 (commit)
  discards  c3c9c3277519a1ed70f3bf4d3d5ae5872cd24c7a (commit)
  discards  771f20df7863d8d805000c19db24ed1e6754e16d (commit)
  discards  e2e60dac618a28a80786d9d29597852e5fab52b5 (commit)
  discards  2159e39353aa5b30b8844921e5987df83b696630 (commit)
  discards  c33fcd7a33d13c5cf15d0b378dc46884e7c01e59 (commit)
  discards  fd0134c8622110678b524ebeb6d889b68d0639c7 (commit)
  discards  3a6aa7f3b7e74bd7ed9089bddb81bd91cd948d81 (commit)
  discards  d3e7137cc2168ec99e4de7a31f481cce6a76da09 (commit)
  discards  74cd902cb970e64f25456f81b1c7899b15b53c4e (commit)
  discards  6a7288c1d6779bf83f005f19f12e446ca17e24b6 (commit)
       via  64d0007aca0bc5f3c53aa3e710d9db6b745af36e (commit)
       via  f0974ae238ee109b046051441a9599076ddcc0f0 (commit)
       via  abf773d673a29b3e927e8151f9d47458bfeb9153 (commit)
       via  3586c3491555ef7e686bcbc145a64ac2d9405710 (commit)
       via  6551d3169355de7e05d94c905014ff1ec9fbf264 (commit)
       via  145d7ddb84ef36122a0bf6a539923b558d0b3928 (commit)
       via  5d2546d1597645a88ef70bcbdc706c9e8a16fb9a (commit)
       via  3ea2200407b17dac2b704979bd92cb7563d51d15 (commit)
       via  1da5e744766689acca7cf5b477a0c9bb4b900c95 (commit)
       via  ea5906244efac709eb1fbcbd021a285fc4cbe7b8 (commit)
       via  82b3b65f217a0432ab88909f9fb29d7aa1856541 (commit)
       via  5b88584b0604269b1a830f8dfe1fec7319f3cdc7 (commit)
       via  4c3cf3701c217c99cdaee39d28ee0acea2e1c374 (commit)
       via  6cc8140d2b32abf620d6b47cd691162728029122 (commit)
       via  018d3829388438da9a7734bcd6972077c8c35090 (commit)
       via  d8abd5df576d0180b2f78009fbe5707094e8aef1 (commit)
       via  b92861ddbc4a559a6e4c82b875a2b1406f6d6279 (commit)
       via  4ffbfd3e4ebf5bfb5fa4ee298d88a6b721525420 (commit)
       via  1d89966270e03838fa61a15a874ceb8be34c3de3 (commit)
       via  98ed3fd8089b26895ef0177f1c1f3567cc3d0871 (commit)
       via  7c998bd273986d5cbd38b93523e8bfd59ec6fbae (commit)
       via  bf6f8f6186f4cbd757e8c72083214493a95bc60d (commit)
       via  e08b0de9fe2d7c0ad60a7eb6b4bdc0af2d62177c (commit)
       via  0b9733bfb3d15544b867210f138367bcf7d25cf5 (commit)
       via  a4c1213fd7b7da144b011a7d74d28f5ba670df1a (commit)
       via  46d6fcc2cc2cbf6ad034a20fc6461f5016aa0bca (commit)
       via  666d4c7f68e56bc380f1028509c60cbec5eebe87 (commit)
       via  7275bd4d224ec91b50c0938d31b602a3c19a9a18 (commit)
       via  13266266b9a8c6518b7153b8fdea4aa2c0328ffe (commit)
       via  83c3f0be610073a5fe001ba39e45d4ac1f925e4c (commit)
       via  3ee9213a6d3131da085f97929051a0b0d3e65d0c (commit)
       via  49bf767559c5beb4d7de145fdbabe02a9cae577f (commit)
       via  dfb5722cf088eca6bc3bcbb74cdf5704b5638568 (commit)
       via  418f6de75d1e54405e4e5501f50b1c4b2146551c (commit)
       via  18e441b66e0a974028f4042141b1a801fb5f21e0 (commit)
       via  71cd89a3e591951e7efd4b7065acfa9b15177ac5 (commit)
       via  91740a67cb090f0507b8c364f8f3950b73a52beb (commit)
       via  d390bc452c0dca5c2e455ccdcda5515204a11413 (commit)
       via  a6cb44ea20e173b50c8197f44f588411c25c1c31 (commit)
       via  33245e2c62baf0569edc53b4d39c3344a0372eba (commit)
       via  2892ebf3ea5ae575c3760fd2fca95b331f9ba9d2 (commit)

This update added new revisions after undoing existing revisions.  That is
to say, the old revision is not a strict subset of the new revision.  This
situation occurs when you --force push a change and generate a repository
containing something like this:

 * -- * -- B -- O -- O -- O (89b5af1f8d450e039602030d06ea6b0bd62925a2)
             N -- N -- N (64d0007aca0bc5f3c53aa3e710d9db6b745af36e)

When this happens we assume that you've already had alert emails for all
of the O revisions, and so we here report only the revisions in the N
branch from the common base, B.

Those revisions listed above that are new to this repository have
not appeared on any other notification email; so we list those
revisions in full, below.

- Log -----------------------------------------------------------------

commit 64d0007aca0bc5f3c53aa3e710d9db6b745af36e
Author: Jacob Erlbeck <jerlbeck at>
Date:   Tue Sep 30 13:53:26 2014 +0200

    sgsn: Added TODOs (TODO)
      - fix and remove TODOs
      - add test cases for Detach(reattach)
      - add test cases for PDP context deletion (OW#1311)
      - add test cases for Cancel pending timer in sgsn_mm_ctx_free
      - implement age based MM context expiry
      - remove this commit
    Sponsored-by: On-Waves ehf

commit f0974ae238ee109b046051441a9599076ddcc0f0
Author: Jacob Erlbeck <jerlbeck at>
Date:   Fri Oct 17 10:08:02 2014 +0200

    sgsn: Do not die in _bssgp_tx_dl_ud on TLLI mismatch (TODO)
    Currently an OSMO_ASSERT fails, if the message's and the mmctx's
    TLLIs do not match.
    This commit turns the assertion into an ERROR log message and uses
    the default values for IMSI, DRX, and RA CAP instead of the MM
    context values in this case.
      - find the real source for the problem
      - add test cases derived from what is being described in the ticket
    Ticket: OW#1322
    Sponsored-by: On-Waves ehf

commit abf773d673a29b3e927e8151f9d47458bfeb9153
Author: Jacob Erlbeck <jerlbeck at>
Date:   Wed Dec 17 14:03:35 2014 +0100

    sgsn: Add PDP info to subscriber data (TODO)
      - commit message
      - raw binary representation ???
      - tests
    Sponsored-by: On-Waves ehf

commit 3586c3491555ef7e686bcbc145a64ac2d9405710
Author: Jacob Erlbeck <jerlbeck at>
Date:   Fri Dec 19 19:11:20 2014 +0100

    sgsn: Add logging to connect_timer_cb (TODO)
      - LOGLevel ?
    Sponsored-by: On-Waves ehf

commit 6551d3169355de7e05d94c905014ff1ec9fbf264
Author: Jacob Erlbeck <jerlbeck at>
Date:   Thu Jan 8 15:29:01 2015 +0100

    gprs: Support the full cancellation procedure (TODO)
      - Commit message
      - Reply to cancellation request (either with RES or ERR)
    Ticket: OW#1338
    Sponsored-by: On-Waves ehf

commit 145d7ddb84ef36122a0bf6a539923b558d0b3928
Author: Jacob Erlbeck <jerlbeck at>
Date:   Thu Jan 8 16:23:25 2015 +0100

    gprs: Block other GSUP procedures during PURGE_MS (TODO)
      - commit message
      - move LOGGSUBSCRP fix in gprs_subscr_purge to HEAD^^
      - testing
    Ticket: OW#????
    Sponsored-by: On-Waves ehf

commit 5d2546d1597645a88ef70bcbdc706c9e8a16fb9a
Author: Jacob Erlbeck <jerlbeck at>
Date:   Thu Jan 8 15:18:39 2015 +0100

    gprs: Retry PURGE_MS procedure after timeout
    Currently, when the PURGE_MS_REQ to the HLR gets lost (e.g. by a
    connection or peer failure), the expired subscriber entry will not get
    This commit adds a retry mechanism then restarts the procedure after
    a timeout (currently 10s). The maximum number of retries is limited
    (currently to 3 PURGE_MS messages). If none of these procedures is
    completed (either with success or error), the subscriber data is
    Sponsored-by: On-Waves ehf

commit 3ea2200407b17dac2b704979bd92cb7563d51d15
Author: Jacob Erlbeck <jerlbeck at>
Date:   Tue Jan 6 16:32:41 2015 +0100

    gprs: Support PURGE MS messages
    When a subscriber entry is going to be deleted by SGSN and when the
    subscriber info has been optained from a remote peer via GSUP, the
    peer should be informed before the entry is really deleted. For this
    purpose, MAP defines the PURGE MS procedure (see GSM 09.02, 19.1.4).
    This patch adds support for the PURGE_MS_REQ/_ERR/_RES messages and
    invokes the procedure when the subscriber entry is going to be
    removed. This only applies if GSUP is being used, the Update
    Location procedure has been completed successfully, and the
    subscriber has not been cancelled. The removal of the entry is
    delayed until a PURGE_MS_RES or PURGE_MS_ERR message is received.
    Ticket: OW#1338
    Sponsored-by: On-Waves ehf

commit 1da5e744766689acca7cf5b477a0c9bb4b900c95
Author: Jacob Erlbeck <jerlbeck at>
Date:   Fri Jan 9 09:46:43 2015 +0100

    sgsn/doc: Add message definitions for PURGE_MS responses
    Currently the definitions of the message PURGE_MS_RES and
    PURGE_MS_ERR are missing in the specification.
    This patch adds those definitions.
    Sponsored-by: On-Waves ehf

commit ea5906244efac709eb1fbcbd021a285fc4cbe7b8
Author: Jacob Erlbeck <jerlbeck at>
Date:   Tue Jan 13 11:46:32 2015 +0100

    sgsn/test: Add checks and reports to detect subscr leakage
    This commit adds talloc reports to log remaining NULL chunks after the
    terminates. It also adds explicit checks for empty subscriber lists.
    Sponsored-by: On-Waves ehf

commit 82b3b65f217a0432ab88909f9fb29d7aa1856541
Author: Jacob Erlbeck <jerlbeck at>
Date:   Fri Jan 9 15:07:16 2015 +0100

    sgsn/test: Add test that intercepts gprs_gsup_client_send
    This test replaces gprs_gsup_client_send by a custom function, that
    emulates a GSUP remote peer by calling gprs_subscr_rx_gsup_message
    with responses for all requests. It then executes a full
    Attach/Detach cycle.
    Sponsored-by: On-Waves ehf

commit 5b88584b0604269b1a830f8dfe1fec7319f3cdc7
Author: Jacob Erlbeck <jerlbeck at>
Date:   Tue Jan 6 16:32:41 2015 +0100

    gprs: Add expiry timeout for subscriber entries
    Set the expiry delay after the subscriber has been deleted (e.g. by
    freeing the MM context). If cancelled, the subscriber will be deleted
    immediately and no timeout will be set. If the expiry time is set to
    SGSN_TIMEOUT_NEVER, no timer will be started and the subscriber entry
    will be kept until it is cancelled.
    The following VTY command is added to the sgsn node:
      - subscriber-expiry-time <0-999999>    set expiry time in seconds
      - no subscriber-expiry-time            set to SGSN_TIMEOUT_NEVER
    The default is an expiry time of 0 seconds, which means that the
    subscriber entries are wiped out immediately after an MM context is
    Note that unused MM contexts are not expired yet. Therefore the
    subscriber will only be expired after a successful MM detach.
    Sponsored-by: On-Waves ehf

commit 4c3cf3701c217c99cdaee39d28ee0acea2e1c374
Author: Jacob Erlbeck <jerlbeck at>
Date:   Fri Jan 9 11:59:50 2015 +0100

    sgsn/test: Refactor subscriber test
    The code sequence that checks, whether a subscriber is still
    reachable for a given IMSI, is repeated several times.
    This patch puts this code sequence into a single function and adds a
    check for the IMSI after the entry has been found. In addition,
    some comments are extended.
    Sponsored-by: On-Waves ehf

commit 6cc8140d2b32abf620d6b47cd691162728029122
Author: Jacob Erlbeck <jerlbeck at>
Date:   Thu Jan 8 14:13:46 2015 +0100

    sgsn/test: Fix subscriber cleanup
    Currently the subscribers are not really deleted by
    cleanup_subscr_by_imsi, but kept in RAM instead.
    This patch fixes this and adds a test to verify, that the subscriber
    is really deleted afterwards.
    Sponsored-by: On-Waves ehf

commit 018d3829388438da9a7734bcd6972077c8c35090
Author: Jacob Erlbeck <jerlbeck at>
Date:   Thu Jan 8 14:08:16 2015 +0100

    gprs/test: Move subscr cleanup code into a separate function
    Refactor several occurences of the same subscriber cleanup code into
    a seperate cleanup_subscr_by_imsi function.
    Sponsored-by: On-Waves ehf

commit d8abd5df576d0180b2f78009fbe5707094e8aef1
Author: Jacob Erlbeck <jerlbeck at>
Date:   Mon Jan 5 16:20:47 2015 +0100

    gprs: Pass GMM causes related to the MSC connection
    Currently the error causes MSC_TEMP_NOTREACH, NET_FAIL, and
    CONGESTION are silently dropped to force the MS to continue. On the
    other hand, GSM 04.08/24.008, in combination with,
    require the MS to retry the attachment procedure for cause codes
    above 15 instead of disabling GPRS. All of the mentioned GMM causes
    have codes above 15, so using a REJECT message including the cause
    code is a better choice. This way, the retry algorithm based on T3311
    (15s, 5 times) and T3302 (default 12min) could be used.
    This patch modifies gprs_subscr_handle_gsup_auth_err and
    gprs_subscr_handle_gsup_upd_loc_err to proceed like when the access
    has beed denied, except that the corresponding subscriber's
    information fields are not cleared.
    This has been successfully tested which an iphone which enters a
    retry loop as it is being described in the specification.
    Sponsored-by: On-Waves ehf

commit b92861ddbc4a559a6e4c82b875a2b1406f6d6279
Author: Jacob Erlbeck <jerlbeck at>
Date:   Tue Dec 23 14:24:16 2014 +0100

    gprs: Use LOGGSUBSCRP and LOGMMCTX for logging in gprs_susbcriber.c
    To unify the layout of the logging messages in gprs_subscriber.c,
    this patch replaces each LOGP by LOGGSUBSCRP, unless a non-NULL
    pointer to a subscr is not available. In those cases, it uses
    LOGMMCTXP if a pointer to an MM context is available or LOGP
    Sponsored-by: On-Waves ehf

commit 4ffbfd3e4ebf5bfb5fa4ee298d88a6b721525420
Author: Jacob Erlbeck <jerlbeck at>
Date:   Thu Jan 8 12:02:12 2015 +0100

    gprs: Add LOGGSUBSCRP macro to log subscriber info
    This patch adds a new logging macro, that logs to DGPRS and provides
    a uniform prefix containing the IMSI without using the mm reference.
    This is an improvement over using LOGMMCTXP, since the new macro also
    provides an IMSI if no MM context is attached.
    Sponsored-by: On-Waves ehf

commit 1d89966270e03838fa61a15a874ceb8be34c3de3
Author: Jacob Erlbeck <jerlbeck at>
Date:   Mon Jan 5 18:57:32 2015 +0100

    gprs: Let GSUP parser functions return GMM causes on errors
    Currently the GSUP message handling function in gprs_subscriber.c and
    the functions in gprs_gsup_messages.c are not consistent with respect
    to the return codes if an error happens. Albeit all error return
    codes are negative, the semantics of the absolute value are not
    clearly defined. In addition, some return codes are not passed to the
    calling function.
    This path changes these functions to always return a negated GMM
    cause value in case of errors. Return values of called parser
    functions are not longer ignored.
    Sponsored-by: On-Waves ehf

commit 98ed3fd8089b26895ef0177f1c1f3567cc3d0871
Author: Jacob Erlbeck <jerlbeck at>
Date:   Mon Jan 5 18:38:41 2015 +0100

    gprs: Use the cause value in  GSUP error messages
    Currently always a cause with the meaning of 'access denied' is
    assumed. gprs_subscr_handle_gsup_auth_err just clears the auth
    triplets and the authorized flag before calling the update function.
    gprs_subscr_handle_gsup_upd_loc_err only clears the authorized flag
    and calls the update function. This means, that an MS will not retry
    to attach even on temporary network errors.
    This patch changes these functions to use the GSUP error cause value
    to decide, whether to clear the corresponding subscriber fields, to
    just continue with the corresponding update function, or to log,
    ignore and not pass the cause to the MS in case the error is directly
    related to the GSUP protocol. The subscriber's error_cause field is
    updated, if the update function is going to be called. The
    error_cause fielt is reset on non-error GSUP messages.
    Sponsored-by: On-Waves ehf

commit 7c998bd273986d5cbd38b93523e8bfd59ec6fbae
Author: Jacob Erlbeck <jerlbeck at>
Date:   Mon Jan 5 17:51:17 2015 +0100

    sgsn: Pass subscriber error causes to the GMM layer
    This patch extends gsm0408_gprs_access_denied and
    gsm0408_gprs_access_cancelled to accept GMM cause codes. These are
    then passed to the MS, unless gsm0408_gprs_access_cancelled is called
    with cause 0 (no error -> updateProcedure).
    Since gsm0408_gprs_access_denied uses GMM_CAUSE_GPRS_NOTALLOWED if
    the cause is not set, and the subscriber's error_cause is never set
    (and thus always 0), the SGSN's behaviour does not change with this
    Sponsored-by: On-Waves ehf

commit bf6f8f6186f4cbd757e8c72083214493a95bc60d
Author: Jacob Erlbeck <jerlbeck at>
Date:   Mon Jan 5 17:30:02 2015 +0100

    gprs: Move protocol value_strings to gsm_04_08_gprs.c
    Currently the mapping between GSM 04.08 (GPRS) protocol specific
    numbers and their textual description was put into gprs_gmm.c and not
    This commit moves the mappings to a new file gsm_04_08_gprs.c,
    renames some of them, and exports them via gsm_04_08_gprs.h.
    The following identifiers are renamed to match the corresponding type
      - gmm_cause_names -> gsm48_gmm_cause_names
      - gsm_cause_names -> gsm48_gsm_cause_names
    Sponsored-by: On-Waves ehf

commit e08b0de9fe2d7c0ad60a7eb6b4bdc0af2d62177c
Author: Jacob Erlbeck <jerlbeck at>
Date:   Fri Dec 19 19:19:46 2014 +0100

    sgsn: Show GSUP client info on 'show sgsn'
    This commit adds a line to the output of 'show sgsn' if the GSUP
    client has been initialized:
      - Remote authorization: [not] connected to HOST:PORT via GSUP
    Sponsored-by: On-Waves ehf

commit 0b9733bfb3d15544b867210f138367bcf7d25cf5
Author: Jacob Erlbeck <jerlbeck at>
Date:   Fri Dec 19 19:18:54 2014 +0100

    gprs: Send PING and eventually reconnect
    Currently, the reconnect mechanism relies on gsup_client_updown_cb
    which in turn gets called based on the OS' view of connection state.
    This patch adds a timer based PING mechanism that regularly sends
    PING messages and forces a reconnect if a PONG message won't be
    received until the next PING message is scheduled. The current ping
    interval is 20s.
    Sponsored-by: On-Waves ehf

commit a4c1213fd7b7da144b011a7d74d28f5ba670df1a
Author: Jacob Erlbeck <jerlbeck at>
Date:   Fri Dec 19 19:15:55 2014 +0100

    gprs: Handle incoming IPA CCM message in gsup_client_read_cb
    Currently the IPA CCM messages are not handled by the GSUP client.
    This means, that the client doesn't answer to PING and ID_GET and
    logs notices when receiving PONG or ID_ACK. At least the PING
    functionality (remotely originated PING) shall be supported.
    This patch extends gsup_client_read_cb by a call to
    ipaccess_bts_handle_ccm. Only when the return code is 0, the message
    is processed further and checked for an OSMO/GSUP message. ID_GET
    messages are answered by a dummy description, where only the
    unit_name is set to 'SGSN'.
    Sponsored-by: On-Waves ehf

commit 46d6fcc2cc2cbf6ad034a20fc6461f5016aa0bca
Author: Jacob Erlbeck <jerlbeck at>
Date:   Fri Dec 19 19:00:56 2014 +0100

    gprs: Handle return code of ipa_client_conn_open correctly
    The ipa_client_conn_open function does not distinguish between a
    connection being already established or waiting for establishment.
    In either case, the application gets informed about the connection
    state via the updown_cb. The 'up' parameter is only set, if
    poll/select consider the socket as writable.
    This patch handles both cases equally and fully relies on the
    updown_cb to adjust the gsupc obejct state.
    Sponsored-by: On-Waves ehf

commit 666d4c7f68e56bc380f1028509c60cbec5eebe87
Author: Jacob Erlbeck <jerlbeck at>
Date:   Fri Dec 19 18:50:05 2014 +0100

    gprs: Avoid sending stale GSUP requests after reconnect
    Currently, messages are added to the tx queue even if the connection
    is down for some reason and all of these messages are eventually sent
    after a re-connect.  The MS has probably sent several Attach Requests
    while the connection was down and will continue doing so. Therefore
    these stored messages could be dropped.
    This patch clears the queue before re-connecting and also extends
    gprs_gsup_client_send to return immediately, when the connection is
    not established instead of calling ipa_client_conn_send.
    Sponsored-by: On-Waves ehf

commit 7275bd4d224ec91b50c0938d31b602a3c19a9a18
Author: Jacob Erlbeck <jerlbeck at>
Date:   Thu Dec 18 15:00:29 2014 +0100

    gprs: Add automatic re-connect if the GSUP connection is down
    Currently the GSUP connection to a server is not restarted if the
    connection cannot be established or is terminated during operation.
    This commit adds a timer based connection mechanism, basically
    consisting of a timer callback that calls gsup_client_connect. The
    timer is eventually triggered (up == 0) or cleared (up != 0) by
    gsup_client_updown_cb. It adds calls to osmo_timer_del() to
    gsup_client_connect and gprs_gsup_client_destroy. The latter is now
    called instead of talloc_free in gprs_gsup_client_create on error to
    be on the safe side.
    Sponsored-by: On-Waves ehf

commit 13266266b9a8c6518b7153b8fdea4aa2c0328ffe
Author: Jacob Erlbeck <jerlbeck at>
Date:   Fri Dec 19 18:30:41 2014 +0100

    sgsn/test: Extend tests to simulate lost GSUP requests
    This patch extends test_gmm_attach to optionally resend Attach
    Requests until the SGSN responds to it instead of calling
    OSMO_ASSERT at a few places. The test_gmm_attach_subscr_gsup_auth
    test optionally uses this feature. It is called once in either mode.
    Sponsored-by: On-Waves ehf

commit 83c3f0be610073a5fe001ba39e45d4ac1f925e4c
Author: Jacob Erlbeck <jerlbeck at>
Date:   Fri Dec 19 18:26:09 2014 +0100

    sgsn/test: Add GMM test for a GSUP based attach procedure
    This test calls test_gmm_attach() where the mocked subscriber
    functions insert GSUP messages instead of manipulating the
    subscriber structure directly.
    Sponsored-by: On-Waves ehf

commit 3ee9213a6d3131da085f97929051a0b0d3e65d0c
Author: Jacob Erlbeck <jerlbeck at>
Date:   Fri Dec 19 18:19:50 2014 +0100

    sgsn/test: Add wrapper for gprs_subscr_rx_gsup_message
    Currently, several lines of boiler plate are needed around a call to
    This patch puts all of this (including the call to
    gprs_subscr_rx_gsup_message) into a separate function.
    Sponsored-by: On-Waves ehf

commit 49bf767559c5beb4d7de145fdbabe02a9cae577f
Author: Jacob Erlbeck <jerlbeck at>
Date:   Fri Dec 19 18:08:48 2014 +0100

    sgsn: Add global require_update_location flag
    This flag is used to determine, whether the Update Location procedure
    shall be invoked. This is currently only set, when the 'remote'
    authorization policy is set. When the flag is set, sgsn_auth_update
    will not never be called directly by sgsn_auth_request, if an Attach
    Request procedure is pending, even if the remote connection fails for
    some reason.
    Sponsored-by: On-Waves ehf

commit dfb5722cf088eca6bc3bcbb74cdf5704b5638568
Author: Jacob Erlbeck <jerlbeck at>
Date:   Wed Dec 17 13:20:08 2014 +0100

    sgsn: Replace subscr.authenticate by global require_authentication flag
    Currently the flag 'authenticate' is managed per subscriber.
    This patch replaces that flag by a global cfg.require_authentication
    flag that enables/disables the use of the Auth & Ciph procedure for
    every subscriber. The flag is set by the VTY, if and only if the
    authorization policy is 'remote'.
    The VTY command
      - update-subscriber imsi IMSI insert authenticate <0-1>
    is removed.
    Sponsored-by: On-Waves ehf

commit 418f6de75d1e54405e4e5501f50b1c4b2146551c
Author: Jacob Erlbeck <jerlbeck at>
Date:   Thu Dec 18 12:46:47 2014 +0100

    sgsn: Integrate the GSUP client into the SGSN
    This commit adds GSUP client configuration (via VTY), connection set
    up, and real message sending.
    The following configuration commands are added:
     - gsup remote-ip A.B.C.D            set server IP address
     - gsup remote-port PORT             set server TCP port
    Ticket: OW#1338
    Sponsored-by: On-Waves ehf

commit 18e441b66e0a974028f4042141b1a801fb5f21e0
Author: Jacob Erlbeck <jerlbeck at>
Date:   Thu Dec 18 12:28:21 2014 +0100

    gprs: Add GSUP client
    This commit adds the client code to get subscriber information from a
    remote server. It provides an IPA over TCP connection to transmit and
    receive GSUP messages.
    Sponsored-by: On-Waves ehf

commit 71cd89a3e591951e7efd4b7065acfa9b15177ac5
Author: Jacob Erlbeck <jerlbeck at>
Date:   Mon Jan 12 13:54:39 2015 +0100

    gprs: Clear GSUP message structures before decoding
    Currently the message structure is not cleared before the message is
    parsed which can cause information leaking between messages if the
    same gprs_gsup_message object is used. Especially list elements (auth
    tuples and pdp info) are not replaced by an IE, but the IE is
    This patch uses the assignment operator to clear gprs_gsup_message,
    gsm_auth_tuple, and gprs_gsup_pdp_info before using them. This also
    replaces the use of memcpy of the latter.
    Sponsored-by: On-Waves ehf

commit 91740a67cb090f0507b8c364f8f3950b73a52beb
Author: Jacob Erlbeck <jerlbeck at>
Date:   Mon Jan 12 13:39:15 2015 +0100

    gprs: Always reset auth tuples/pdp infos in gprs_gsup_decode
    Currently auth tuples are always appended to the lists when
    gprs_gsup_decode is called with a gsup_msg structure. This makes a
    test case fail, where the same gsup_msg is used again and again
    without clearing it after each use.
    Ticket: OW#????
    Sponsored-by: On-Waves ehf

commit d390bc452c0dca5c2e455ccdcda5515204a11413
Author: Jacob Erlbeck <jerlbeck at>
Date:   Mon Jan 12 13:23:05 2015 +0100

    gprs: Handle empty GSUP messages correctly
    Currently, the gprs_gsup_decode function doesn't check the return
    value of gprs_shift_v_fixed before using the value pointer. The
    function fails, if the GSUP message length (not including IPA
    headers) is 0. In this case, a segfault can happen, depending on the
    value of the uninitialized 'value' pointer. The test case doesn't
    trigger a segfault, but valgrind complains about reading
    uninitialized data.
    This patch adds a check for the return value that would return with
    an error code if the shift function failed.
    Sponsored-by: On-Waves ehf

commit a6cb44ea20e173b50c8197f44f588411c25c1c31
Author: Jacob Erlbeck <jerlbeck at>
Date:   Mon Jan 12 11:14:18 2015 +0100

    gprs/test: Add tests for invalid GSUP messages
    Currently, the tests only use valid GSUP messages. This does not
    check the robustness of the parser.
    This commit adds a test for truncated messages and another for
    modified messages.
    Thanks to Holger for the basic truncation test.
    Sponsored-by: On-Waves ehf

commit 33245e2c62baf0569edc53b4d39c3344a0372eba
Author: Jacob Erlbeck <jerlbeck at>
Date:   Tue Jan 13 11:56:28 2015 +0100

    sgsn/test: Fix msgb freeing in mocked bssgp_tx_dl_ud
    Currently the mocked bssgp_tx_dl_ud function used for testing in
    sgsn_test.c does not free the msg as it is done by the original
    function. This leads to a msgb leak in the test.
    This patch fixes the test code accordingly and uncomments the
    assertion that checks for left-over msgbs.
    Sponsored-by: On-Waves ehf

commit 2892ebf3ea5ae575c3760fd2fca95b331f9ba9d2
Author: Jacob Erlbeck <jerlbeck at>
Date:   Tue Jan 13 11:46:32 2015 +0100

    sgsn/test: Add checks and reports to detect msgb leakage
    This commit adds talloc reports to log remaining chunks after the
    testing has been finished. It also adds a (currently disabled) check,
    that no msgbs are allocated after running the tests.
    Sponsored-by: On-Waves ehf


Summary of changes:
 openbsc/src/gprs/gprs_gsup_client.c | 2 +-
 openbsc/tests/sgsn/sgsn_test.c      | 3 +--
 2 files changed, 2 insertions(+), 3 deletions(-)

The OpenBSC GSM Base Station Controller (+MSC/HLR/SGSN)

More information about the osmocom-commitlog mailing list