AgeCommit message (Collapse)AuthorFilesLines
2018-06-03Re-introduce support for IPA-encapsulated MGCPlaforge/sccpliteHarald Welte7-20/+161
Old osmo-bsc-sccplite already supported this, but in the migration over to libosmo-sigtran and to real 3GPP AoIP, this functionality got lost. We now crate a UDP proxy socket. Any MGCP commands received via IPA from MSC (or rather: bsc_nat) are retransmitted to the MGW via UDP on this socket. Any responses back from the MGW received on the UDP socket are retransmitted back to MSC/bsc_nat as MGCP inside the IPA multiplex. Closes: OS#2536 Change-Id: I38ad8fa645c08900e0e1f1b4b96136bc6d96b3ab
2018-06-03LCLS: add VTY config to enable/disable LCLS on per-MSC basisHarald Welte3-0/+34
The user might not want to enable LCLS support for administrative reasons. So let's keep it disabled by default, until somebody explicitly enables it with "lcls-mode mgw-loop". In the future, we may want to introduce a "lcls-mode bts-loop" where we don't loop at the BSC-colocated MGW, but where we instruct the BTSs to feed the RTP directly to each other. This would require a falt/transparent IP routing between the BTSs in the RAN. Change-Id: Ied7985056c8cd182bf16119007a08cc5be14459b Related: OS#1602
2018-06-02Add initial 3GPP LCLS support to OsmoBSCHarald Welte11-1/+981
This code contains the following code: * receive/parse/interpret LCLS specific BSSMAP IEs and PDUs * osmo_fsm handling the various states and their transitions * call leg correlation (finding the other subscr_conn with same GCR) * communication between the two call-leg LCLS FSMs * detection of supported / unsupported LCLS configurations * display of GCR / LCLS information in "show conns" * switch the media streams locally using MDCX to the MGW Closes: OS#1602 Change-Id: I614fade62834def5cafc94c4d2578cd747a3f9f7
2018-06-02VTY: Print some more information in "show conns"Harald Welte1-2/+13
We now print information such as * SCCP connection ID * MSC number * handover decision2 fail count * channel mode (SIGN/SPEECH) * MGW endpoint * secondary lchan (if any, e.g. during assignment) * don't crash if conn->lchan == NULL Change-Id: I2f8000844afc9da93ca39976399e5f76a45530de
2018-05-30abis_rsl: rsl_rx_chan_rqd: Format bts log string as in everywhere elsePau Espin Pedrol1-3/+3
Change-Id: I7928d78e2ce3e6eca545455d08bc89ca67e01f9f
2018-05-30abis_rsl.c: Fix whitespacePau Espin Pedrol1-8/+8
Change-Id: I67659275beda1b6a89bdf11eb0231438b0c8b912
2018-05-30chan_alloc: Print bts nr on chan alloc failurePau Espin Pedrol1-3/+3
Also change format of related debug message to match similar format than gsm_ts_and_pchan_name(). Change-Id: Id4f875944c40fd862fc63feec82216c45d2e41c2
2018-05-28doc: add msc charts on Assignment/Handover internalsNeels Hofmeyr3-0/+357
In doc/, add two message sequence charts with lots of implementation specifics, to clarify the current code state. Mark various problems in red notes. This chart and others should help to illustrate future code changes as I go along refactoring handover, lchan allocation and adding inter-BSC handover. Change-Id: I20999e938441d4fed2d37462b262b74a696f616d
2018-05-28acc_ramp: Increase log level of some messagesPau Espin Pedrol1-4/+4
Right now, it's impossible to see any ACC Ramping information unless RSL category is set to DEBUG. Barring and Allowing Access Control Class is an important event which should be printed in most cases. Increase log levels of messages printed during some error conditions to be handled as errors. Change-Id: I5404e91fe7c0b3209e8033594e659387bbc924d6
2018-05-27move 'extern struct gsm_network *bsc_gsmnet" to header fileHarald Welte8-20/+8
It's not a good idea to keep extern declarations copied over half a dozen C files. Let's move it to a header. Change-Id: I6f643f1393ba0955d9c0cf1cf78d5c604e7b9451
2018-05-27Remove 'struct bsc_msc_connection' + fix IPA-encapsulated CTRLHarald Welte16-506/+182
The bsc_msc_connection dates back to the old pre-libosmo-sigtran days, and 90% of the field members weren't used at all (even the new sigtran specific ones!). Let's merge what remains into struct bsc_msc_data. As a side effect, the already dysfunctional "dest A.B.C.D" VTY command has been removed from the MSC node. There's quite a bit of fall-out in the CTRL interface, which was the code with strongest ties to bsc_msc_connection. This was resolved by properly porting CTRL handling over to libosmo-sigtran, meaning that an IPA/SCCPlite connected MSC can now again send CTRL GET/SET commands, and can also receive those selective few TRAPs that old osmo-bsc-sccplite also sent to its MSC[s]. Change-Id: I6b7354f3b23a26bb4eab12213ca3d3b614c8154f Related: OS#2012
2018-05-27remove remaining bits of osmo-bsc_natHarald Welte39-10806/+2
osmo-bsc_nat is too heavily tied into legacy SCCPlite code, as it is not using libosmo-sigtran/osmo_ss7 so far. It's also full of customer-specific code and it's shared use of some libbsc code here has been complicating osmo-bsc development. The current plan is to continue to use osmo-bsc_nat from openbsc.git for those legacy users that need it, and not use osmo-bsc_nat in new 3GPP AoIP setups. Should we ever get a strong demand for an AoIP based bsc_nat, we can still revisit this later. Change-Id: Ia05dc76336a64a7f08962843b9a7cc19f2c83387
2018-05-25bsc: Don't create MSC-side MGCP connection in IPA/SCCPlite caseHarald Welte1-9/+19
In IPA/SCCPlite, the MSC is handling MGCP to the BSC-MGW directly and the BSC only takes care of the BTS-side MGCP connection. We achieve this by transitioning directly from WAIT_MDCX_BTS into ACTIVE in this case. Change-Id: I96179b4324b976bded36023a8ccbdc007b6b3e05 Related: OS#2544
2018-05-25bsc: Don't include AoIP IEs in ASSIGNMENT COMPLETE over SCCPliteHarald Welte1-6/+12
Change-Id: I6ace15f368b5ba6e5c8845ea21d17dd3a7a15be7
2018-05-25bsc: Don't reject ASSIGNMENT for Audio in IPA/SCCPlite caseHarald Welte1-13/+1
Change-Id: If362a0084de452727cd063063dfb645eca2f9beb
2018-05-25bsc: Use correct MGCP endpoint name for IPA/SCCPliteHarald Welte1-1/+20
In IPA/SCCPlite, we have to use the CIC to construct the MGCP endpoint name instead of the usual dynamic endpoint allocation. Change-Id: I03e2cdbc8e40169e52df3720c40b66734e880525
2018-05-25bsc: Add mgcp_port_to_cic() to determine CIC from RTP PortHarald Welte2-13/+22
Also: Move mgcp_timeslot_to_port() next to it, as they are more or less the inverse transformation of each other. Change-Id: Ica908e2bb8fc4e59e0d146b428c93a9efc385688
2018-05-25vty: Permit selection of other ASP protocol than M3UAHarald Welte4-1/+21
We used to have hard-coded M3UA. Let's allow the user to configure this per MSC using a new "asp-protocol (m3ua|sua|ipa)" VTY command. For SUA this should just work 1:1 without any trouble. For IPA, this of course only changes the underlying transport without reflecting the various differences in terms of BSSMAP ASSIGNMENT, MGCP handling, etc. Change-Id: I0800c709e574cedd7f5dd98be81c78782245cd13 Related: OS#2544
2018-05-25bsc: Fix check for MSC-side FSM allocation failureHarald Welte1-1/+1
In gscon_fsm_wait_mdcx_bts() we try to allocate conn->user_plane.fi_msc but then check whether conn->user_plane.fi_bts is set, possibly due to a copy+paste error. Let's fix that. Change-Id: I1f515910f67492257866791588f32b350fadf815
2018-05-24fix misaligned memory write access in abis_nm_ipaccess_rsl_connect()Stefan Sperling1-1/+1
Use memcpy() to avoid unaligned access, instead of writing through a pointer cast to uint32_t. Problem spotted by address sanitizer: abis_nm.c:2802:24: runtime error: store to misaligned address 0x7ffc95396706 for type 'uint32_t', which requires 4 byte alignment 0x7ffc95396706: note: pointer points here 81 0b bb 80 00 00 00 00 ed 79 28 56 00 00 e0 9c 00 00 a0 61 00 00 ... ^ Related: OS#3196 Change-Id: I8e591a56ae522b371da01ea968151a7e6fa24bb9
2018-05-23gscon: remove dead codePhilipp Maier2-45/+1
The state ST_WAIT_MODE_MODIFY_ACK can never be reached by the current FSM model. - Remove ST_WAIT_MODE_MODIFY_ACK and all related code Change-Id: Iacaae2ee50ca1956066b7dce4517bbc9c2b0897e Related: OS#2762
2018-05-23cosmetic: logging and ordering in handle_ass_compl()Neels Hofmeyr1-18/+20
* Check the message length once at the start, before any other actions. * Use only one local gsm48_hdr pointer. * Read the cause value once near the top, re-use it. * Log "ASSIGNMENT COMPLETE" always, not only during handover. * Fully initialize local struct lchan_signal_data. Change-Id: Idcfd932d3dfb0b621ed6d8c4f92c0231abcdcec8
2018-05-23cosmetic: define TCH_F_PDCH_PENDING_MASK as actual bitwise orNeels Hofmeyr1-2/+1
Change-Id: I5266d6e1ea0ea89c66fa481136c31e2f84bcb313
2018-05-23cosmetic: bsc_dyn_ts.c: make local functions staticNeels Hofmeyr1-2/+2
Change-Id: Iee47b4ffefedf9532e11f17d5f9b603fc8872cb9
2018-05-23abis_rsl.h: drop unused rsl_chan_activate() declarationNeels Hofmeyr1-6/+0
Change-Id: I28a48dc9a1719793d369468aa9a4bdcf460d201c
2018-05-23bsc_api.c: actually log with contextNeels Hofmeyr2-43/+58
bsc_api.c notoriously lacks log context. Provide gsm_lchan_name() and/or bsc_subscr_name() in roughly a million instances, using new LOGPLCHAN macro. Add LOGPLCHAN() to gsm_data.h, to encourage use of it in other .c files. Change-Id: If469defcc6fe8950dac5df61db3f39d297893318
2018-05-22nat: Add jitter buffer on the uplink receiverPau Espin Pedrol3-0/+96
Default usage values are defined in mgcp node, and can be per-BSC overriden on each bsc node. This commit is a forward-port of openbsc.git Change-Id Ibf3932adc07442fb5e9c7a06404853f9d0a20959. Depends on osmo-mgw.git Change-Id Ie19a64ac09f9d51f2434ad0d7925610fc919a90e. Change-Id: Ie07b8a577caf731d59d68e3b3510ae2f9fd3dc93
2018-05-22add counter for connection attempts from BTS with unknown unit idStefan Sperling2-0/+4
Add a global counter to the BSC which shows the number of failed connections attempts due to a unit_id mismatch between the BSC and the BTS. Change-Id: I58866aff36a1c8463bf84b4392a5124ffeaa32ea Related: OS#3245
2018-05-18show all global counters of osmo-bsc in vtyStefan Sperling2-7/+11
The 'show statistics' VTY command was not showing all counters maintained by osmo-bsc. Instead of printing just two counters related to paging, print all available counters in a generic way. Adjust descriptions of some counters for nicer display. After startup (all counters are zero) is now looks like this: OsmoBSC# show statistics handover:attempted: 0 Received handover attempts. handover:no_channel: 0 Sent no channel available responses. handover:timeout: 0 Timeouts of timer T3103. handover:completed: 0 Received handover completed. handover:failed: 0 Received HO FAIL messages. paging:attempted: 0 Paging attempts for a subscriber. paging:detached: 0 Paging request send failures because no responsible BTS was found. paging:responded: 0 Paging attempts with successful response. OsmoBSC# Change-Id: I58ae04e1960774d760e3ebb54a4f307c9f753655 Related: OS#3245
2018-05-17a_reset: cleanup + remove dead codePhilipp Maier7-105/+77
The function a_reset_free() is not used anywhere at the code. The reason for this is that a BSC instance is never cleared once it is started up. Also the timer number is not according to the spec. - Remove a_reset_free() - Fix timer identification number (T4) - use fi->priv to hold context info - Fix sourcecode formatting Change-Id: I72095d52304c520e383755eee6c889bce492cbd4 Related: OS#3102
2018-05-15Bump version: → Espin Pedrol1-0/+23
Change-Id: I5f6135b7e16cac95c58a1bbc147234e10ee37574
2018-05-14cosmetic: dyn ts init: undup logging for gprs = noneNeels Hofmeyr2-22/+20
Reshuffle the decision not to activate PDCH when GPRS is off: Even though all current callers should avoid passing a PDCH activation in case GPRS is off, it's a better idea to not assert on it and crash osmo-bsc. Move the decision to omit PDCH activation and logging about it into the actual functions that do PDCH activation. If PDCH activation is skipped, the lchan then just stays as it was, and that's what it should anyway be doing. Change-Id: Ib26642f08044d71a2469e6dbabf1e6fbcb02044d
2018-05-14log: indicate hr/fr in audio_support_to_gsm88() errorNeels Hofmeyr1-2/+2
Change-Id: Iaaa710d0274fe813d227cb658d82a24db68c4161
2018-05-14fix default fallbacks in audio_support_to_gsm88()Neels Hofmeyr1-2/+2
For audio->hr == true, use HR1, and for hr == false use FR1; not vice versa. Change-Id: Ifb4dba7c8e9c1d0a22a007355fbd2eda57e789d3
2018-05-14deprecate dyn_ts_allow_tch_f and by default allow all TCHNeels Hofmeyr2-3/+7
In osmo-nitb, the way TCH lchans were assigned often resulted in mismatching TCH kinds, causing problems in the lack of transcoding. Hence dyn_ts_allow_tch_f was introduced as a workaround. Now however, we always assign an SDCCH to a requesting MS first, and only later assign a TCH channel, which then adheres to the codec list configured at 'msc' in the vty config. Hence it is now considerably harder to obtain a mismatch. Furthermore, forcing specific codecs is possible by simply omitting the unwanted ones from the msc config's codec-list. The equivalent of 'dyn_ts_allow_tch_f 0' could be e.g. 'codec-list hr3 hr2 hr1'. Change-Id: Ib2335d02ea545aff837aadd49f15b2fdb418c46e
2018-05-10debug log: verbosely log all lchan alloc choicesNeels Hofmeyr1-8/+40
Change-Id: If0070b8f7b6cc7fffbc582cf9f472e60f3096000
2018-05-10dyn TS: allow any pchan type changes, fix for gprs mode noneNeels Hofmeyr2-68/+96
In rsl_chan_activate_lchan(), remove a condition to also allow switching pchan modes when not in PDCH mode, which is actually not needed and would hinder switching from pchan=NONE or between TCH/F <-> TCH/H. Refactor the part where lchan_alloc() decides to switch a pchan mode into a separate function, ts_usable_as_pchan(), which transparently checks both dyn TS kinds for: - Already in switchover? (missing check for ip.access style dyn TS) - Is the lchan->state in error? (missing check for ip.access style dyn TS) - Switch from pchan=NONE? (missing feature for Osmocom style dyn TS, for proper handling with gprs mode none) - Switch between TCH/F <-> TCH/H when all subslots are unused? (missing feature for Osmocom style dyn TS, also useful for gprs mode none) Always pass the desired pchan in the dyn_as_pchan argument to the _lc_find_* functions to make switchover decisions transparent. Use the _lc_dyn_find_bts() function for ip.access style dyn TS for the same reason. Related: OS#3244 Change-Id: I72d5d833b186b1e1925d513885b405d8c19aa496
2018-05-10dyn TS: init only when both RSL and the Channel OM are establishedNeels Hofmeyr8-3/+73
Recent Icf6e25ff068e8a2600562d52726ead65e864ec02 changed the dyn_ts_init() hook from bootstrap_rsl() to the Channel OPSTART ACK, but this is not sufficient. Now RBS2k never calls dyn_ts_init(), and we may need to wait for RSL: Dyn TS should actually be initialized only when *both* OML opstart and RSL link are established. To that end, introduce a generalized API to query OML and RSL status and to trigger a timeslot init at the appropriate time. Add gsm_ts_check_init() to be called both when RSL and OML opstart are established: trigger gsm_ts_init() only when both are given. Add gsm_bts_trx_ts->initialized flag to mark whether initialization has already taken place. Add gsm_bts_mark_all_ts_uninitialized() to conveniently clear this flag for all TS in a BTS. Add gsm_bts_model.oml_is_ts_ready() callback so that each BTS implementation can return the OML status of a timeslot in its own OML implementation. Actually, currently all BTS models that need this init mechanism store the TS' OML status in ts->mo.nm_state. While we would in practice correctly init dyn TS by just looking at ts->mo.nm_state, semantically, the decision whether the TS is ready is up to the BTS models' specific OML implementations. From bootstrap_rsl(), call gsm_ts_check_init(), in case the TS OML Opstart has happened before RSL is established -- applies to all BTS models. For all BTS models: - call gsm_{bts,trx}_mark_all_ts_uninitialized() when OM is torn down, to make sure the TS init mechanism will work a second time. For all BTS models supporting dyn TS, i.e. osmo-bts, nanobts and RBS2k: - implement oml_is_ts_ready(). - call gsm_ts_check_init() when a Channel OM is taken into operation. Any BTS models that don't set oml_is_ts_ready() will see a ts init as soon as RSL is bootstrapped (incidentally, the old dyn TS behavior before recent Icf6e25ff068e8a2600562d52726ead65e864ec02). This firstly fixes dyn TS for RBS2k by re-adding the initial switch to PDCH, and furthermore does so only after both OML TS opstart and RSL are through. This fixes the ttcn3-bsc-tests around dyn TS, since for the osmo-bts-virtual, the RSL is established only after OML opstart on the TS, which was broken by Icf6e25ff068e8a2600562d52726ead65e864ec02. Nokia Site and Siemens BS11 practically do not require this init mechanism, since all that happens there so far is dyn TS init, and these BTS models do not support dyn TS of any kind. A future patch may add oml_is_ts_ready(). Related: OS#3205 Change-Id: I99f29d2ba079f6f4b77f0af12d9784588d2f56b3
2018-05-08osmo_bsc_vty.c: fix: write MGW configurationVadim Yanitskiy1-0/+3
Previously the MGW configuration was ignored during writing of the MSC configuration. Let's fix this by calling the mgcp_client_config_write() function. Change-Id: I7d1eedb782a4f30bd089838969ce54f27cde060d
2018-05-05dyn TS, assignment: allow switch from PDCH with associated connNeels Hofmeyr1-2/+6
Typically, an lchan that is released should no longer be associated with subscriber connection. If that is the case, an S_LCHAN_UNEXPECTED_RELEASE is triggered, which aborts, e.g., an ongoing assignment. However, with dynamic timeslots, we may set lchan->conn and then start to switch over from PDCH to a TCH mode, in which case it is perfectly fine to release an lchan that is associated to a conn. In lchan_free(), do not fire S_LCHAN_UNEXPECTED_RELEASE for a dyn TS that is currently in switchover. This is the second and last part to fix dynamic timeslots handling of the gscon. Related: OS#3211 Change-Id: Id7d9dd06451722eb328db77bb586826c954bd85c
2018-05-05dyn TS, assignment: set lchan state to LCHAN_S_ACT_REQ in the proper placeNeels Hofmeyr2-3/+4
Set lchan->state to LCHAN_S_ACT_REQ in rsl_chan_activate_lchan(), not in handle_new_assignment(). This is the first part of a fix for dynamic timeslots handling in the gscon. Rationale: In rsl_chan_activate_lchan(), we may choose to set the lchan state to LCHAN_S_REL_REQ and wait for dyn TS switchover from PDCH. So the caller from bsc_api.c handle_new_assignment() must not bluntly set the state to LCHAN_S_ACT_REQ, which is not accurate in the case of dyn TS switchover. In case of dyn TS switchover, a later release ack received from the BTS will cause rsl_chan_activate_lchan() to be called again, at which point we may accurately set state LCHAN_S_ACT_REQ, and continue the Assignment. Related: OS#3211 Change-Id: Iedb4fb63bf1959d5f1d2c6edb6a7f5097ff16bd7
2018-05-05dyn ts, bts_ipaccess_nanobts.c: init PDCH on Chan OPSTART ACKNeels Hofmeyr4-2/+54
Sending PDCH activation upon RSL bootstrap is too early. Introduce OPSTART ACK handling to call dyn_ts_init() only when the dynamic timeslot is indeed ready to receive a PDCH activation. Related: OS#3205 Change-Id: Icf6e25ff068e8a2600562d52726ead65e864ec02
2018-05-04resurrect meas_feed.c: vty, vty-testNeels Hofmeyr3-3/+69
At this point, meas-feed is usable again, however, osmo-bsc is not able to include the IMSI in every report like osmo-nitb did. In consequence, the meas-vis and meas-web tools are unable to handle the current measurement reports: these so far use the IMSI to list reports, and all reports without an IMSI are collapsed onto the same line, swapping values. So though osmo-bsc now sends usable measurement reports via meas-feed, two avenues to improve should be pursued: OS#3192: the visualization tools should use bts,ts,ss numbers, not IMSI. OS#2969: osmo-bsc should always know a mobile identity. Related: OS#2968 Change-Id: I186c7a995dd2b81746c32a58b55da64ed195a1ce
2018-05-04bsc_api/GSCON: prevent unnecessary channel mode modificationsPhilipp Maier2-9/+19
gsm0808_assign_req() checks if the new channel mode is compatible with the new mode. If it is, it does a gsm48_lchan_modify(), but it does not actually check if the new mode is equal to the current mode. - skip when the channel is compatible and the new mode is equal to the old mode. - send the ASSIGNMENT COMPLETE directly from ST_ACTIVE when no mode modify was necessary. Change-Id: I86a2d52836c54d2dbd77441b182f757327ec7262 Related: OS#2936
2018-05-03Bump version: → Espin Pedrol2-10/+397
Change-Id: Ie6ad5c769dd11c79b2bfd0d19f0feda8416e09cd
2018-05-03resurrect meas_feed.c: make it compile, add loggingNeels Hofmeyr2-16/+34
Change-Id: I34ac25bcc460e87e813090b6d7c5085cffa2d78f
2018-05-03resurrect meas_feed.c from openbsc.git historyNeels Hofmeyr1-0/+168
meas_feed.c used to live in libmsc, to send out measurement reports to external entities for evaluation. When splitting osmo-bsc and osmo-msc from openbsc.git, meas_feed.c should have moved to osmo-bsc.git, but was dropped with libmsc. Re-add the old meas_feed.c now into libbsc. This is the latest version that existed in libmsc, and will not compile as-is here. Modifications to incorporate in the osmo-bsc build will follow with subsequent patches. Change-Id: Ic070d82e61c122061fe7297a8c5aabbbcef6b301
2018-05-02cosmetic: Add note about libosmo-legacy-mgcp to configure.acPhilipp Maier1-0/+3
Change-Id: Ia496bd9ec02c0028aa541a798b1358a68ad18e21 Closes: OS#3226
2018-04-24flush paging when RSL link is droppedStefan Sperling1-0/+6
Flush the paging queue if the link to TRX 0 is dropped from a BTS. This should prevent stale paging requests sent to the BTS when it disconnects or reconnects, as seen in the TTCN3 BSC_Tests test suite. Also, add entries to the log when RSL or OML links are dropped so that related error messages in the log can be interpreted in context. Change-Id: If4401c1139cd01faf5ff374301a9a701898c3777 Related: OS#2901
2018-04-24extend documentation of paging_flush_bts()Stefan Sperling1-1/+1
Document the semantics of a NULL msc argument. Change-Id: Ib3f6fd4e4e6108f03d688eba2416c0714393e6e9 Related: OS#2901