summaryrefslogtreecommitdiffstats
path: root/bsc
AgeCommit message (Collapse)AuthorFilesLines
4 daysbsc: f_ipa_unknown_unit_id: use different idOliver Smith1-1/+1
Change the unit_id value, that is supposed to be unknown at BTS, from 0/0/0 to 99/0/0 to make TC_oml_unknown_unit_id pass again. The test was failing, because a new bts that matches 0/0/0 was added to osmo-bsc.cfg in [1], [2] and [3]. [1] Ia4ba0e75abd3d45a3422b2525e5f938cdc5a04cc [2] I1c57a04747f5ec004ccf4657954dcb0b003c24fc (docker-playground) [3] I00547dccf8908d46df4730cb513fe577956d7683 (docker-playground) Related: OS#4156 Change-Id: I585ff825c7b6bca17e5c6fb6d670a649965a1653
5 daysBSC_Tests: update number-of-bts to 4Oliver Smith1-1/+1
Fix failing TC_ctrl test for ttcn3-bsc-test and -latest. Depends: I00547dccf8908d46df4730cb513fe577956d7683 (docker-playground) Related: OS#4156 Change-Id: Ie2c664ba0f845da644e20e2c919c12d8fc2af6ba
6 daysBSC_Tests: Change test-expectation of TC_paging_resp_unsolPhilipp Maier1-6/+23
The testcase TC_paging_resp_unsol expects the BSC to close the channel when an unsolicit paging response is received. This expectation is due to the fact that osmo-bsc supports multiple MSC, which is not specified in 3gpp specs. Therefore the BSC needs to know which MSC is in charge. However, with MT-CSFB calls it is a normal situation that the BSC receives a paging response on abis without sending a paging first since the MSC has paged the UE via the SGs interface. In those cases we expect the BSC to forward the paging response to the first configured MSC. Related: SYS#4624 Change-Id: I5562cbf61a2aa42e6950860bc0f9c6c20c61a9fe
9 daysfix TC_ho_int; make neighbor config more robustNeels Hofmeyr2-2/+5
Since osmo-bsc commit "neighbor config: allow re-using ARFCN+BSIC pairs" (I29bca59ab232eddc74e0d4698efb9c9992443983), osmo-bsc considers only those cells as neighbors that are explicitly listed, or all local cells if none are listed. This lead to breaking TC_ho_int, because the osmo-bsc.cfg has only one remote-cell neighbor for bts 0, and hence a handover to local cell bts 1 is now regarded as invalid. The remote-cell neighbor is needed for inter-BSC handover tests; also consider that the TC_ho_neighbor_config_* tests each place individual neighbor configuration by live VTY interaction. Hence make all of these tests more robust: remove the neighbor config from the osmo-bsc.cfg file, and instead include VTY interaction for each test case that sets the particularly needed neighbor configuration at runtime. An analogous osmo-bsc.cfg change in docker-playground is in change If44dd6b578cdc55076c8180707d1c2d69fe5f2a8. (It is not actually harmful to leave the neighbor config in osmo-bsc.cfg, but remove that since it is also not needed anymore.) Change-Id: If44dd6b578cdc55076c8180707d1c2d69fe5f2a8
10 daysbsc: add TC_ho_neighbor_config_1 thru _7Neels Hofmeyr2-1/+394
Add tests to play through various neighbor configurations. Tests will pass as soon as osmo-bsc I29bca59ab232eddc74e0d4698efb9c9992443983 is merged. Add RSL2 to allow triggering handover to BTS 2. Adjust osmo-bsc.cfg to match the new tests. Also applied in docker-playground I1c57a04747f5ec004ccf4657954dcb0b003c24fc. - Actually enable handover. - Add bts 3 Depends: osmo-bsc I8623ab581639e9f8af6a9ff1eca990518d1b1211 ('no neighbors') Related: OS#4056 Change-Id: Ia4ba0e75abd3d45a3422b2525e5f938cdc5a04cc
2019-07-14rename TC_ho_out_fail_no_ho_detect to TC_ho_out_fail_no_result_after_ho_cmdNeels Hofmeyr2-9/+6
Change-Id: If84a1d4535ee12e33a50588e178bf1ebb6fb2cd5
2019-07-09f_tc_ho_out_fail_no_ho_detect: stricter orderingNeels Hofmeyr1-7/+7
Make sure that the RR is released only after the MSC has sent the Clear Command: - first expect a Clear Request from osmo-bsc and return a Clear Command, - only then accept RR release messages. See 3GPP TS 48.008 3.1.5.3.3 "Abnormal Conditions": "The terrestrial resource in the old BSS shall remain assigned until a CLEAR COMMAND message is received from the MSC" osmo-bsc already complies, the test should continue to pass. Change-Id: Iba05336d3c4af8a1c57cdc828dae464eae3510b9
2019-07-09comments: BSC_Tests: clarify f_tc_ho_out_fail functionsNeels Hofmeyr1-5/+9
Change-Id: Ia0584324476d36077e92d1564d9d7e79a069addf
2019-07-03bsc: Do proper chan release in TC_ho_out_fail_no_ho_detectPau Espin Pedrol1-11/+13
BSC waits to receive a ClearCommand in response to its ClearRequest before it starts tearing down the MGCP conn on the MSC-side of the MGW endpoint. As a result, expected DLCX was not being sent which made test fail. However, currently test still fails because current osmo-bsc master sends a repeated ClearRequest message in this scenario. Related: OS#4078 Change-Id: Ic398896147a0b6b04ffeae56a23d25783b2b17fe
2019-07-01cosmetic: bsc: Document timer timing out in f_tc_ho_out_fail_no_ho_detectPau Espin Pedrol1-1/+1
Change-Id: I4d06fcff63d82f849da08f509b007cb19e809fb0
2019-06-24bsc: Test MGCP-over-IPA forwarding in SCCPlite testsPau Espin Pedrol2-18/+135
* MGCP-over-IPA handling in MSC_ConnectionHandler means we need to use the new MGCP_CLIENT_MULTI port since we'll be managing MGCP messages from 2 different UDP connections, and we need to be able to route answers correctly. As a result, parameter multi_conn_mode is enabled for SCCPlite and all code adapted to use that port in that type of scenario. * iDuring calls when on SCCPlite, send a full (all-required-params-in) CRCX through the MGCP-over-IPA connection towards the BSC in order to emulate the MSC, and expect the correct answer back. This way we test BSC funcionality to forward MGCP messages coming from MSC works as expected. Related: OS#2536 Depends: osmo-bsc.git I38ad8fa645c08900e0e1f1b4b96136bc6d96b3ab Change-Id: I31fed700772dd0b063f913b1e1639fd428c46e7d
2019-06-24lib/mgcp: Add new port with support to handle multiple MGCP socketsPau Espin Pedrol1-1/+2
* Some scenarios like MGW BSC-attached in SCCPlite require handling of 2 MGCP-over-UDP sockets in MGCP Emulation: 1 for regular libosmomgcp-client from osmo-bsc and another one from the forward socket from osmo-bsc (of MGCP-over-IPA messages communicated with MSC). * Old port is kept for backward compatibility with other tests and enabled by default. It's also interesting to keep it because it makes tests without special needs (2 sockets) to use the old port/API which produces simpler code to read and mantain. * Users of the new port have to enable multi_conn_mode parameter and expect to interact with port MGCP_CLIENT_MULTI instead of MGCP_CLIENT, which will offer messages containing information about the UDP connection being used by that message. Change-Id: Ic0ba8c5cde068c07671512a83095d83e28b86746
2019-06-24bsc: Refactor altstep as_Media_mgw()Pau Espin Pedrol1-52/+66
Move logic handling CRCX and MDCX to function, so they can be reused for other ports in forthcoming commits. Change-Id: I07344657c5d1465a8e0c278adb76150ca7f449ba
2019-06-19bsc: Move hardcoded ip addr to variablePau Espin Pedrol1-1/+3
This way it's more clear what are those IP addresses for. Change-Id: I3600fd0d0486205e5bdc487fa75e9038d3bfb52c
2019-06-19bsc-sccplite: Fix expected DLCX in HO testsPau Espin Pedrol1-64/+31
* TTCN3 code was not ACKing the DLCXs, and as a result retransmitted DLCX BSC->MGW were being counted as 2nd DLCX. * In SCCPLite, only 1 DLCX is expected BSC->MGW, because the BSC only takes care of the BTS-side conn in the endpoint, while MSC takes care of the MSC-side conn (which is not sent in this case because doesn't really involved the BSC other than forwarding the message, which will already be tested in other places in forthcoming commits). * Getting rid of retransmissions by ACKing the DLCX, it unconvers a bug in TC_ho_out_fail_no_ho_detect when on AoIP, where BSC only deletes one of the 2 previously created connections. * Code is refactored into the function because its logic is made more complex, and may be even more complex in forthcoming commits when we add MGCP-over-IPA forwarding verification support. Change-Id: Ia1d0db9af073760105cc8509e228e317dbea2268
2019-06-13bsc: Introduce test TC_ctrl_locationPau Espin Pedrol3-2/+35
Related: OS#2012 Change-Id: I3640f6889e3b6adc3b4601082f948543c64a2fe3
2019-06-13bsc: Fix trailing whitespacePau Espin Pedrol2-3/+3
Change-Id: I97c6cc290d6cc23611ac38e7e370b5b0b88cca1d
2019-06-06bsc: Only append Osmux IEs in BSSAP AoIPPau Espin Pedrol1-2/+4
Previous to this commit, BSSAP Reset (Ack) messages contained Osmux Support IE even if transport was SCCPLite, where those IEs are actually meaningless. Change-Id: If6cc0f65a0f273297a4523e5d6a7564d966f0aa6
2019-06-06Add and set transport field for RanOps structurePau Espin Pedrol1-0/+1
This will allow RAN_Emulation to have better knowledge on the protocol stack in use, and behave differently based on that information. For intance, forthcoming commit will append OsmuxSupport IE only if transport is BSSAP AoIP. Change-Id: Ife62e328af2d3f2475ff93249f2138820c7ddabb
2019-06-06bsc: Avoid appending Osmux CID IE in AssignReq/Compl with SCCPLitePau Espin Pedrol1-10/+2
That IE is not required when on SCCPlite, only for AoIP. Change-Id: Iff939e431fa3da9fef648242140fdb8d0dc92e52
2019-05-29bsc: Add module param to disable osmux (fix bsc-latest jenkins job)Pau Espin Pedrol1-2/+11
ttcn3-bsc-test-latest currently fails on most tests because it tries to use "osmux off" VTY param and only current osmo-bsc master supports it. Change-Id: I61e4c59b2926f3f70cb6d0190a8683861e54179a
2019-05-27cosmetic: Update copyright statement, license notice and SPDXHarald Welte3-0/+15
Some of our files didn't have a copyright notice at all, let's add it. Also, update the notices in other files and ensure a SPDX identifier is present in all but the most trivial files. Change-Id: If7fa19ce484b415bc645e39b3d0d666b44b5f0fd
2019-05-27bsc: Introduce Osmux infra and one test for osmo-bscPau Espin Pedrol3-28/+106
Test verifies once osmux is enabled in osmo-bsc, BSSMAP RESET (ACK) contains Osmux Support IE and that it correctly handles BSSMAP ASsign Req with Osmux CID. Related: OS#2551 Depends: osmo-bsc 6de754cdde5319af3059d8fc6abf85037ec7eacc Depends: titan.ProtocolModules.BSSMAP Iaf1e137269c0da20b2c96fd104b57edf336693af Change-Id: If69c716dc06d61d810c32d1720a237c7535baca8
2019-05-21Depend on our own fork of titan.ProtocolModules.BSSMAPPau Espin Pedrol1-1/+1
Since we use some BSSMAP extensions to signal Osmux, we need to maintain our own fork of BSSMAP_Types in order to supports those IEs in BSSMAP RESET and BSSMAP Assin Req/Compl. Hence, switch all build componenets to fetch and use our fork. Depends: titan.ProtocolModules.BSSMAP Iaf1e137269c0da20b2c96fd104b57edf336693af Change-Id: Ic8debe5f3ffe8e1d4258fa6b4632a3871b99af40
2019-05-07bsc: deterministic TC_ho_in_fail_msc_clears: wait for MGW setupNeels Hofmeyr1-0/+3
Randomly, this test issues the Clear Command so fast that the MGW has not yet been set up, and hence the test cannot possibly see DLCX for endpoint conns that haven't been set up at all. So, simply wait a bit before clearing. Change-Id: Idd0b3810916efd02a499e0ac060a6a275265b8c3
2019-04-25RAN_Adapter: Rename functions from f_bssap_* to f_ran_adapter_*Harald Welte1-4/+4
Change-Id: I73818247f1dfc71c8ece11660e6c18f5f153d186
2019-04-25RAN_Emulation: Modularize protocol supportHarald Welte3-2/+3
The RAN_Emulation currently unconditionally provides BSSAP and MGCP support. Let's re-structure the code so that support for those protocols is now possible to enable/disable at compile time. This patch is in preparation of introducing RANAP support in RAN_Emulation. Change-Id: Id53ba3ff05f9946230e0e4a759245de14a0f9fbd Related: OS#2856
2019-04-21Rename BSSMAP_Emulation -> RAN_EmulationHarald Welte4-36/+36
So far, BSSMAP_Emulation supported only a transport over BSSMAP. However, we soon intend to merge support for RANAP in order to simulate RANAP/Iu connections as well as BSSMAP. Let's start by renaming some of the existing types/functions/ports/modules without introducing any functional changes just yet. Related: OS#2857, OS#2856 Change-Id: Iecbcb0c6c136baad9460eca40606bb4010d8882d
2019-03-27BSC_Tests: execute S15-S0 related tests only on AoIPPhilipp Maier1-17/+20
Make sure that the S15-S0 related testcases are only executed when AoIP is used as transport. Change-Id: I62d51bbe4b1f089ded6c271b68414a4a37d509d8 Related: OS#3864
2019-03-26bsc: add TC_chan_rel_hard_clear_csfb to expected-results.xmlPhilipp Maier1-0/+1
The testcase TC_chan_rel_hard_clear_csfb is missing from the expected-results.xml. Lets add it. Change-Id: I087e6715364f72944cb47fa64d96eab6538afda5
2019-03-20expected-results: add new S15-S0 related testscasesPhilipp Maier1-0/+17
The S15-S0 related testscases are missing from the expected-results.xml file. Lets add them. Change-Id: I28a853ec25feb723b2d333061e1ae4d2e0b34d86 Related: SYS#4470
2019-03-19BSC_Tests: add testcases to verify S15-S0 handlingPhilipp Maier2-0/+224
The handling of the AMR rate configuration bits S15-S0 is currently only superficially checked. Lets add more some more elaborated testcases to check through varios different situations. Also make sure that the resulting mr configuration IE is verified Change-Id: Ica323deb9836deea72982e093c9cb31deb5a216b Related: SYS#4470
2019-03-19BSC_Tests: fix TC_assignment_codec_amr_f/hPhilipp Maier1-2/+2
The testcase TC_assignment_codec_amr_f uses a combination of S-Bits that has S1 which configures a set of four rates at once. This is quite a complex situation and since the BSC was upgraded with new features affecting the behavior in this special case lets simplify this testcase for now. depends: osmo-bsc Ie52376b51fe07ed07056e8df2e9557293ff67a78 Change-Id: Ibf730f76947cdeed23eb3119167450e3b7a9b314 Related: SYS#4470
2019-02-05bsc: Test CSFB "Fast Return" in new TC_chan_rel_hard_clear_csfbHarald Welte1-2/+31
When a MSC releases a connection using the BSSMAP CLEAR CMD, it can specify that this call was part of CSFB. The BSC is then supposed to add a special IE to the RR RELEASE message which will help the phone to switch back to LTE as soon as possible. This commit adds a new test case testing for exactly that behavior. The test does *not* verify if the EARFCN information contained is actually correct, only that the IE is present in the RR RELEASE. Change-Id: I7501fb25578412c882ff92da5d388f3079bbce7f Requires: osmo-bsc Ibfbb87e2e16b05032ad1cb91c11fad1b2f76d755 Related: OS#3777
2019-02-05bsc: replace octet string with decmatch when matching RR RELEASEHarald Welte2-4/+3
The 'decmatch' keyword allows us to match the decoded version of some octetstring, which is very useful in the situations where we have the L3 message only as octetstring but want to check if it matches some L3 template. Change-Id: I0a91e067f7e8062bf991fef8b0d4d8da740bfafc
2019-02-05BSC_Tests: Don't make invalid assumptions about RR RELEASEHarald Welte2-6/+4
The RR RELEASE message does not always have to be '060D00'O, which constrains it to: * not having any optional IEs * not having a cause value != 00 Let's relax the matching to accept any RR RELEASE message, whatever the cause may be, and whether or not there are any optional IEs at the end. At the same time, also remove some copy+pasting but rather have one template that gets used everywhere. Change-Id: I4b9d078c9b66f040fe673b5d957cf8e2c6d5892c
2019-01-29BSC_Tests: fix vty interface for channel allocator testspmaier/challocPhilipp Maier1-12/+12
The tests presented in Change I109d986dd7ece1a56422a669ca64353ed46f7ed6, are using a slightly outdated vty syntax and fail because of that. Lets update the VTY syntax. Change-Id: I302e8872dbdc8e65d51902a881f777863de22915 Related: OS#3503
2019-01-29BSC_Tests: Add tests to check channel allocatorPhilipp Maier2-0/+304
When a channel is assigned via the assignment request throught the A interface, the MSC may offer either FR, HR or both. When FR and HR are permitted, a preference is set on one of the two. At the moment we do not check how the bsc is reacting to those preferences and its also not checked how the behavior is when the preferred rate is not available because all lchan of that type are already in use. Lets add a set of tests to verify this. Change-Id: I109d986dd7ece1a56422a669ca64353ed46f7ed6 Depends: osmo-bsc I397e68e26d6a1727890353fa34f4897b54795866 Related: OS#3503
2019-01-23update expected resultsNeels Hofmeyr1-1/+7
Change-Id: Idacaf8343bed4a37878eacdf338c4d5eb46bf7a7
2018-12-10bsc: TC_chan_exhaustion: Send correct RA to alloc all different channelsPau Espin Pedrol1-2/+4
Previous RA value (23, Establishment cause = 0010XXXX) meant MS was dual rate capable but was asking speciifically for only TCH/F channel. As a result, TCH/H was not being allocated and an immediate assignment reject was sent. Change-Id: I3e58592c661fc004e648dbe46b67a3b3f5a20bc8
2018-12-08BSC_Tests: expect a HANDOVER PERFORMED message on internal handoverPhilipp Maier1-0/+4
When an internal handover is performed, the BSC is expected to inform the MSC about the event by sending a BSSMAP HANDOVER PERFORMED message. This feature was missing in the BSC and has now been added. The tests need to be upgraded in order to handle the additional message. - Upgrade f_tc_ho_int so that it expects a HANDOVER PERFORMED message Change-Id: I10f4e578c96a90317939ba49b61b14a3c7e488a7 Depends: osmo-bsc If26e5807280e0f75a423b3b04f8e3c698c82a351 Related: OS#3645
2018-12-07bsc: Fix TC_paging_resp_unsolPau Espin Pedrol1-1/+1
TC_paging_resp_unsol spent some time in gerrit before being merged. As a result, other commits were merged in between the test was submitted (tested) and merged. As a result, commit a5302c8151d1da2e43ed52efc0544d70bffab911 was merged while this test was still in gerrit and thus was not updated accordingly. Similar stuff happened with the osmo-bsc commit fixing the scenario this test was showcasing: The osmo-bsc patch (77cd1129931928d2a6e7667d0374feeeed71b0ce) had merge conflict with other osmo-bsc commits merged in-between, and was merged even later than the commit introducing this TTCN3 test, so failures were expected for this test for a while. Change-Id: I933cba41912640eb7e15be4a27bda5b4dd489962
2018-11-27bsc: Start the timer at the beginning of f_initDaniel Willmann1-2/+3
This avoids a possible indefinite timeout if osmo-bsc crashes during the test. Ticket: OS#3713 Change-Id: I88f6254d9d2d90ab49a3c3d8255af25ca6ffe624
2018-11-27BSC_Tests.ttcn: introduce TC_chan_deact_silenceVadim Yanitskiy2-0/+35
The idea of this test case is to verify channel deactivation procedure due to no response to Immediate Assignment. Change-Id: I00b0838c9f919303aef72280248b0d1317f42b3b Related: OS#3709
2018-11-22bsc: Introduce test TC_paging_resp_unsolPau Espin Pedrol2-1/+27
With this test we want to verify that channels are released if BSC fails to complete an L3 request, for instance because no pending Paging CMD is found for a received Paging Response. Related: OS#3680 Change-Id: Iabe8a51aa13d2fcfec4500cf7aab47d60cc138ce
2018-11-19bsc: add inter-bsc ho incoming failure testsNeels Hofmeyr1-0/+348
Change-Id: I849e4c0a14cc091195d948adb8df7a0b7414ecfe
2018-11-19BSC: log BTS number on failureMax1-1/+1
It's useful to see which BTS exactly has failed the test in configuration with multiple BTS on single BSC. Change-Id: Ib813635862c04d51a30e7bbcca4ec05ce664f7e9
2018-11-19BSC LCLS: add bts-loop testsMax3-12/+127
Add basic establishment and teardown tests for 'bts-loop' mode of LCLS: * add explicit vty init for desired LCLS kind * add necessary IPA RSL MDCX functions * explicitly pass LCLS kind as a parameter to shared functions (defaulting to 'mgw-loop') Change-Id: I40e786b430591899c722d99d685db26efa868508 Related: OS#3659
2018-11-18LCLS: explicitly check for *CX adjustmentMax1-3/+6
Previously the expectations for number of CRCX and MDCX messages from MGW was adjusted unconditionally for LCLS tests. However this is only necessary for MGW-loop type of LCLS. Use explicit variable (with default value preserving current behavior) to decide whether to apply this adjustment or not. This simplifies support for other kinds of LCLS loops. Change-Id: I07b2c56991977b5e80c372a5b8338f348f14c076 Related: OS#3659
2018-11-18BSC LCLS: restructure testsMax1-60/+52
Move duplicated code into separate functions to simplify adding test cases which do not use MGW MDCX methods for closing the loop. Change-Id: Idf9fc8e639b0ece662ab26c481899ef39fb7edfe Related: OS#3659