https://osmocom.org/https://osmocom.org/favicon.ico?16647414092019-04-15T07:54:24ZOpen Source Mobile CommunicationsOsmoPCU - Bug #3925: Missing PCU_Tests.ttcn UL TBF testshttps://osmocom.org/issues/3925?journal_id=139672019-04-15T07:54:24Zlaforge
<ul><li><strong>Priority</strong> changed from <i>Normal</i> to <i>High</i></li></ul> OsmoPCU - Bug #3925: Missing PCU_Tests.ttcn UL TBF testshttps://osmocom.org/issues/3925?journal_id=139682019-04-15T07:55:29Zlaforge
<ul><li><strong>Tags</strong> set to <i>TTCN3</i></li></ul> OsmoPCU - Bug #3925: Missing PCU_Tests.ttcn UL TBF testshttps://osmocom.org/issues/3925?journal_id=156592019-08-21T12:47:45Zosmith
<ul><li><strong>Assignee</strong> changed from <i>lynxis</i> to <i>osmith</i></li></ul>There are two existing tests:
<ul>
<li>TC_ul_tbf</li>
<li>TC_ul_tbf_single_llc_sizes</li>
</ul>
<p><a class="external" href="https://git.osmocom.org/osmo-ttcn3-hacks/tree/pcu/PCU_Tests.ttcn?id=92c2bdb00fd840ec5fca6e49c12776146b2f4ef4">https://git.osmocom.org/osmo-ttcn3-hacks/tree/pcu/PCU_Tests.ttcn?id=92c2bdb00fd840ec5fca6e49c12776146b2f4ef4</a></p>
<p>According to Harald, they were already passing at some point: <a class="external" href="https://lists.osmocom.org/pipermail/openbsc/2019-July/012992.html">https://lists.osmocom.org/pipermail/openbsc/2019-July/012992.html</a></p>
<p>However, we are not running them in jenkins so far and the configs that used to work with the tests are not saved anywhere. I'm trying to get them running again, first locally, then in docker with jenkins for the nightly tests, and then extend the tests from there.</p>
<p>This is the test setup:<br /><pre>
testsuite ---/tmp/osmocom_l2--- virt_phy ---GSMTAP UDP multicast--- osmo-bts-virtual ---/tmp/pcu_bts--- osmo-pcu
|
| A-bis OML
|
osmo-bsc
</pre></p>
I've extended my existing ttcn3.sh script to run the above setup locally. WIP branches:
<ul>
<li>osmo-ttcn3-hacks.git: osmith/pcu-tests</li>
<li>osmo-dev.git: osmith/ttcn3.sh</li>
</ul>
<p>After resolving various issues with my configuration and finding out that the cell_identity in BVCI only gets matched by the testsuite if it is byte-swapped in the config (I'll investigate this further once I have the tests running...<a href="https://gerrit.osmocom.org/c/osmo-pcu/+/13369" class="external">related</a>), the current error comes from f_establish_ul_tbf() and says:</p>
<blockquote>
<p>Timeout establishing UL TBF</p>
</blockquote>
<p>From reading the test, this should happen:<br /><pre>
test -> virtphy tune_req
test -> virtphy est_req
test <- virtphy est_res *message should come in, but we get a timeout*
</pre></p>
<p>I'm still investigating where the tune_req/est_req/est_res messages get lost... if somebody has a hint where to look it would be great (<a class="user active" href="https://osmocom.org/users/67">fixeria</a>, <a class="user active" href="https://osmocom.org/users/7">laforge</a>, <a class="user active" href="https://osmocom.org/users/1741">lynxis</a> maybe?). I can't find any of them in wireshark, so I'm assuming that virtphy loses them. I've also tried to sniff the first socket with udtrace, but for some reason, during most runs, the ttcn3 dissector fails to output anything (sometimes it had worked though).</p> OsmoPCU - Bug #3925: Missing PCU_Tests.ttcn UL TBF testshttps://osmocom.org/issues/3925?journal_id=156602019-08-21T12:50:15Zosmith
<ul></ul> OsmoPCU - Bug #3925: Missing PCU_Tests.ttcn UL TBF testshttps://osmocom.org/issues/3925?journal_id=156712019-08-22T12:43:54Zosmith
<ul><li><strong>File</strong> <a href="/attachments/3833">PCU_Tests.TC_ul_tbf.merged.log</a> <a class="icon-only icon-download" title="Download" href="/attachments/download/3833/PCU_Tests.TC_ul_tbf.merged.log">PCU_Tests.TC_ul_tbf.merged.log</a> added</li></ul><p>The timeout does not happen because of any components outside of the testsuite, but because of the code in the testsuite itself, it seems. I've read through the TTCN3 logs again (decided to use the titan eclipse log viewer after a while to make it more feasible), and this is what happens:</p>
<pre>
MTC -> 5: BCCH_tune_req (pcs=false, arfcn=871, combined_ccch=true)
MTC -> 5: TBF_UL_establish_req (tbf_nr=0, ra=123)
MTC <- 5: BCCH_tune_req (pcs=false, arfcn=871, combined_cch=true, id=1)
5 -> socket: L1CTL_FBSB_REQ (pcs=false, arfcn=871, ...)
5 <- socket: L1CTL_FBSB_CONF (pcs=false, arfcn=871, ...)
MTC <- 5: TBF_UL_establish_req (tbf_nr=0, ra=123, id=2)
5 -> socket: L1CTL_RACH_REQ (ra=123, combined=1, offset=0, ...)
5 <- socket: L1CTL_RACH_CONF
after RACH_REQ, DATA IND like this start coming in (and they are not answered).
5 <- socket: L1CTL_DATA_IND (ch0=18, tn=0, ...)
5 <- socket: L1CTL_DATA_IND (ch0=24, tn=7, ...)
</pre>
<p>I've traced it in the code and found that it is stuck here:<br /><a class="external" href="https://git.osmocom.org/osmo-ttcn3-hacks/tree/library/L1CTL_PortType.ttcn?id=92c2bdb00fd840ec5fca6e49c12776146b2f4ef4#n129">https://git.osmocom.org/osmo-ttcn3-hacks/tree/library/L1CTL_PortType.ttcn?id=92c2bdb00fd840ec5fca6e49c12776146b2f4ef4#n129</a></p>
<p>After LAPDm_RAW_PT.ttcn:f_establish_tbf() receives L1CTL_FBSB_CONF, it calls f_L1CTL_WAIT_IMM_ASS() to wait for a L1CTL_DATA_IND that has an immediate assignment for the correct RACH.</p>
<pre>
timer T := 10.0;
T.start;
alt {
[] pt.receive(tr_L1CTL_DATA_IND(t_RslChanNr_PCH_AGCH(0))) -> value dl {
rr := dec_GsmRrMessage(dl.payload.data_ind.payload);
log("PCH/AGCH DL RR: ", rr);
if (match(rr, t_RR_IMM_ASS(ra, rach_fn))) {
log("Received IMM.ASS for our RACH!");
} else {
repeat;
}
};
[] pt.receive { repeat };
[] T.timeout {
setverdict(fail, "Timeout waiting for IMM ASS");
mtc.stop;
}
}
</pre>
<p>The "match(rr, t_RR_IMM_ASS(ra, rach_fn)" is always failing, hence it just repeats and never exits this function. The timeout here is set to 10 seconds, but the other timeout that results in "Timeout establishing UL TBF" is only 2 seconds. That is why "Timeout waiting for IMM ASS" is never printed to the log. I'm attaching the current log file ("PCU_Tests.TC_ul_tbf.merged.log") in case somebody wants to take a look at this.</p> OsmoPCU - Bug #3925: Missing PCU_Tests.ttcn UL TBF testshttps://osmocom.org/issues/3925?journal_id=156832019-08-23T07:28:46Zosmith
<ul></ul><p>I found that this the IMM ASS not getting sent is caused by a change (regression?) in osmo-bts. With old version osmo-bts 1.0.1 it gets past that part.</p>
<p>So I did a bisect. Rebasing on 1.0.1 before the bisect is necessary to have this patch in the tree and therefore prevent build failures: <a class="external" href="https://gerrit.osmocom.org/c/osmo-bts/+/13425/">https://gerrit.osmocom.org/c/osmo-bts/+/13425/</a><br /><pre>
$ git checkout master
$ git checkout -b bisect-me
$ git rebase 1.0.1
$ git bisect start HEAD 1.0.1
$ git bisect run ~/code/temp/3925/bisect_imm_ass.sh
</pre></p>
<p>This is the commit that caused the behavior change:<br /><a class="external" href="https://gerrit.osmocom.org/c/osmo-bts/+/14687">https://gerrit.osmocom.org/c/osmo-bts/+/14687</a> ("Move Access Burst link quality handling to L1SAP")</p> OsmoPCU - Bug #3925: Missing PCU_Tests.ttcn UL TBF testshttps://osmocom.org/issues/3925?journal_id=156842019-08-23T07:47:37Zosmith
<ul></ul><p>The above patch moves the check for link quality. And indeed, the bts log says:</p>
<blockquote>
<p>20190823093233821 DL1C <0006> ../../../../src/osmo-bts/src/common/l1sap.c:1279 000881/00/23/14/09 Ignoring RACH request: link quality (0) below the minimum (50)</p>
</blockquote>
<p>So the testsuite needs to be adjusted to send a proper link quality :)</p> OsmoPCU - Bug #3925: Missing PCU_Tests.ttcn UL TBF testshttps://osmocom.org/issues/3925?journal_id=156862019-08-23T10:19:47Zosmith
<ul></ul><p>The change needs to be done in osmo-bts actually, and it is a regression from above commit (where the necessary change was made for all other osmo-bts-* variants, but not for virtual). This patch fixes it, and makes the IMM ASS part pass: <a class="external" href="https://gerrit.osmocom.org/c/osmo-bts/+/15271">https://gerrit.osmocom.org/c/osmo-bts/+/15271</a></p> OsmoPCU - Bug #3925: Missing PCU_Tests.ttcn UL TBF testshttps://osmocom.org/issues/3925?journal_id=156872019-08-23T10:30:02Zfixeria
<ul></ul><p>Hi Oliver,</p>
<blockquote>
<p>This is the commit that caused the behavior change:<br /><a class="external" href="https://gerrit.osmocom.org/c/osmo-bts/+/14687">https://gerrit.osmocom.org/c/osmo-bts/+/14687</a> ("Move Access Burst link quality handling to L1SAP")</p>
</blockquote>
<p>Sorry, my bad.</p>
<blockquote>
<p>So the testsuite needs to be adjusted to send a proper link quality :)</p>
</blockquote>
<p>Link quality is defined by C/I ratio. By default, the test cases should be using 60 cB:</p>
<p><a class="external" href="https://gerrit.osmocom.org/c/osmo-ttcn3-hacks/+/14961">https://gerrit.osmocom.org/c/osmo-ttcn3-hacks/+/14961</a></p>
<blockquote>
<p>[...] Ignoring RACH request: link quality (0) below the minimum (50)</p>
</blockquote>
<p>It's interesting why the test case is using 0... The recent Wireshark can help you to investigate (dissectors osmo_trxd and osmo_trxc).</p> OsmoPCU - Bug #3925: Missing PCU_Tests.ttcn UL TBF testshttps://osmocom.org/issues/3925?journal_id=156882019-08-23T10:30:38Zfixeria
<ul></ul><p>Ah, you're not using fake_trx/trxcon, nevermind.</p> OsmoPCU - Bug #3925: Missing PCU_Tests.ttcn UL TBF testshttps://osmocom.org/issues/3925?journal_id=156892019-08-23T12:57:43Zosmith
<ul></ul><p>Thanks for the input, Vadim!</p>
So with the osmo-bts-virtual fix, these issues are resolved:
<ul>
<li>IMM ASS not received</li>
<li>Timeout establishing UL TBF</li>
</ul>
<p>TC_ul_tbf is <em>almost</em> completely running through now. The next issue is this one:</p>
<blockquote>
<p>Timeout waiting for { pDU_BSSGP_UL_UNITDATA := { bssgpPduType := '01'O, tLLI := '452BD97E'O, qoS_Profile := ?, cell_Identifier := ?, pFI := *, lSA_Identifier_List := *, redirect_Attempt_Flag := *, iMSI_BSSGP := *, unconfirmed_Send_State_Variable := *, selected_PLMN_ID := *, selected_Operator := *, cS_Registered_Operator := *, alignment_octets := *, lLC_PDU := { iEI := '0E'O, ext := ?, lengthIndicator := ?, lLC_PDU := '01C0...'O } } }</p>
</blockquote>
<p>The error comes from here in PDU_Tests.ttcn:<br /><pre>
function f_bssgp_wait_ul_ud(template PDU_BSSGP exp) runs on dummy_CT {
timer T := 5.0;
T.start;
alt {
[] BSSGP[0].receive(exp) {
log("found matching BSSGP UL-UNITDATA PDU");
}
[] T.timeout {
setverdict(fail, "Timeout waiting for ", exp);
mtc.stop;
}
}
}
</pre></p>
<p>Wireshark shows this when filtered for bssgp:</p>
<pre>
...
FLOW-CONTROL-BVC
FLOW-CONTROL-BVC-ACK
SAPI: LLGMM, UI, unprotected, non-ciphered information, N(U) = 0(DTAP)
SAPI: LLGMM, UI, unprotected, non-ciphered information, N(U) = 1(DTAP)
SAPI: LLGMM, UI, unprotected, non-ciphered information, N(U) = 2(DTAP)
SAPI: LLGMM, UI, unprotected, non-ciphered information, N(U) = 3(DTAP)
SAPI: LLGMM, UI, unprotected, non-ciphered information, N(U) = 4(DTAP)
FLOW-CONTROL-BVC
FLOW-CONTROL-BVC-ACK
SAPI: LLGMM, UI, unprotected, non-ciphered information, N(U) = 5(DTAP)
</pre>
<p>I'll continue to investigate next week.</p> OsmoPCU - Bug #3925: Missing PCU_Tests.ttcn UL TBF testshttps://osmocom.org/issues/3925?journal_id=156962019-08-26T09:12:15Zosmith
<ul><li><b>Checklist item</b> <input type='checkbox' class='checklist-checkbox' disabled> <i>Proper fix for byte-swapped cell-id in testsuite</i> added</li><li><b>Checklist item</b> <input type='checkbox' class='checklist-checkbox' disabled> <i>Proper fix for f_bssgp_wait_ul_ud()</i> added</li><li><b>Checklist item</b> <input type='checkbox' class='checklist-checkbox' disabled> <i>Get TC_ul_tbf running in jenkins</i> added</li><li><b>Checklist item</b> <input type='checkbox' class='checklist-checkbox' disabled> <i>Fix running multiple tests after another</i> added</li><li><b>Checklist item</b> <input type='checkbox' class='checklist-checkbox' disabled> <i>Get TC_paging running in jenkins</i> added</li><li><b>Checklist item</b> <input type='checkbox' class='checklist-checkbox' disabled> <i>Get TC_selftest_ns running locally</i> added</li><li><b>Checklist item</b> <input type='checkbox' class='checklist-checkbox' disabled> <i>Get TC_selftest_ns running in jenkins</i> added</li><li><b>Checklist item</b> <input type='checkbox' class='checklist-checkbox' disabled> <i>Get TC_ul_tbf_single_llc_sizes running locally</i> added</li><li><b>Checklist item</b> <input type='checkbox' class='checklist-checkbox' disabled> <i>Get TC_ul_tbf_single_llc_sizes running in jenkins</i> added</li><li><b>Checklist item</b> <input type='checkbox' class='checklist-checkbox' disabled> <i>Develop new tests (using H/L annotation in RAW codec for CSN.1, this checklist item will be split up further)</i> added</li></ul><p>The above "Timeout waiting for { pDU_BSSGP_UL_UNITDATA" error was caused by a regression from: <a class="external" href="https://gerrit.osmocom.org/c/osmo-ttcn3-hacks/+/13865">https://gerrit.osmocom.org/c/osmo-ttcn3-hacks/+/13865</a> ("BSSGP_Emulation: Abandon "BssgpDecoded" intermediate structure")</p>
<p>After reverting it, I have <ins>TC_ul_tbf test passing for the first time</ins> \o/</p>
So there are four existing tests in total, and now two of them are passing:
<ul>
<li>TC_ul_tbf (OK)</li>
<li>TC_selftest_ns (FAIL)</li>
<li>TC_ul_tbf_single_llc_sizes (FAIL)</li>
<li>TC_paging (OK)</li>
</ul>
<p>Unfortunately, when running multiple tests after another, the tests will also fail (so the shutdown isn't done properly it seems).</p>
<p>I think it makes sense to create proper fixes for the two workarounds I have right now, and then get TC_ul_tbf running in jenkins. That way what I have locally won't divide too much from what we have working in jenkins. From there on, I'll get the two other existing tests running, and then develop new tests. Checklist updated accordingly.</p> OsmoPCU - Bug #3925: Missing PCU_Tests.ttcn UL TBF testshttps://osmocom.org/issues/3925?journal_id=157022019-08-26T13:36:50Zosmith
<ul><li><b>Checklist item</b> changed from <i>Proper fix for byte-swapped cell-id in testsuite</i> to <i>Proper fix for byte-swapped cell-id (OS#3854)</i></li></ul><p>The byte-swapped cell-id is not an issue in the testsuite, but in another component in the chain (likely osmo-bts with code from libosmocore?). See <a class="issue tracker-1 status-3 priority-2 priority-default closed" title="Bug: OsmoPCU uses wrong CellID in BSSGP (Resolved)" href="https://osmocom.org/issues/3854">#3854</a> for details.</p> OsmoPCU - Bug #3925: Missing PCU_Tests.ttcn UL TBF testshttps://osmocom.org/issues/3925?journal_id=157042019-08-26T13:37:23Zosmith
<ul><li><strong>Related to</strong> <i><a class="issue tracker-1 status-3 priority-2 priority-default closed" href="/issues/3854">Bug #3854</a>: OsmoPCU uses wrong CellID in BSSGP</i> added</li></ul> OsmoPCU - Bug #3925: Missing PCU_Tests.ttcn UL TBF testshttps://osmocom.org/issues/3925?journal_id=157432019-08-29T12:43:48Zosmith
<ul><li><b>Checklist item</b> <input type='checkbox' class='checklist-checkbox' checked disabled> <i>Proper fix for byte-swapped cell-id (OS#3854)</i> set to Done</li></ul> OsmoPCU - Bug #3925: Missing PCU_Tests.ttcn UL TBF testshttps://osmocom.org/issues/3925?journal_id=157502019-08-29T15:46:26Zosmith
<ul><li><strong>Related to</strong> <i><a class="issue tracker-1 status-3 priority-2 priority-default closed" href="/issues/4180">Bug #4180</a>: Regression in osmo-ttcn3-hacks broke f_bssgp_wait_ul_ud()</i> added</li></ul> OsmoPCU - Bug #3925: Missing PCU_Tests.ttcn UL TBF testshttps://osmocom.org/issues/3925?journal_id=157522019-08-29T15:46:46Zosmith
<ul><li><b>Checklist item</b> changed from <i>Proper fix for f_bssgp_wait_ul_ud()</i> to <i>Proper fix for f_bssgp_wait_ul_ud() (OS#4180)</i></li><li><b>Checklist item</b> <input type='checkbox' class='checklist-checkbox' checked disabled> <i>Proper fix for f_bssgp_wait_ul_ud() (OS#4180)</i> set to Done</li></ul> OsmoPCU - Bug #3925: Missing PCU_Tests.ttcn UL TBF testshttps://osmocom.org/issues/3925?journal_id=158562019-09-04T10:49:46Zosmith
<ul></ul><p>I have gotten the TC_ul_tbf test running docker with jenkins.sh. The biggest obstacle was not having multicast support in docker across multiple containers... but I've worked around that by running virtphy and osmo-bts-virtual in the same container. Pushed to docker-playground.git, osmith/pcu-tests branch. It works, when just running this test.</p>
<p>However, running the existing PCU_Tests_RAW before this test, it gives a "Timeout establishing BSSGP connection" error again. So I will look into proper reset before starting tests next, and when this is resolved, we can run the two working tests in jenkins.</p> OsmoPCU - Bug #3925: Missing PCU_Tests.ttcn UL TBF testshttps://osmocom.org/issues/3925?journal_id=158612019-09-04T14:01:47Zosmith
<ul></ul><p>Regarding the proper reset, I've taken a look at the SGSN code (as suggested by <a class="user active" href="https://osmocom.org/users/1741">lynxis</a>, thanks!) and found that it issues a reset command in f_init_vty:<br /><pre>
f_vty_transceive(SGSNVTY, "reset sgsn state");
</pre></p>
<p>I'm trying to run similar reset commands now with the PCU, before running each test. These look promising (1234 would be inserted from the test config):<br /><pre>
nsvc nsei 1234 reset
nsvc nsvci 1234 reset
</pre></p>
<p>The PCU tests can't access the OsmoPCU VTY so far, so I've added that first (adjusted the gen_links.sh, regen_makefile.sh etc.). Using the VTY did not work out of the box, because the VTY prompt of OsmoPCU is "Osmo-PCU> " and not "OsmoPCU> ", which does not match the "\w+" pattern used in the TTCN-3 VTY code. After patching it in OsmoPCU, talking to the VTY is working. The reset commands above are not enough it seems, but I'll continue here tomorrow.</p> OsmoPCU - Bug #3925: Missing PCU_Tests.ttcn UL TBF testshttps://osmocom.org/issues/3925?journal_id=158762019-09-05T15:28:50Zosmith
<ul><li><strong>File</strong> <a href="/attachments/3850">running-multiple-tests-broken.tar.xz</a> <a class="icon-only icon-download" title="Download" href="/attachments/download/3850/running-multiple-tests-broken.tar.xz">running-multiple-tests-broken.tar.xz</a> added</li></ul><p><a class="user active" href="https://osmocom.org/users/7">laforge</a>, as discussed, here are logs and pcaps of running multiple tests after one another:</p>
<p>1. running just TC_ul_tbf (OK)<br />2. running TC_ul_tbf twice (NOK)<br />3. running PCU_Tests_RAW.TC_pcuif_suspend, then TC_ul_tbf (similar to how it would run in jenkins) (NOK)</p> OsmoPCU - Bug #3925: Missing PCU_Tests.ttcn UL TBF testshttps://osmocom.org/issues/3925?journal_id=158772019-09-05T16:37:35Zlaforge
<ul></ul>The log file is quite clear:
<ul>
<li>the PCU continues "as usual" and sends FLOW-CONTROL-BVC</li>
<li>there is an ICMP destination unreachable which appears to be ignored by the PCU
<ul>
<li>normally I'd expect this translates to some kind of error behavior on send/recv on the UDP socket? If the PCU would get notified by this, it could BLOCK the NS-VC and BVC and start sending NS_RESET again</li>
</ul>
</li>
<li>thre is no response to the FLOW-CONTROL-BVC
<ul>
<li>the PCU doesn't appear to check if it gets ACKs, or at least not detect the absence of ACKS</li>
<li>detecting the absence of ACKs could also be used to BLOCK the NS-VC + BVC and start sending NS_RESET again</li>
</ul>
</li>
<li>there is no NS_ALIVE procedure within the ~ 20 seconds of the test. This procedure is responsible for checking if the link is still active.
<ul>
<li>not sure what the default timeout is, but I remember those are tuneable, so the PCU config file could probably set a lower timeout, causing NS_ALIVE to be sent e.g. every 10s, which would allow the PCU to detect that the link is down / SGSN is restarted</li>
</ul></li>
</ul> OsmoPCU - Bug #3925: Missing PCU_Tests.ttcn UL TBF testshttps://osmocom.org/issues/3925?journal_id=158782019-09-05T17:08:38Zlaforge
<ul></ul>Quick look at the PCU and libosmogb reveals:
<ul>
<li>GPRS-NS-ALIVE timer can be configured via VTY</li>
<li>expiration of NS-ALIVE would trigger S_NS_BLOCK to be dispatched to PCU</li>
<li>PCU would handle S_NS_BLOCK with an internal reset of most of its Gb related state
<ul>
<li>however, no NS-RESET would be sent,as the only code path to this is when PCU calls gprs_ns_nsip_connect()</li>
</ul></li>
</ul>
So I guess the PCU should be starting to send NS-RESET in suhc situations. Basically, there should be a state machine where
<ul>
<li>it first sends a NS-RESET, wait for NS-RESET-ACK, and keeps re-transmitting NS-RESET until it receives an ack</li>
<li>once it has received the RESET-ACK it transitions into ALIVE, BLOCKED and starts the NS-ALIVE procedure</li>
<li>it then tries to UNBLOCK it and when it's unblocked, it continues with BSSGP bring-up.</li>
</ul>
<p>If the PCU and/or libosmogb would stick to that proper state machine, I would guess running multiple tests wouldn't be as much of a problem.</p>
<p>What's a bit annoying about all of this is that if using the more modern and spec-compatible IP-SNS (which we only support on the PCU side so far, but not on the SGSN), all of those procedures don't exist anymore anyway. So we could simply chose to run the PCU tests in the IP-SNS dialect of Gb, which should save us from a lot of trouble. But then we still don't have proper functionality of the PCU in the (by now mostly used) non-SNS case.</p> OsmoPCU - Bug #3925: Missing PCU_Tests.ttcn UL TBF testshttps://osmocom.org/issues/3925?journal_id=167522019-12-05T18:23:15Zpespin
<ul></ul><p>I think this ticket links with my last comment in <a class="external" href="https://gerrit.osmocom.org/c/osmo-ttcn3-hacks/+/16321">https://gerrit.osmocom.org/c/osmo-ttcn3-hacks/+/16321</a>:<br />Do we want to drop old test infra?</p> OsmoPCU - Bug #3925: Missing PCU_Tests.ttcn UL TBF testshttps://osmocom.org/issues/3925?journal_id=167582019-12-06T07:33:42Zosmith
<ul></ul><p>pespin wrote:</p>
<blockquote>
<p>I think this ticket links with my last comment in <a class="external" href="https://gerrit.osmocom.org/c/osmo-ttcn3-hacks/+/16321">https://gerrit.osmocom.org/c/osmo-ttcn3-hacks/+/16321</a>:<br />Do we want to drop old test infra?</p>
</blockquote>
<p>As written in gerrit, I think it makes sense.</p>
<p><a class="user active" href="https://osmocom.org/users/7">laforge</a>, what do you think?</p> OsmoPCU - Bug #3925: Missing PCU_Tests.ttcn UL TBF testshttps://osmocom.org/issues/3925?journal_id=182852020-05-12T12:18:42Zlaforge
<ul><li><strong>Assignee</strong> changed from <i>osmith</i> to <i>pespin</i></li></ul><p><a class="user active" href="https://osmocom.org/users/30187">pespin</a> please check how much of this is still relevant and change checklist accordingly and/or resolve ticket</p> OsmoPCU - Bug #3925: Missing PCU_Tests.ttcn UL TBF testshttps://osmocom.org/issues/3925?journal_id=189552020-06-26T12:42:14Zpespin
<ul><li><strong>Status</strong> changed from <i>New</i> to <i>Closed</i></li></ul><p>Closing this ticket now as it's too generic and we already have some infra working and test passing.<br />Let's create more specific tickets about this topic from now on.</p>