Just now I looked at the failures in ttcn3-bsc-test-latest.
https://jenkins.osmocom.org/jenkins/job/ttcn3-bsc-test-latest/
TL;DR: we should roll a release.
Two real failures happen that are actual bugs in the latest release,
and the rest of the failures are due to the lingering subscriber leak.
BSC_Tests.TC_ho_into_this_bsc
fails because osmo-bsc-latest omits the 'Codec List (BSS Supported)' in the
Handover Request Acknowledge message, which is mandatory on AoIP. This error is
fixed on osmo-bsc master. It makes no sense to me to punch a hole in the tests
for latest -- this failure shows a real failure of the current osmo-bsc
release.
BSC_Tests.TC_cm_service_during_lcs_loc_req
shows a subscriber leak, where a bsc_subscr is not cleaned up. This bsc_subscr
lingers all the way to the end of the BSC_Tests run, and all other current
failures of ttcn3-bsc-test-latest are due to that. (The only one not failing is
TC_refuse_chan_act_to_vamos(), because it is the only one not verifying that no
bsc_subscr stay behind.)
So this also shows a real error in the current osmo-bsc release.
There is nothing I would do at this point to make bsc-test-latest appear to be
passing, because the errors shown are errors.
The conclusion is that it might be time to roll another release one of these
days.
Related: important fixes I recently submitted are all merged to master, so from
my side it would be a good time.
~N
--
- Neels Hofmeyr <nhofmeyr(a)sysmocom.de>
http://www.sysmocom.de/
=======================================================================
* sysmocom - systems for mobile communications GmbH
* Alt-Moabit 93
* 10559 Berlin, Germany
* Sitz / Registered office: Berlin, HRB 134158 B
* Geschäftsführer / Managing Directors: Harald Welte