#opnfv-testperf: Functest weekly meeting February 16th
Meeting started by jose_lausuch at 08:04:53 UTC
(full logs).
Meeting summary
-
- Juha Kosonen (juhak,
08:05:09)
- Jose Lausuch (jose_lausuch,
08:05:13)
- action points follow-up (jose_lausuch, 08:07:05)
- previous minutes
http://ircbot.wl.linuxfoundation.org/meetings/opnfv-testperf/2016/opnfv-testperf.2016-02-09-08.00.html
(jose_lausuch,
08:07:16)
- jose_lausuch AP1: vPing troubleshooting on
remaining failed scenario: still ongoing, troubleshooting done on
Apex installer, where it fails, for the other scenarios with ODL-L3
looks good (jose_lausuch,
08:08:11)
- jose_lausuch AP2: jose_lausuch create a summary
at the end of rally run. DONE. (jose_lausuch,
08:08:31)
- juhak AP1: patch for heat to use good ext
network (all the installers have different names) (jose_lausuch,
08:08:43)
- Viktor Tikkanen (viktor_nokia,
08:09:24)
- DONE:
https://gerrit.opnfv.org/gerrit/#/c/9625/ (jose_lausuch,
08:09:26)
- juhak AP2: exclude 3 test cases related to
live migration and document in the user guide .... (jose_lausuch,
08:09:42)
- DONE:
https://gerrit.opnfv.org/gerrit/#/c/9867/ (jose_lausuch,
08:10:52)
- morgan_orange: config of gsutils not correct on
Orange POD 2 ... there is an ongoing discussion about if we should
enable that on non CI PODs... for orange-pod2 to be discussed with
trevor (jose_lausuch,
08:11:57)
- pending patch to avoid pushing if the POD is
not a CI POD : https://gerrit.opnfv.org/gerrit/#/c/10025/
(jose_lausuch,
08:12:46)
- morgan_orange AP2: test workaround no
parrallel on apex/LF POD1 (jose_lausuch,
08:13:13)
- --serial argument added to run_functest to run
tempest in serial instead of parallel. (jose_lausuch,
08:13:51)
- morgan_orange AP3: troubleshoot ODL scenario to
better undestand the difference of errors. Not sure of the progress
here.. maybe he can comment next week (jose_lausuch,
08:15:58)
- morgan_orange AP4: based on this summary pushed
in DB add rally in dashboard. DONE (jose_lausuch,
08:18:41)
- viktor_nokia AP1: troubleshooting Tempest, with
at least access to logs on the artifact (jose_lausuch,
08:18:56)
- https://jira.opnfv.org/browse/FUNCTEST-136 and
https://jira.opnfv.org/browse/FUNCTEST-137 to track the tempest
progress (jose_lausuch,
08:21:05)
- Release B (jose_lausuch, 08:24:19)
- B release date 25th Feb (next week)
(jose_lausuch,
08:24:40)
- the move to ODL Be give in general better
results than Li (jose_lausuch,
08:25:11)
- new Dashboard look:
http://testresults.opnfv.org/dashboard/ (jose_lausuch,
08:25:45)
- almost perfect results for Compass
installer (jose_lausuch,
08:26:51)
- https://build.opnfv.org/ci/view/functest/job/functest-compass-huawei-us-deploy-bare-1-daily-brahmaputra/
(jose_lausuch,
08:27:06)
- for joid, almost perfect results on orange
pod2, but its not an "official" CI POD .. so we should rely on
intel-pod5 in theory, but the results there are so bad..and I think
they are due to some pod missconfiguration or something (jose_lausuch,
08:28:25)
- http://artifacts.opnfv.org/logs/functest/huawei-us-deploy-bare-1/brahmaputra/2016-02-16_06-09-04/brahmaputra/tempest/tempest.log
(jose_lausuch,
08:30:13)
- 10 tempest cases fail always with
os-onos-nofeature-ha scenario in compass (viktor_nokia,
08:31:05)
- One theory is that this is because "user data"
is not supported there (viktor_nokia,
08:31:56)
- tempest 0 failures in compass when using ODL
scenario, but those 10-11 errors for ONOS scenario (jose_lausuch,
08:33:57)
- ACTION: viktor_nokia:
check with ONOS experts the failed tempest results, probably due to
user-data support (jose_lausuch,
08:36:19)
- fuel: almost green, but need to see if the
dea.yaml change to increase the subnet IP range fixes the Tempest
failed test cases (jose_lausuch,
08:38:18)
- apex: installer not stable, problems with
external network creating. To be tested again today onwards.
(jose_lausuch,
08:38:52)
- summary of all the tests/scenario/installers
here: https://wiki.opnfv.org/functextnexttaks (jose_lausuch,
08:39:19)
- Functest meeting in Helsinki (jose_lausuch, 08:41:18)
- date 30th March in Espoo, Finland (jose_lausuch,
08:42:12)
- the number of places is limited to 15, but
contact viktor_nokia or juhak for more details (jose_lausuch,
08:42:45)
- Topics to be discussed:
https://etherpad.opnfv.org/p/functest_release_C (viktor_nokia,
08:43:02)
- anyone is invited, the goal is to decide the
Release C Functest goals and milestones (jose_lausuch,
08:43:04)
- anyone who cannot attend, can post also the
ideas in the etherpad or by email, we will discuss it for
sure (jose_lausuch,
08:45:49)
- Berlin (jose_lausuch, 08:47:33)
- feel free everyone to post a
presentation (jose_lausuch,
08:47:42)
- possible JAM session to be prepared this time
for the summit. Any volunteer at the durms/bass/vocals? :)
(jose_lausuch,
08:51:14)
Meeting ended at 08:53:41 UTC
(full logs).
Action items
- viktor_nokia: check with ONOS experts the failed tempest results, probably due to user-data support
Action items, by person
- viktor_nokia
- viktor_nokia: check with ONOS experts the failed tempest results, probably due to user-data support
People present (lines said)
- jose_lausuch (119)
- viktor_nokia (35)
- juhak (7)
- collabot` (6)
- Serena-zte (5)
- raghavendrachari (3)
Generated by MeetBot 0.1.4.