08:00:55 <ollivier> #startmeeting Functest weekly meeting 17 July. 2018
08:00:55 <collabot> Meeting started Tue Jul 17 08:00:55 2018 UTC.  The chair is ollivier. Information about MeetBot at http://wiki.debian.org/MeetBot.
08:00:55 <collabot> Useful Commands: #action #agreed #help #info #idea #link #topic.
08:00:55 <collabot> The meeting name has been set to 'functest_weekly_meeting_17_july__2018'
08:01:09 <ollivier> #info Cédric Ollivier
08:01:22 <viktor_t> #info Viktor Tikkanen
08:03:40 <ollivier> depo: depo_ : Hello.
08:05:42 <ollivier> #topic Action point follow-up
08:05:59 <ollivier> #info all testcases were declared in the DB
08:06:20 <ollivier> #info new bugs opened by end users
08:06:33 <ollivier> #link https://jira.opnfv.org/browse/FUNCTEST-992
08:06:41 <ollivier> #link https://jira.opnfv.org/browse/FUNCTEST-991
08:06:49 <ollivier> #link https://jira.opnfv.org/browse/FUNCTEST-993
08:07:38 <ollivier> We are asked to support vmdk. It should already work for all testcases except vnfs.
08:08:45 <ollivier> The other part is to support vnfs in ovs dpdk scenarios.
08:09:45 <ollivier> Then it makes sense to update our filters (for the time being vnfs are executed when nofeature)
08:10:40 <ollivier> I'm updating cloudify ims to leverage on Cloudify (then its config can be easily patched for vmdk)
08:13:58 <ollivier> #topic Functest Fraser
08:14:36 <ollivier> #info it's working well except the new bug about juju_epc (this morning)
08:15:00 <ollivier> #link https://build.opnfv.org/ci/view/functest/job/functest-compass-baremetal-daily-master/185/console
08:15:28 <ollivier> I will ask support for Rebaca team
08:16:06 <viktor_t> OK. Do we have any other Fraser activities in functest?
08:17:39 <ollivier> No. We may simply backport the next changes about vnfs. It would allow patching image fomat format, floavor specs, etc.
08:18:15 <ollivier> The issue about juju_epc is coming from upstream (ubuntu/juju).
08:19:47 <ollivier> #topic Functest Gambia
08:20:06 <ollivier> #info several api_check and snaps_smoke subtests have been temporarily disabled
08:20:34 <ollivier> Steven fixed few of them yesterday's evening and we may enable them asap.
08:21:01 <ollivier> #link https://jira.opnfv.org/browse/SNAPS-320
08:21:15 <ollivier> #link https://jira.opnfv.org/browse/SNAPS-322
08:21:24 <ollivier> #link https://gerrit.opnfv.org/gerrit/#/c/59801/
08:22:38 <ollivier> #info new testcases: shaker and vmtp are working well
08:22:50 <ollivier> #link http://artifacts.opnfv.org/logs/functest/huawei-pod1/2018-07-16_11-03-58/shaker/report.html
08:23:04 <ollivier> #link http://artifacts.opnfv.org/logs/functest/huawei-pod1/2018-07-16_11-03-58/vmtp/vmtp.html
08:24:42 <ollivier> #info Functest verified Compass successfully (all errors are about to Compass)
08:25:08 <ollivier> #link https://build.opnfv.org/ci/view/functest/job/functest-compass-baremetal-daily-master/lastFailedBuild/console
08:25:25 <ollivier> #link https://jira.opnfv.org/browse/COMPASS-605
08:25:38 <viktor_t> I've checked briefly COMPASS-605 issue
08:25:53 <viktor_t> There is an existing open bug report for tempest: https://bugs.launchpad.net/tempest/+bug/1710845
08:27:13 <ollivier> ok. In that case, we do skip it (and in Fraser as well).
08:28:39 <ollivier> It's not fully confirmed. Do we blacklist it?
08:29:04 <viktor_t> I'm not sure...
08:29:21 <ollivier> ok. Then we can wait. I think it only fails for Compass.
08:30:12 <viktor_t> But it looks like tempest doesn't wait for some status info (like if some port is UP) before proceeding
08:30:42 <ollivier> Compass is known as slower than the other installers.
08:30:48 <ollivier> (POD related?)
08:30:54 <viktor_t> is ir permanent failure (occurs every time)?
08:30:58 <ollivier> It's working very well here.
08:31:51 <ollivier> I think it failed for nofeature and worked for bar scenario.
08:32:41 <ollivier> no I'm mixing with rally_sanity
08:33:08 <viktor_t> I think it makes sense to blacklist it until the bug is fixed or marked as invalid
08:33:24 <ollivier> yes but the bug is not really confirmed.
08:34:06 <ollivier> We have to check if it runs in OpenStack gates.
08:34:18 <viktor_t> I can try to debug it though I need access to the POD
08:36:06 <ollivier> Thank you. Harry Huang may help you regarding the access
08:36:33 <viktor_t> OK, got it
08:37:10 <ollivier> If you confirm it, be free to blacklist it by yourself during my holidays.
08:37:55 <ollivier> #info All testcases are being verified in Orange
08:38:09 <ollivier> only juju_epc and tempest_full may be buggy
08:38:15 <ollivier> lots of tempest tests stay in init state
08:39:09 <ollivier> (about half of the testcase list)
08:39:40 <ollivier> It would be great if you could have a look to tempest_full as well.
08:40:15 <viktor_t> OK, I will check it
08:40:24 <ollivier> thank you.
08:40:45 <ollivier> #info Tempest-based testcases can be dynamically skipped (services or neutron extensions)
08:41:03 <ollivier> It mainly allows removing INSTALL_TYPE dependencies.
08:42:25 <ollivier> Else Barbican ran out of OPNFV installers even if uninstall.
08:42:52 <ollivier> shaker could be updated as well (it has to be skipped if only one compute)
08:43:26 <depo_> #info Delia Popescu
08:43:31 <depo_> Hi! Sorry for beig late
08:43:49 <ollivier> depo_: hello
08:46:03 <ollivier> Then our master branch is already working very well. We are reusing it internally (Orange ONAP Openlab)
08:47:00 <ollivier> depo_: AlexAvadanii: could you please have a look to the MTU if Fuel. Vtmp is failing when packet size is 1500.
08:47:15 <depo_> I will look into it
08:47:37 <ollivier> thank you. I think the testcase is working very well.
08:47:54 <ollivier> #topic AoB
08:48:18 <ollivier> I would propose to reschedule the meeting. Maybe Valentin can join us (Canada)
08:48:38 <ollivier> Nobody from Asia is contibuting.
08:50:50 <ollivier> Nothing else from my side. I will be on holidays for the next weeks.
08:51:05 <ollivier> I will do my best to update cloudify_ims and chack cinder ut before leaving :)
08:51:31 <depo_> ollivier: thank you very much!
08:51:42 <viktor_t> Do you have any proposals for the new time? Tue 14.00 UTC?
08:52:52 <ollivier> yes. It could be great. I will send an email to all (and I would like to wait Juha too)
08:53:21 <viktor_t> ok
08:53:40 <ollivier> Anything else?
08:53:51 <viktor_t> Not here...
08:54:01 <depo_> no
08:54:19 <ollivier> #endmeeting