09:11:34 <gmann> #startmeeting qa 09:11:35 <openstack> Meeting started Thu Dec 6 09:11:34 2018 UTC and is due to finish in 60 minutes. The chair is gmann. Information about MeetBot at http://wiki.debian.org/MeetBot. 09:11:36 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 09:11:38 <openstack> The meeting name has been set to 'qa' 09:11:43 <gmann> who all here today? 09:14:06 <mbindlish> hi 09:14:27 <mbindlish> @gmann 09:14:28 <gmann> hi 09:15:21 <gmann> mbindlish: seems 2 of us 09:15:27 <mbindlish> hmm 09:15:49 <gmann> we can discuss on bug things if you are ready with that ? 09:16:01 <mbindlish> okk 09:16:06 <mbindlish> sharing link 09:16:15 <mbindlish> https://etherpad.openstack.org/p/QA-Stein-Bug-Triage 09:16:54 <mbindlish> There are multiple issues, there is no update on those, even after asking for update 09:17:02 <mbindlish> So what to do with those cases? 09:17:06 <gmann> seems like number are increasing instead of decreasing :) 09:17:18 <mbindlish> yes 09:17:20 <gmann> we can discuss here 09:17:34 <mbindlish> so we can go one by one 09:18:01 <mbindlish> https://bugs.launchpad.net/tempest/+bug/1676207 09:18:03 <openstack> Launchpad bug 1676207 in tempest "test_create_router_set_gateway_with_fixed_ip is flaky" [High,New] 09:18:50 <mbindlish> No update on this 09:20:00 <mbindlish> @gmann , so what can we do with such item? 09:20:10 <frickler> slawaq last commented on that one, maybe he wants to continue? 09:20:14 <frickler> also hi ;) 09:20:43 <gmann> yeah, mbindlish what needed on this? 09:21:01 <gmann> i think we need modification in tests to avoid that race 09:22:39 <mbindlish> hmmm 09:22:50 <mbindlish> its a race condition here 09:23:06 <mbindlish> we can add some wait 09:23:43 <gmann> wait will create more time for other tests to get that ip 09:23:54 <frickler> well, the "long" solution would be not to take an IP from the public pool, but use some dedicated network 09:24:13 <frickler> that would avoid any possible race 09:24:43 <frickler> slaweq seems to be offline, I'll discuss with him once he returns 09:24:44 <gmann> you mean create network in test itself and create port from it 09:24:49 <frickler> gmann: yes 09:25:00 <gmann> work for me. 09:25:17 <frickler> gmann: should be possible to take the new network as gateway net instead of the default public 09:25:36 <gmann> frickler: thanks we can wait for slaweq opinion 09:25:55 <gmann> yeah, that should be ok for this tests scope 09:26:21 <mbindlish> 2nd: https://bugs.launchpad.net/tempest/+bug/1672613 09:26:22 <openstack> Launchpad bug 1672613 in tempest "Read or write object with rights tests broken" [Low,New] 09:26:49 <mbindlish> I tried to reproduce it but it was not reproducible to me 09:27:06 <gmann> mbindlish: which one - second is L21 - https://bugs.launchpad.net/tempest/+bug/1806089 09:27:08 <openstack> Launchpad bug 1806089 in tempest "tempest cleanup fails on py35" [Undecided,New] 09:27:35 <gmann> sorry that is first one 09:27:46 <gmann> did you find why it is failing on y35 ? 09:28:03 <gmann> it means we are missing this CLI unit tests 09:28:31 <mbindlish> not checked this issue yet 09:28:34 <gmann> it should be easy to fix 09:28:38 <gmann> ok 09:29:48 <frickler> mbindlish: for 1672613 I'd set it to incomplete, then it will expire automatically if no response is received withing 60 days 09:30:10 <mbindlish> frickler: fine 09:30:13 <frickler> I do the same for devstack bugs regularly, cleans up about 99% of them 09:30:26 <mbindlish> :+1 09:31:06 <gmann> did you check this https://review.openstack.org/#/c/445358/ 09:31:16 <gmann> this bug has fix and then it got abandon 09:32:19 <gmann> if acl is not maintained with v2 and v3 token and it is supposed to work in compatible way then, we can raise issue swift side 09:32:40 <frickler> oh, seems that was fixed in I79e5754379e3cf72d2a9858e6e69e8c190b0b2ed 09:33:15 <frickler> that's why it cannot be reproduced anymore probably 09:33:50 <gmann> frickler: nice. that i remember now 09:34:30 <gmann> commented on bug 09:34:33 <gmann> next 09:34:40 <mbindlish> Should we mark as fix to this bug? 09:35:56 <mbindlish> Next: https://bugs.launchpad.net/tempest/+bug/1703405 09:35:57 <openstack> Launchpad bug 1703405 in tempest "Snapshot Backup fails with remotefs based Drivers" [Undecided,New] 09:36:09 <gmann> done 09:36:48 <gmann> mbindlish: what is on 1703405 09:36:53 <mbindlish> thanks 09:37:11 <mbindlish> This issue seems a valid issue, but no assignee yet 09:38:01 <frickler> seems to affect only proprietary drivers, so difficult to tackle upstream 09:38:38 <gmann> yeah and not sure we should introduce feature flag here for snapshot backup 09:38:59 <gmann> mbindlish: do we have that flag in our config ? 09:41:19 <mbindlish> @gmann we have "snapshot" flag in config for "VolumeFeaturesGroup" 09:42:44 <gmann> not snapshot, "snapshot a back" 09:42:46 <gmann> backup 09:43:03 <mbindlish> seems not like this 09:43:38 <gmann> i commented on bug about feature is common or not. based on that we can introduce the flag as last option otherwise let driver skip this tests as they are missing this feature 09:44:16 <frickler> maybe also ping smcginnis about it? would be interesting to know what cinder has to say on this 09:44:32 <gmann> +1 09:44:45 <mbindlish> next: https://bugs.launchpad.net/tempest/+bug/1708550 09:44:46 <openstack> Launchpad bug 1708550 in tempest "Some network tests do not use preprovisioned networks" [Undecided,New] 09:45:14 <mbindlish> Some tests doesn't use pre-provisioned n/w 09:45:26 <gmann> done. pinged them on cinder channel 09:46:56 <frickler> the network bug I can take up with slaweq, too, as we two seem to be having different opinions there 09:48:50 <mbindlish> Harbott's comment looks valid to me here 09:49:37 <frickler> mbindlish: that's me, btw ;) 09:49:49 <mbindlish> ohh, I am sorry :) 09:50:08 <frickler> np 09:50:14 <gmann> i agree with frickler here. those tests either should use mentioned network from account.yaml or fixed network otherwise skip 09:50:29 <gmann> create_isolated_networks is more for dynamic cred. 09:50:36 <gmann> i will comment on bug 09:50:57 <gmann> ok so we are 10 min left for office hour. 09:51:18 <gmann> i will open that for other discussion if you have 09:51:44 <gmann> mbindlish: thanks. i think we need to include all QA programs bugs here not just tempesst 09:52:08 <frickler> so anything left to do regarding the bionic switch? I'm still unsure about the status for neutron 09:52:27 <frickler> sadly that would also need feedback from slaweq 09:52:40 <gmann> frickler: i did not check for neutron. but i sent reminder yesterday on ML 09:52:56 <gmann> yeah, we need to catch up slaweq on that 09:53:03 <frickler> o.k., I'll try to ask mlavalle later 09:53:14 <gmann> thanks 09:53:45 <gmann> i will also try to catch later 09:55:16 <gmann> frickler: there is "install under python3 by default when enabled" patch from dhellmann. you might have seen the mail - https://review.openstack.org/#/c/622415 09:55:45 <frickler> gmann: yes, I saw that patch and like it, will review soon 09:55:49 <gmann> we need to see what effect on devstack plugins projects 09:56:31 <gmann> nice thanks. we might need to hold that to give time for testing on project side. which is what dhellmann asked in ML 09:56:52 <gmann> i am +1 on that and +2 when we are ready to merge it 09:57:01 <frickler> o.k., fine 09:57:30 <gmann> other thing i wanted to ask from tosky about grenade zuulv3 job 09:58:09 <gmann> https://review.openstack.org/#/c/548936/ 09:58:33 <gmann> there are few patches up to move the neutron-grenade job from neutron 09:58:54 * frickler moves to another meeting now, will check back later 09:59:08 <gmann> frickler: thanks 09:59:28 <gmann> tosky: any update on that or any blocking. is it only test running things pending 10:02:32 <gmann> let's close office hour and will catch up on grenade thing later 10:02:39 <gmann> thanks all for joining 10:02:41 <gmann> #endmeeting