17:01:21 <andreaf> #startmeeting qa 17:01:22 <openstack> Meeting started Thu Jul 7 17:01:21 2016 UTC and is due to finish in 60 minutes. The chair is andreaf. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:01:23 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 17:01:26 <openstack> The meeting name has been set to 'qa' 17:01:26 <hogepodge> o/ 17:01:38 <andreaf> hello - who's here today? 17:01:48 <andreaf> hi hogepodge 17:01:59 <dpaterson> o/ 17:02:03 <hogepodge> Hi andreaf! 17:02:50 <andreaf> let's wait another minute to see if anyone else joins 17:03:05 <andreaf> mkoderer, afazekas: around? 17:03:21 <hogepodge> andreaf: thanks for the awesome test results for defcore btw, they're really useful for current work happening in DefCore. 17:04:30 <andreaf> hogepodge: do you mean those I submitted for helion? 17:04:41 <andreaf> anyways, time to start the meeting now 17:05:01 <andreaf> Today's agenda: #link https://wiki.openstack.org/wiki/Meetings/QATeamMeeting#Agenda_for_July_7th_2016_.281700_UTC.29 17:05:04 <hogepodge> andreaf: Yeah, I compared them against the proposed guidelines and they really shone a light on potential problem areas. 17:06:13 <andreaf> #topic Newton priorities 17:06:40 <andreaf> Does anyone have updates on newton priority items? 17:06:56 <andreaf> #link https://etherpad.openstack.org/p/newton-qa-newton-priorities 17:08:02 <hogepodge> I finished up the work on documenting negative tests in the hacking rules. 17:08:24 <andreaf> hogepodge: cool, do you have a link? 17:08:39 <hogepodge> #link https://review.openstack.org/#/c/333618/ 17:09:03 <andreaf> thanks for that 17:09:37 <andreaf> on tempest.lib, I'm doing some progress on having a stable client manager, and I did test integration with a couple of plugins 17:10:04 <andreaf> the cleanup part (prio1) is still lacking contributors unfortunately 17:10:13 <dpaterson> hogepodge: looks like it was merged already? 17:10:53 <andreaf> ok let's move on 17:10:59 <hogepodge> dpaterson: yeah, I was just updating status, didn't mean to imply it needed review. sorry. 17:11:12 <andreaf> #topic Specs Reviews 17:11:23 <dpaterson> hogepodge: no worries 17:11:38 <andreaf> #link https://review.openstack.org/#/q/status:open+project:openstack/qa-specs,n,z 17:11:55 <andreaf> anything to discuss on specs? There's nothing new in the list 17:12:49 <andreaf> Next topic I will work on is make credential providers stable, and I'm thinking that may warrant a spec 17:13:18 <andreaf> I want to setup something like this #link http://paste.openstack.org/show/526609/ 17:13:40 <andreaf> So it's probably worth collecting some feedback in gerrit on it first 17:14:12 <andreaf> ok, next topic 17:14:14 <andreaf> #topic Tempest 17:14:21 <andreaf> Anything to discuss on tempest? 17:14:57 <hogepodge> yeah 17:15:09 <hogepodge> #link https://review.openstack.org/#/c/329220/ 17:15:27 <hogepodge> This review came into DefCore about one of the tests for listing servers 17:16:06 <andreaf> looking 17:16:42 <hogepodge> I'm wondering if there's room for a refactor that assumes less information. We don't need action today, just something for the team to review. I'm happy to take any bug action items on it, as I don't think dwalleck has the bandwith to work on it. 17:18:11 <andreaf> hogepodge: well I will be working on splitting cred providers from network resources 17:18:27 <andreaf> hogepodge: so that only tests that need network details will invoke that code path 17:18:40 <andreaf> hogepodge: but that solves only part of the issue I guess 17:19:18 <andreaf> hogepodge: I need to read those comments in more details though, I will leave a comment on the spec 17:20:09 <hogepodge> andreaf: thanks 17:20:12 <andreaf> hogepodge: the reason I'm using the nova API there is because that will work for both neutron and non-neutron jobs 17:20:48 <andreaf> hogepodge: so in case of neutron jobs it may be that the nova proxy is not there, but that would be strange 17:21:00 <andreaf> hogepodge: anyways, I need to check this in more detail 17:21:03 <fnaval> The support for filtering by name feature for Tempest is somewhat incomplete. I still need to make changes so that it can be configured as a parameter in config.py. Will work on that when I have a chance. 17:21:15 <hogepodge> andreaf: ok. dwalleck says the ip information is there in the response and doesn't need network information. Is there space to write a new test that tries to use that information? 17:23:48 <andreaf> hogepodge: right I'm not sure about the specific test case - my comment was in general about resolving network names to IDs 17:25:06 <hogepodge> andreaf: I think the complaint is about the names->ids resolution in general too, if I'm reading it correctly. I guess it's a "problem" at both scopes. The local test and the larger network framework. 17:25:55 <eantyshev> Hello! May we discuss our issue with AttachVolumeShelveTestJSON in our CI: http://paste.openstack.org/show/527014 17:26:10 <andreaf> hogepodge: I will have a look at the code and leave a comment on the review 17:26:58 <andreaf> eantyshev: I can try, I guess you would have more luck with gmann or oomichi on that one 17:27:10 <eantyshev> AFAIU AttachVolumeShelveTestJSON requires v2.20 nova api, but schema is not adjusted for additional properties in os-extended-volumes:volumes_attached which may exist for v2.20, and listing server details fails 17:28:41 <andreaf> eantyshev: that sounds like a bug in tempest 17:29:24 <andreaf> eantyshev: do you run a cinder 3rd party CI? is it possible that your backend only triggers that extra attribute in the response? 17:30:09 <eantyshev> andreaf: yes, I'm responsible for Virtuozzo Storage CI 17:31:20 <eantyshev> andreaf: but this is hardly related to our backend, IMHO. Upstream doesn't run validation and skips this TC 17:31:29 <andreaf> eantyshev: there was a change recently in tempest to control the 'delete_on_termination' flag in the create_test_server helper when using a volume backed VM 17:31:56 <eantyshev> andreaf: Icf6148017cdbc4f48d3ef9a964ee39fedda72b4e ? 17:32:04 <andreaf> eantyshev: yes that makes more sense, it shouldn't really be backend specific 17:33:14 <andreaf> eantyshev: so what you're saying is that we changed the schema but the only test that checks that upstream is skipped? 17:35:30 <eantyshev> andreaf: seems so 17:36:31 <andreaf> eantyshev: looking at that test, it seems to me that it could run without ssh validation, and only the ssh checks should be skipped - it's much less powerful without ssh, but better than nothing 17:36:43 <andreaf> eantyshev: can you file a lp bug on this please? 17:37:17 <eantyshev> andreaf: namely, tempest.api.compute.volumes.test_attach_volume.AttachVolumeShelveTestJSON.test_attach_detach_volume requires v2.20 API. tempest.api.compute.volumes.test_attach_volume.AttachVolumeTestJSON.test_attach_detach_volume uses v2.19 and doesn't fail 17:37:43 <andreaf> eantyshev: you can see that this has broken the neutron-ssh job as well indeed: http://logs.openstack.org/01/337601/1/check/gate-tempest-dsvm-neutron-full-ssh/4e7392b/logs/testr_results.html.gz 17:37:51 <andreaf> eantyshev: so we should run that test ssh or not 17:38:11 <andreaf> eantyshev: thanks for bringing this up 17:38:44 <andreaf> eantyshev: I can make the change to the test, but I will need gmann / oomichi to do the schema update change - unless you want to do it? 17:39:37 <eantyshev> andreaf: I would be glad to try by myself 17:39:49 <andreaf> eantyshev: cool, thank you! 17:40:00 <hogepodge> eantyshev: ping me if you need help also, I've done quite a bit with the schemas 17:40:25 <andreaf> hogepodge, eantyshev: thanks 17:40:30 <andreaf> let's move on 17:40:53 <andreaf> #topic DevStack-Grenade 17:41:30 <andreaf> Does anyone have anything to discuss on these? 17:42:05 <andreaf> I'll take that as no 17:42:22 <andreaf> #topic OpenStack-Health 17:42:37 <andreaf> Any topic for discussion on the dashboard? 17:43:34 <andreaf> #topic Critical Reviews 17:43:43 <andreaf> Any critical review that requires attention? 17:44:22 <andreaf> On my side this series: https://review.openstack.org/#/q/status:open+branch:master+topic:bp/client-manager-refactor 17:44:32 <andreaf> #link https://review.openstack.org/#/q/status:open+branch:master+topic:bp/client-manager-refactor 17:45:34 <andreaf> #topic Open Discussion 17:45:52 <andreaf> Let's open the floor to any other topic you may have for discussion 17:46:19 <andreaf> 3... 17:46:21 <andreaf> 2.. 17:46:23 <andreaf> 1. 17:46:42 <andreaf> ok, let's get 14min back today 17:46:45 <fnaval> thanks! 17:46:57 <andreaf> thanks for joining, have a nice rest of your day 17:47:12 <andreaf> #endmeeting