09:00:35 #startmeeting qa 09:00:36 Meeting started Thu Oct 6 09:00:35 2016 UTC and is due to finish in 60 minutes. The chair is masayukig. Information about MeetBot at http://wiki.debian.org/MeetBot. 09:00:37 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 09:00:39 The meeting name has been set to 'qa' 09:01:06 Who's here for today? 09:02:35 or should I skip this meeting? 09:03:53 ping dmellado , jordanP ? 09:04:03 I am here :) 09:04:03 o/ 09:04:10 phew 09:04:40 masayukig: got caught 'interoping' xD 09:04:43 but I'm here 09:05:02 dmellado : heh, hi :) 09:05:39 ok, so, let's get started 09:05:52 #link https://wiki.openstack.org/wiki/Meetings/QATeamMeeting#Agenda_for_October_6th_2016_.280900_UTC.29 09:06:07 Today's agenda ^^ 09:06:23 so, first topic is 09:06:35 #topic Newton Priorities 09:06:36 #link https://etherpad.openstack.org/p/newton-qa-newton-priorities 09:06:45 But I don't think we need to track one by one. 09:07:02 Does anyone have something to say? 09:07:03 I don't think so either 09:07:30 there are a few missing unit tests for service clients only 09:07:39 and a few service clients to migrate, maybe 1 or 2 09:07:44 but that part should be about done 09:08:01 anything from os-health, masayukig ? 09:08:51 dmellado : thanks, actually no about o-h thing, But I saw some patches landed. 09:08:53 and 09:09:03 I actually have one, https://review.openstack.org/#/c/355666/ is for "The last piece: Use 'tempest init' and 'tempest run' on the gate" 09:09:33 I'm now tackling with this. This is still SUPER WIP, but baby steps. 09:09:56 * andreaf o/ 09:10:05 hi andreaf ! 09:10:05 andreaf: hey ;) 09:10:13 welcome back 09:10:30 hi - thanks 09:10:42 andreaf : any update from you about https://etherpad.openstack.org/p/newton-qa-newton-priorities ? 09:11:33 masayukig: no - I'm just back from holidays - so work done in the past two+ weeks 09:11:58 andreaf : ah, cool, I see :) 09:12:35 anything else, guys? 09:13:20 I guess that's it 09:13:24 ok, 09:13:29 next topic 09:13:35 I think I'll give Cleanup client and manager aliases (the manager aliases) a shot soon 09:13:58 jordanP : oh, nice :) 09:15:04 #topic Next Summit 09:15:36 so, we have some QA slots: 3fb 4wr 09:16:07 But I think the proposed topic is still not enough: https://etherpad.openstack.org/p/ocata-qa-summit-topics 09:16:15 masayukig: andreaf 09:16:33 I'd like to propose a topic about how to handle plug-ins 09:16:42 but maybe that'd be related to mtreinish's TC proposal 09:16:51 and andreaf work on Manila 09:17:03 I'm aware that we do treat them as some kind of 'third party' 09:17:05 dmellado : yeah, but please feel free to add your thoughts 09:17:30 but I've spending a ton of time cleaning up them, and was wondering if having a discussion on how to proceed with them 09:17:37 either the forceful repo 09:17:43 or another options 09:17:45 would be great 09:18:01 btw /me got added to discuss that for manila too 09:18:12 so I think that spending a session for that could be good 09:18:35 wdyt guys? 09:18:36 manila? what kind of thing? 09:19:02 masayukig: on the dev session from manila they asked me to join for a topic on how to stop pinning tempest 09:19:06 and properly handle the plugin 09:19:43 so instead of doing also that, maybe we could put up a session on the plugins and have folks from other projects come 09:20:14 hmm, we should add more docs or should have a hands-on session or something? 09:20:19 as far as I know they have a problem with authentication of tenant-isolation 09:21:10 mmm masayukig jordanP maybe both 09:21:16 jordanP: also, there were quite a few of tempest imports 09:21:19 there's a blueprint for that 09:21:28 mkoderer___: put up a patch and we were discussing that at the mid-cycle 09:22:15 but if there are available slots and not too many topics, I think it'd be worthwhile 09:22:31 andreaf: you were also working on manila to avoid pinning, weren't you? 09:22:50 anyway there's not a lot we can do, they should step up and learn how to use the framework/plugins 09:23:01 there's always a bit of a learning curve 09:23:29 dmellado: I was working on the manila plugin to use the latest stable interface (clients) - which requires using a more recent version of tempest 09:23:33 ideally no pinning 09:23:58 if all unstable interfaces are dropped so can pinning, but there's more used by manila right now 09:25:27 jordanP: yep, for instance I've been seeing (and putting patches) 09:25:37 against wrong usage of service_available CFG 09:25:45 in quite a bit of projects 09:26:14 It might be good to have a slot for it. But anyways, I think it's good to add your ideas on the etherpad. We will be able to discuss them later. 09:26:24 masayukig: will do 09:26:34 let's discuss it there and with a broader audience ,) 09:26:35 dmellado : cool, thanks :) 09:26:37 so please continue 09:26:45 yeah 09:27:10 anything else on the next summit topic? 09:27:29 ok 09:27:31 #topic Specs Reviews 09:27:58 we have some patches on the gerrit 09:28:03 #link https://review.openstack.org/#/q/status:open+project:openstack/qa-specs,n,z 09:28:35 I actually leave them these days... 09:29:04 andreaf : you have some patches, do you want to bring something? 09:30:13 masayukig: nothing from my side right now 09:30:20 ok 09:30:51 I'll review them, anyways 09:31:12 so, next topic is 09:31:20 #topic Tempest 09:31:57 do you have anything to bring up? 09:33:08 so, if nothing, I have one :) 09:33:14 go for it ;) 09:33:36 the patch https://review.openstack.org/#/c/358149/ was approved 09:33:49 'Add "daily" option to bugdaystats.py' 09:34:35 This makes daily bug triage counts in JSON for a half year 09:34:51 * dmellado needs to go for a doctor checkup, will follow-up later with you guys 09:34:51 like oomichi's graph 09:35:06 dmellado : oh, take care :) 09:35:09 masayukig: I'll look for the link on the spec later and will review it 09:35:11 see you later guys 09:35:23 thanks! 09:35:29 \o/ 09:35:31 I have a couple of reviews to drop sahara tests, negative auto tests and stress tests. I'd like them to be merged right after we tag tempest 13. that would be great :) 09:35:50 jordanP: just before I go, did you sync with tosky on the sahara drop? 09:35:54 I kinda recall that yes 09:35:56 but just in case 09:36:03 yeah he +1 the patch 09:36:07 jordanP : wow, nice 09:36:10 ack then, thanks! 09:36:29 when do we plan to tag 13, btw? 09:36:33 this week? next? 09:37:00 I have no idea, oomichi or mtreinish knows it? 09:37:13 * dmellado will read up later, going ;) 09:37:23 * masayukig waves 09:37:38 so, anything else on Tempest? 09:37:57 if not, next topic is 09:38:07 #topic DevStack + Grenade 09:38:23 hi, sorry for stumbling in late, but I'd like to draw some attention to 09:38:28 #link https://bugs.launchpad.net/devstack/+bug/1629133 09:38:29 Launchpad bug 1629133 in devstack "New neutron subnet pool support breaks multinode testing." [Critical,In progress] - Assigned to Miguel Angel Ajo (mangelajo) 09:38:40 frickler : hi :) 09:38:44 this is being pushed around for a week now 09:39:33 and seems to still be a cause for a lot of gate jobs failing 09:41:42 frickler : hrm, it's too bad, but it seems Miguel are doing for it? 09:42:35 I actually don't know the bug details, though.. 09:42:39 https://review.openstack.org/381965 has -W, maybe it would help if some QA core could join the effort 09:44:06 frickler : ok, I'm actually not a devstack core but a tempest. But I hope oomichi , mtreinish , sc68cal , or some other guys probably help it. 09:44:55 masayukig: k, lets hope they take a look at the log later 09:45:05 yeah.. 09:45:14 anything else? 09:45:22 (15 mins) 09:45:40 if not, the next topic is 09:45:42 #topic openstack-health 09:46:10 But I already told about this before. I have nothing more. 09:46:35 ok, move forward 09:46:45 #topic Critical Reviews 09:46:56 as a general note, I've seen a couple of migration tests trying to be added lately. Be sure that they are run somewwhere, in some jobs. 09:48:05 jordanP : thanks, you added it for the REVIEW docs 09:48:34 and as usual, pay also attention to code duplication. We can't have 10 different ways to create a volume or a server 09:49:00 * masayukig keeps in mind 09:49:22 finally, when a new test is beeing added, ask yourself whether it is not already tested somewhere else 09:49:50 it's easier to add a test for an untested feature than to go over all the tests and look for duplicates 09:49:57 once a test is in, it super hard to remove it 09:50:45 jordanP : yeah, agree 09:51:32 I think it's important, There's very few people in this meeting, though.. 09:51:49 ok, 9 mins left 09:52:14 do you have anything to bring up for critical reviews? 09:52:19 Is it possible to discuss a Tempest spec I created during this part of the meeting? 09:52:26 Or should I save it for open discussion? 09:52:50 DavidPurcell : hi, 09:52:58 hello :) 09:53:04 you can go ahead :) 09:53:08 yeah 09:53:20 Okay. I brought it up at last week's meeting 09:53:22 https://review.openstack.org/#/c/382672/ 09:53:47 #link https://review.openstack.org/#/c/382672/ 09:54:32 Thanks. 09:54:35 DavidPurcell : you need more reviewers for this, right? 09:54:50 I assume so, I'm not 100% sure on the process 09:56:35 ok, I think if this is complicated or needs to face to face discussion, and you'll come to the summit, you can add this as a design summit topic. 09:57:26 I won't be there, but some other representatives from my company (AT&T) will be. 09:57:57 DavidPurcell : ok, you can feel free to add it if you need it. https://etherpad.openstack.org/p/ocata-qa-summit-topics 09:58:15 But the slots are limited, though... 09:58:51 DavidPurcell : thanks :) 09:58:56 so, 2 mins 09:59:11 anything else, guys? 09:59:28 #topic Open Discussion 09:59:38 ok, i just changed the topic 09:59:59 do you have anything to chat? 10:00:14 if not, time's up! 10:00:48 so, thanks guys!!! 10:01:08 if you have something, bring it to qa channel :) 10:01:09 #endmeeting