08:00:01 #startmeeting qa 08:00:01 Meeting started Thu Jun 29 08:00:01 2017 UTC and is due to finish in 60 minutes. The chair is gmann. Information about MeetBot at http://wiki.debian.org/MeetBot. 08:00:02 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 08:00:04 The meeting name has been set to 'qa' 08:00:10 hi, who all here today? 08:00:13 o/ 08:00:14 hi 08:00:14 gmann: \o/ 08:00:15 o/ 08:00:42 hi 08:01:20 o/ 08:01:20 #link https://wiki.openstack.org/wiki/Meetings/QATeamMeeting#Agenda_for_June_29th_2017_.280800_UTC.29 08:01:25 ^^ today agenda 08:01:26 o/ 08:01:55 #topic Previous Meeting Action review 08:01:57 o/ 08:02:06 #link http://eavesdrop.openstack.org/meetings/qa/2017/qa.2017-06-15-08.00.html 08:02:24 gmann to check with samP about HA testing spec 08:02:35 we discussed this in last meeting only 08:02:42 gmann to discuss plugin clients in tempest with team and get consensus 08:02:47 this one i did not check yet 08:03:12 anyone knows which patch moving keystone plugin service client to tempest ? 08:03:31 we can take that example to decide the strategy on this 08:03:59 gmann: nope 08:04:10 ok let me find later 08:04:16 i am continuing on this AI 08:04:28 #action gmann to discuss plugin clients in tempest with team and get consensus 08:04:37 gmann: no idea but last night they have remove tempest plugin code from keystone project completely 08:04:38 #topic PTG 08:04:55 chandankumar, removed? 08:05:12 gmann: https://review.openstack.org/#/c/471060/ 08:05:35 chandankumar, it is separate repo - http://git.openstack.org/cgit/openstack/keystone-tempest-plugin 08:05:41 gmann: as tempest plugin moves to anew repo https://github.com/openstack/keystone-tempest-plugin 08:05:44 which is nice 08:05:51 yea 08:06:05 on PTG 08:06:07 #link http://lists.openstack.org/pipermail/openstack-dev/2017-May/117369.html 08:06:45 need to check with andreaf if any action item he has to do, room booking etc but we have time for that 08:07:16 but all can plan their travel in advance and there is discount for hotel also (~149USD) 08:07:47 anything else to discuss or query on PTG 08:08:20 is there any travel support for PTG? 08:08:30 masayukig, yes 08:08:59 masayukig, #link https://www.openstack.org/ptg#tab_travel 08:08:59 gmann: thanks, cool. I'm actually not sure I'll get the budget, yet. 08:09:08 masayukig, you should fill soon 08:09:18 i think last date is july 2nd for first round 08:09:39 Where can I see more info about travel support ? 08:09:46 budget things is for everyone seems 08:09:56 afazekas, here - #link https://www.openstack.org/ptg#tab_travel 08:09:57 gmann: oh, wow, I didn't know that.. 08:10:13 afazekas, masayukig there is form to fill for requesting travel suppoprt 08:10:38 "July 2, 2017: Final deadline to submit applications for Round One approvals" 08:10:55 afazekas, are you planning to come so that i can finally meet you face to face :) 08:10:57 gmann: thanks, I need to do, probably. 08:11:03 yea please 08:11:57 #topic Gate Stability - status update 08:11:59 gmann, I am planning, I try to not miss the deadlines .. 08:12:10 afazekas, nice. 08:12:30 gate seems stable #link https://goo.gl/ptPgEw 08:12:43 #link http://status.openstack.org/elastic-recheck/gate.html 08:12:59 there was issue on newton job which andreaf fixed and merged 08:13:28 other was for shelve tests where tests expect image to be deleted immediately after unshelve 08:14:18 which failed on ceph but tests expectation was not right and written anywhere, so fixing that in https://review.openstack.org/#/c/477821/ 08:15:01 yeah I've left comment on it, it's a good solution 08:15:11 zhufl, not sure about yoour comment, 08:15:25 anyways let';s discuss after meeting 08:15:31 yeah 08:15:52 any other issue/things to discuss on gate stability ? 08:16:46 ok, let's move next 08:17:06 #topic Specs Reviews 08:17:15 #link https://review.openstack.org/#/q/status:open+project:openstack/qa-specs,n,z 08:17:36 i did not get time to review spec 08:18:04 may be next week i put some in my review queue 08:18:21 please review and provide feedback 08:18:50 OK 08:19:25 if anyone want to bring or discuss any spec, go ahead 08:20:29 let's move if nothing 08:20:31 #topic Tempest 08:20:45 2 mailing thread on tempest last week 08:21:33 #link http://lists.openstack.org/pipermail/openstack-dev/2017-June/118427.html 08:21:50 this was not generic issue, it is some tests failing on ceph 08:22:14 this time unshelve tests which i mentioned earlier and its tests issue and fix is up 08:22:36 but from that mail it seems ceph job still not happy with many tests 08:23:30 in previous discussion we agreed on having ceph job skipping the tests which fail on ceph because either feature is not implemented or unstable 08:24:01 which is in progress #link https://review.openstack.org/#/c/459774/ 08:24:09 but not yet merged 08:24:50 but now ceph job is all n-v everywhere 08:25:30 2nd mail was tempest plugin scope from chandankumar #link http://lists.openstack.org/pipermail/openstack-dev/2017-June/118835.html 08:25:45 hope it is clear now about what goes in plugin and what in tempest 08:26:03 gmann: yes, Thanks for the input :-) 08:26:31 i want to ask, do we have well document about Tempest scope (what all tests tempest should have and what all project side etc) ? 08:26:45 or we just assume that as we all know the scope in our mind 08:27:00 i remember i added the clear scope for microversion tests 08:27:34 gmann: probably, we should have that document. 08:27:34 #link https://docs.openstack.org/developer/tempest/microversion_testing.html#tempest-scope-for-microversion-testing 08:27:51 gmann: but, I'm actually not sure that the DefCore test must be in Temepst repo. 08:28:12 masayukig, ok, a very simple doc is needed which is always difficult 08:28:13 gmann: i think we need a doc difference between tests in tempest and tempest plugin for core projects 08:28:53 chandankumar: ++ 08:28:57 gmann: this time there was a defcore thread also regarding tempest tests recently. 08:29:06 masayukig, as per TC decision it should be in tempest but as defcore cover more projects as interop that is still undecided 08:29:28 chandankumar, +1 that will be nice 08:29:38 i will take AI on mine 08:29:49 chandankumar, great, about to ask you :) 08:30:04 #action chandankumar to add a doc stating difference between tests in tempest and tempest plugin for core projects 08:30:20 chandankumar, thanks 08:30:38 gmann: my understanding is "in tempest" doesn't equal to "in tempest tree". 08:30:52 masayukig, for defcore? 08:31:19 gmann: yeah, let me find the ml thread... 08:31:23 masayukig, it is not yet decided 08:31:39 masayukig, problem will be same either it is tempest tree or another plugin 08:32:11 main thing is we need to solve the man power issue to own more and more projects tests under Tempest team 08:33:23 #link http://lists.openstack.org/pipermail/openstack-dev/2017-June/117756.html 08:33:31 masayukig, this one ^^ 08:33:57 gmann: yeah 08:34:01 anyways that is future things 08:34:37 next is : Bug Triage & Integrated gate issue categorization: 08:34:37 gmann: yeah, fyi, you may know, http://lists.openstack.org/pipermail/openstack-dev/2017-June/117808.html 08:34:56 masayukig, yea 08:35:13 gmann: just one update from tripleo ping test thread, we will be merging these some pieces of two scenario tests into new scenario test in tempest which will implement boot an instance from volume then attach flotaing ip ssh/ping ot it an ddestroy it 08:35:55 https://github.com/openstack/tempest/blob/master/tempest/scenario/test_volume_boot_pattern.py and https://github.com/openstack/tempest/blob/master/tempest/scenario/test_network_advanced_server_ops.py 08:35:57 chandankumar, ok, we do not have any tests doing that? 08:36:02 chandankumar: it seems like test_minimum_basic.py one. 08:36:10 before we move away from the tempest topic: is it time to tag a new tempest, so that projects (plugins) can depend on it? 08:36:31 gmann: nope, based on my investigation, we need a test for the same 08:36:45 ltosky[m], need to check how much we have since 16.0.0 08:36:54 for example there is a deprecation warning about test.attr moved to tempest.lib 08:37:11 chandankumar, that can be extended in existing one or you are thinking new one? 08:37:29 ltosky[m], i see, attr and related_bug are not yet tagged 08:37:34 gmann: thinking of a new one, 08:37:47 gmann: let me get the patch up and we can take it from thetre 08:38:03 chandankumar: ++ 08:38:10 chandankumar, only "attach flotaing ip ssh/ping" will be extra what we are already doing in volume pattern tests 08:38:19 yep, patches for attr (seen one in the mistral plugin) can't be merged without a new tag 08:38:21 chandankumar, yea that will be fine 08:38:28 gmann: yes 08:39:03 ltosky[m], i know i think i have also one for mistral 08:39:05 #link https://review.openstack.org/#/c/456823/ 08:39:23 ltosky[m], ll check with andreaf if there is enough things for new tag 08:39:45 #action gmann to check with andreaf about new tag 08:39:58 thanks! 08:40:04 let's discuss bug 08:40:26 blancos, turn this week 08:40:35 #link https://etherpad.openstack.org/p/pike-qa-bug-triage 08:40:52 #link https://etherpad.openstack.org/p/tempest-weekly-bug-report 08:41:02 I wasn't able to triage the one new bug as it came in a little later yesterday 08:41:03 blancos, is this report up to date? ^^ 08:41:06 Yes 08:41:11 blancos, ok 08:41:35 There are a few incomplete bugs, but they're currently counting down to expire (except the one to remove Glance v1 tests) since no one is currently working on them 08:41:52 we didnt have meeting last week, so bug triage for last week is also yet to be discussed 08:41:54 blancos, ok 08:42:26 blancos, glance v1 tests yes we need to wait more may be but need to check ML thread before that 08:42:42 prateek, oh, please update if you have any key finding 08:43:56 gmann, a cinder ceph issue https://bugs.launchpad.net/tempest/+bug/1687533 08:43:57 Launchpad bug 1687533 in tempest "Regression issue of adding force delete backup support" [Undecided,Confirmed] - Assigned to Fei Long Wang (flwang) 08:44:15 maybe this is the reason the ceph team was talking about failing tempest tests 08:44:58 prateek, that is already reverted 08:44:59 #link https://review.openstack.org/#/c/461625/ 08:45:27 mail thread also - #link http://lists.openstack.org/pipermail/openstack-dev/2017-May/116172.html 08:45:33 ok, do we have any decorator for skipping a specific test for a particular backend like ceph ? 08:45:35 what is issue on this? 08:45:52 prateek, not actually and honestly we do not want to do that 08:46:01 gmann, ok 08:46:10 that was all i wanted to ask with reference to this bug 08:46:40 prateek, if you read that mailing thread we should mainly concentrate on mandatory features and if any backend does not implement those then their CI can skip those 08:46:52 prateek, ok, thanks for bug triage 08:47:03 blancos, thanks for this week triage 08:47:39 we have 62 in-progress which is doubtable actually :) we should keep cleaning those 08:48:06 mkopec turn next week 08:48:14 gmann I noticed some of them are attached to patches that haven't been updated in >6 months 08:48:14 gmann, ack 08:48:34 martinkopec, thanks, you changed your irc name :) 08:49:10 gmann, sorry, I always forgot that I have a different nic on freenode 08:49:16 blancos, yea, not sure what we should do on those, some are really good fix but no progress 08:49:41 may be we can unassign such bug so that people can take up even old patches 08:49:50 martinkopec, no issue 08:50:03 chandankumar, seems like you volunteered for https://etherpad.openstack.org/p/pike-gate-issue-categotisation 08:50:21 gmann: i didnot worked on this week 08:50:26 anything else on tempest (10 min left) 08:50:35 chandankumar, your turn is next week 08:50:46 gmann: i will working on gates this week will catch sdague and mtreinish 08:50:53 for learning on how to do that. 08:50:59 #topic Patrole 08:51:13 blancos, your turn 08:51:16 chandankumar, sure, thanks 08:51:29 gmann Nothing new to report 08:51:31 chandankumar, andreaf mentioned in mail about expectation on those 08:51:53 will take a look gmann 08:51:59 blancos, i reviewed the nova spec for gating patrole on nova #link https://review.openstack.org/#/c/471145/ 08:52:11 i am +1. ll ask matt to check this 08:52:11 I saw, thank you :) 08:52:31 i forgot to bring this in yesterday nova api meeting, ll add for next week 08:53:03 #topic Critical Reviews 08:53:19 any critical reviews to bring on priority 08:53:47 gmann: https://review.openstack.org/477549 need +w 08:54:03 and followed up https://review.openstack.org/474164 08:54:17 i need this for speeding up tempest plugin sanity gate 08:54:37 https://review.openstack.org/#/c/478730/ I'd like to know your opinions. 08:54:39 chandankumar, ok ll check. did not follow up those though 08:55:00 masayukig, ll add my comment after meeting 08:55:09 gmann: thanks :) 08:55:24 masayukig, not sure we need to keep release named dire as well release note name also 08:55:25 gmann, https://review.openstack.org/#/c/445419/ 08:55:51 prateek, sure, 08:55:56 gmann: yeah, but I feel the directory is just a mess.. 08:55:58 gmann, thanks 08:56:13 #topic Documentation 08:56:55 anything specific to discuss on documentation. one we already covered about scope of tests in tempest and plugin 08:57:37 #topic Open Discussion 08:57:40 3 min left 08:57:51 anything more to discuss related to anything 08:58:26 related to documentation: https://review.openstack.org/#/c/472275/ 08:58:43 https://review.openstack.org/#/c/478546/ 08:59:17 We should change the doc directory. 08:59:38 masayukig, yea that is global work going on 08:59:39 * asettle waves 08:59:40 thanks 08:59:44 If you have any questions about docs 08:59:45 Lemme know :) 08:59:53 asettle, sure, thanks for that 09:00:02 yeah 09:00:16 thanks all for joining 09:00:21 Thanks!! 09:00:26 it time to close 09:00:28 #endmeeting