14:00:05 #startmeeting trove 14:00:06 Meeting started Wed Oct 3 14:00:05 2018 UTC and is due to finish in 60 minutes. The chair is dkrol. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:07 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:09 The meeting name has been set to 'trove' 14:00:20 hi! 14:00:32 lets wait a few more minutes 14:00:33 #topic roll call 14:00:36 pgodek: Hello 14:00:45 Hello everyone 14:00:47 hello 14:01:10 Hello everyone 14:01:33 I see we have more people then last time 14:02:24 Marcin has problems with connection 14:02:30 Let's give him a minute 14:02:54 sure thing 14:06:12 hello everyone 14:06:42 hello 14:06:44 mpiwowarczy: Hello 14:06:46 sorry for late, I had a connection issues - looks like ip quotas 14:06:46 hi 14:06:55 so maybe let's start 14:07:17 #topic Core team members 14:07:59 bzurkowski: you proposed the topic 14:08:20 Right 14:08:49 During last meeting we considered adding Marcin to the core team 14:09:35 Today we have more people participating in the meeting 14:10:02 ok 14:10:33 so basically you would like to propose mpiwowarczy to be joined to the core contributor's list 14:10:54 Yes, definitely 14:10:55 I've sent an email to all missing core contributors about our meetings 14:11:19 unfortunately, I see there was very little response rate 14:11:37 last time I suggest we should contact them first as a reminder 14:11:41 btw., what are the duties of core team members? (in nutshell/briefly) More impact on code review? 14:11:53 dkrol: was there anyone objected? 14:12:22 mpiwowarczy: no, though I didn't stated your case explicitly 14:12:54 cezary_zukowski: Core member has permissions to add +2 for changes and setting +1 for workflow 14:13:04 dkrol: oh, ok 14:13:11 bzurkowski: okay, more power :) 14:13:13 That means that they have more impact on the decision whether given change should be submitted to the mainline 14:13:18 IMHO we can put mpiwowarczy candidacy for voting 14:13:43 surely we can 14:14:19 #startvote (Do you accept mpiwowarczy candidacy for trove core contributor ? ) ? YES, NO 14:14:20 Begin voting on: (Do you accept mpiwowarczy candidacy for trove core contributor ? ) ? Valid vote options are YES, NO. 14:14:21 Vote using '#vote OPTION'. Only your last vote counts. 14:14:32 #vote YES 14:14:40 #vote YES 14:14:43 #vote YES 14:14:52 #vote YES 14:15:09 #endvote 14:15:10 Voted on "(Do you accept mpiwowarczy candidacy for trove core contributor ? ) ?" Results are 14:15:12 YES (4): cezary_zukowski, bzurkowski, pgodek, dkrol 14:15:32 I didn't say it before but only core contributors should vote :) 14:15:48 dkrol: okay, so I am out :) 14:16:11 looks it doesn't matter :) 14:16:12 The remaining voices have symbolic meaning ;) 14:16:27 anyway I think the output is obvious that mpiwowarczy should be added as core contributor to Trove due to his work on the project in recent months. 14:17:02 great 14:17:12 as bzurkowski said the last time, having more core contributors should facilitate trove development 14:17:19 thanks guys 14:17:20 mpiwowarczy: congratulations 14:17:21 I am very pleased :) 14:17:31 thanks 14:17:34 congrats 14:17:35 Congrats Marcin 14:18:12 we can move to the next topic on our agenda 14:18:13 I will do mybest as new core member 14:18:31 #topic Recent changes and reviews 14:19:15 there were a few new changes and bugfixes merged 14:19:30 by external contributors which is great 14:20:28 https://review.openstack.org/606320 14:20:34 https://review.openstack.org/571652 14:20:39 https://review.openstack.org/598412 14:20:47 https://review.openstack.org/604854 14:20:52 https://review.openstack.org/601886 14:21:17 I hope these people will maintain active - maybe we could invite them to our meetings 14:22:02 these are mainly small changes and fixes 14:22:14 nothing major but still it is great to see it 14:22:38 do you have anything to share on this ? 14:23:28 dkrol: It is a good idea to encourage those people to co-operate 14:24:24 Especially the ones that contributed changes that were database-related 14:24:42 https://review.openstack.org/#/c/601886/ being an example 14:25:10 jian.song looks more involved to trove project https://review.openstack.org/#/q/owner:jian.song+status:open 14:26:22 this can be one of our action points regarding trove revitalization 14:27:05 mpiwowarczy: His changes look very interesting 14:27:20 It would be great to bring him back 14:27:36 I agree 14:28:01 #action send an email to possible trove contributors 14:28:56 anything else on this topic or can we move to the next 14:29:47 Let's move on 14:31:10 #topic Trovestack redesign 14:32:35 https://review.openstack.org/606960 14:33:07 I made this change as a first step 14:33:56 it is about adding image setup to trove devstack plugin 14:33:58 I saw this change, a lot of good work 14:34:23 there is one more thing about this 14:34:59 do we want build these images in devstack for now? 14:35:21 I don't follow 14:35:25 if not, we could disable it temporarily 14:35:28 what do you mean ? 14:35:40 isn't needed for CI? 14:36:22 for now we have images built in trovestack 14:37:17 mpiwowarczy: If not now, how to test it in the CI? 14:37:22 hmm... we don't need them for now in CI 14:38:06 mpiwoarczy: so it follows a trend of running away from trovestack (as with tests, i.e. rewriting to Tempest)? 14:38:21 In my opinion we cannot just leave a bulk snippet of bash code turned off, outside of the testing scope 14:38:29 are these images used anywhere now? 14:39:21 bzurkowski: great idea 14:39:38 mpiwowarczy: no, besides development environment setup 14:39:46 mpiwowarczy: It's rather about users evaluating new devstack feature in their environment 14:40:50 bzurkowski: could you list some examples of such features? 14:41:00 ok, got it 14:42:02 I mean, we are providing new version on Trove devstack plugin to the users 14:42:21 do you think we should switch to this image in trovestack tests ? 14:43:15 And user will be able to run devstack with image building turned on 14:43:47 Therefore I suggest having a testing job in the CI pipeline that evaluates this use case 14:43:59 bzurkowski: so, it's about preparing customized images rather than downloading them from some repo (artifactory)? 14:44:55 so such a test should be run only if this image was created 14:45:27 lets build image for now, it will tests building image code 14:45:33 normally when we have integration tests, we build an image first within a test job 14:45:53 later we can consider to use images from devstack plugin 14:46:17 dkrol: Yes, that is a one way to test it 14:46:34 mpiwowarczy: So you suggest building an image twice per job? 14:47:08 looks we are forced to do that 14:47:17 hmmm 14:47:26 Once during devstack installation, second time on trovestack kickstart 14:47:40 we have a flag to disable it 14:47:59 so in most integration testing we can turn this flag on 14:48:08 dkrol: Good point 14:48:16 and then we will not have double image build 14:48:55 and we can add another integration test which will build this image and then have a tempest test to test it 14:49:47 testing is on topic, btw., is it a long process of building an image? 14:50:05 I consider to create separate job in gate section 14:50:08 we can also reverse the flag to build an image only if a flag is turned on 14:50:14 cezary_zukowski: Depending on the datastore, around 30 minutes 14:50:19 @cezary ~20 min 14:50:25 thanks for that we will test this piece of code on each merge 14:50:32 not bad :) 14:50:44 mpiwowarczy: I like this approach 14:50:54 then, build time will be not a issue 14:51:00 We could keep it experimental until rewriting trovestack and functional jobs 14:51:30 there is a question what kind of test we should have 14:51:48 let's have an offline discussion about it since we are running out of time 14:52:04 #topic Python 3 first support 14:52:33 last week I had a discussion about supporting stable branches by python3 14:52:46 it seems there was a misunderstanding 14:53:21 and there is a task in python3-first goal to migrate zuul configuration to each project instead of having them centralized 14:53:26 or something like this 14:53:39 this is not related to python3 to my best knowledge 14:53:49 and this change should be merged to stable branches 14:54:04 as it will break tests otherwise 14:54:25 do we have anything about this topic ? 14:54:42 dkrol: We need to have appropriate permissions to submit remaining changes 14:55:11 to stable branches ? 14:55:12 Currently, we can only mark them with +1 14:55:16 dkrol: Yes 14:55:26 you are correct 14:55:56 unfortunately for now my request for adding us to stable branch maintainers was declined 14:56:19 as for master branch, the current status is that we have two open changes https://review.openstack.org/604854 and https://review.openstack.org/#/c/604816/ 14:56:34 first is almost done 14:56:48 will be fine if anyone will review this change 14:56:55 we should contact people from the list for help 14:57:15 mpiwowarczy: great 14:57:20 keep up the good work 14:57:45 we have only 2 more mites so probably we should finish now 14:57:53 thanks all for being here 14:58:04 we had a great week 14:58:22 Thanks, it was great to see now people in the conversation 14:58:26 see you next time 14:58:31 Bye 14:58:32 thanks, bye 14:58:36 bye 14:58:44 #endmeeting