18:02:54 #startmeeting trove 18:02:55 Meeting started Wed Aug 17 18:02:54 2016 UTC and is due to finish in 60 minutes. The chair is amrith. Information about MeetBot at http://wiki.debian.org/MeetBot. 18:02:56 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 18:02:58 The meeting name has been set to 'trove' 18:03:09 o/ 18:03:09 o/ 18:03:14 sorry I'm late folks 18:03:23 thanks for the reminder peterstac 18:03:29 np ;) 18:03:39 #agenda https://wiki.openstack.org/wiki/Meetings/TroveMeeting 18:03:49 o/ 18:03:55 o/ 18:03:57 o/ 18:05:14 having issues with agenda, one second 18:05:26 😏/ 18:05:54 ./ 18:06:07 ok, sorry for that, just had a quick update to the agenda 18:06:14 and had to reconcile mine with Peter's recent edit 18:06:18 #topic Action Items 18:06:33 peterstac, yours was "peterstac to propose changes to add non-voting jobs for mariadb, postgresql, pxc, and cassandra" 18:06:38 I think that's been proposed, yes? 18:06:52 yes, but it's just sitting there 18:07:01 I may have to poke someone in infra 18:07:14 they are a bit busy today 18:07:26 lots of stuff seems to have started to let out magic smoke 18:08:02 you can ping them. 18:08:13 pmalik had "pmalik to work with infra to see if we can get different machines for gate" 18:08:21 pmalik ... have you? 18:08:53 let's skip over that one if pmalik isn't around 18:08:56 peterstac, "peterstac to review py35 failure" 18:09:11 I am waiting for the above patch set to merge first to have something in hand before going to them. 18:09:28 ok 18:09:32 I haven't have much time to look at that (py35) - I'll see if I can get to it this week 18:09:42 #action peterstac to review py35 failure 18:09:48 #action pmalik to work with infra to see if we can get different machines for gate 18:10:06 #action amrith and peterstac to ping infra re: https://review.openstack.org/#/c/354881/3 18:10:16 sounds good 18:10:18 #topic Trove pulse update 18:10:24 I've updated the numbers 18:10:34 still a backlog but the cleanup from last week helped a bit. 18:10:40 #link https://docs.google.com/spreadsheets/d/1vJxNaoR3VVNS1Cpiz7U--1zyJRZ6ybMxzJoayrjdduo/edit#gid=1104331956 18:11:11 we've seen a spurt of new patches (#noqa, ...) 18:11:20 over the past couple of days. 18:11:31 that's skewing the numbers a bit. 18:11:48 to those who've not had a chance to do any reviews 18:11:50 #link https://gist.github.com/anonymous/1513fafa647403c9d8d494d2c42ac57f 18:11:58 the help would be appreciated ... 18:12:23 anyone have anything to add? 18:13:42 We should be seeing a new type from oslo soon that we can use to validate our tcp/udp lists in cfg.py :) 18:13:54 #topic Newton release schedule 18:14:13 hi trevormc 18:14:18 go ahead 18:14:19 we dont use oslo's tip though do we? 18:14:34 we would only see that next release 18:14:44 or am I mistaken... (amrith, you should know this) 18:14:48 could we hold that conversation till open discussion please ... 18:15:02 #link http://releases.openstack.org/newton/schedule.html 18:15:06 I will be marking a test client branch tomorrow, and once next week. The absolute deadline for all client changes is week R-5. After that, no new features, no bug fixes till AFTER Newton is cut. So it is important that ALL changes that must go into python-troveclient make it by Wednesday of week R-5. The release team wants clients tagged by Thursday. 18:15:13 everybody OK with this ... 18:15:23 well, there's not a lot I can do if you aren't ok with it. 18:15:56 I've put a list of all the client changes that I think still need to go in 18:16:07 If anyone wants to add to the list, feel free 18:16:15 (it's in the agenda) 18:16:16 #topic Client reviews that need to be merged 18:16:23 good introduction to that topic 18:16:26 peterstac, all yours ... 18:16:29 and there we are :D 18:17:20 Those six reviews are ones that I found that I think need to go into Newton 18:17:58 I didn't add the one for backup scheduling, but it should probably be there too 18:20:05 A couple of them still need some TLC, but I'd be good if we could get them reviewed before Friday in case they need tweaking 18:20:22 sure, will help with reviews 18:20:23 (for example release notes) 18:20:32 yup, will review today 18:21:36 so there are some others on the list 18:21:48 I've put a note on the ones without release notes - will bug the authors to get those up today 18:21:53 I assume we're not going to hold our breaths for https://review.openstack.org/356293 18:22:13 I'll try and get to https://review.openstack.org/296667 18:22:19 as soon as I finish api-ref 18:22:55 I'll add that to the list if you like amrith 18:23:09 I'm hoping not 18:23:12 let's focus on these ones 18:23:25 ok 18:23:28 if we get the others to the point where we can get them working, let's add later 18:23:31 if that works for all. 18:24:13 sure 18:24:20 ok, ready to move on? 18:24:30 (I added sched backups to the list though, since it'd be good to have that in) 18:24:31 johnma, ... 18:24:37 ok, thx peterstac 18:24:39 yep, I'm good 18:24:40 yes amrith 18:24:55 #topic Open Discussion 18:24:58 I had one on there 18:25:04 and I know trevormc has something 18:25:11 trevormc, go ahead 18:25:18 Hey sorry for speaking out of turn. 18:25:23 no such thing :) 18:25:30 Range type is making good progress in oslo now 18:25:31 https://review.openstack.org/#/c/339787/ 18:26:00 so to mvandijk's question, yes 18:26:03 once this merges in oslo 18:26:08 It's being co-authored by a core and it looks like it will be merged now. It converts integers like 1-8 to a python range(1, 8) 18:26:14 it'll be the next release before we get to see this in the wild 18:26:25 it'll be Ocata before we get to see this 18:26:30 requirements are now frozen 18:26:56 well, maybe not 18:27:03 Final release for non client libs is R-6 18:27:13 if oslo pulls this one out of a hat next week, it may make it to newton 18:27:25 so trevormc did they give any indication when they'd merge this? 18:27:39 ahh ok, I'm not familiar with they're cycles. This is just an update because one of the bugs I'm working on is dependent on this. 18:28:08 As for when they will merge it, I said soon because of the activity from the core and a previous +2 I think they're just doing some finishing touches 18:28:20 so I really don't know exactly when 18:28:28 I just asked harlowja on #openstack-oslo 18:28:45 he's in this room as well, so if he sees this, he may respond here. 18:28:51 anyway, good deal trevormc 18:28:54 nice work! 18:29:05 mvandijk, did that answer your question (at least in principle) 18:29:28 I'll assume that this is a yes ... 18:29:30 yes 18:29:38 OK, I wanted to also bring up https://review.openstack.org/#/c/316381/ 18:29:57 and now https://review.openstack.org/#/c/356631/ which has replaced that one. 18:30:10 finally got it patched to the point where the failing tests were all fixed 18:30:13 but it is messy 18:30:21 hence the respin https://review.openstack.org/#/c/356631/ 18:30:28 that one is a little cleaner 18:30:38 but it just failed py27, have to see why. it'll be something silly. 18:30:46 but we have to get this merged in the next couple of days. 18:30:55 will need some help once the gate passes. 18:31:15 that's all I had 18:31:28 anyone have anything else they want to discuss ... 18:31:49 nothing from me 18:31:53 trevormc, harlowja just said the change could merge (on #openstack-oslo) 18:31:58 about the tempest tests 18:32:06 johnma, yes please go ahead 18:32:11 EmilienM, mentioned that 18:32:18 and yes, I said I'd discuss at the meeting today 18:32:20 he pushed up a change. 18:32:26 sorry I was on vacation last week, so couldnt attend the meeting 18:32:38 #link https://review.openstack.org/#/c/356587/ 18:32:47 yup 18:32:50 that one 18:32:54 hope you had a good vacation 18:33:29 hello ! 18:33:31 right, that one is different from why the gate-tempest-trove job is still failing 18:33:52 we (puppet) are testing trove and our jobs are failing 18:34:10 I think my patch might help though I haven't tested it yet, I'm working on it 18:35:18 right. we need to update the devstak-gate.yaml and update the job definition to include the Trove tempest plugin is installed in the tempest venv. Matthew pointed me to how we could do this. 18:36:20 so I am going to put out a patch for it today. its going to be a patch under openstack-infra, and not trove 18:37:57 thx johnma 18:38:08 so Emilien, the gate-tempest-dsvm-trove is not detecting any trove tests to run and looks like this is because the Trove tempest plugin is not getting installed in the tempest venv 18:38:30 once thats fixed, along with Emilien's change all the tests should pass 18:38:37 ok, puppet CI doesn't have the same error 18:38:45 but yeah 2 changes are required I guess :) 18:39:34 #action johnma to push infra change to enable tempest plugin install ... 18:39:43 I will put out a patch today itself and we will try to get this resolved for you EmilienM. 18:39:56 awesome thx folks 18:40:01 right. Thanks for your patience EmilienM 18:40:19 thx EmilienM 18:40:52 anything else folks? 18:41:42 going once ... 18:41:54 going twice 18:42:04 enjoy the extra 15m 18:42:06 #endmeeting