15:00:13 #startmeeting openstack search 15:00:16 Meeting started Thu Sep 22 15:00:13 2016 UTC and is due to finish in 60 minutes. The chair is TravT. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:17 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:20 The meeting name has been set to 'openstack_search' 15:00:25 o/ 15:00:31 o/ 15:00:52 o/ 15:00:56 o/ 15:00:57 we'll give a moment or two 15:01:18 morning 15:01:48 * matt-borland is taking notes of the SL-UI study going on right now 15:01:50 'eloo 15:02:17 i'm gonna log in as well to that... but won't be note taking 15:03:14 o/ 15:03:19 ok 15:03:22 let's get rolling 15:03:27 actually have quite a few things today 15:03:43 https://etherpad.openstack.org/p/search-team-meeting-agenda 15:04:33 #topic Newton release status 15:04:43 rc1 is tagged 15:05:16 this means we are in bug/fix mode 15:05:31 and should be looking to get particularly useful fixes in 15:05:39 and backport any to stable/newton 15:05:52 i’ve been tagging stuff newton-backport-potential where i think it’s appropriate 15:06:06 https://review.openstack.org/#/q/project:%255E.*searchlight.*+status:merged 15:06:13 those are all that have already merged... 15:06:54 i cherry picked this one 15:06:54 https://review.openstack.org/#/c/374533/ 15:07:08 but please backport any asap that we need to 15:07:38 * TravT sorry i'm a little distracted 15:07:39 what date did you have in mind for tagging the final rc? 15:07:51 watching an operator work with searchlight ui 15:07:52 i think technically we’ve got next week but shouldn’t go to friday 15:08:10 i think we should tag by EOD thursday 15:08:12 next week 15:08:32 that works 15:09:12 Question 15:09:17 should we go ahead with this one: 15:09:18 https://review.openstack.org/#/c/348701/ 15:09:24 just has a merge conflict 15:09:42 i’m a little hesitant because of the potential for introducing bugs 15:09:56 i have tested it a fair amount 15:10:42 but it’s a pretty big patch. i guess if we get it in today or tomorrow there’s some time next week to fix anything that comes up 15:11:06 I'll fix the merge conflicts today. 15:11:13 lei-zh: yingjun any chance you guys could test it? 15:12:05 sure, I'll take a look tomorrow 15:12:08 I can send folks my turn-on scripts if thjat would help get some testing started. 15:12:24 RickA-HP: I'd be happy for you to add test scripts to the repo 15:12:37 https://github.com/openstack/searchlight/tree/master/test-scripts 15:12:48 steve gets a little cranky about such a thing 15:12:58 Sure 15:13:02 So I've heard. :) 15:13:59 thanks yingjun lei-zh for looking at it 15:14:10 if you guys +2 it with sjmc7 then let's put it in 15:14:59 i'm also inclined to push the ref docs through sooner than later. 15:15:00 https://review.openstack.org/#/c/357287/ 15:15:17 i’ll take a look at those today 15:15:38 It's on my list to review today. 15:15:46 thanks rosmaita for the diligence on them 15:15:53 np 15:16:00 #topic matt-borland for searchlight ui core 15:16:06 as i sent out on the ML 15:16:13 he's up for core on searchlight-ui 15:16:40 i’m good with that. i htink most of us have trouble reviewing those patches 15:16:43 this will require a new grouping in gerritt where we can put allow UI only cores 15:17:16 I endorse both proposals! 15:17:28 okay, i'll make that change later today or tomorrow. 15:17:40 or if you want to do it sjmc7, that'd be very nice as well. 15:17:45 i can do it 15:17:50 awesome, thank you 15:17:57 Consider it PTL training! 15:18:07 speaking of which 15:18:08 just for the record, i am good with both proposals (as indicated on the ML) 15:18:22 #topic ocata 15:18:27 Congratulations to our new Ocata PTL 15:18:32 sjmc7 15:18:48 Yay! Good job Steve. 15:18:50 thanks! 15:18:55 congrat! 15:18:59 with newton wrapping in the next week, i believe the baton will officially pass to him starting when we tag rc2 15:19:08 my reign of terror will begin 15:19:13 i'm already afraid 15:19:16 ;) 15:19:31 cong 15:19:52 i have to say, i'm really enjoying watching this guy use the searchlight UI... 15:20:11 he's seems rather pleased with it 15:20:12 is he in the room? or on video? 15:20:19 he's on the hangout right now 15:20:55 i think it is going to be pretty hard to find anyone who thinks the searchlight UI is not an improvement 15:21:02 haha 15:21:16 so, for ocata 15:21:26 Nova Specs From PTL on ML: 15:21:50 any that were filed previously for notifications need to be resubmitted for ocata 15:22:08 hahaha, the cycle continues! 15:22:36 ocata’s a short release - did they say when the freeze for nova is? 15:22:44 i would guess december? 15:22:50 i'm not sure 15:22:57 feature or spec 15:23:09 spec will be soon 15:23:11 the spec freeze is pretty early 15:23:21 before the summit, i think 15:23:22 I can continue the nova work now:) 15:23:37 feature will be until o3 15:23:45 both priority 15:23:50 and non 15:24:50 wow. ok, so we should get the ones in we know about 15:25:32 are we missing any specs over there? 15:26:01 i know we have flavors, server groups plugins in sl without notifications now 15:26:20 i will file bps tomorrow 15:27:01 we also need hypervisors 15:27:41 thanks Kevin_Zheng 15:28:02 #topic summit 15:28:24 :) 15:28:31 first of all, who's going? 15:28:54 i got approval and bought tickets yesterday 15:29:17 i will be there 15:29:29 I'm booking tickets 15:30:02 o/ 15:30:06 Even though he's busy watching the UI focus group right now, Matt B is going. 15:30:10 awesome 15:30:14 our final list didnt came out 15:30:20 yes matt-borland is going 15:30:28 i know david-lyle is going as well 15:31:13 that's pretty good represenation for searchlight at the summit 15:31:31 yes, i'm very sad sjmc7 can't make this one 15:31:42 yeah :( 15:31:48 i'm basically going on his behalf this time... 15:32:06 sjmc7: i may have to mail you a consolation pie 15:32:15 the best kind! 15:32:28 so then we have to talk sessions. 15:32:35 we were allocated 1 fishbowl 15:32:50 and 2 working sessions 15:33:00 so i started an etherpad 15:33:00 https://etherpad.openstack.org/p/searchlight-ocata-summit 15:33:04 #link https://etherpad.openstack.org/p/searchlight-ocata-summit 15:33:51 need to figure out best usage of that time 15:34:18 please jump out there and start adding ideas 15:34:33 i'm particularly worried about the fishbowl session 15:34:44 that there will be too few fish? 15:34:58 lol 15:35:01 well, we could do a sort of repeat from Austin 15:35:18 sardine pie 15:35:37 we might get quite a few horizon folks 15:35:46 their fishbowl is immediately proceeding in the same room 15:35:56 my biggest concern remains the state of notifications 15:36:07 but it’s not exactly a topic that brings in a lot of interest 15:36:15 unless rosmaita gets hold of a REALLY good pie 15:36:47 not sure barcelona is a pie kind of town 15:37:44 we could do something like a mini-training session.... 15:37:53 overview on creating plugins 15:38:02 + overview on integrating into horizon 15:38:11 training on aggregation query abilities 15:38:55 we could try to do something with nova notification team... 15:39:03 our sessions overlap with nova a bit... 15:39:22 it would be nice to try to persuade them not to trim down their notifications further 15:39:58 or alternatively investigate whether hitting the API on some/all of their notifications is actually the problem i think it is 15:40:48 i really will need help with session planning this 15:41:09 i will cast my mind to it. when’s the deadline on submitting them? 15:41:18 i'm not sure exactly... 15:41:21 let me try to find it 15:41:26 hahahaha 15:41:48 the operator just said "it seems like we could replace most of the other dashboards in horizon with this search page" 15:42:09 you should take him to barcelona with you 15:42:10 lol 15:42:26 * TravT imagines david-lyle bristling... ;) 15:43:02 lei-zh: do you want to propose a working session on the pipeline architecture? 15:43:13 if so, please add to the etherpad 15:43:56 sure, I can add that 15:44:32 that’s not a bad idea. maybe see what other uses we have for it 15:45:08 "I'm looking forward to having this in our cloud" 15:45:11 another quote 15:46:07 I'm involving in another project, find polling api is really annoying yet common way to do things. 15:46:49 yeah, although i’ve seen other projects that use polling but with persistent connections which works better 15:47:58 if zaqar can do that kind of push notification i can see it being useful in a number of places 15:49:16 yes, I'll refine the pipeline patch after stable release 15:49:22 ok, well please everybody 15:49:34 just add any brainstorm ideas you can for sessions 15:49:55 if we don't come up with enough good ideas for sessions, i'll want to donate them back to the pool 15:50:06 ok 15:50:39 next week this should be our primary focus 15:50:44 in the meeting 15:50:53 thank you everybody 15:51:10 #topic open discussion 15:51:17 anything else? 15:51:56 no… just keep an eye on reviews. if you are fixing a bug that’s a candidate for backporting, please either mark it in the comments or propose the backport yourself once it’s merged 15:53:00 all right everybody. thank you very much 15:53:00 sure 15:53:09 see you next week 15:53:16 Bye 15:53:21 bye 15:53:30 #endmeeting