15:00:26 #startmeeting openstack search 15:00:27 Meeting started Thu Jan 5 15:00:26 2017 UTC and is due to finish in 60 minutes. The chair is sjmc7. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:28 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:30 The meeting name has been set to 'openstack_search' 15:00:46 o/ 15:00:51 o/ 15:00:53 Good morning 15:01:11 o/ 15:01:15 o/ 15:01:46 I don’t have much on the agenda so if anyone has anything they want to discuss, please bring it up 15:02:01 We’ll get started in another minute 15:02:50 #topic Reviews 15:03:16 The only big outstanding reviews are the pipeline one (https://review.openstack.org/#/c/359972/) and the ironic plugin (https://review.openstack.org/#/c/364462/) 15:03:35 They’re both really close and I think we can get them in in the next few days 15:03:47 cool 15:04:37 I had a look through the blueprint list yesterday and I didn’t see any big things we wanted in ocata. Versioned notifications jumped out; I don’t know if anyone’s looked at those 15:05:47 Worked 15:05:55 On some of those 15:06:09 But didn't got very deep 15:06:31 Me neither. I think we will definitely need to start supporting them in Pike 15:07:02 Btw, I just noticed this today 15:07:18 Do we handle instance tags ? 15:07:59 they are mapped 15:08:06 I mean, just noticed that update/create/delete tags don't sent out notification 15:08:21 hmm, that's not good 15:08:22 We got it by instance 15:08:33 we have that problem in a number of places 15:08:35 do they need a defect on nova? 15:09:00 I planning to add it in nova next release 15:09:11 Talked to some cores today 15:10:20 I guess if you file a bug against nova mark it as also affects searchlight 15:10:28 Yeah 15:11:10 It's actually our ceilometer product team reported to me :) 15:11:16 It’d be much nicer for us if the notifications didn’t mirror the database schema; we have a lot of problems resulting from one-to-many fields sending very limited notifications 15:11:44 Well, at least someone’s looking :) 15:12:06 I had a look through our bug list too; there wasn’t anything that really jumped out 15:13:33 #topic ocata-3 15:13:41 O-3 is the week of January 23rd 15:13:54 this has been a very short cycle 15:14:08 yeah, it really snuck up on us! 15:14:20 Especially with the holidays in the US 15:14:56 yeah, and chinese new year happens at late january : ( 15:15:16 Ah, when is that? 15:15:35 y 15:15:48 probally from jan 27th I guess 15:16:14 Maybe i’ll celebrate it, i could do with a couple more days off 15:16:16 we will have a week holiday then 15:16:33 There’s also the PTG in late February; I will almost certainly not be able to make it 15:17:27 short otaca collides with two big holidays 15:17:55 Yeah. It’ll balance out better with the following release cycles; this one was because of fixed dates for the conference I think 15:18:15 Yeah, following cycles will be back to normal length 15:18:37 I have something i'll be looking at for searchlight ui. however, in trying to verify on new devstack i hit a hiccup 15:18:46 a hiccup in devstack? never 15:18:48 #topic Open discussion 15:19:09 yeah, i couldn't install the java package for ES. 15:19:19 looks like it is for 16.06 15:19:29 or 16.04.. 15:19:35 You couldn’t install java? 15:19:50 trying to remember (was before holidays) 15:19:53 Check which version you/it was trying to install 15:20:01 They don’t backport old JRE packages 15:20:02 but I think it is trying to install default-jdk 15:20:03 I met before, are you installing openjdk8? 15:20:54 lemme check... 15:21:08 but basically, the install script was looking for something not in the 14.04 repos 15:21:38 File a bug for it 15:22:00 i thought lei-zh fixed this 15:22:29 https://github.com/openstack/searchlight/blob/master/devstack/plugin.sh#L294 15:22:39 yeah, i think he fixed it so it works on 16.04 15:22:47 ah, that doesn’t exist on 14? 15:22:56 yeah, I have a patch to use default jdk instead 15:22:56 i think so.... 15:23:30 devstack officially supports 14 as we as 16 so we should make it work 15:23:55 previous script couldn't work on 16.04 15:24:24 i'm using 14.04.4 and didn't get the issue yet 15:24:30 hmmm... 15:24:41 well, i tried updating repos 15:24:44 TravT: make sure you also have updated apt list 15:24:46 but kept hitting it 15:24:51 ah :) 15:25:04 maybe i did it on a bad day 15:26:11 I don't think you *have* to support 14.04, fwiw... 15:26:28 okay, well, i actually have an upgrade to 16.04 running as we speak. 15:26:42 Devstack says it aims to support 14.04. shouldn’t be a big deal to support both 15:27:37 Fair enough 15:27:38 Anything else anyone wanted to discuss? 15:27:51 Did I miss the pipeline stuff? 15:27:57 I can check logs if so 15:28:07 * robcresswell thinks pipeline is the right term 15:28:17 It’s really close to merging 15:28:35 I think lei-zh put up a patch last night (my time), gonna take a final look today 15:28:49 Hoping to get it merged by tomorrow 15:29:43 Awesome, thanks 15:30:23 Sorry it’s taken a bit long 15:30:31 December’s always a terrible time to try to get things in 15:30:47 robcresswell: have your colleagues started doing any more with sockets in horizon? 15:31:49 TravT: Got a meeting with them in an hour or so 15:31:56 ok. cool 15:32:31 They've been looking into it for a few weeks, so should be demoing today 15:33:06 very cool 15:33:40 after pipeline, I could work on zaqar publisher 15:33:44 Yeah. I'll let you know when I have more solid info 15:34:42 lei-zh: Anything that helps plug everything together would be great 15:35:11 But I would imagine you have other priorities than Horizons experiments :) 15:36:43 robcresswell: yeah, I'll balance them : ) 15:38:24 yeah, be interesting to see that 15:38:30 Anything else? 15:39:46 Ok. If that’s it, I’ll close out. Thanks everyone 15:40:07 thanks! 15:40:13 bye 15:40:14 have a good evening! 15:40:31 Thanks all 15:40:34 #endmeeting