14:01:21 #startmeeting tripleo 14:01:22 Meeting started Tue Dec 8 14:01:21 2015 UTC and is due to finish in 60 minutes. The chair is dprince. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:01:23 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:01:25 The meeting name has been set to 'tripleo' 14:01:41 o/ 14:01:43 o/ 14:01:44 o/ 14:01:49 hiya! 14:02:23 ohai \o 14:02:34 hello all 14:02:34 o/ 14:03:02 #topic agenda 14:03:02 * bugs 14:03:02 * Projects releases or stable backports 14:03:02 * CI 14:03:04 * Specs 14:03:11 * confirm that people are okay with *not* renaming tripleo-common just yet, even though we're placing the API code in there (tzumainn) 14:03:17 * open discussion 14:03:38 ^^^ that looks like the agenda today (from the wiki) 14:04:59 okay, lets get started 14:05:04 #topic bugs 14:05:14 #link https://bugs.launchpad.net/tripleo/+bug/1521922 14:05:14 Launchpad bug 1521922 in tripleo "CentOS delorean has migrated to aodh but we haven't yet" [Critical,In progress] - Assigned to John Trowbridge (trown) 14:05:40 This one is the only thing (as of last week anyways) blocking working trunk delorean 14:06:10 for RDO anyways, where we dont have the luxury of pulling some subset of packages from current 14:06:39 trown|BOS: I like that you are a BOS(s) today 14:07:12 trown|BOS: I saw your patch yesterday and was waiting for the -ha CI to pass before sending it 14:07:13 EmilienM is working on actually replacing the alarm functionality, but I put up two patches to do the removal part since it does not currently work anyways 14:07:20 trown|BOS: https://review.openstack.org/#/c/253716/ 14:07:28 dprince: there is a similar patch needed for THT 14:07:36 o/ 14:07:44 this morning I sent a patchset to make the upgrade working from ceilo to aodh 14:07:44 https://review.openstack.org/#/c/253717/ 14:07:51 it should pass CI 14:08:11 https://review.openstack.org/#/c/241412/ 14:09:08 EmilienM: thanks, will review these things 14:09:45 trown|BOS: so it sounds like RDO is always going to be at risk for breakage from us because we treat some packages as "special" and build them inline 14:10:26 dprince: that is fine though... RDO current does not need to be stable 14:10:45 dprince: as long as we can get a stable snapshot, and keep moving it forward 14:10:53 trown|BOS: okay 14:11:13 we just dont have a starting point yet for trunk 14:11:15 #link https://bugs.launchpad.net/tripleo/+bug/1522943 14:11:15 Launchpad bug 1522943 in tripleo "UpdateDeployment on non-controller roles can leave packages not updated" [High,In progress] - Assigned to James Slagle (james-slagle) 14:11:33 looks like slagle is working on that critical bug ^^^ 14:11:45 i've posted a patch and would like some reviews 14:11:48 would be good to review that soon too 14:11:59 slagle: cool, thanks 14:12:40 any other bugs to highlight this week? 14:13:24 #topic Projects releases or stable backports 14:13:47 anyone have an update on stable branch progress this week 14:14:08 ya I do 14:14:10 i wanted to mention that we're going through and doing releases from the stable branches 14:14:33 and as part of that, will be bumping the major versions on the master branches 14:14:39 slagle: cool, for all the tripleo projects? 14:14:47 which means releasing a 1.0.0 for some of the projects 14:14:55 dprince: yea, eventually 14:14:58 slagle, are you doing the releases? 14:15:20 rhallisey: for some, not all of them 14:15:36 slagle, who's doing tripleo-heat-templates and tripleo-image-elements? 14:15:48 I just want those to be picked up in the next puddle 14:16:23 marios was going to start looking at tht 14:16:29 there is no timeframe on when it will be done 14:17:01 I'll volunteer to help to speed it up 14:17:09 if that's ok 14:17:16 sur 14:17:17 e 14:17:22 cool 14:17:57 dprince, I have a bunch of backports for stable liberty. Should I mention the patch set? 14:18:13 all container related 14:18:32 rhallisey: probably wait for open discussion for the containers patches... 14:18:35 kk 14:18:42 rhallisey: any other updates here though? 14:18:48 dprince, nothing else 14:19:09 #topic CI 14:19:40 we look to be running fairly smoothly. I noticed a few HA job failures which might need a closer look 14:23:21 #topic Specs 14:24:33 The composable roles spec is still under review... good feedback last week but we are still discussing the possability of using Heat resource chains to implement some of the internal features there 14:24:55 * dprince would like the external (Heat parameters) interface to be very simple though 14:26:44 I noticed Steve Baker posted a nice spec about enhancing tripleoclient (or whatever) with extra output about heat software deployment failures 14:26:47 #link https://review.openstack.org/#/c/251587/ 14:27:04 I noticed this last Monday... but worth mentioning here 14:28:21 any other specs to highlight this week? 14:29:19 #topic one off agenda items 14:29:32 tzumainn: your up 14:30:03 hiya! okay, I just wanted to confirm that people are okay with not renaming the tripleo-common repo 14:30:40 from recent IRC conversations, it seems like most people think that renaming is premature, and I agree that it just seems to add extra work at a time when we're trying to add a bunch of new code 14:31:13 er, that's all I had 14:31:34 tzumainn: cool, yeah I tend to agree 14:32:15 yea +1 for not renaming just yet 14:32:24 risk v. reward seems to favor not renaming at this point 14:34:06 okay, awesome 14:34:25 cool, moving on 14:34:31 #topic open discussion 14:34:45 rhallisey: you mentioned some containers stuff? 14:35:05 dprince, ya I have a bunch of backports 14:35:06 https://review.openstack.org/#/c/254408/1 14:35:18 that patch chian 14:35:21 chain* 14:35:25 sweet 14:36:11 just a gentle reminder about the ui spec, it hasn't been getting any attention recently (sorry I missed to mention in the spec topic) 14:36:13 #link https://review.openstack.org/#/c/239056/ 14:37:01 akrivoka: ack, thanks for mentioning this 14:37:45 dprince: on CI, sorry missed topic change while in a meeting 14:37:50 The rh1 overcloud is in a funny state, I've spent a little time trying to figure it out but havn't managed it yet, out of office all week so wont have a chance to look this week. 14:37:54 too many instances are going to the error state, nodepool can't delete them so they are using up nodepools quota. 14:37:59 things are only working because I'm running "nova reset-state" on instacnes in the error state every day or two(this allows nodepool to delete them). 14:38:39 derekh: oh snap, I didn't realize we had this happening 14:39:01 derekh: any pattern, like possibly a bad compute node or two? 14:39:15 dprince: yup, its been happening since we had the nodpool regression a week two ago 14:39:44 dprince: that regression caused lots of VM's (LOTS) to constantly get created and destroyed, 14:39:46 derekh: would a stale (perhaps invalid) floating IP be the cause 14:39:54 dprince: im thinking it provolked something, 14:40:00 haven't noticed a pattern 14:40:04 derekh: yeah, nodepool is aggressive 14:40:51 provolked? 14:41:05 Anyways, I can get back to digging into it next week, but in the mean time if anybody else wants to take a look... 14:41:08 trown|BOS: yeah, provoked. It picked a fight with us man 14:41:16 ah... haha 14:41:22 trown|BOS: *provoked 14:41:39 * derekh can't spell 14:42:19 derekh: thanks for the update 14:43:29 np 14:46:05 I mentioned this in 14:46:30 the workflow email list this week... but I posted a Mistral demo yesterday 14:46:36 #link https://www.youtube.com/watch?v=bnAT37O-sdw 14:47:22 13 minutes if you are at all interested in the TripleO workflow discussions 14:48:16 havent watched the demo yet, but looked over the bulk introspection POC 14:48:19 * derekh downloads for air plane viewing 14:48:23 I think that will be a big improvement 14:48:32 I think the ui people were looking into it as well 14:48:46 right jtomasek ^ ? 14:48:58 * trown|BOS pictures derekh watching that 32 times in a row 14:49:17 derekh: I hope you like my voice a lot 14:49:48 trown|BOS: I'll put the speakers on to see what everybody else thinks 14:50:00 :) 14:50:01 jrist, dprince: yeah, we're looking forward to the meeting in 40 minutes 14:51:50 okay, It sounds like it is all crickets now so lets call it 14:52:16 thanks everyone 14:52:17 thanks all! 14:52:20 \o 14:52:24 thanks 14:52:26 o/ 14:52:31 #endmeeting