19:02:06 #startmeeting infra 19:02:07 Meeting started Tue Oct 11 19:02:06 2016 UTC and is due to finish in 60 minutes. The chair is pleia2. Information about MeetBot at http://wiki.debian.org/MeetBot. 19:02:08 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 19:02:10 The meeting name has been set to 'infra' 19:02:14 #link https://wiki.openstack.org/wiki/Meetings/InfraTeamMeeting#Agenda_for_next_meeting 19:03:29 #topic Announcements 19:03:38 so I'm here chairing because 19:03:42 #info reminder: fungi is on vacation October 8-16, ask pleia2 if you need things done prior to his return 19:03:54 fungi ran away from hurricane olaph 19:04:04 to Canada 19:04:05 :S 19:04:06 and in other news... 19:04:08 on a boat 19:04:10 #info Our infrastructrure made it through another OpenStack release, hooray Newton! 19:04:15 \o/ 19:04:16 so congrats everyone :) 19:04:22 yay 19:04:26 any other announcements? 19:04:46 the summit is in 2 weeks 19:04:53 can't wait ^ 19:04:56 :-) 19:05:16 #info The Summit is coming up in 2 weeks 19:05:29 (we have an agenda item on that later) 19:05:47 #topic Actions from last meeting 19:05:50 #link http://eavesdrop.openstack.org/meetings/infra/2016/infra.2016-10-04-19.03.log.html 19:05:55 1. (none) 19:05:59 done and done ;) 19:06:05 #topic Specs approval 19:06:11 #link https://review.openstack.org/381330 Zuul v3: correct vagueness in describing job inheritance 19:06:34 since this was a "sync with reality change" it was only supposed to stay open until Thursday, October 6, but I think between hurricane prep and trip prep, fungi didn't get to it, folks can still chime in if they want 19:06:46 I imagine he'll approve when he gets back 19:07:00 #link https://review.openstack.org/381329 Zuul v3: update with Ansible role information 19:07:31 on behalf of jeblair, this one is also open from the last meeting, but no deadline was set on it so just know that it's still out there if you haven't looked at it yet 19:07:44 any other specs to look at? 19:08:34 and it doesn't look like we have anything from the Priority Efforts this week, so I'll just move on to the other topics unless anyone comes up with anything in the next couple minutes 19:08:56 infracloud is related to my cacti item, so i'll just wait 19:09:06 sounds good 19:10:40 alright then, we'll move along to the other topics 19:10:44 #topic Ocata Summit Planning (pleia2, c/o fungi) 19:10:58 fungi sent an email last week about firming up our design summit sessions 19:11:07 #link http://lists.openstack.org/pipermail/openstack-infra/2016-October/004768.html 19:11:52 he made a proposal there as to what to do, please read it if you haven't yet, there were crickets on the mailing list :) 19:12:03 ah, sorry, it worked for me so I just didn't say anything 19:12:08 me too 19:12:35 I have keys to add them to the schedule, just want to make sure we have some vague consensus here 19:12:36 erm , yeah 19:12:37 me also 19:12:44 i could have just nod about th einfracloud items myself 19:12:49 Maybe fungi needs to do a worse job so we have something to complain about next time. :) 19:12:55 pleia2: I'm not attending so I have no opinion 19:13:00 i'll reply afterwards 19:13:02 ya I didn't really have any opposition to what was there 19:13:10 nibalizer, jhesketh? 19:13:12 bkero: haha, that'll be a tough one :-) 19:14:26 pleia2: sadly I won't be at the summit :-( 19:14:45 jhesketh: :( 19:14:46 a lot of that this time around 19:14:54 :( 19:14:58 jhesketh: we can be not at the summit together 19:15:00 ill be at the summit, yes 19:15:12 but cant see the sechedule becauae phone-- 19:15:31 nibalizer: your apple device can't look at http://lists.openstack.org/pipermail/openstack-infra/2016-October/004768.html ? 19:15:34 anteaya: oh exciting, I'll get to see Canada instead :-) 19:15:45 jhesketh: would love to have you here 19:16:03 clarkb: you and your fancy web 19:16:22 I just have to give nibalizer a hard time for his apple device >_> 19:16:27 hehe 19:16:32 Next time maybe a device that can open URLs. 19:16:55 its more loke i switched to weechat 19:17:09 ans cant get the nicklist to go away, so i cant click the url 19:17:13 alright, that's consensus enough and we all did collaborate on the etherpad, I'll work on figuring out the UI for this scheduling thing this afternoon and add stuff 19:17:22 this worked fine on my irssi >.> 19:17:32 nibalizer: I can help after meeting if you like 19:17:36 kk 19:17:36 nibalizer: that large nicklist 19:17:56 pleia2: thanks 19:18:15 #action pleia2 to add infra summit schedule 19:18:26 ok, let's move along then 19:18:31 #topic Replacement of cacti node (rcarrillocruz) 19:18:37 rcarrillocruz: floor is yours :) 19:18:38 so, our cacti is sad 19:18:43 we need a bigger instance 19:18:47 which I have created 19:18:50 i have two changes for that 19:18:50 yay 19:18:53 oh good 19:19:00 https://review.openstack.org/#/c/383761/ 19:19:05 https://review.openstack.org/#/c/378996/ 19:19:13 #link https://review.openstack.org/#/c/383761/ 19:19:15 #link https://review.openstack.org/#/c/378996/ 19:19:22 and i'd like to hear how people want to proceed 19:19:31 should we schedule a schdule window or we just do it? 19:19:49 i believe it's struggling regardless, not that we will miss much functionality 19:19:52 also , INFRACLOUD 19:20:00 afaict it's useless at the moment 19:20:01 rcarrillocruz: on the firewall front I think you should keep both IPs in the fireawll to start 19:20:06 then remove the old one when things are happy 19:20:11 i've held up chocolate setup on nodepool just to not worsen things 19:20:24 clarkb: good call 19:20:38 I'm inclined to suggest we do this ASAP 19:20:46 but we do have a pretty lean crew this week 19:20:51 so as soon as we get a new cacti, we can get chocolate rolling (and start looking at newton redeploy) 19:21:15 rcarrillocruz: I will leave comments, as for scheduling we don't have any major outages planned between now and summit where we might want to avoid cacti outages so I think we can just do it 19:21:30 the process would be stop cacti, umount fs, detach cinder volume, attach on new machine, mount fs and start cacti as far as i can see 19:21:34 rcarrillocruz: basically get the two needed changes in then when you have some time go for it 19:21:43 sweet 19:22:01 i can do that tonight then, since tomorrow is bank holiday in Spain 19:22:10 yeah, tomorrow is a holiday for me too 19:22:24 I think there was some gotcha related to removing the cinder volume from the old host 19:22:32 i'll ping folks for final reviews after i push a patchset for the iptables change 19:22:33 where if you don't do some step it takes horus to timeout? 19:22:33 permissions is the big one, right? 19:22:40 oh, that 19:23:28 (thanks everyone who has pmmed me the answer to my weechat issue btw) 19:24:29 rcarrillocruz: we don't want it to be broken on the holiday, so want to do it tonight, or Thursday? 19:24:47 is getting late 19:25:00 yeah, thursday works too 19:25:12 maybe just write up a quick plan tonight in case we remember cinder weirdness issues 19:25:13 I will be around thursday (all week too) 19:25:18 me too 19:25:24 ++ thx 19:25:29 lbragstad, Gerrit is unclear in the patch ordering. THe first of your patches is "Make sure all v3 tokens are validated the same" right? 19:25:31 thanks rcarrillocruz 19:25:48 and the last is "One validate method to rule them all..." 19:25:52 ayoung: you are in the infra meeting 19:25:55 sorry 19:26:05 ok, sounds like we have a plan for Cacti 19:26:17 #topic Open discussion 19:26:25 look at all this time for open discussion :) 19:26:32 heh 19:27:39 we are running with tls-proxy enabled on a few tempest jobs now. We still need to sort out why apache restarts and reloads break on centos but not fedora or ubuntu. Also there is an issue with nova api talking to neutron for port listings being tracked by e-r 19:27:48 (more of a general FYI than anyone needing to take action) 19:28:17 i don't remember now, but did we agree on waiting until after openstack release or after summit to try gerrit version with fixed jgit? 19:28:43 zaro: after summit because we weren't erally comfortable cramming it in before summit especialyl with all the travel/vacation people seem to be doing beforehand 19:29:02 ok. 19:29:14 how people feels about looking at redeploying chocolate infracloud on newton? 19:29:16 pre-summit 19:29:27 or a nice item to hack on *during* summit 19:29:29 ? 19:30:11 i could try to help with that during the summit 19:30:27 I dno't have a ton of time between nwo and summit 19:30:39 k, i'll just look at other stuff then 19:30:41 but would likely help with that during summit sprint day 19:31:00 same, time until summit is busy for me 19:31:04 rcarrillocruz: that mgiht be a good excuse to do that "how to deploy" infracloud thing that we have talked about? 19:31:04 but at summit is good 19:31:33 yeah 19:33:00 any other things? 19:33:16 yes, something that just popped up in -infra 19:33:21 neutron + multinode testing is still broken 19:33:30 and I think it would be good to fix that before adding new multinode jobs 19:33:51 yeah 19:38:34 while in open discussion, I have a review to promote 19:38:47 https://review.openstack.org/#/c/382447/ 19:38:48 hah, ok :) 19:39:07 ah, a spec 19:39:20 It's something I planned to work on but wanted some quick confirmation that I crossed all ts 19:40:09 basically pushing TC documents under /tc and setting up redirects, so that all three sites (tc, uc, election) are served from a neutral / index page 19:40:20 (without breaking existing links) 19:40:38 so if you have a few spare cycles, please consider reviewing 19:40:50 19:41:14 probably would have made sense to talk about the structure of things more formally earlier, but fixing it now is good 19:42:00 ok, I think we'll wrap up early if no one has anything else 19:43:11 i'm good 19:43:18 nothing more from me 19:43:30 alright, thanks everyone 19:43:35 #endmeeting