21:00:51 <mriedem> #startmeeting nova 21:00:51 <openstack> Meeting started Thu Apr 14 21:00:51 2016 UTC and is due to finish in 60 minutes. The chair is mriedem. Information about MeetBot at http://wiki.debian.org/MeetBot. 21:00:52 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 21:00:55 <openstack> The meeting name has been set to 'nova' 21:00:59 <jroll> hai 21:00:59 <tonyb> o/ 21:01:01 <dansmith> o/ 21:01:03 <diana_clarke> o/ 21:01:03 <melwitt> o/ 21:01:05 <knikolla> o/ 21:01:08 <sdague> o/ 21:01:09 <bauzas> \o 21:01:09 <sarafraj> o/ 21:01:09 <edleafe> \o 21:01:16 <mriedem> #link agenda https://wiki.openstack.org/wiki/Meetings/Nova#Agenda_for_next_meeting 21:01:18 <scottda> hi 21:01:22 <rdopiera> o/ 21:01:24 <auggy> \o/ 21:01:47 <mriedem> #topic bugs 21:01:49 <DuncanT> Lo 21:02:02 <mriedem> gate status looks ok 21:02:20 <mriedem> i don't really have any status on third party ci, not sure if the mellanox ci is fixed yet 21:02:26 <mriedem> lennyb: ^ any idea? 21:02:30 <mriedem> it was hitting merge conflicts 21:02:40 <alaski> o/ 21:02:47 <dansmith> I don't see it commenting even lately 21:02:48 <mriedem> i pinged tjones this morning about vmware ci not running on some changes after multiple rechecks 21:03:02 <dansmith> 2 days since the last report from mellanox that I see 21:03:41 <mriedem> i haven't checked that ci watch page in awhile 21:04:24 <mriedem> on a related note for CI 21:04:28 <mriedem> #info There is a proposal to make the ceph plugin job voting and gating (again, kind of): https://review.openstack.org/#/c/306009/ 21:04:28 <patchbot> mriedem: patch 306009 - openstack-infra/project-config - Make the ceph devstack-plugin-based job gating on ... 21:04:41 <dansmith> I love that proposal 21:04:59 <cburgess> late but here 21:05:11 <mriedem> i guess i never heard what kind of fallout there is from not having the mirroring done? 21:05:13 <mriedem> dansmith: do you know? 21:05:28 <dansmith> mriedem: just us controlling our own destiny if the ceph site goes down 21:05:31 <dansmith> which is a good plan for sure 21:05:33 <dansmith> but... 21:05:40 <sdague> it's not just if the site goes down 21:06:03 <sdague> connectivity from our providers across the wider internet goes squishy quite often 21:06:10 <dansmith> well, heh 21:06:10 <dansmith> okay 21:06:20 <dansmith> " if the ceph site appears down" 21:06:33 <mriedem> but that hasn't been a problem since red hat has been monitoring has it? 21:06:39 <mriedem> which has been ~1 month now? 21:06:44 <dansmith> nope, more than a month 21:06:47 <dansmith> but it could happen any time 21:06:56 <dansmith> so a month is probably not really a useful metric 21:07:01 <dansmith> we know the internet is not perfect 21:07:06 <mriedem> it's not? 21:07:15 <dansmith> yeah, sorry to burst your bubble 21:07:42 <sdague> anyway, it's probably worth the risk 21:07:52 <sdague> we can always make it NV if there are big issues 21:08:00 <sdague> +2ed it right now 21:08:07 <mriedem> is pabelanger going to work on the AFS mirror stuff? 21:08:23 <dansmith> mriedem: yeah, he is in the next few days it seems 21:08:29 <mriedem> ok, cool 21:08:35 <dansmith> I'm pro going forward without the mirror so the pressure is on though 21:08:40 <dansmith> because I htink it's a safe bet 21:08:48 <sdague> https://review.openstack.org/#/c/293155/ seems to be rebased 21:08:48 <patchbot> sdague: patch 293155 - openstack-infra/system-config - Apt mirror for ceph hammer release 21:08:52 <mriedem> just like the panthers in the super bowl 21:08:53 <mriedem> what could go wrong 21:09:06 <mriedem> related to the ceph job 21:09:07 <mriedem> #help Need someone to convert gate-tempest-dsvm-full-ceph-src-glance_store to using the devstack plugin so we can make that job voting for glance_store and os-brick. 21:09:18 <mriedem> ^ for newcomers 21:09:31 <mriedem> should be some trivial copy/paste/tweak in project-config 21:09:37 <dansmith> I think melwitt said she was bored earlier today 21:09:46 <mriedem> orly, i still have those charts to work on... 21:09:59 <mriedem> auggy: anything on bugs? 21:10:01 <melwitt> hah 21:10:21 <auggy> mriedem: i don't have much to report, i think people have been doing a great job helping out with the skimming and triage 21:10:21 <bauzas> nothing really huge I saw recently 21:11:03 <mriedem> #link untriaged nova bugs https://bugs.launchpad.net/nova/+bugs?search=Search&field.status=New&orderby=-datecreated&start=0 21:11:14 <mriedem> yeah, it's pretty simple, i just triaged this https://bugs.launchpad.net/nova/+bug/1570241 21:11:15 <openstack> Launchpad bug 1570241 in OpenStack Compute (nova) "ERROR (ClientException): Unexpected API Error during type the command: nova virtual-interface-list" [Undecided,Invalid] 21:11:20 <mriedem> by triage i mean marked it invalid 21:11:38 <mriedem> #topic reminders 21:11:45 <mriedem> #info Launchpad bug spring cleaning day Monday 4/18: http://lists.openstack.org/pipermail/openstack-dev/2016-April/091456.html 21:11:57 <tonyb> mriedem: we coud probably write a script to do that kind of "triage" 21:12:18 <mriedem> tonyb: meh 21:12:19 <auggy> one thing i do try to keep in mind, if an error message could be more helpful, that could be the focus of the bug 21:12:20 <bauzas> sssssht, don't awake the beast 21:12:24 <bauzas> tonyb: ^ 21:12:32 <auggy> or if we see several where people were following documentation instructions, that could be a doc bug 21:12:41 <mriedem> tonyb: or one of the 377 people in the nova channel could look at a new bug reported once per day 21:12:49 <auggy> the same documentation instructions rather, resulting in a config error 21:13:11 <tonyb> mriedem: true dat I shoudl do that thing 21:13:21 <mriedem> auggy: yeah, that reminds me of the neutron connection error bug where the person didn't have credentials setup properly for nova talking to neutron 21:13:30 <mriedem> auggy: and we talked about wrapping that so it's not a 500 in the nova api 21:13:45 <bauzas> well 21:13:48 <auggy> mriedem: yeah, the bug had already been closed as invalid so i wasn't sure if i should reopen it with that new info or file a new bug 21:13:53 <bauzas> sometimes it's documented 21:14:03 <bauzas> it's just that people don't really look at those 21:14:18 <mriedem> auggy: we should probably have a bug for that regardless, i hit a similar thing last week with listing instances and neutron server was down 21:14:20 <mriedem> we get a 500 21:14:31 <auggy> mriedem: kk i'll go ahead and file it and reference that invalid bug 21:14:42 <auggy> i still have notes from the conversation we had about steps towards fixing it 21:14:47 <mriedem> ok, thanks 21:14:58 * mriedem golf claps for auggy 21:15:00 <auggy> i believe you had volunteered rlossit so i guess i can just assign it to him?? ;P 21:15:11 <mriedem> eh, maybe 21:15:13 <mriedem> #link Newton review focus list: https://etherpad.openstack.org/p/newton-nova-priorities-tracking 21:15:26 * auggy kids 21:15:33 <mriedem> ^ mostly a reminder to keep that fresh 21:15:41 <mriedem> as things tend to build up that are already merged 21:16:25 <mriedem> #link Draft Newton release schedule is up: http://releases.openstack.org/newton/schedule.html 21:16:33 <mriedem> #link Open re-proposed specs: https://review.openstack.org/#/q/project:openstack/nova-specs+status:open+previously-approved 21:17:04 <mriedem> of those 4, 3 are going to be discussed at the summit 21:17:21 <mriedem> would be good if we were happier with the glance v2 spec before the summit 21:17:25 <mriedem> i need to go through the recent updates 21:17:46 <mriedem> the volume multiattach spec is probably not ready yet given the nova/cinder meeting we had yesterday 21:17:59 <mriedem> starting to see some progress on the design there at least 21:18:21 <mriedem> We already have 44 approved blueprints: https://blueprints.launchpad.net/nova/newton - 4 are completed, 6 have not started 21:18:30 <mriedem> #link https://blueprints.launchpad.net/nova/newton 21:19:26 <mriedem> the ones that dont have code up yet, 21:19:29 <mriedem> i'm starting to worry about 21:19:36 <mriedem> because they didn't have code up last week either 21:19:57 <mriedem> #help https://wiki.openstack.org/wiki/Nova/BugTriage#Weekly_bug_skimming_duty Volunteers for 1 week of bug skimming duty? 21:20:36 <mriedem> looks like we actually have one person for this week 21:20:37 <mriedem> cihand 21:20:59 <mriedem> #topic stable branch status 21:21:04 <mriedem> #link Stable branch status: https://etherpad.openstack.org/p/stable-tracker 21:21:19 <mriedem> so far so good for nova 21:21:24 <mriedem> we released liberty 12.0.3 today 21:21:54 <mriedem> #topic stuck reviews 21:22:06 <mriedem> there was nothing in the agenda, does anyone have anything they want to bring up here? 21:22:35 <mriedem> #topic instance users 21:22:44 <mriedem> The Cross Project workshops: Instance Users session needs Nova representation. Nova plays a part in the problem and needs to be part of the solution. Can some representatives from Nova please attend so that we can find a supportable solution that everyone will be happy with? Thanks - kfox1111 21:22:53 <mriedem> #link https://www.openstack.org/summit/austin-2016/summit-schedule/events/9485 21:23:18 <mriedem> is anyone planning to be in that session? i'm going to be in something else at the same time 21:23:22 <mriedem> hence asking for volunteers 21:23:38 <dansmith> I feel like the overwhelming feedback (and the last time we discussed this) was "do it outside of nova" so I will probably choose to attend something else during that slot 21:23:45 <mriedem> part of me is thinking about volunteering johnthetubaguy since he's (1) on vacation and (2) has replied to the spec before 21:24:33 <mriedem> alright, moving on...if you plan on being there, please let me or kfox1111 know 21:24:40 <auggy> i can attend and take notes but i can't really speak for nova 21:24:56 <alaski> I probably need to be in the quota one at the same time 21:25:09 <mriedem> alaski: yeah, there was that too 21:25:20 <bauzas> I would honestly prefer the quota discussion... 21:25:24 <tonyb> I'll be there but as with auggy can't really speak for nova 21:25:47 <mriedem> auggy: tonyb: ok 21:26:09 <mriedem> #topic open discussion 21:26:12 <mriedem> #link Austin Design Summit sessions: https://www.openstack.org/summit/austin-2016/summit-schedule/global-search?t=Nova%3A+ 21:26:30 <bauzas> I have a point 21:26:31 <mriedem> #action mriedem still needs to create some etherpads for the nova design summit sessions 21:26:40 <mriedem> bauzas: hold up 21:26:50 <mriedem> any questions on the design summit sessions? 21:27:00 <bauzas> yup 21:27:05 <mriedem> oh, about this :) 21:27:08 <mriedem> go ahead then 21:27:38 <bauzas> I saw that feature classif session was on Thurs (AFAIR) but I was thinking about discussing about feature classif also in the performance VMs session 21:27:45 <bauzas> which is on Wed 21:28:09 <bauzas> so, wondering if it was okay for us 21:28:22 <bauzas> I don't really want to swap the sessions 21:28:26 <dansmith> I'd like to focus on verification stuff during the one on thursday 21:28:35 <mriedem> we wanted feature classification after the API sessions 21:28:40 <dansmith> which I can see spending most of the time on anyway 21:28:40 <mriedem> because of it's relation to API discoverabliilty 21:28:51 <bauzas> okay, let's leave that way 21:28:52 <mriedem> wow, more i's and l's please 21:29:21 <mriedem> yeah there were 3 other xp sessions on thursday afternoon which are strict on scheduling since they are with other projects 21:29:57 <mriedem> anything else on the summit? 21:30:11 <mriedem> #link Newton midcycle planning: http://lists.openstack.org/pipermail/openstack-dev/2016-April/092014.html 21:30:18 <mtreinish> mriedem: https://www.openstack.org/summit/austin-2016/summit-schedule/events/9081?goback=1 overlaps with the qa track 21:31:03 <tonyb> mtreinish: you're not the PTL anymore so you can choose ;P 21:31:46 <mriedem> i can see if anything could be shuffled after the meeting 21:31:51 <mriedem> but it's tight 21:32:10 <mtreinish> mriedem: swap it with unconference #2 right after it and the conflict goes away :) 21:32:29 <mriedem> mtreinish: i won't be in that one 21:32:33 <mriedem> the thurs 1:30 21:32:48 <mriedem> anyway, can look later 21:33:14 <mriedem> #action mriedem to see if anything can be moved around so QA can be at the feature classification session 21:33:31 <mriedem> for the midcycle 21:33:32 <mriedem> It's sounding like 7/19-7/21 at the Intel campus in Beaverton, OR is what we'll be doing, but it's not confirmed yet. 21:34:08 <mriedem> i never heard back about vmware 21:34:24 <tonyb> mriedem: did they address mikal's query? 21:34:25 <sdague> PDX is good 21:34:30 <dansmith> that's cool with me.. I don't remember issues at intel 21:34:33 <mriedem> does anyone have any major issues with that date and location? besides the chaperoned bathroom runs? 21:34:34 <dansmith> I distinctly remember peeing alone 21:34:45 <tonyb> mriedem: do we have feedback from horizon on how ther last mid-cycle wemt? 21:34:51 <bauzas> mriedem: I commented something 21:35:02 <bauzas> mriedem: about the flight prices 21:35:05 <sdague> Intel's network is pretty port locked down, but I thought it ended up fine 21:35:17 <tonyb> sdague: ssh out? 21:35:24 <david-lyle> IRC won't go through on the guest net, ssh works fine 21:35:29 <dansmith> I vpn out of everything anyway, so I didn't notice 21:35:37 <sdague> oh, maybe that was it, IRC 21:35:39 <mriedem> bauzas: ah, so flights from europe are considerably cheaper in R-15 rather than R-11? 21:35:51 <mtreinish> david-lyle: does ssh have to go over port 22? 21:35:53 <bauzas> I was only checking on my own 21:35:55 * tonyb does email/irc/git over ssh anyway so cool 21:36:00 <bauzas> so that could be only me 21:36:05 <david-lyle> mtreinish: I believe so 21:36:06 <bauzas> just raising a flaf 21:36:08 <bauzas> flag 21:36:38 <mriedem> i really have no idea how many other people from europe are planning on making the midcycle 21:36:45 <bauzas> but anyway, I don't want to stop things 21:36:49 <mriedem> or asia for that matter 21:37:00 <bauzas> just expressing that point as worth it 21:37:07 * tonyb will be there 21:37:17 <mriedem> bauzas: yeah, thanks for looking it up 21:37:35 <mriedem> personally i like the R-11 date for when it falls in the schedule 21:37:50 <bauzas> i agree 21:38:14 <bauzas> R-11 showed that it's a good time for sprinting against our left prios 21:38:14 <mriedem> any other stuff about the midcycle? 21:38:44 <david-lyle> Intel is trying to make sure we have an appropriate room available, WIP 21:38:51 <auggy> i'm also local PDX so if we do go with the Intel option here, I'm happy to help out with any local organization stuff 21:39:16 <david-lyle> but a date helps 21:39:28 <dansmith> david-lyle: we were already told we're good 21:39:37 <mriedem> david-lyle: for either date, yeah 21:40:03 <david-lyle> dansmith, mriedem ok, I missed that 21:40:17 <sdague> I think R-11 is a better point in our schedule to do it 21:40:47 <mriedem> yeah, so i'll follow up with them next week 21:40:57 <mriedem> #action follow up with Intel on the midcycle date 21:41:12 <jlvillal> o/ 21:41:15 <mriedem> last item in the agenda is should we have the nova meeting next week? 21:41:40 <mriedem> i don't expect to make any ground breaking decisions 4 days before the summit 21:41:40 <sdague> it's the early one right? I think it's fine for sorting any last minute summit item 21:41:53 <mriedem> it is the early one for US people yeah 21:42:03 <edleafe> yeah, a quick meeting would be good 21:42:04 <mriedem> i don't mind holding it and it just ends fast if nothing to talk about 21:42:07 <mriedem> ok 21:42:08 <sdague> yep 21:42:16 <mriedem> i'll find some stuff to fill it out for dansmith 21:42:25 <mriedem> anyone else have anything for open discussion? 21:42:31 <tonyb> 1 thing 21:42:49 <tonyb> do we know which (if any cores) will *not* be in Austin? 21:43:07 <mriedem> i know of at least 2 that one be 21:43:14 <mriedem> *won't be 21:43:26 <mriedem> tonyb: why? 21:43:48 <tonyb> mriedem: just thinking about in-person discussions 21:44:01 <mriedem> ok... 21:44:02 <tonyb> mriedem: theres stuff I'll do more prep work for if they're goign to happen 21:44:24 <mriedem> well danpb won't be there which is why i asked him to write up the os-vif status in the ML 21:44:39 <mriedem> tdurakov won't be there so i asked him to write up the status on the live migration testing 21:44:58 <mriedem> tonyb: anyway, if you have specific stuff, hit me up in -nova 21:45:09 <tonyb> okay thanks 21:45:17 <mriedem> alright, anything else? 21:45:24 <mriedem> no?! 21:45:24 <mriedem> ok 21:45:26 <mriedem> thanks 21:45:28 <mriedem> #endmeeting