14:01:31 <dprince> #startmeeting tripleo
14:01:32 <openstack> Meeting started Tue Jan 19 14:01:31 2016 UTC and is due to finish in 60 minutes.  The chair is dprince. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:01:33 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
14:01:33 <jdob> o/
14:01:36 <openstack> The meeting name has been set to 'tripleo'
14:01:36 <derekh> hi
14:02:24 <tzumainn> hiya
14:03:35 <dprince> #topic agenda
14:03:35 <dprince> * bugs
14:03:35 <dprince> * Projects releases or stable backports
14:03:35 <dprince> * CI
14:03:35 <dprince> * Specs
14:03:38 <dprince> * one off agenda items
14:03:39 <trown> o/
14:03:40 <dprince> * open discussion
14:03:53 <dprince> no on-off agenda items this week I think
14:04:29 <dprince> lets get started
14:04:35 <dprince> #topic bugs
14:06:02 <dprince> any bugs to highlight this week?
14:06:02 <trown> It looks like bnemec has solved the neutron-nova bug
14:06:22 * derekh is testing it at the moment
14:06:22 <trown> still waiting on https://review.openstack.org/#/c/269370/ to pass HA CI
14:06:27 * trown too
14:07:00 <trown> if that works it will make my proposal in the CI topic much less controversial
14:07:15 <dprince> trown: cool, thanks for the update
14:07:43 <trown> #link https://launchpad.net/bugs/1519525
14:07:44 <openstack> Launchpad bug 1519525 in tripleo "Cannot boot instance due to Neutron-Nova notifications" [Critical,In progress] - Assigned to Ben Nemec (bnemec)
14:07:56 <trown> for posterity sake
14:10:14 <dprince> okay, I saw a few IPv6 releated bugs coming in too. I will try and look over some of those reviews today
14:11:40 <dprince> #topic Projects releases or stable backports
14:13:08 <dprince> I have a note from Dmitry that we need to release DIB to fix an ironic bug
14:13:20 <trown> dprince: I think marios did that
14:13:33 <dprince> trown: ah, cool. :)
14:14:07 <dprince> okay, so in general where do we stand as far as stable backports go
14:14:34 <trown> I think we are in the same place for stable backports as we are with master
14:14:55 <trown> that being, that without overcloud validation our CI is essentially nonsense
14:14:58 <slagle> we had most everything backported, ergo ipv6 and ssl
14:15:11 <slagle> not sure why i said "had"
14:15:15 <slagle> we still do
14:15:40 <trown> slagle: maybe the couple of reverts where we 'had' them for a few weeks
14:15:42 <slagle> there have been some recent ipv6 and ssl fixes that need to get backported still
14:15:54 <dprince> okay, cool
14:15:56 <slagle> and maybe a handful of bugfixes
14:17:45 <d0ugal> o/
14:18:26 <dprince> #topic CI
14:18:34 <dprince> trown: you wanted to say something here?
14:19:16 <trown> dprince: yes, I would like to propose that we do not merge anything to either master or stable/liberty until we have the ping test merged (with the obvious exeption of things needed to make the ping test pass)
14:19:48 <dprince> trown: fine with me
14:19:50 <trown> without that, we will continue to merge broken things
14:19:54 <bnemec> +1
14:19:56 <slagle> +1
14:19:58 <dprince> trown: I would have agree to this long ago FWIW :)
14:20:15 <bnemec> Long ago we were still broken. ;-)
14:20:18 <dprince> trown: like before we switched to instack ;)
14:20:22 <bnemec> It's been almost two months.
14:20:36 <trown> dprince: well I would use my time machine for better purposes :)
14:20:48 <trown> but we can do something now
14:21:13 <trown> thanks to bnemec, it should not be too painful
14:21:15 * slagle pockets that quip for future reuse
14:21:49 <dprince> trown: is the etherpad updated? https://etherpad.openstack.org/p/delorean_master_current_issues
14:22:14 <trown> dprince: ya, and the sahara package one I do not think affects tripleoci
14:22:41 <trown> dprince: so I think we should just have the neutron-nova thing to get back on master/current
14:22:45 <derekh> +1, when we switched to instack I opened cards for a bunch of things that needed to be done including the pingtest, hardly anybody has picked un on any of the card since then,
14:23:38 <bnemec> We have the ping test proposed now.  It just hasn't been able to merge because we had no way of knowing if it worked.
14:23:57 <derekh> yup
14:24:21 <dprince> so if we land the neutron-nova fix and then the ping test passes we merge both we are good right?
14:24:23 <bnemec> It is also super important that we get our current-tripleo symlink updated.  I believe it's still pointing at a version of Ironic that is broken on CentOS 7 hosts.
14:24:30 <dprince> I mean until the next breakage...
14:24:40 <bnemec> dprince: Sounds like it.
14:24:48 <derekh> bnemec: you patch gets me further, I still can't ping my instances though still trying to find the porblem (it could be local)
14:25:22 <derekh> bnemec: yup, it has a inc command not found problem
14:25:46 <dprince> derekh: appoligies if my comment about "since instack" seemed overly harsh. I just meant to say the idea of a ping test before landing code is something we've done before...
14:25:51 <trown> derekh: bnemec, tempest smoke is passing with that patch in RDO, so I think we are good
14:26:08 <dprince> so agreeing not to land code if it breaks is I think reasonable again
14:27:11 <derekh> dprince: no prob, just wanted to make sure eveybody wasn't looking at me, I never wanted use to go 4 months without testing the overcloud, we just havn't done it
14:27:26 <derekh> trown: ack, thanks
14:27:37 <dprince> derekh: wasn't even looking in your direction :)
14:27:50 <bnemec> derekh: I blame downstream, not you. :-)
14:28:15 * derekh goes back to his happy place
14:28:28 <derekh> btw, I rebuilt the TE hosts for CI on sunday
14:28:43 <derekh> our underclouds now have 2xvCPU and 5G RAM
14:28:48 <bnemec> \o/ memory
14:28:52 <derekh> in ci
14:29:07 <dprince> are we noticing that this helped the HA jobs pass more often?
14:29:41 <bnemec> Looks like it's passing about 50%, which is way better than before.
14:30:02 <bnemec> I've seen a lot of db issues in the bogus failures lately, so I think we may have an actual bug there.
14:30:14 <derekh> dprince: things appear to be running faster at least when CI isn't under load, but were now more overcommited so not sure what will happen when CI is busy
14:30:40 <dprince> derekh: I guess we cross our fingers and see
14:30:54 <derekh> as for failure rate, Its probably a little to early to tell but it looks better
14:30:57 <derekh> dprince: yup
14:33:05 <dprince> okay, lets move on
14:33:12 <dprince> #topic Specs
14:35:10 <dprince> Oh, before I forget. I think we had 100% agreement on the ping test passing. So probably worth sending this out on the list too right?
14:35:17 <dprince> just so everyone knows
14:35:28 <dprince> not to merge code unless it passes
14:35:36 <trown> dprince: +1
14:35:52 <slagle> yes, i'd say no concensus needed on the list
14:36:02 <dprince> slagle: just a notification
14:36:02 <slagle> just send it out as "the way"
14:36:23 <dprince> okay I can do that
14:36:28 <dprince> back to specs
14:36:32 <dprince> anything new this week?
14:40:18 <dprince> okay, so no specs updates for this week I think
14:40:24 <dprince> #topic open discussion
14:45:18 <trown> hmm... maybe we can just close early
14:45:21 * bnemec has nothing and has another meeting starting right now
14:45:24 <dprince> sounds like everyone is either tired or busy
14:45:53 <bnemec> Last week was kind of brutal for a lot of us.
14:45:54 <dprince> okay, we can end early then
14:46:00 <dprince> thanks everyone
14:46:01 <trown> thanks for chairing dprince
14:46:12 <dprince> #endmeeting