15:06:06 <johnthetubaguy> #startmeeting XenAPI 15:06:07 <openstack> Meeting started Wed Jul 2 15:06:06 2014 UTC and is due to finish in 60 minutes. The chair is johnthetubaguy. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:06:08 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:06:10 <openstack> The meeting name has been set to 'xenapi' 15:06:40 <BobBall> I'm about 2 seconds ahead of the reminder even though I'm 5 minutes late for the meeting johnthetubaguy :) 15:06:50 <johnthetubaguy> lol 15:06:54 <johnthetubaguy> hows things? 15:06:59 <BobBall> good good 15:07:05 <BobBall> and you? 15:07:10 <johnthetubaguy> not too bad thanks 15:07:15 <johnthetubaguy> #topic CI 15:07:22 <johnthetubaguy> hows the CI world? 15:07:42 <BobBall> stopped listening to gerrit events over the w/e 15:07:44 <johnthetubaguy> (I actually have some updates this week...) 15:07:45 <BobBall> not sure why 15:07:56 <BobBall> (it just couldn't connect to gerrit) 15:08:04 <BobBall> so it might have been a temporary network thing 15:08:15 <johnthetubaguy> hmm, you have hardcoded the IP right, I tried using the hostname 15:08:21 <BobBall> but apart from that we're all good 15:08:31 <BobBall> interesting threads on the ml though 15:08:50 <BobBall> about whether we can combine results in gerrit or mandate the use of zuul+jenkins 15:09:09 <johnthetubaguy> oh, I probably missed those, on the main dev rather than tagged nova I guess? 15:09:18 <BobBall> yeah 15:09:50 <BobBall> there's lost of traffic on CIs atm 15:10:43 <BobBall> Topic was '3rd Party CI vs. Gerrit' 15:10:57 <johnthetubaguy> OK, cool 15:11:05 <BobBall> Have a butchers 15:11:20 <BobBall> but I suspect it'll mean that we need to move towards gerrit 15:11:24 <BobBall> -gerrit+zuul 15:11:28 <BobBall> which we're planning to do anyway 15:11:33 <BobBall> but it makes it more required 15:11:51 <johnthetubaguy> anyways, there is more traction on getting a new CI that tested the combination we used internally, so fingers crossed on that front 15:12:13 <johnthetubaguy> so, I would avoid deploying zuul yourself, we can probably get you into the turbohipster deploy 15:12:29 <BobBall> is the t-h deploy maintained by rackspace? 15:12:34 <johnthetubaguy> we have that team hopefully giving us a hand with this, although the timezone is not great 15:13:05 <johnthetubaguy> yeah, turbohipster is michael still's team, and we are trying to get some of there time to help us with XenServer testing, but not certain at this point 15:13:22 <BobBall> understood 15:13:33 <BobBall> although I wouldn't want to dilute the t-h tests as such 15:13:52 <BobBall> i.e. it'd be fine if we could comment as a seperate account + slower than the t-h comments 15:14:11 <BobBall> dunno if the standard stack lets you use multiple accounts / parallel tests rather than running them all in one job lot 15:14:32 <johnthetubaguy> right, its might be a parallel stack, but yeah, all to be worked out 15:15:51 <johnthetubaguy> OK… 15:16:07 <johnthetubaguy> #topic Open Discussion 15:16:13 <johnthetubaguy> any more for any more? 15:16:38 <BobBall> oh - was that your update? :P 15:16:46 <BobBall> I was hoping you'd played with your deployment with the fixed config 15:17:12 <johnthetubaguy> oh yeah 15:17:19 <johnthetubaguy> I did try that, but nodepool broke 15:17:29 <BobBall> in what way? 15:17:33 <johnthetubaguy> one image "succeeded" then but then all the builds failed 15:17:37 <johnthetubaguy> the other image builds got stuck 15:17:45 <johnthetubaguy> then yeah, nodepool just wouldn't start 15:18:04 <johnthetubaguy> so I kinda had to move on to other things, as I just threw another few hours down the drain on that damm thing :( 15:18:07 <BobBall> the notepool not starting issue is a bug that's been fixed upstream now 15:18:15 <BobBall> you need to rm /var/run/nodepool/nodepool.pid 15:18:24 <BobBall> because if you kill it the PID is stale 15:18:31 <BobBall> and nodepool didn't check the PID for staleness 15:18:35 <johnthetubaguy> oh, boy, that makes sense 15:18:47 <BobBall> I pushed a fix upstream but I don't think it's in our nodepool branch 15:19:01 <johnthetubaguy> I did run it manually, again, and that didn't seem to work quite right, then the ssh timedout, etc 15:19:11 <johnthetubaguy> I need to get back to that one 15:19:17 <BobBall> fair enough 15:19:20 <johnthetubaguy> it didn't quite "just work" is the summary 15:19:30 <johnthetubaguy> it kinda all fell apart in a heap again 15:19:40 <BobBall> heh :) 15:19:59 <johnthetubaguy> anyway, thanks for the pid thing, thats probably the last straw that made me stop messing with that 15:20:06 <johnthetubaguy> I should take another peek soon 15:20:09 <BobBall> it's a very frustating one 15:20:13 <johnthetubaguy> yeah 15:20:17 <BobBall> 1 sec 15:20:19 <BobBall> I'll get the fix 15:20:23 <johnthetubaguy> thats OK 15:20:33 <johnthetubaguy> any more for any more? 15:20:50 <BobBall> #link https://review.openstack.org/#/c/90052/ 15:20:57 <johnthetubaguy> ah, cool, thanks 15:21:16 <BobBall> you can cherry-pick it easily enough 15:21:23 <BobBall> no more from me 15:22:20 <johnthetubaguy> cool 15:22:25 <johnthetubaguy> thanks for the chat 15:22:29 <johnthetubaguy> catch you next week 15:22:33 <johnthetubaguy> #endmeeting