15:02:15 <johnthetubaguy> #startmeeting XenAPI 15:02:16 <openstack> Meeting started Wed Oct 1 15:02:15 2014 UTC and is due to finish in 60 minutes. The chair is johnthetubaguy. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:02:17 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:02:19 <openstack> The meeting name has been set to 'xenapi' 15:02:56 <johnthetubaguy> hello hello 15:03:59 <johnthetubaguy> BobBall: matel: hows things> 15:04:22 <johnthetubaguy> #topic CI 15:04:36 <matel> hi, I'm on a meeting, mainly eavesdropping. 15:04:47 <johnthetubaguy> I see the CI has started voting again 15:04:51 <johnthetubaguy> hows things going? 15:04:58 <BobBall> soprry 15:04:59 <BobBall> sorry* 15:05:04 <BobBall> Got pulled away for an urgent meeting 15:05:08 <BobBall> I am back now 15:05:12 <BobBall> https://review.openstack.org/#/c/125346/ 15:05:17 <BobBall> Looking good! \o/ 15:05:32 <BobBall> matel fixed the CI 15:05:40 <johnthetubaguy> sweet 15:06:05 <BobBall> upgrading to Trusty needed a few tricky things to happen - in part because we were running on branches of a few things and haven't kept up with infra's progress 15:06:11 <BobBall> but also in part due to Trusty bugs 15:06:28 <BobBall> The key thing is it's working again 15:06:30 <johnthetubaguy> thats the latest LTS right? 15:06:33 <BobBall> Yes 15:06:35 <matel> Yep, it was quite tricky to find out that the kernel needs to be updated. 15:06:48 <BobBall> Hopefulyl we'll be able to stick on Trusty for at least 3 weeks before another upgrade is needed... 15:06:52 <johnthetubaguy> matel: ouch, good working fixing it 15:07:24 <johnthetubaguy> BobBall: yeah, joy, do you not to an apt-get upgrade thing when preping the nova-compute 15:07:57 <BobBall> yes 15:08:02 <BobBall> but then that didn't reboot the host 15:08:06 <BobBall> so the new kernel wasn't being used... 15:08:11 <BobBall> and you needed a dist-upgrade anyway 15:08:21 <BobBall> which didn't work first run, and you needed to re-run it 15:08:31 <johnthetubaguy> oops 15:08:33 <johnthetubaguy> gotcha 15:08:36 <BobBall> but all of that had to happen before uploading the image (because of the reboot issue) so we had to do it in the XVA 15:08:51 <johnthetubaguy> nasty 15:08:55 <BobBall> but the xva was built with debootstrap which has no way (that we can find) to use trusty-updates and can only install the base system 15:09:10 <johnthetubaguy> yuck 15:09:14 <BobBall> Which we then have to boot + dist-upgrade twice etc etc 15:09:17 <johnthetubaguy> anyways, good work 15:09:32 <johnthetubaguy> hows everything else doing 15:09:41 <johnthetubaguy> I guess we are now on more recent branches and stuff 15:09:44 <johnthetubaguy> whats next? 15:09:53 <BobBall> Good. I think. No new bugs raised so that's good 15:10:16 <BobBall> next step for the CI is to look @ zuul, jjb and other upstream repos so we can just do what infra are doing more 15:10:38 <BobBall> also look at uploading the XenServer image to RAX so we don't have to build it online 15:10:50 <BobBall> but it's not clear if we'll be able to do that... We'll see 15:11:38 <johnthetubaguy> let us know how that goes 15:11:43 <johnthetubaguy> a single VHD should do 15:12:22 <BobBall> Indeed 15:12:27 <BobBall> It's a nice theory 15:12:29 <BobBall> if it works 15:12:33 <BobBall> but it's currently untested 15:13:10 <johnthetubaguy> well, customers have used that feature, but yeah 15:13:17 <johnthetubaguy> anyways, sounds good 15:13:24 <BobBall> I didn't mean the RAX feature 15:13:32 <BobBall> but the XenServer / nodepool integration points 15:14:21 <johnthetubaguy> heh OK 15:14:44 <johnthetubaguy> #topic Open Discussion 15:14:54 <johnthetubaguy> anyone got anything else to raise? 15:14:58 <BobBall> What's the status of the vhd issue for Juno? 15:15:06 <johnthetubaguy> we don't have a summit session in paris, depends what people want 15:15:10 <BobBall> You had two fixes to try and get in 15:15:12 <johnthetubaguy> good question 15:15:16 <BobBall> are they both in? 15:15:34 <johnthetubaguy> yes, I think they are 15:15:47 <johnthetubaguy> all three 15:15:53 <johnthetubaguy> we clean up the VDI chain 15:16:03 <johnthetubaguy> cookie issues are auto-repaired where possible 15:16:06 <BobBall> wonderful 15:16:19 <johnthetubaguy> and the fix for the collase logic itsself 15:16:21 <BobBall> Have you tested these changes in your env? Where you saw the cookie issues? 15:16:29 <johnthetubaguy> we tested it in a test env, and it seems to be working nicely 15:16:37 <johnthetubaguy> yet to be rolled out 15:16:51 <johnthetubaguy> got delayed by doing lots of reboots... 15:16:54 <BobBall> Perfect 15:17:01 <BobBall> Reboots? can't imagine what they were for... 15:17:06 <johnthetubaguy> quite 15:17:27 <johnthetubaguy> #link https://review.openstack.org/#/c/121177/ 15:17:30 <johnthetubaguy> thats the big one 15:17:41 <BobBall> great - glad it's in 15:17:45 <johnthetubaguy> me too 15:17:53 <BobBall> Should improve the reliability of the CI too 15:18:01 <johnthetubaguy> yeah 15:18:12 <BobBall> (just checking - we haven't cut Juno yet, have we?) 15:18:27 <johnthetubaguy> its been cut 15:18:38 <BobBall> oh... Was the cut after the fixes or before? 15:18:43 <johnthetubaguy> well after 15:18:50 <johnthetubaguy> was cut last night / this morning 15:18:53 <BobBall> oh good. that's all I care about :D 15:19:04 <johnthetubaguy> cool, any more for any more? 15:19:12 <BobBall> Nah 15:19:45 <johnthetubaguy> cool, thanks for your good work fixing the CI :) 15:19:48 <johnthetubaguy> #endmeeting