13:00:02 <esberglu> #startmeeting powervm_driver_meeting 13:00:03 <openstack> Meeting started Tue Oct 10 13:00:02 2017 UTC and is due to finish in 60 minutes. The chair is esberglu. Information about MeetBot at http://wiki.debian.org/MeetBot. 13:00:04 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 13:00:06 <openstack> The meeting name has been set to 'powervm_driver_meeting' 13:00:23 <esberglu> https://etherpad.openstack.org/p/powervm_driver_meeting_agenda 13:00:27 <efried> \o 13:00:45 <edmondsw_> o/ 13:00:58 <mdrabe> o/ 13:01:11 <esberglu> #topic In-Tree Driver 13:01:49 <esberglu> Waiting on reviews for the Config Drive patch 13:02:22 <esberglu> OVS is now passing the jenkins jobs, I'm planning on testing that live this week 13:02:34 <esberglu> I might repurpose one of the staging CI neo's for that 13:03:00 <esberglu> Nothing else IT from me 13:03:45 <edmondsw> we need to work on the request thorst brought up 13:03:55 <edmondsw> oh, nm, that's OOT 13:04:07 <edmondsw> will get to that in a minute 13:04:18 <edmondsw> I don't think we've got anything else going on IT at the moment 13:04:35 <esberglu> #topic Out-of-Tree Driver 13:04:38 <esberglu> edmondsw: Go ahead 13:04:55 <edmondsw> we've been asked to port host-level CPU util metrics down to pypowervm instead of nova-powervm 13:05:31 <edmondsw> we're probably going to be asked for an update on this in 2 weeks, so we should get going on that as soon as we can 13:05:54 <edmondsw> I added it to the TODO list a couple weeks ago, but haven't had a chance to dig into it 13:06:06 <edmondsw> anybody have cycles to go do that? 13:06:10 <edmondsw> maybe mdrabe? 13:06:36 <mdrabe> Not at the moment 13:06:49 <edmondsw> esberglu? 13:07:14 <esberglu> edmondsw: I could possibly do it. I agreed to help out with some storage UT work that has a big gap so I'm a little stretched out 13:07:31 <esberglu> edmondsw: This stuff here? 13:07:32 <esberglu> https://github.com/openstack/nova-powervm/blob/master/nova_powervm/virt/powervm/host.py 13:08:14 <edmondsw> yeah, I think that's it, about line 166 13:08:39 <esberglu> edmondsw: Sure I can work on that 13:08:51 <edmondsw> esberglu tx 13:09:26 <edmondsw> we'll need to get the changes into pypowervm first, then spin a release of that before updating nova-powervm to use it... same old drill 13:09:40 <esberglu> edmondsw: Yep 13:10:01 <edmondsw> we've got some other things we're doing that will also require a pypowervm release, so we can try to bundle those together 13:10:05 <edmondsw> e.g. iscsi 13:10:31 <edmondsw> tjakobs pinged me that he's got iSCSI tested now and ready for serious reviews 13:11:07 <edmondsw> I'll be looking at that, and dragging in efried and gfm to look as well 13:11:35 <efried> Good, I was just joking with my reviews up to this point. 13:11:50 <edmondsw> efried I didn't mean to imply you're not already looking 13:12:00 <edmondsw> tx for that 13:12:04 <efried> I'm just being hilariously funny, dude. No offense taken. 13:12:12 <edmondsw> har har :) 13:12:43 <edmondsw> tjakobs also has a couple reviews up for volume resize 13:13:12 <edmondsw> has a couple more comments to address there, but that should finish up soon 13:13:38 <edmondsw> efried we should be careful not to merge those before the pypowervm release that adds what they are dependent on... 13:13:49 <edmondsw> just realized we almost goofed on that yesterday 13:14:24 <edmondsw> I think that's all I had for OOT 13:14:39 <esberglu> #topic Device Passthrough 13:14:43 <esberglu> You're up efried 13:15:26 <efried> I've been spending most of my time in blueprint-land, moving Queens toward being something we could actually use 13:15:30 <efried> i.e. not have to wait for Rocky. 13:15:36 <efried> And I think it just might make it. 13:15:46 <efried> If all the blueprints actually get implemented. 13:16:34 <efried> No update on the Pike-based hackup - I'm waiting for a comeback from jwcroppe on the requirements before going further with that. 13:17:11 <efried> I guess you guys could review this blueprint... 13:17:21 <efried> https://review.openstack.org/510244 13:17:32 <edmondsw> efried ack 13:18:04 <efried> If you want to get a better sense of where we stand today, you could also review the ones listed in the Dependencies section. 13:19:01 <esberglu> efried: ack 13:19:10 <esberglu> #topic PowerVM CI 13:19:29 <esberglu> CI results are looking pretty good, mostly just the Map LU issue that nvcastet is looking into 13:19:43 <esberglu> However one of the SSP groups (the same one) is down again 13:19:54 <esberglu> 3 of the 4 systems I can't ssh into 13:20:10 <esberglu> Planning on debugging that today to get the system back up to full strength 13:20:44 <edmondsw> when do you write the book on "fun with SSPs!" ? 13:21:38 <esberglu> I'm working on ch. 3 right now :) 13:22:29 <esberglu> Other than that I was trying to get a pike undercloud working on the staging env, still just battling devstack there 13:22:53 <esberglu> The title of my second book will be "fun with devstack!" 13:23:01 <esberglu> That's all for CI 13:23:02 <edmondsw> sounds like a great read 13:23:24 <esberglu> #topic Open Discussion 13:23:32 <esberglu> Anything else today? 13:23:33 <edmondsw> the pike undercloud work is just an "as able" effort, right? 13:23:45 <edmondsw> anything really driving that? 13:23:45 <esberglu> edmondsw: Yep 13:23:50 <edmondsw> k 13:24:07 <edmondsw> good to stay current, of course 13:24:19 <esberglu> We've just been trying to lag 1 release behind master as able 13:24:24 <edmondsw> yep 13:24:26 <esberglu> Nothing driving 13:24:57 <edmondsw> there are these emails about debian packaging of pypowervm... anything to say there? 13:26:05 <esberglu> edmondsw: We determined that it is still being used? 13:26:22 <edmondsw> adreznec says he thinks so 13:26:23 <efried> Yeah, that's how they're packaging it up to put in our apt repos 13:26:42 <efried> it was a hack to begin with - minhn had to do some goofy stuff to get the dependencies working right. 13:26:52 <efried> And yeah, they bear little resemblance to what's in requirements at this point. 13:27:10 <efried> But the ask from the issue report was that we split it out into a separate packaging project 13:27:16 <efried> which seems to be how the openstack projects do stuff. 13:27:44 <edmondsw> yeah, that makes sense to me 13:28:01 <efried> I don't even know if minhn still works for us. 13:28:12 <efried> But yeah, he or jfoliva should do that split. 13:28:14 <edmondsw> oh really 13:28:32 <edmondsw> yep... I'm replying to the email agin 13:28:35 <efried> I mean on our team 13:28:50 <edmondsw> that's all I had 13:29:20 <esberglu> Alright thanks for joining have a good week all 13:29:31 <esberglu> #endmeeting