21:01:15 <johnthetubaguy> #startmeeting nova 21:01:16 <openstack> Meeting started Thu Jul 16 21:01:15 2015 UTC and is due to finish in 60 minutes. The chair is johnthetubaguy. Information about MeetBot at http://wiki.debian.org/MeetBot. 21:01:17 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 21:01:19 <openstack> The meeting name has been set to 'nova' 21:01:31 <jroll> ohai 21:01:41 <bauzas> o............./ 21:01:43 <edleafe> o/ 21:01:45 <gabriel-bezerra> Good evening! 21:01:47 <xyang2> hi 21:01:47 <dansmith> -o- 21:01:47 <earlephilhower> Howdy 21:01:48 <n0ano> o/ 21:01:48 <tjones1> hi 21:01:58 <mriedem> hi 21:01:58 <johnthetubaguy> #topic Release status 21:02:02 <ctrath> \o 21:02:02 <tpatil_> Hi 21:02:04 <melwitt> o/ 21:02:17 <johnthetubaguy> so I send an ML message, but today was non-priority feature proposal freeze 21:02:34 <johnthetubaguy> #info today was non-priority feature proposal freeze 21:02:54 <johnthetubaguy> #info liberty-2 is july 30th and non-priority feature freeze 21:03:03 <johnthetubaguy> #link https://etherpad.openstack.org/p/liberty-nova-non-priority-feature-proposal-freeze 21:03:13 <johnthetubaguy> thats got all the details, if people are interested in what happened 21:03:22 <johnthetubaguy> #link https://wiki.openstack.org/wiki/Nova/Liberty_Release_Schedule 21:03:40 <johnthetubaguy> ^ that describes why we are doing all this process, or it tries to 21:03:47 <johnthetubaguy> any questions on process stuff? 21:04:25 <johnthetubaguy> #action johnthetubaguy needs to go through nova-specs reviews and make sure people know about the move to the backlog 21:04:44 <johnthetubaguy> OK, either IRC has netsplit without telling me, or there are now questions, so moving on I guess... 21:04:49 <bauzas> :) 21:04:51 <johnthetubaguy> #topic Bugs 21:04:51 <mriedem> no questions 21:04:55 <mriedem> however 21:04:58 <mriedem> sec 21:04:59 <mriedem> on backlogs 21:05:03 <gabriel-bezerra> is that backlog in the links? 21:05:13 <mriedem> we keep saying to move things that aren't making it to the backlog, 21:05:17 <johnthetubaguy> ah, yes, backlogs, yes, lets do that 21:05:19 <johnthetubaguy> yeah 21:05:21 <bauzas> gabriel-bezerra: http://specs.openstack.org/openstack/nova-specs/specs/backlog/ 21:05:33 <mriedem> but if you read the docs on the backlog, it's about dumping ideas for people to go work on, not a place to defer things that didn't make n and need to go to n+1 21:05:50 <mriedem> we probably need to add something to the docs explaining that if backlog is going to become a deferral limbo place 21:05:58 <dansmith> yeah, 21:06:08 <dansmith> I was not in favor of making it a missed-deadline wasteland 21:06:12 <mriedem> +1 21:06:17 <johnthetubaguy> mriedem: yeah, so I think I want to discuss the plans at the midcycle 21:06:26 <mriedem> in the past people just move their stuff when they repropose 21:06:29 <gabriel-bezerra> thanks, bauzas 21:06:33 <mriedem> alright 21:06:47 <mriedem> so maybe we should hold off on telling people to start moving specs to backlog until after the meetup? 21:06:55 <johnthetubaguy> so I had some ideas, but they are a little crazy 21:06:57 <johnthetubaguy> #link https://wiki.openstack.org/wiki/Nova/Liberty_Release_Schedule#Stop_having_to_submit_a_spec_for_each_release 21:07:06 <mriedem> oh, runways 21:07:07 <mriedem> i see 21:07:12 <mriedem> RIP jogo 21:07:20 <johnthetubaguy> yeah, heading that way at least 21:07:34 <gabriel-bezerra> so the backlog is a directory in the specs repository? 21:07:43 <johnthetubaguy> gabriel-bezerra: yes 21:07:46 <bauzas> gabriel-bezerra: yep 21:07:47 <gabriel-bezerra> ok 21:07:52 <mriedem> gabriel-bezerra: yes, it's talked about in the nova-specs readme 21:08:03 <mriedem> we can move on, i'll add a thing to the etherpad for the meetup to talk about htis 21:08:07 <johnthetubaguy> gabriel-bezerra: basically merge in the backlog, rather than a specific release 21:08:24 <johnthetubaguy> mriedem: yeah, thats cool, lets bring this up at the end if that makes sense 21:08:32 <johnthetubaguy> so I should have used the undo, there, my bad 21:08:34 <johnthetubaguy> anyways, bugs 21:08:45 <gabriel-bezerra> got it 21:08:48 <johnthetubaguy> anything people want to talk about? 21:08:56 <bauzas> one criitcal 21:09:05 <bauzas> probably not critical 21:09:20 <bauzas> https://bugs.launchpad.net/nova/+bug/1475202 21:09:20 <openstack> Launchpad bug 1475202 in OpenStack Compute (nova) "Snapshot deleting of attached volume fails with remotefs volume drivers" [Critical,In progress] - Assigned to Dmitry Guryanov (dguryanov) 21:09:29 <johnthetubaguy> yes, feels like a high is the max there 21:09:48 <johnthetubaguy> bauzas: you OK to update the priority and add an educational comment? 21:09:49 <bauzas> changing the prio unless someone yells 21:09:55 <johnthetubaguy> cool, anything else 21:09:56 <bauzas> johnthetubaguy: sure thing 21:10:07 <bauzas> that's it for bugs - none for the client 21:10:23 <johnthetubaguy> mriedem: man of the gate, hows things looking right now, I saw lots of reverts being discussed that fixed a few things, I guess? 21:10:29 <mriedem> it's fine 21:10:31 <mriedem> backed up 21:10:32 <mriedem> but fine 21:10:41 <johnthetubaguy> cools 21:10:47 <dansmith> "man of the gate" ... I'll have to remember that 21:11:11 <johnthetubaguy> heh 21:11:18 <johnthetubaguy> #topic regular reminders 21:11:49 <johnthetubaguy> so its probably a good time to review low priority specs, give people have been really quite good at getting code up to review 21:11:52 <bauzas> mriedem: 16 patches in my series, no Jenkins false positives, so yeah it's cool :p 21:11:53 <johnthetubaguy> but ... 21:11:55 <johnthetubaguy> #link https://etherpad.openstack.org/p/liberty-nova-priorities-tracking 21:12:19 <johnthetubaguy> please to try and review these things too, ideally ^ 21:12:20 <johnthetubaguy> anyways 21:12:34 <johnthetubaguy> I commented on unshelf as added in the agenda 21:12:45 <johnthetubaguy> #link https://etherpad.openstack.org/p/YVR-nova-liberty-summit-action-items 21:12:55 <johnthetubaguy> please do work on post summit actions^ 21:13:02 <tpatil_> johnthetubaguy: Yes, we will move the specs to backlog 21:13:06 <johnthetubaguy> #link https://etherpad.openstack.org/p/liberty-nova-midcycle 21:13:17 <tpatil_> We have simplified specs by removing complex deployments strategies (host aggregate groups) and also addressed all comments received from Alaski and Matt in PS 15 21:13:21 <tpatil_> After moving specs to the backlog, request everyone to please review the specs. 21:13:23 <johnthetubaguy> tpatil_: thanks, hopefully we can get a way forward agreed really soon 21:13:36 <johnthetubaguy> tpatil_: yeah, it feels like we are getting closer now 21:13:42 <johnthetubaguy> #link https://etherpad.openstack.org/p/liberty-nova-midcycle 21:13:53 <tpatil_> johnthetubaguy: Thanks, looking forward to it. 21:13:53 <johnthetubaguy> so ideas for next week, feel free to add them ^ 21:14:06 <johnthetubaguy> #topic Open Discussion 21:14:11 <johnthetubaguy> I don't see any stuck reviews 21:14:14 <bauzas> saw a google hangout link, thanks ! 21:14:15 <johnthetubaguy> so straight to open 21:14:40 <johnthetubaguy> bauzas: yeah, it might not work, but we can try a few things 21:14:44 <johnthetubaguy> so we have agenda stuff... 21:14:47 <gabriel-bezerra> I have a stuck review, but mriedem advised me to put it in Open Discussion 21:14:57 <bauzas> johnthetubaguy: sure thing 21:14:57 <gabriel-bezerra> it's the first topic there 21:15:09 <johnthetubaguy> so yeah, we have: https://blueprints.launchpad.net/nova/+spec/pass-down-instance-name-to-ironic-driver 21:15:17 <gabriel-bezerra> That one 21:15:25 <gabriel-bezerra> the code is ready, just 7 lines 21:15:34 <johnthetubaguy> now, the problem here, is the blueprint freeze was at the end of June 21:15:48 <johnthetubaguy> and the exception process ended last week 21:15:58 <mriedem> mikal said in the spec that it probably doesn't need a spec https://review.openstack.org/#/c/189850/ 21:16:03 <mriedem> so i think that's why it's 'stuck' 21:16:07 <gabriel-bezerra> it just adds the instance/display_name to the message nova sends to ironic 21:16:07 <mriedem> got caught in process limbo 21:16:15 <johnthetubaguy> oh dear, I see 21:16:18 <mriedem> seems trivial enough to not require a spec 21:16:21 <johnthetubaguy> yeah 21:16:25 <mriedem> we have a couple of these 21:16:31 <johnthetubaguy> gabriel-bezerra: do you have you code up for review now? 21:16:39 <gabriel-bezerra> yes, it is in the second link 21:16:40 <mriedem> https://review.openstack.org/#/c/190675/ 21:16:43 <mriedem> ^ is the code 21:16:46 <jroll> gabriel-bezerra: so the idea is to pass instance display_name to ironic node.instance_info? 21:16:48 <jroll> aha 21:17:01 <gabriel-bezerra> yes, jroll 21:17:06 <dansmith> I violently disagree with this code 21:17:14 <gabriel-bezerra> so we can improve the UX of the driver 21:17:18 <jroll> ..... that code needs a spec? 21:17:28 <johnthetubaguy> dansmith: in a block the blueprint kind of way? 21:17:35 <dansmith> johnthetubaguy: no in a completely sarcastic way 21:17:45 <johnthetubaguy> dansmith: :) 21:17:52 <mriedem> so i move to trivially approve the bp 21:17:54 <mriedem> drop the spec 21:17:56 <mriedem> merge the code 21:17:59 <johnthetubaguy> yep 21:18:02 <gabriel-bezerra> :) 21:18:04 <jroll> mriedem: +1 21:18:35 <johnthetubaguy> gabriel-bezerra: so to make this happen next time, basically just add the blueprint to the nova-meeting, attempting to document and streamline all this, we had a few fall between the gaps 21:18:54 <bauzas> oh, sooooo many co-authors :) 21:19:03 <johnthetubaguy> ah, so we have a few specs that are bug fixes 21:19:05 <dansmith> yeah, how is that possible? 21:19:07 <johnthetubaguy> they need a reveiw 21:19:11 <dansmith> that's more than one author per line I think 21:19:13 <edleafe> bauzas: they each wrote one character 21:19:13 <mriedem> must be an edu thing 21:19:17 <tonyb> bauzas: Yeah one line each? 21:19:19 <jroll> johnthetubaguy: I'm shocked this even requires a blueprint 21:19:29 <johnthetubaguy> jroll: its only for tracking 21:19:34 <gabriel-bezerra> johnthetubaguy: sure. thanks. 21:19:37 <jroll> ok 21:19:38 <mriedem> moving on? 21:19:38 <johnthetubaguy> jroll: so we can tell users we added a feature 21:19:41 <bauzas> tokyo is close, I guess 21:19:44 <gabriel-bezerra> yes, we are a university 21:19:47 <jroll> johnthetubaguy: heard 21:19:49 <johnthetubaguy> #link https://review.openstack.org/#/c/198622 21:19:54 <gabriel-bezerra> ufcg.edu.br 21:19:56 <johnthetubaguy> #link https://review.openstack.org/#/c/198184/ 21:20:03 <johnthetubaguy> I think those ones are API changes 21:20:21 <johnthetubaguy> but are bug fixes that spiralled, so some eyes on those would be good, I should take a look too 21:20:33 <johnthetubaguy> there is a review request on the agenda too 21:20:37 <johnthetubaguy> look there if you fancy 21:20:55 <johnthetubaguy> ah, so the next one is some brick stuff 21:21:13 <mriedem> yeah, so this is another case of process limbo, 21:21:26 <mriedem> when those were proposed we weren't sure if they would be os-brick or old style volume drivers, 21:21:33 <mriedem> so that held them up even though the code was up, 21:21:43 <mriedem> now hemna's os-brick change has a +2 and should get in, 21:21:47 <johnthetubaguy> yeah, they feel like tiny things we can just document the blueprints for 21:21:51 <mriedem> and the hgst and scaelio changes are on top of htat 21:21:54 <mriedem> right 21:21:57 <johnthetubaguy> yeah, I hoped to do the brick review today, but it didn't happen 21:22:09 <hemna> mriedem, thanks for all the help on that 21:22:12 <mriedem> so, another case where i don't think we should block those 21:22:22 <mriedem> the nova code is just a shim layer that talks to brick 21:22:40 <johnthetubaguy> yeah, seems reasonable 21:22:44 <mriedem> we should actually probably add these kinds of things to the spec template or readme about what doesn't need a spec 21:22:51 <mriedem> which i could do 21:23:14 <johnthetubaguy> we can do those kinds of blueprints on IRC I think, rather than wait for the meeting, if that works better 21:23:22 <mriedem> sure 21:23:32 <johnthetubaguy> mriedem: yeah, we lots some of the good examples in the doc clean up, these would be good adds 21:23:33 <mriedem> just airing it out here 21:23:45 <johnthetubaguy> mriedem: yeah, good idea 21:23:58 <mriedem> #action mriedem to add volume driver examples about what doesn't need a spec 21:24:11 <johnthetubaguy> the spec vs blueprint guidelines are very poor right, now, I think we are better practiced now, so we can document that 21:24:17 <johnthetubaguy> mriedem: awesome, thank you 21:24:26 <johnthetubaguy> so.. now its open open discussion 21:24:32 <johnthetubaguy> any more for any more? 21:24:39 <earlephilhower> Just checking, for those 2 BPs 21:24:48 <earlephilhower> do the authors need to do anything (Xyang and I)? 21:24:51 <mriedem> https://blueprints.launchpad.net/nova/+spec/emc-sdc-libvirt-volume-driver 21:24:54 <mriedem> https://blueprints.launchpad.net/nova/+spec/emc-sdc-libvirt-volume-driver 21:25:05 <mriedem> https://blueprints.launchpad.net/nova/+spec/add-os-brick-volume-driver-hgst-solutions 21:25:17 <johnthetubaguy> right, I think we can just approve those, as trivial things 21:25:24 <mriedem> sounds like abandon the specs and approve the bps 21:25:26 <johnthetubaguy> mostly because they are already up for review, and agreed 21:25:32 <johnthetubaguy> yeah 21:25:35 <earlephilhower> Cool, thanks. 21:25:37 <xyang2> thanks a lot! 21:25:51 <johnthetubaguy> I will do the approve and abandon in the morning 21:25:57 <johnthetubaguy> unless someone beats me to it 21:26:03 <mriedem> i can do it 21:26:42 <johnthetubaguy> mriedem: cool 21:26:48 <johnthetubaguy> anyways, any more for any more? 21:27:21 <melwitt> https://blueprints.launchpad.net/nova/+spec/service-version-number spec merged and can be approved I noticed 21:27:52 <johnthetubaguy> melwitt: ah, I missed that, oops, yeah 21:28:12 <melwitt> it wasn't there before this afternoon, I noticed it when I clicked the link in a patch about it 21:28:14 * johnthetubaguy should really automate that... 21:28:16 <dansmith> melwitt: the check is in the mail, thanks 21:28:25 <melwitt> :) 21:29:15 <johnthetubaguy> any more for any more? 21:29:25 <mriedem> nope 21:29:27 <mriedem> end it 21:29:30 <dansmith> +1 21:29:46 <johnthetubaguy> cool, so I am busy approving blueprints 21:29:53 <johnthetubaguy> see some of you on tuesday 21:30:01 <mriedem> o/ 21:30:02 <johnthetubaguy> see some of you on IRC tomorrow 21:30:04 <johnthetubaguy> thanks all 21:30:07 <johnthetubaguy> #endmeeting