21:00:43 <mikal> #startmeeting nova
21:00:43 <openstack> Meeting started Thu Jun  4 21:00:43 2015 UTC and is due to finish in 60 minutes.  The chair is mikal. Information about MeetBot at http://wiki.debian.org/MeetBot.
21:00:44 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
21:00:47 <openstack> The meeting name has been set to 'nova'
21:00:56 <mikal> Ok, so who is around for a nova meeting?
21:00:58 <dpaterson> 0/
21:00:58 <edleafe> o/
21:01:00 <mriedem> hi
21:01:00 <n0ano> o/
21:01:01 <bauzas> \o
21:01:02 <beagles> o/
21:01:02 <melwitt> o/
21:01:04 <jlvillal> o/
21:01:05 <neiljerram> o/
21:01:07 <tjones1> o/
21:01:08 <mikal> abhishekk mikal tjones cburgess jgrimm adrian_otto funzo mjturek jcookekhugen irina_pov krtaylor danpb alexpilotti flip214 jaypipes gilliard garyk edleafe dims moshele anteaya Nisha sileht claudiub lxsli neiljerram markus_z swamireddy alevine tonyb andreykurilin ndipanov sc68cal akuriata artom jlvillal: ping
21:01:16 <anteaya> o/
21:01:17 <dansmith> o/
21:01:18 <sc68cal> o/
21:01:25 <mikal> #topic Release Status
21:01:40 <alexpilotti> o/
21:01:49 <mikal> Ok, so John emailed out a set of Liberty deadlines and dates last night
21:01:54 <sdague> o/
21:01:57 <mikal> Some of those dates are proposed and open for discussion
21:01:59 <alaski> o/
21:02:03 <mikal> For example, there is a spec review day proposed
21:02:12 <mikal> #link http://lists.openstack.org/pipermail/openstack-dev/2015-June/065819.html
21:02:28 <mikal> So, I'd encourage people with thoughts on that to just reply to his email instead of discussing it here
21:02:53 <jaypipes> o/
21:02:56 <mikal> John also wanted to remind reviewers and sub-teams about the priority tracking etherpad
21:03:03 <mikal> #link https://etherpad.openstack.org/p/liberty-nova-priorities-tracking
21:03:23 <mikal> At the moment it has trivial bugs on it mostly, but we'd like to see priorities add things for review as they become ready
21:03:29 <mikal> i.e. what we did in kilo, but in a new etherpad
21:03:36 <dansmith> well,
21:03:42 <dansmith> some of the priority owners are already on top of that
21:03:48 <dansmith> like that upgrades guy
21:03:51 <bauzas> yeah johnthetubaguy recently added it
21:04:08 <bauzas> he moved the scheduler section on top
21:04:11 <mikal> Yeah, the ordering of the page is a bit odd to me too
21:04:17 <mikal> Like, having the trivial bugs in the middle
21:04:23 <mikal> But whatevers, at least we're tracking _something_
21:04:39 <bauzas> mikal: prios first, then trivial bugs, then subteams which are not priority
21:04:50 <mikal> Oh, I see
21:04:52 <sdague> dansmith: I can list my blog post for review :)
21:04:52 <dansmith> yeah
21:04:54 <mikal> That makes sense
21:05:18 <dansmith> sdague: well, if you have nothing else to show for work, I guess :)
21:05:38 <mikal> So I guess in all that, the most obvious thing to remind people of is that there is a spec approval deadline at liberty-1, which isn't that far away
21:05:46 <mikal> So reviewers and authors of specs need to be on the ball there
21:06:28 <sdague> dansmith: better watch out or I'll sign you up for writing a ton of english
21:06:33 <mikal> There is also a nova-specs review out to document the priorities, but it just matches the summit so no surprises there
21:06:42 <tonyb> mikal: Are we going to have an international specs review day again?
21:06:48 <dansmith> sdague: you *know* I will be glad to nit out all your typos, you just told me to hold off :)
21:07:06 <mikal> tonyb: so, what's happened in the past is we declare a date and the follow the sun with it
21:07:08 <bauzas> tonyb: June 12th is proposed
21:07:12 <sdague> edleafe already signed up, I just need to apply his edits
21:07:17 <mikal> tonyb: i.e. June 12th we'd all spend in our local timezone
21:07:18 <dansmith> sdague: I know that comment this morning was really prefixed with "dansmith: " :)
21:07:28 <tonyb> Ahh I missed that date
21:07:38 <mikal> tonyb: that's from John's email that happened over night
21:07:39 <tonyb> mikal, bauzas: Thanks
21:07:46 <mikal> tonyb: so, its understandable if you haven't seen it yet
21:07:51 <mikal> Ok, next...
21:08:03 * tonyb wonders how to gurn on IRC?
21:08:07 <mikal> John has also gone mad with power and is attempting to document all the action items from summit sessions
21:08:15 <mikal> #link https://etherpad.openstack.org/p/YVR-nova-liberty-summit-action-items
21:08:25 <mikal> It is a very very long list
21:08:30 <mikal> And I don't think he's done every session yet
21:08:49 <mikal> So, if you think you had an action item from the summit, or that you tricked dansmith into having an action item, it might be worth checking
21:08:52 <bauzas> mikal: he said tonight that he was done
21:09:03 <mikal> Oh really?
21:09:04 <mikal> Cool
21:09:21 <sdague> yeh, he did the big push today to get it pulled together
21:09:23 <mikal> So, the other thing is if you have 47 action items, now would be a good time to be honest about how much you can get done in a single cycle
21:09:35 <mriedem> whew, only 46
21:09:48 <mikal> This johnthetubaguy person has way too many things for example
21:09:56 <sdague> yeh, I think that's my only concern on the action plan. johnthetubaguy has way too many things on that list
21:10:10 <mikal> So, I think we could volunteer to rescue him if we wanted
21:10:16 <mikal> Unless watching him drown is funnier
21:10:28 <dansmith> I'm quite sure he'll delegate some of those things
21:11:06 <mikal> So that moves us on to non-spec blueprint discussion unless there's anything else to cover here first
21:11:14 <mikal> Anything on dates, deadlines, action items, etc etc?
21:11:35 <mikal> Ok, moving on
21:11:40 <mikal> #topic Non-spec blueprints
21:11:53 <mikal> First off, there is a request for the vmware console thing to get a trivial approval
21:12:00 <mikal> #link https://blueprints.launchpad.net/nova/+spec/vmware-console-log
21:12:48 <mikal> So...
21:12:50 <sdague> no API change right?
21:12:52 <mikal> Sound of crickets
21:12:57 <mikal> I don't think there's one
21:13:12 <mikal> Anyone from vmware around to talk to this?
21:13:15 <mriedem> the old change https://review.openstack.org/#/c/149203/
21:13:17 <tjones1> oops
21:13:18 <sdague> if so I say fine
21:13:20 <tjones1> yes i am here
21:13:24 <mriedem> it's pretty small
21:13:30 <mriedem> can be tested with tempest once it's in
21:13:40 <tjones1> rado is in europe so he is alseep
21:13:47 <mikal> It complicates vmware deploys, but in a way which doesn't worry anyone else
21:13:47 <mriedem> looks fine
21:13:50 <mikal> So I think I am ok with it
21:14:09 <mikal> Does anyone oppose a trivial approval?
21:14:18 <tonyb> mikal: seems okay to me.
21:14:21 <alaski> if it implements existing apis I'm good with it
21:14:41 <cfriesen> I just posted https://review.openstack.org/#/c/188575/ for review as requested by johnthetubaguy....that's dropping support for booting multiple instances in one REST API call.  It's mostly there as a place for having the discussion about whether or not we should do it at all.
21:14:41 <mriedem> i'd only ask if it requires a min version of vcenter
21:14:47 <mriedem> but that could be handled in the review
21:14:53 <mikal> Ok, approved
21:15:02 <mikal> cfriesen: that sounds like open discussion at the end please
21:15:06 <cfriesen> sure
21:15:09 <bauzas> cfriesen: open discussion
21:15:11 <bauzas> oops
21:15:15 <tjones1> thanks!!
21:15:26 <mikal> I hope I did the approval right, we shall see
21:15:35 <mikal> The other thing I wanted to discuss is config drive for PCS
21:15:45 <mikal> So I realized yesterday that they want to add a new format for config drives
21:15:48 <mikal> Namely ext4
21:15:59 <cfriesen> PCS?
21:16:08 <mikal> Which worries me because I worry that operators and users will then expect other hypervisors to support ext3 as a config drive format
21:16:14 <mikal> Parrallels Cloud Server
21:16:22 <mikal> A parallels container thingie
21:16:35 <mikal> ext4 would be hard for say hyper-v to support
21:16:42 <dansmith> mikal: is there a reason not to support vfat or iso9660?
21:16:44 <mriedem> you mean virtuozzo
21:16:45 <dansmith> because yeah
21:16:46 <mikal> The problem being that PCS only supports ext4 filesystems
21:16:48 <alexpilotti> yep, I’d -1 on ext4
21:16:57 <dansmith> yeah, me too
21:17:02 <alexpilotti> it’s not about Hyper-V, it’s about Windows images
21:17:19 <alexpilotti> running on whatever hypervisor
21:17:26 <mikal> Yeah, they'd point out PCS can't use windows images, but I think we need to maintain a coherant set of options for all drivers here
21:17:34 <sdague> alexpilotti: but those aren't going to run on PCS
21:17:34 <dansmith> +1
21:18:03 <sdague> I'm not sure I get why it's an issue, config drive generation is completely hidden inside of nova, and can be specific per hypervisor right?
21:18:04 <mikal> So, if people don't think I'm totally bonkers I might just start a mail thread about it
21:18:16 <mikal> sdague: config drive is consumed by the instance
21:18:24 <mikal> sdague: so if instances start assuming things about the filesystem
21:18:30 <mikal> sdague: then that leaks across to other drivers
21:18:54 <dansmith> yep, wouldn't want kvm to start exposing ext3 config drives
21:18:55 <mikal> sdague: i.e. say a tiny linux instance which doesn't want to include iso9660 or vfat support
21:18:56 <alexpilotti> I’m perfectly fine, just asking not to implement it where Windows images are supported
21:19:42 <mikal> So the filesystem format is currently an admin falg
21:19:53 <mikal> So I think the counter argument is that users are just used to getting whatever they're handed
21:19:57 <mikal> And probably don't think about it much
21:20:11 <sdague> yeh, I'd be really surprised if they care
21:20:31 <tonyb> How is this different to an ephemeral disk wheer we only support ext* and xfs?
21:20:40 <dansmith> tonyb: the guest owns the ephemeral
21:20:56 <mikal> Yeah, the config drive is populated by us and then handed over, its a data transmission format
21:21:15 <mikal> So, this now sounds like a mailing list thread to me
21:21:22 <tonyb> okay.
21:21:23 <mikal> Unless anyone hates that plan
21:21:34 <sdague> ML thread sounds right
21:21:34 <dansmith> I hate it, but that's fine
21:21:40 <mikal> #topic Stuck spec reviews
21:21:55 <mikal> So there is this placeholder heading where I think I am meant to ask about stuck spec reviews
21:22:21 <mikal> Where stuck probably means "we will never agree in the review"
21:22:25 <mikal> So, any of those?
21:22:38 <mikal> ...
21:22:48 <mriedem> no
21:22:52 <mriedem> we all agree all the time
21:22:55 <mikal> Yay!
21:22:59 * cfriesen snickers
21:23:02 <mikal> Moving on then before someone argues against that statement
21:23:11 <mikal> #topic Stable branch status
21:23:15 <mikal> The agenda says we're fine?
21:23:35 <mikal> Sounds like we are...
21:23:42 <mikal> #topic Gate status
21:23:52 <mikal> What is our current doom factor?
21:24:08 <mriedem> it's fine
21:24:13 <mikal> Yay!
21:24:24 <mikal> #topic Critical bugs
21:24:28 <bauzas> nope
21:24:30 <bauzas> zero
21:24:37 <mriedem> bugs are fine
21:24:38 <mikal> Even in novaclient?
21:24:49 <mriedem> we released novaclient 2.26.0 yesterday with great success
21:24:56 <mikal> Oh, look at that
21:24:58 <mikal> So that's nice
21:25:00 <mriedem> which deprecated the volume crap
21:25:08 <sdague> woot
21:25:09 <mikal> Fine then
21:25:14 <mikal> #topic Open Discussion
21:25:14 <bauzas> zero too
21:25:19 <mikal> cfriesen: you had a thing?
21:25:22 <mrda> Nothing critical in the Ironic driver, fwiw
21:25:51 <cfriesen> yeah, not critical, but there was a request for a spec to have a discussion around removing support for min_count and max_count from booting instances
21:25:59 <cfriesen> so I wrote something up
21:26:12 <mikal> cfriesen: cool, I will take a look later today
21:26:32 <mikal> In other news, there is some talk of the M mid-cycle being in Europe somewhere
21:26:45 <bauzas> yup
21:26:49 <mikal> I think John is looking for thoughts on if that's a terrible idea or a fantastic idea
21:27:05 <dansmith> I probably won't be there if so :(
21:27:05 <alexpilotti> mikal: +1
21:27:13 <sdague> fine by me
21:27:24 <neiljerram> Not sure how to comment on that.... but I'm in Europe, and it's quite nice sometimes.
21:27:31 <edleafe> +1 to the idea, but I probably couldn't make it
21:27:34 <mikal> So, I think people should tell John that
21:27:39 <mikal> Especially if they think its a problem
21:27:43 <alaski> seems like a good idea, but I can't say whether or not I could make it
21:28:10 <neiljerram> I guess it depends if Nova folk are very predominantly outside Europe.
21:28:31 <bauzas> it is also about costs
21:28:40 <mikal> I think that's true
21:28:45 <jlvillal> Low cost area is good.  Prague, Budapest, Bucharest :)
21:28:57 <mikal> And my impression is that Europe is generally more expensive for more people
21:29:00 <mriedem> i won't be there, but it's a decent idea
21:29:04 <bauzas> I can try to see for some room blocks if possible
21:29:05 <anteaya> budapest has good chocolatae
21:29:05 <alexpilotti> we can happily host it in Transilvania
21:29:24 <bauzas> (speaking of Grenoble, by the winter season y'know)
21:29:40 * bauzas bribes
21:29:43 <mikal> Anyways, something for people to ponder
21:29:54 <mikal> John is also surveying people about the time for this meeting
21:30:02 <mikal> He sent email, but the punch line is there is a doodle survey
21:30:09 <mikal> #link http://doodle.com/eyzvnawzv86ubtaw
21:30:25 <mikal> So if you care about this meeting it might be a good idea to do one of those
21:30:35 <mikal> Although obviously I am only saying this to the people who can amke this time
21:30:40 <mikal> Which is a bit self-reinforcing
21:30:52 <anteaya> mikal: the email went out too
21:31:02 <mikal> The final thing on the agenda is that John also wants bug tag owners to be more active
21:31:18 <mikal> Either by doing that thing, or by some new people volunteering if people need a break
21:31:28 <mikal> #link https://wiki.openstack.org/wiki/Nova/BugTriage
21:31:54 <mikal> So if you're looking to add more bugs to your day, that might be a thing to sign up for
21:32:04 <jlvillal> Question: As an Ironic liaison.  I am assuming that someone on Nova would tag a bug as Ironic related.  Not our job to look at every Nova bug to figure out if it is Ironic related.
21:32:09 <jlvillal> Is that correct?
21:32:18 <bauzas> jlvillal: yup
21:32:18 <mikal> jlvillal: correctish
21:32:19 <sdague> jlvillal: yes
21:32:28 <mikal> The idea is someone is reading incoming bugs and tagging them
21:32:30 <bauzas> jlvillal: provided someone does that :)
21:32:30 <sdague> please just further triage that tag
21:32:31 * jlvillal Whew!
21:32:34 <mikal> But there is nothing stopping you from also doing that thing
21:32:54 <sdague> https://bugs.launchpad.net/nova/+bugs?field.tag=ironic currently 29 bugs
21:32:58 * jlvillal thinks, "Only hours in the day..."
21:33:05 <bauzas> 38 new bugs need triage :)
21:33:05 <mikal> Yeah, so I'd start with those...
21:33:10 <mrda> sdague: jlvillal and I are working through that
21:33:19 <sdague> mrda: cool, awesome
21:33:26 <mikal> And that's all we have on the agenda
21:33:31 <mikal> Do people want an Early Mark?
21:33:39 <sdague> yes
21:33:40 <mikal> Or shall I waste 27 minutes rantinga bout how my feet are cold?
21:33:46 <mikal> Cause they are
21:33:48 <mikal> Even with socks
21:34:04 * jlvillal waits to hear more ;)
21:34:10 <mikal> Don't encourage me
21:34:11 <mrda> mikal: ugg boots are the answer :)
21:34:15 <mikal> Going...
21:34:18 <mikal> going...
21:34:29 <mikal> gone.
21:34:35 <neiljerram> Bye!
21:34:35 <mikal> #endmeeting