17:00:02 <jroll> #startmeeting ironic
17:00:03 <openstack> Meeting started Mon Apr 17 17:00:02 2017 UTC and is due to finish in 60 minutes.  The chair is jroll. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:00:04 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
17:00:06 <openstack> The meeting name has been set to 'ironic'
17:00:11 <NobodyCam> o/
17:00:11 <jlvillal> o/
17:00:12 <crushil> \o
17:00:12 <jroll> anyone around?
17:00:13 <mjturek> o/
17:00:16 <ricardoas> o/
17:00:18 <fellypefca> o/
17:00:22 <xavierr> o/
17:00:22 <rpioso> o/
17:00:23 <jason77> o/
17:00:39 <krtaylor> o/
17:00:46 <nicodemos> \o/
17:00:59 <jroll> ok, let's roll
17:01:03 <jroll> #topic announcements
17:01:10 <jroll> apparently mitaka EOL is this week
17:01:32 <rajinir> o/
17:01:34 <sauloaislan> o/
17:01:42 <rloo> o/
17:01:45 <jroll> looks like nothing to merge there, except a devstack patch
17:01:51 <jroll> which isn't necessary at EOL
17:01:53 <jroll> :D
17:01:57 <jroll> any other announcements?
17:02:09 <rloo> jroll: next week meetup
17:02:17 <jroll> ah yes!
17:02:26 <jroll> oh and I have an announcement I forgot
17:02:29 <jroll> but first, meetup link
17:03:01 <jroll> we're having a voip meetup april 25, details here:
17:03:04 <jroll> #link http://lists.openstack.org/pipermail/openstack-dev/2017-April/115488.html
17:03:20 <jroll> and then my thing
17:03:30 <jroll> so uh, my employment situation is changing
17:03:39 <jroll> I haven't decided on a new thing yet
17:03:51 <jroll> may or may not end up working on ironic or openstack in general
17:04:05 <jroll> but I'm job hunting for the next few weeks, so don't expect much from me here :(
17:04:23 <jroll> I will be around IRC, though, happy to help out if anyone needs anything
17:04:26 <jlvillal> Best of luck jroll. That sucks :(
17:04:44 <jroll> life moves on, I'll be okay :)
17:04:59 <jroll> anything else?
17:05:21 <NobodyCam> Best wishes jroll
17:05:30 <jroll> <3
17:05:33 <rloo> jroll: good luck, i have no doubt that you'll do something great. /me hopes you stay in openstack though, for selfish reasons :)
17:05:51 <jroll> heh, thanks
17:05:54 <jroll> #topic subteam status reports
17:06:07 * jroll moves on, condolences can be elsewhere :P
17:06:13 <jroll> so these are here:
17:06:17 <jroll> #link https://etherpad.openstack.org/p/IronicWhiteBoard
17:06:21 <joanna> jroll: good luck!
17:06:25 <jroll> starting at line 79
17:06:46 * jroll hasn't been working upstream for about 2 weeks so has no idea what is updated and what is not
17:06:52 <rloo> oh, give me 5 minutes and i'll update the bugs stats...
17:07:06 * JayF == jroll re: job situation
17:07:17 <jroll> rloo: wanna just update after the meeting before rama_y sends the report?
17:07:33 <rloo> jroll: ah, yeah, will do that instead.
17:07:51 <jroll> I see lots of april 10 updates
17:09:19 * jroll has no questions here, and isn't sure people are around to answer very many questions anyway
17:09:24 <rloo> vsaienk0: i saw a WIP patch for changing grenade to only upgrade conductor, how is that going?
17:12:08 <rloo> jroll: wrt xproject nova, we can delete that?
17:12:30 <rloo> jroll: nova feature freeze was last week right?
17:12:31 <jroll> rloo: done :P
17:12:36 <jroll> I even looked for it and didn't see it
17:12:42 <jroll> yeah, spec freeze, not feature
17:12:51 <rloo> jroll: thx. right, sorry for starting a rumour :)
17:13:02 <jroll> heh
17:13:28 <jroll> anything else?
17:13:38 <rloo> wrt SNMP, is someone looking into it?
17:13:43 <rloo> gate broken
17:14:21 <rloo> hmm, the fix has merged. is it still broken? https://review.openstack.org/#/c/456476/
17:14:40 <rloo> L301 for those that are following
17:15:09 <jroll> I asked in channel a few minutes ago
17:15:35 <jroll> http://status.openstack.org/openstack-health/#/?searchProject=ironic
17:15:41 <jroll> looks like not everything failing right now
17:17:08 <jroll> shall we move on?
17:17:21 <rloo> ++
17:17:35 <jroll> #topic Deciding on priorities for the coming week
17:17:47 <jroll> like I said, haven't been around much for 2 weeks, no clue what's going on
17:17:56 <jroll> any problem with leaving these as-is?
17:18:06 <rloo> it doesn't seem to matter, who is around to do any work? (ok, i'll be more positive tomorrow)
17:18:14 <jroll> >.>
17:18:49 <jroll> ok, let's leave it
17:18:53 <jroll> #topic open discussion
17:18:56 <jroll> anything else?
17:18:59 <rloo> jroll: well, just need to update it a bit but i can do that.
17:19:07 <jroll> rloo: okay cool, thanks
17:19:36 <wanyen> jroll, I have questions regarding resource provider/class & ironic
17:19:38 <rloo> jroll: ok, updated to reflect this week's stuff
17:19:58 <jroll> wanyen: sure, go ahead and ask
17:20:03 <wanyen> will ironic uses standard resource classes
17:20:30 <jroll> not this cycle, no
17:20:39 <jroll> that's in the specs that have been written about this
17:20:46 <wanyen> yes.  what about queens?
17:20:54 <jroll> I'm not sure, we haven't discussed it
17:21:04 <jroll> I think long term, we would like to be able to use them
17:21:26 <wanyen> ok.  My question is how inspection can populate teh discoveredinfo into resource classes?
17:21:53 <jroll> we haven't discussed that, either. I think it will need to be a deployer-specific inspection rule
17:22:54 <wanyen> I am not very comforatble about Ironic ust use one single custom rsc class.  I am not sure that's teh plan for queens.
17:23:10 <JayF> it's per-node, not "a single custom rsc class"
17:23:18 <JayF> you could set all your nodes to a separate resource class if you wanted.
17:23:38 <wanyen> so each node has one single custom rsc class?
17:23:44 <jroll> wanyen: I'm not sure what you mean about the plan for queens - we don't have a specific plan for queens
17:23:46 <jroll> yes...
17:23:52 <jroll> have you read the specs about this?
17:24:29 <wanyen> yes, I have read a few specs about rsc class and make cutom rsc class for scheduling, etc
17:25:14 <jroll> okay, those mention the resource class is per-node
17:25:26 <wanyen> I am hopng in Queens , ironic can use standard rsc class and inspection can populate the discovered info into standard rsc classes.
17:25:54 <jroll> I hope so too - when those discussions arise, please contribute to them. you might even start this discussion :)
17:26:32 <wanyen> Yes. I would like to participate these discussion.
17:26:54 <jroll> great!
17:27:00 <jroll> anything else for open discussion?
17:27:19 <wanyen> It seems to me that we can do something in get_discovery().
17:28:09 <jroll> maybe, I don't know inspector enough to say how we would do it
17:28:11 <wanyen> get_discovery can return the correct standard rsc classes inventory to nova
17:28:47 <jroll> please do send an email to the mailing list if you want to discuss the long term plans. nova team will need to discuss as well.
17:29:40 <wanyen> jroll, when will ironic discuss these topics: rsc provider, rsc classes, inspection,e tc?
17:30:07 <jroll> wanyen: I can't predict the future. I don't know. probably when someone sends an email to the mailing list about it.
17:30:23 <jroll> right now we've been focusing on pike
17:30:44 <wanyen> I meant get_inventory not get_discovery in the previous msg
17:30:47 <jroll> queens planning will probably start shortly before the PTG
17:31:57 <wanyen> jroll, ok.   The reason taht I am looking into these is because it has impact on oob inspection spec.
17:32:36 <jroll> wanyen: sure
17:32:47 <jroll> I left a comment on there with what I think the best way forward is
17:32:52 <jroll> "The fastest way to get my +2 on this spec would be to just drop the non-boolean capabilities from this spec, though I've left comments on a few about moving them or making them boolean. :)"
17:33:22 <jroll> I think we should have a larger discussion about the non-boolean capabilities and how they fit into the bigger picture
17:33:28 <wanyen> jroll, Nisha and myself would like to add more things to discover for inspections.  Some of them are quantative proerties.
17:33:35 <jroll> but I haven't had time to start that conversation
17:34:07 <jroll> if you want to start that conversation, it needs to start on the mailing list
17:34:21 <wanyen> jroll, ok.
17:35:15 <jroll> thanks
17:35:17 <jroll> anything else?
17:36:02 <wanyen> jroll, the fundamental concern that I have is that I am not sure how inspection can populate custom rsc classes but I think it's easier to populate the standard rsc classes.
17:36:22 <jroll> wanyen: sure. that isn't a concern I can address in this meeting.
17:36:41 <jroll> wanyen: first we need to agree that ironic can or should use standard resource classes
17:36:47 <rloo> i think we/you should take this offline now
17:36:52 <jroll> yes, please
17:37:01 <wanyen> jroll, ok. so much for this discussion today. Thanks!
17:37:14 <jroll> you're welcome
17:37:23 <jroll> thanks for coming, y'all
17:37:25 <jroll> #endmeeting