17:00:06 <NobodyCam> #startmeeting Ironic
17:00:07 <NobodyCam> #chair devananda
17:00:07 <NobodyCam> Welcome everyone to the Ironic meeting.
17:00:08 <openstack> Meeting started Mon Apr 20 17:00:06 2015 UTC and is due to finish in 60 minutes.  The chair is NobodyCam. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:00:09 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
17:00:12 <openstack> The meeting name has been set to 'ironic'
17:00:12 <jlvillal> o/
17:00:13 <openstack> Current chairs: NobodyCam devananda
17:00:20 <NobodyCam> Of course the agenda can be found at:
17:00:20 <NobodyCam> #link https://wiki.openstack.org/wiki/Meetings/Ironic#Agenda_for_next_meeting
17:00:24 <devananda> o/
17:00:32 <NobodyCam> #topic Greetings, roll-call and announcements
17:00:32 <NobodyCam> Roll-call: Who's here for the Ironic Meeting?
17:00:42 <NobodyCam> howdy all
17:00:42 <jlvillal> o/
17:00:42 <rloo> me
17:00:44 <pshige> o/
17:00:45 <vdrok> o/
17:00:56 <Nisha> o/
17:00:58 <Shrews> o/
17:01:05 <TheJulia> o/
17:01:19 <dtantsur> o/
17:01:25 <NobodyCam> great to see you all :)
17:01:30 <BadCub> hiya folks
17:01:49 <NobodyCam> looks like one one Announcement
17:02:02 <NobodyCam> Big Thank you to Lucas and Chris Dent for picking up WSME
17:02:08 <NobodyCam> as core members
17:02:20 <NobodyCam> for more info see the link on the agenda
17:02:44 <jroll> \o
17:02:51 * BadCub applauds
17:03:19 <NobodyCam> any other Announcements
17:03:23 <devananda> while its awesome that we can breathe a litlte life into WSME to fix bugs
17:03:42 <JayF> p/
17:03:55 <devananda> I'd like us to also begin thinking about moving off of WSME soon, as the project isn't maintained by anyone else, and I dont think there's benefit (long term) in us taking that on
17:04:05 <devananda> something to discuss at the summit, perhaps :)
17:04:07 <NobodyCam> devananda: I think I know where your going I also have a item under open discussion sectiom
17:04:17 <jroll> cdent being on it means ceilometer will probably continue to work on it no?
17:04:44 <devananda> just noting that here so that the announcement doesn't look like we're taking it on long-term
17:05:32 <NobodyCam> ahh
17:05:34 <NobodyCam> :)
17:06:07 <NobodyCam> ok then we can move on to :
17:06:16 <NobodyCam> #topic SubTeam: status report
17:06:44 <NobodyCam> white board seems like only few sections are updated
17:06:57 <NobodyCam> #link https://etherpad.openstack.org/p/IronicWhiteBoard
17:07:27 <jroll> I don't really expect a ton of updates this week
17:07:41 <rloo> no news is good news :)
17:07:45 <NobodyCam> :) ya .. :)
17:08:02 * devananda notes that stackforge doesn't have any info for driverlog + ironic + kilo -- http://stackalytics.com/report/driverlog?project_id=openstack%2Fironic
17:08:04 <pshige> :)
17:08:15 <NobodyCam> dtantsur: posted bug numbers https://etherpad.openstack.org/p/IronicWhiteBoard
17:08:24 <rloo> do we want to mention what is kilo-rc-potential material?
17:08:40 <jroll> rloo: +1
17:09:13 <devananda> rloo: yes, i was just looking at that.
17:09:14 <jroll> hm, I don't see kilo-rc-potential tag
17:09:23 <devananda> #link https://bugs.launchpad.net/ironic/+bugs?field.tag=kilo-rc-potential
17:09:28 <dtantsur> it's not official, so it's not in the list
17:09:41 <rloo> dtantsur: what's not official?
17:09:46 * devananda adds it to official bug list
17:09:47 <dtantsur> the tag
17:10:04 <dtantsur> rloo, launchpad has notion of "official" tags for a project
17:10:09 <rloo> dtantsur: oh. guess it'll be official very soon. thx ;)
17:10:21 <NobodyCam> so currentlly only one open bug
17:10:28 <jroll> well
17:10:41 <jroll> the "fix committed" bug still has the kilo backport patch open
17:10:55 <jroll> ttx has a -2 on it, I guess the RC window isn't open yet or something?
17:11:03 <NobodyCam> ahh
17:11:11 <jroll> https://review.openstack.org/#/c/174122/ fwiw
17:11:14 <rloo> so if someone thinks/wants something to go to kilo, they tag it, and we approve (or not) the patch, and then ttx has final approval?
17:11:15 <devananda> correct -- the rc2 window isn't open yet
17:11:24 <NobodyCam> #link https://review.openstack.org/#/c/174122
17:11:46 <devananda> process is outlined on the wiki page for stable branches, though it was just changed slightly last week (see emails from dhellmann and ttx)
17:12:10 <jroll> JoshNang: do we have a server patch for https://bugs.launchpad.net/ironic/+bug/1441445 ?
17:12:12 <openstack> Launchpad bug 1441445 in Ironic "As per the state machine, ironic doesn't provide an option to move from CLEANFAIL to CLEAN" [High,Triaged]
17:12:23 <jroll> or am I thinking of a rescue mode thing
17:12:42 <devananda> tldr; file bug, tag with kilo-rc-potential (or kilo-backport-potential after the release), propose fix to master. fix must be merged in master. cherry-pick -x the commit back to stable/kilo and propose that.
17:12:55 <JoshNang> jroll: that was rescue
17:12:58 <jroll> k
17:13:08 <rloo> the wiki page for stable branches: https://wiki.openstack.org/wiki/StableBranch
17:13:25 <devananda> then ttx and I decide if and when to merge it, based on release cycle, severity of bug, risk of the change, etc
17:13:33 <NobodyCam> # link https://wiki.openstack.org/wiki/StableBranch
17:13:48 <rloo> thx devananda
17:13:49 <devananda> note that we also have a stable branch for the client now, too
17:14:08 <rloo> devananda: oh yeah. the client. was that branch cut on a package release?
17:14:09 <devananda> iow, we need to follow the same process there, which it looks like dtantsur has done for the retry patch
17:14:13 <devananda> rloo: yes
17:14:49 <devananda> 0.5.x will be consisered stable/kilo series, and 0.6.x (not yet tagged) is what we'll develop on going forward
17:15:16 <dtantsur> why not starting giving client names like "Kilo" and versions like 2014.1? :)
17:15:27 <dtantsur> looks like they follow usual openstack process now
17:15:37 <devananda> well, not exactly
17:15:42 <devananda> we can still tag clients any time
17:16:18 <dtantsur> right
17:16:39 <devananda> there seem to be some more changes happening around this stable branches of libraries and clients thing
17:16:50 <devananda> like not having a proposed/kilo branch any more
17:17:18 <devananda> anyway, we've got 2 bugs right now tagged for backport potential
17:17:31 <devananda> any others that folks want to bring up?
17:17:33 <NobodyCam> devananda: is there info we can review on the changes forthcomming?
17:17:55 <devananda> NobodyCam: aside from reading the mailing list, not that I know of.
17:18:02 <NobodyCam> ack :)
17:18:19 <devananda> i'm still digesting it, which is partly why I haven't done an announce of how it affects us yet
17:18:37 <NobodyCam> oh we should prob change the topic
17:18:47 <NobodyCam> #topic Open Discussion
17:19:00 <NobodyCam> as that is the only place we have items on the agenda
17:19:25 <NobodyCam> devananda: :) ahh Thank you
17:20:14 <NobodyCam> rloo: your up 1st with utf-8 encoding
17:20:42 <rloo> so this bug/patch has been around for awhile. I'm into 'spring-cleaning' mode. wanted to 'get rid of it'.
17:20:48 <rloo> https://bugs.launchpad.net/ironic/+bug/1325193
17:20:49 <openstack> Launchpad bug 1325193 in python-ironicclient "Some Python files do not specify the file encoding" [Low,In progress] - Assigned to Martin Geisler (mgeisler)
17:20:59 <rloo> patch: https://review.openstack.org/#/c/97027/
17:21:16 <rloo> the patch basically adds a utf-8 coding line to every file.
17:21:54 <rloo> I discussed it briefly with lucas, and he would like that line added, because it will help eg Asian folks.
17:22:19 <rloo> Anyone against adding it?
17:22:54 <jlvillal> rloo: Why doesn't it use:  # coding=utf-8 ?
17:23:21 <dtantsur> why should it? :)
17:23:26 <jlvillal> rloo: But no opposition here
17:23:40 <jroll> I have no issues
17:23:46 <jlvillal> dtantsur: Well it is using the emacs version ;)
17:23:47 <dtantsur> IIRC this form was chosen as both Emacs and Vim understand it
17:23:47 <rloo> jlvillal: that's what it is proposing, to add coding: utf-8.?
17:23:51 <pshige> no opposition
17:24:04 <rloo> jlvillal: oh, that's what you meant. guess dtantsur answered :)
17:24:49 <jlvillal> dtantsur: If it works for both, it works for me :)
17:24:55 <rloo> ok, so can we do some #agree thingy saying it is OK, and I'll get in touch with Martin to see if he wants to finish it.
17:25:02 <devananda> rloo: my only opposition to this is that it's adding cruft that humans shouldn't have to maintain.. we're going to add files and forget to add this header to them
17:25:17 <devananda> rloo: it should be an editor setting, not a hint in every single file
17:25:30 <dtantsur> the opposite is valid too: people will add files with this line :)
17:25:39 <rloo> devananda: just like the other cruft we have. I'm fine not putting it in either. We already have some files with it.
17:25:50 <NobodyCam> can we add a hacking check for it?
17:25:56 <devananda> look at other projects -- I do not see this in every single file in nova, swift, keystone, ..
17:25:58 <NobodyCam> just enforce its there?
17:26:38 <rloo> I think there are lots of other things we should be spending time on deciding. I just wanted to 'do something' and be done with it. Note that the client files have it. (or had it; maybe some new ones don't now!)
17:27:02 <devananda> I understand the reason: some developers don't have their project (or their editor) configured in the same way that I do
17:27:10 <dtantsur> I agree that it should be done via hacking, whichever way we prefer
17:27:14 <devananda> but those developers are asking the project to assume responsibility to tell their editor what to do
17:27:25 <devananda> and i don't think it's the project's responsibility to carry the hint in every single file
17:27:39 <JayF> devananda: +1 I agree
17:27:45 <devananda> and then add a hacking rule to reject code that doesnt include a # -*- coding: utf8 -*- line
17:27:47 <pshige> devananda: +1
17:27:52 <jlvillal> I think it is only needed if there is non-ASCII code in the file.  From my reading.  Which should then fail the gate, which would be a hint to add it.
17:27:57 <jroll> isn't this also a signal to python?
17:28:06 <jlvillal> jroll: Yes
17:28:10 <jroll> to parse the file as utf-8, not ascii
17:28:13 <jroll> right, ok
17:28:15 <jlvillal> https://www.python.org/dev/peps/pep-0263/
17:28:27 <devananda> right. so if we HAVE non-ascii chars in a file, we should clearly add this hint for python
17:29:14 <devananda> but i still object to seeing it in every single file
17:29:16 <rloo> Martin's choice " c)  Remove the lines where they aren't strictly needed (Python requires the line if the file has non-ASCII characters)."
17:29:30 <devananda> so, let's vote
17:29:32 <NobodyCam> so I take this as add to a file if its needed
17:29:32 <rloo> but I don't think we want to enforce that *only* the files that need it, should have it.
17:29:50 <jlvillal> NobodyCam: I think if a file needs it, it should already be there.
17:30:31 <dtantsur> tbh I'm not sure we should have files with non-ASCII
17:30:36 <devananda> ugh. /me llooks up how to run a vote
17:31:12 <natorious> python 3 defaults to utf-8 too
17:31:12 * jroll votes for apathy
17:31:21 <devananda> #startvote should ironic require all files to have an encoding hint? Yes, No, Only, Abstain
17:31:21 <openstack> Begin voting on: should ironic require all files to have an encoding hint? Valid vote options are Yes, No, Only, Abstain.
17:31:22 <openstack> Vote using '#vote OPTION'. Only your last vote counts.
17:31:31 <devananda> "only"  = only when needed by python
17:31:43 <devananda> well. that's redundant.
17:31:45 <devananda> #undo
17:31:46 <openstack> Removing item from minutes: <ircmeeting.items.Link object at 0x98d9810>
17:31:54 <devananda> #startvote should ironic require all files to have an encoding hint? Yes, No, Abstain
17:31:55 <openstack> Already voting on 'should ironic require all files to have an encoding hint'
17:32:00 <devananda> bah
17:32:13 * rloo thinks we don't vote enough ;)
17:32:17 <devananda> #endvote
17:32:18 <openstack> Voted on "should ironic require all files to have an encoding hint?" Results are
17:32:24 <devananda> #startvote should ironic require all files to have an encoding hint? Yes, No, Abstain
17:32:24 <openstack> Begin voting on: should ironic require all files to have an encoding hint? Valid vote options are Yes, No, Abstain.
17:32:25 <openstack> Vote using '#vote OPTION'. Only your last vote counts.
17:32:25 <dtantsur> looks like #undo worked for the last link
17:32:32 <devananda> dtantsur: yea :(
17:32:37 <devananda> #vote no
17:32:54 <TheJulia> #vote no
17:32:55 <dtantsur> #vote No
17:32:56 <BadCub> #vote no
17:32:57 <rloo> #vote yes (on lucas' behalf)
17:32:58 <openstack> rloo: yes (on lucas' behalf) is not a valid option. Valid options are Yes, No, Abstain.
17:33:04 <dtantsur> lol
17:33:07 <rloo> #vote yes
17:33:11 <NobodyCam> #vote no
17:33:16 <jlvillal> #vote no
17:33:17 <Shrews> #vote no
17:33:24 <pshige> #vote no
17:33:25 <natorious> #vote no
17:33:37 <JayF> #vote no
17:33:49 <JoshNang> #vote no
17:34:01 <devananda> will give it another 30 seconds
17:34:06 <jroll> #vote abstain
17:34:07 <vdrok> #vote no
17:34:18 <NobodyCam> I do under stand lucas' point, but agree with devananda it should n't required for every file
17:34:25 <Shrews> the change needs a better story, IMO
17:34:51 <devananda> #endvote
17:34:52 <openstack> Voted on "should ironic require all files to have an encoding hint?" Results are
17:34:53 <openstack> Yes (1): rloo
17:34:54 <openstack> Abstain (1): jroll
17:34:55 <openstack> No (12): TheJulia, NobodyCam, pshige, jlvillal, BadCub, devananda, JayF, JoshNang, dtantsur, Shrews, natorious, vdrok
17:35:15 <rloo> sweet. thx everyone!
17:35:16 <NobodyCam> the no's have it
17:35:20 <devananda> rloo: cheers. there you go :)
17:35:49 <NobodyCam> ok to move on.
17:35:57 <devananda> I'd like to take a little collective time to look at the ideas we've got for the summit
17:36:20 <NobodyCam> :)
17:36:20 <pshige> devananda: +1
17:36:24 <devananda> we have about 3 weeks to organize them, which'll go pretty fast
17:36:38 <jroll> before we move on, dare I ask if we should remove the encoding hints from ironicclient now?
17:36:43 <NobodyCam> #link https://etherpad.openstack.org/p/liberty-ironic-design-summit-ideas
17:36:58 <devananda> jroll: heh. I'd +2 that patch, fwiw
17:37:02 <jroll> :)
17:37:08 <devananda> but also don't care /that/ much
17:37:16 <NobodyCam> ++
17:38:06 <rloo> jroll: being a consistency freak, I'll remove them :)
17:38:31 <jroll> heh, thanks rloo :)
17:38:47 <devananda> we've got 3 types of sessions, so I think i'll start (after the meeting) reorganizing this 'pad based on that
17:39:04 <NobodyCam> lots of good topics on the etherpad
17:39:13 <rloo> devananda: do/should we have a deadline for taking suggestions?
17:39:16 <devananda> 2 big room slots, which is more a place to gather feedback from the community
17:39:50 <devananda> several fishbowl / small room things to sort stuff out ourselves
17:39:56 <devananda> and then a full day friday to do what ever we need to
17:40:19 <devananda> also, if anyone is proposing sessions with other service teams,w e should cross-link from this pad
17:41:22 <NobodyCam> devananda: do have sessions inmind for the larger "big room " slots
17:41:54 <devananda> rloo: it'd help me out if all the suggestions are in by May 7 at the latest
17:42:21 <NobodyCam> I'd love to bring up third party testing requirments in fromt of a large group but I also it it easly going sideways
17:42:48 <rloo> devananda: so after you reorg the etherpad, maybe add a section/area for new suggestions?
17:42:49 <NobodyCam> s/it it/see it/
17:42:54 <devananda> NobodyCam: yah. I think 3rd party CI is a thing we need to start being more explicit about, but I dont think a whole session on that is going to produce anything
17:43:09 <devananda> rloo: ++
17:43:21 <NobodyCam> devananda: ya
17:43:58 <devananda> things that might need a big room, on my mind ...
17:44:06 <devananda> - networking stuff
17:44:10 <rloo> is there an operator's track where we might get feedback for ironic. last summit, I don't recall much feedback at all.
17:44:13 <devananda> - general ops feedback
17:44:19 <NobodyCam> would love smaller group session on : Driver composition matrix
17:44:21 <devananda> rloo: there is an ops track. we should propose sessions there
17:44:39 <devananda> NobodyCam: +1 for small group session for the driver matrix
17:46:16 <devananda> #link https://wiki.openstack.org/wiki/Design_Summit/Planning
17:46:26 <devananda> all the projects' etherpads for proposing things are there ^
17:46:27 <NobodyCam> any reason to cover any of the micro version stuff in the big rooms? (i thinking no but want to hear others thoughts)
17:47:13 <rloo> NobodyCam: don't think so, but we might want to discuss 'finishing' it, maybe on Friday?
17:47:30 <rloo> NobodyCam: and/or talk to Nova folks to see how it is working for them :)
17:47:53 <devananda> I think we need to have two conversations with nova this cycle
17:48:16 <rloo> NobodyCam: unless one of the big room sessions is to discuss what is new in kilo.
17:48:17 <devananda> one around how they will continue handling federated / clustered hypervisors
17:48:36 <devananda> another for the networking issues around assuming that we have only one network per port
17:48:46 <devananda> does anyone want to take the lead on either of those?
17:49:02 <NobodyCam> devananda: maybe also some chatting on the topic of testing?
17:49:08 <jroll> I definitely do not want the lead on the former :P
17:49:31 <jroll> I could take the latter if necessary, though I'm not sure the best way to fix it
17:49:42 * rloo guesses that non-ironic folks will be asking about federated/clustered hypervisors
17:50:00 <devananda> NobodyCam: keeping everyone on the same page about testing is good. maybe a slot in our meetup on friday?
17:50:11 <NobodyCam> oh ++
17:50:12 <devananda> rloo: nova folks are, by and large, quite familiar with the problem space by now
17:50:23 <devananda> rloo: they seem to differ in how they want to address it, though
17:50:40 <devananda> jroll: if you can, that'd be great
17:50:41 <rloo> devananda: yeah, so they need to go into a small room and duke it out :)
17:50:44 <jroll> so really this is about getting them all in a room and making them figure it out :P
17:51:02 <devananda> #link https://etherpad.openstack.org/p/liberty-neutron-summit-topics
17:51:09 <jroll> devananda: I guess what does 'take the lead' entail here? I'm not sure I want to run that session, but I can try to make sure it happens
17:51:09 <devananda> #link https://etherpad.openstack.org/p/liberty-nova-summit-ideas
17:51:25 <NobodyCam> * fyi Nine minutes *
17:51:28 <devananda> jroll: where 'them' includes some of us
17:51:35 <jroll> right
17:53:22 <rloo> is it on purpose that people put suggestions in etherpad w/o their names?
17:53:53 <devananda> rloo: harder to read with names every 20 chars?
17:54:11 <devananda> rloo: but seriously, if someone wants to run a session, they should put their name up, otherwise I'm going to run them all :P
17:54:33 <rloo> devananda: sometimes I wonder/want more details, can't ping someone if you don't know who they are.
17:54:59 * rloo is fine with devananda running all the sessions :)
17:55:15 <JayF> devananda: Here's the question: Do you want the person running the session to have an opinion on it?
17:55:26 <devananda> honestly, rather than proposals by a single person // pushing a single new feature, we should really be discussing things that most (if not all) of us are already familiar with
17:55:34 <NobodyCam> rloo: if users set their name the solor will tell you who it was
17:55:36 <JayF> devananda: I'd think an 'impartial moderator' could be more interesting in some of these formats
17:55:38 <devananda> feature proposals should be going through the spec process
17:56:04 <devananda> JayF: I'm hardly impartial, but I am happy to moderate :)
17:56:04 <NobodyCam> s/solor/color/
17:56:05 <rloo> devananda: but eg the state machine stuff got somewhat hashed out at the last summit.
17:56:43 <devananda> rloo: right. that's a great example of the sort of high bandwodth collaboration I'd like to see more of htis time
17:56:47 <rloo> devananda: although I guess it was something that more than one person was interested in.
17:57:13 <JayF> devananda: well I'm saying generally I'm willing to take that role in sessions as needed;
17:57:22 <devananda> JayF: awesome, thanks.
17:57:47 * rloo wonders when JayF has ever been impartial, but am happy to have JayF do it too :-)
17:57:57 <NobodyCam> lol
17:58:01 <JayF> rloo: I usually only talk when I have an opinion on things :)
17:58:09 <NobodyCam> * Two Minutes Left *
17:58:09 <JayF> rloo: something like functional testing ... my only opinion is that we do it :D
17:58:42 <rloo> JayF: opinions are good ;)
17:59:10 <rloo> by the way, I did want to mention/confirm. we are NOT supporting nova baremetal migration outside of juno, right?
17:59:18 <devananda> rloo: correct
17:59:27 <devananda> that code was dropped from nova already anyway
17:59:42 <rloo> devananda: right, but we forgot to drop the migration code from ironic/kilo.
17:59:43 <devananda> and we're not supporting a migration that skips a release at this time
17:59:49 <devananda> rloo: ugh
18:00:02 <rloo> devananda: i have a patch for L*, you want that backported to kilo?
18:00:07 <NobodyCam> * beep * thats time
18:00:09 <devananda> rloo: wait that's right. it should be in kilo, not in liberty
18:00:12 <devananda> anyway, thanks all
18:00:18 <devananda> i'll clean up this pad before next week
18:00:19 <NobodyCam> thank you all
18:00:21 <dtantsur> thanks
18:00:24 <devananda> hopefully some of you want to help add more things :)
18:00:26 <devananda> cheers!
18:00:27 <pshige> thanks
18:00:32 <NobodyCam> #endmeeting