21:00:51 <tonyb> #startmeeting stable
21:00:51 <openstack> Meeting started Mon Aug 29 21:00:51 2016 UTC and is due to finish in 60 minutes.  The chair is tonyb. Information about MeetBot at http://wiki.debian.org/MeetBot.
21:00:52 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
21:00:55 <openstack> The meeting name has been set to 'stable'
21:01:32 <tonyb> So I don't have an agnda really because I'm slack
21:01:41 <tonyb> 2 things that are worth talking about are
21:01:52 <mriedem> o/
21:01:56 <tonyb> 1) space requirements in barcelona
21:02:33 <tonyb> I was thinkign we only need on fb session to discuss the EOL date for newton
21:02:56 <tonyb> and possibly one to discuss the cinder stuff
21:03:08 <mriedem> do you need a fb session for talking about newton EOL?
21:03:22 <mriedem> isn't there just like a general formula for deciding that?
21:03:30 <tonyb> mriedem: only because I thought that was the "tradition"
21:03:40 <mriedem> i guess there is a wrinkle with ocata being a short cycle
21:03:44 <mriedem> i'm not sure how that affects things
21:03:57 <mriedem> short by what 1 month?
21:04:11 <tonyb> I think the policy says 12months post release so I don't think it does
21:05:30 <mriedem> so oct 12 2017
21:05:48 <mriedem> done!
21:05:55 <mriedem> one day after my birthday
21:06:10 <mriedem> maybe we should EOL on my bday as a nice gift
21:06:29 <tonyb> mriedem: awww I was going to push it back to the 16th but now I'm not sure ;P
21:06:41 <mriedem> i'd rather have more costco cake anyway
21:06:54 <tonyb> :)
21:07:49 <tonyb> I'll check with fungi as he has a lot of experience here and then fil in the google form
21:08:12 <mriedem> ok i wasn't sure if it needed to happen after the summit around that same time or not
21:08:24 <mriedem> which would be sydney
21:09:18 <tonyb> yeah.
21:10:16 <tonyb> the other thing I wanted to talk abotu today/ at Barcelona was the whole cinder/cinder-drivers thing.  but I think that shoud be in the cinder track really
21:10:57 <mriedem> well,
21:11:07 <mriedem> i guess i didn't follow the entire ML thread,
21:11:25 <mriedem> but i'd think you'd want the same cast of characters from infra and VMT and QA that were at my xp session in austin
21:11:29 <mriedem> because the concerns are going to be same
21:12:17 <tonyb> Yeah that's true.
21:12:43 <tonyb> We could frame it as a generic discussion but with cinder as the exmaple under consideration.
21:14:04 <mriedem> sure, who is going to run that session?
21:14:09 <mriedem> you? walter? sean?
21:14:16 <mriedem> sean = smcginnis
21:14:34 <tonyb> we can decide closer to the time but if noone better shows up then I'll do it.
21:14:45 <tonyb> mriedem: Yeah that was my assumption
21:14:47 <mriedem> i think cinder should drive this myself
21:14:58 <mriedem> make the proposal and be ready for that at the summit
21:15:05 <mriedem> with input before hand from the various teams mentioned above
21:15:52 <mriedem> i don't have great answers though, on the one hand teams should be free to do what they want, but i don't think they should be able to do that at the expense of other projects or maintainers
21:16:11 <tonyb> Right.
21:16:32 <mriedem> and part of me says boohoo for people having to carry patches out of tree for their distro support branches
21:16:34 <mriedem> everyone does that
21:17:00 <mriedem> i really just wonder how much of this is coming from helion
21:17:11 <mriedem> because the other distros have already handled this problem i think
21:17:14 <tonyb> Yeah.  I guess it's a question of volume of patches
21:18:05 <mriedem> i don't know what the proposal is honestly
21:18:11 <mriedem> someone needs to clearly communicate that
21:18:29 <mriedem> if it means we have to keep around a bunch of old infra just for a stable/liberty branch for cinder, i don't think that's cool
21:19:03 <tonyb> well right now I think the proposal is let cinder do what they've suggested (nuke 3rd party drivers) and removes tags from the team
21:19:23 <mriedem> nuke 3rd party drivers?
21:19:35 <mriedem> what does that mean? when their ci fails or doesn't report for awhile?
21:19:43 <tonyb> Yeah drivers that don't have workgin CI are being git rm'd from master
21:19:46 <mriedem> note that there is no gating job for NFS, which is not 3rd party
21:20:06 <tonyb> that's a good point but different
21:20:09 <mriedem> so there are some double standards
21:20:24 <tonyb> true.  I didn't think of it like that.
21:20:36 <mriedem> still, what does removing poorly behaving drivers have to do with keeping around older stable branches?
21:20:55 <tonyb> nothing they have 2 things going at the same time
21:20:59 <mriedem> anyway, needs a FB session for sure
21:21:15 <mriedem> i'll be front row to nit pick :)
21:21:23 <mriedem> i think mtreinish will like that one
21:21:28 <tonyb> the first was "policy relaxation for stable/* and drivers which implied long lived branches"
21:21:42 <tonyb> the 2nd was/is the removal of "non-ci drivers"
21:21:52 <tonyb> :)
21:22:03 <mriedem> oh right they want to allow backporting any non-feature things to the oldest stable branch as long as it was for volume drivers
21:22:14 <mtreinish> mriedem: ?
21:22:19 <tonyb> okay I'll put it in stable but sugest to ttx that it is probably more appropriate for cross-project
21:22:29 <mriedem> mtreinish: discussing summit session about the cinder stable branch stuff
21:22:36 <tonyb> Oh no you woke him ;P
21:23:17 <mtreinish> oh, yeah I assumed there was going to be a session about that and/or extending the eol window again
21:23:18 <tonyb> I know that cinder have problems with it but I kinda think that the least painful thing is to spin out cinder-drivers and move on
21:23:24 <mtreinish> just like at every summit before it
21:23:28 <tonyb> but that's probably over-simple from my POV
21:24:00 <tonyb> mtreinish: We;; we both said pony up people and we'll talk ... crickets
21:24:02 <mriedem> anyway, maybe we should move on, and agree there will be a session for that
21:24:07 <tonyb> okay
21:24:13 <tonyb> last thing for today is ....
21:24:37 <tonyb> mriedem: If we moved this meetign to weekly ar 8am (Wednesday) for you woudl that suck?
21:25:04 <mtreinish> tonyb: 8am for you, utc, for me, or for mriedem?
21:25:13 <mriedem> my attendance record at this meeting isn't great lately so i wouldn't worry about me
21:25:26 <tonyb> mtreinish, mriedem 8am Central 2300 Canberra
21:25:38 <mriedem> i'm usually in the office by 8:30am CT
21:25:45 <tonyb> mriedem: neither is mine but I don
21:25:53 <tonyb> t want to exclude you if we move
21:25:55 <mriedem> so i just show up late
21:26:26 * mtreinish is normally still asleap at 9am edt
21:26:41 <tonyb> mtreinish: hmm okay.
21:27:09 <mtreinish> but I often forget about this meeting, so don't worry about excluding me
21:27:17 <tonyb> mtreinish: :)
21:27:27 <tonyb> mtreinish: we summon you as required ;P
21:27:55 <mriedem> you have to make a sacrifice to summon the beast
21:27:59 <mriedem> or roll a saving throw
21:28:07 <tonyb> :)
21:28:33 <mriedem> i was going to mention this again if you're out of topics https://review.openstack.org/#/c/327624/
21:28:38 <mriedem> since it was brought up to me again today
21:28:51 <mriedem> and this https://review.openstack.org/#/c/337277/
21:29:08 <mriedem> seems everyone conveniently forgot about that, i'm assuming rhel just patched it as they needed to
21:29:24 <tonyb> hehe, I didn't forget I got distracted :(
21:29:57 <mriedem> since i'm not in the business of handling backports for the ibm distro anymore, i don't know how the team handled that here
21:30:03 <mriedem> or if they have even gotten to it yet
21:30:33 <tonyb> So all the options suck it's a question of which sucks the least.
21:31:12 <tonyb> I don't like Sean's idea of createing a 2.3.1 and settign that as the minumum as that leaves a whole bunch of valid versions that are broken
21:31:43 <tonyb> which leaves a pretty big bump in g-r *or* a hack to nova to work with old (and vilnerable) or new.
21:32:58 <tonyb> I guess the rigth thing for affected users if the g-r bump but then we're going back on our word and invalidate a whole bunch of history :(
21:34:58 <tonyb> I guess we could do the 2.3.1 release and also blacklist the otehrs but that's really just the big minimum bump by stealth
21:37:36 <tonyb> Well that killed the conversation ;P
21:38:02 <mriedem> i got distracted :)
21:38:07 <tonyb> :)
21:38:16 <mriedem> let's just table it, i have to head out to get my kid
21:38:23 <tonyb> okay
21:38:24 <mriedem> let's resume the discussion after 11/17
21:38:31 <mtreinish> mriedem: ++ :)
21:38:40 <mriedem> ha
21:38:44 <tonyb> lol
21:38:50 <mriedem> mtreinish: you have that date memorized?
21:39:24 <tonyb> mriedem: He has a countdown going on his phone's home screen
21:39:33 <mtreinish> hah
21:40:09 <tonyb> #action tonyb to re-evaluate options for fixing the nova CVE than needs an oslo bump
21:40:37 <tonyb> mriedem: needs to leave so ... # endmeeting?
21:40:50 <mriedem> yes please
21:41:24 <tonyb> #endmeeting