22:01:25 <armax> #startmeeting neutron_drivers
22:01:26 <amuller> hiho
22:01:27 <openstack> Meeting started Thu Mar 10 22:01:25 2016 UTC and is due to finish in 60 minutes.  The chair is armax. Information about MeetBot at http://wiki.debian.org/MeetBot.
22:01:28 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
22:01:30 <openstack> The meeting name has been set to 'neutron_drivers'
22:01:30 <dougwig> sure
22:01:52 <njohnston> o/
22:01:54 <armax> ok, before we dive in
22:01:59 <armax> a brief announcement
22:02:10 <armax> mestery has decided to step down due to other committments
22:02:11 <armax> #link https://review.openstack.org/#/admin/groups/464,members
22:02:20 <kevinbenton> :'(
22:02:21 <armax> we all hail to Cesar
22:02:31 <armax> praise his legacy
22:02:33 <carl_baldwin> :(
22:02:45 <dougwig> from drivers, core, release manager, ?
22:02:48 <dougwig> neutron?
22:02:50 <armax> this comes at a good time
22:02:58 <armax> dougwig: don’t freak out
22:03:01 <armax> just the drivers team
22:03:05 <dougwig> no, it was just a question. :)
22:03:06 <armax> for now, I guess anyway
22:03:08 <kevinbenton> oh, good
22:03:13 <armax> sorry, I posted a link
22:03:14 <kevinbenton> i thought he was quitting everything
22:03:20 <armax> to the drivers team
22:03:38 <armax> I would not make that announcement here now would I?
22:03:41 <armax> come on guys
22:03:42 <armax> be sharp!
22:03:48 <armax> :)
22:03:54 <dougwig> oh, it's our fault you were unclear.  i see.  :)
22:03:55 <kevinbenton> well if it was from everything this would be a good place to announce it anyway :)
22:03:59 <armax> :)
22:04:14 * armax rolls back
22:04:29 <armax> mestery has decided to step down from the drivers team due to other committments
22:04:36 <kevinbenton> :'(
22:04:38 <armax> better?
22:04:59 <dougwig> much
22:05:01 <jreeves> precision is a virtue
22:05:02 <ihrachys> breaking change. someone please revert.
22:05:04 <carl_baldwin> Still not good news but not as bad
22:05:19 <armax> someone goes another one comes
22:05:32 <armax> we’ll look into extending the team
22:05:58 <kevinbenton> i would like to nominate armax
22:06:03 <armax> and if any of you on the drivers team as of today wants to take a leave of absence
22:06:07 <armax> please reach out to me
22:06:11 * armax too
22:06:38 <kevinbenton> ack
22:06:46 <armax> bear in mind that if the drivers team dries up, this means we switchs to bug fixes
22:06:56 <armax> and no longer accept features :)
22:07:08 <armax> I leave you to figure out whether I am kidding or not
22:07:23 <dougwig> i've lost the ability to tell when you're kidding.
22:07:32 <armax> I am coy that way
22:07:35 <kevinbenton> dougwig: i think armax has too
22:08:00 <armax> kevinbenton: PTL-dom does that to you
22:08:04 <armax> anyhoo
22:09:15 <armax> one thing I wanted to mention today was handling deprecation of config variables and behavior in a way that’s more trackable
22:09:51 <armax> right now the process is pretty loose
22:10:39 <armax> in that we may decide to issue a deprecation warning, we usually follow up one/two cycles for removal
22:10:47 <armax> we release note etc
22:11:03 <armax> and we sometimes reach out for operator/user feedback
22:11:50 <armax> but it’s difficult right to see really quickly what we decided to deprecate in any given cycle and what countermeasure we provide
22:12:26 <amuller> #link https://review.openstack.org/#/c/291360/
22:12:34 <armax> amuller: attaboy
22:12:54 <armax> you must have though: ‘dude get to the point'
22:12:58 <amuller> :)
22:13:00 <armax> thought
22:13:27 <dougwig> as someone who recently upgraded a cloud from juno -> kilo -> liberty, i will add that we openstack devs deprecate config flags WAY TOO FRIGGING OFTEN. it's a royal pain.
22:13:29 <armax> but that’s only the first step
22:13:54 <armax> right, and we need to manage the communication and feedback a little better
22:14:05 <dougwig> and avoid just doing it for cosmetic reason.
22:14:08 <armax> we were almost dropping the ball on this one
22:14:11 <armax> https://bugs.launchpad.net/neutron/+bug/1502356
22:14:11 <openstack> Launchpad bug 1502356 in neutron "max_fixed_ips_per_port appears to be unnecessary" [Low,Fix released] - Assigned to Kevin Benton (kevinbenton)
22:14:27 <armax> and I stumbled upon it by ‘accident’
22:14:45 <armax> now we’re still trying to figure out what to do on this one, but for starter this didn’t have a release note
22:14:55 <kevinbenton> release notes didn't exist
22:15:00 <armax> kevinbenton: they did
22:15:07 <armax> kevinbenton: it’s mitaka
22:15:09 <kevinbenton> only in an abstract sense :)
22:15:14 <armax> kevinbenton: and reno was already in place
22:15:26 <ihrachys> kevinbenton: then that should have been partial fix ;)
22:15:28 <armax> kevinbenton: I am not blaming you
22:15:38 <armax> kevinbenton: we are only human
22:15:58 <armax> kevinbenton: we’re bound to make mistakes, that’s why we have code janitors and trivial fixers
22:16:07 * carl_baldwin thought kevinbenton was more than human.
22:16:27 <kevinbenton> https://github.com/openstack/neutron/commit/271d3c1faed1103cc7489323728edaa037709e0d
22:16:34 <kevinbenton> i request the court heretoforth read the commit date
22:16:38 <armax> and I think that in the best possible way
22:16:39 <kevinbenton> on said patch
22:16:49 <kevinbenton> and then the merge date of this https://review.openstack.org/#/c/230696/
22:16:54 <kevinbenton> :)
22:17:23 <kevinbenton> I PLEAD THE 5TH!
22:17:34 <armax> kevinbenton: dude, no-one is putting you to trial!
22:17:37 <armax> :)
22:17:46 <amuller> kevinbenton: I think the 5th was removed from the stadium
22:17:46 <armax> kevinbenton: you’re the one pleading guilty with all this fuss!
22:17:51 <armax> :)
22:17:57 <armax> I am only taking that as an example
22:17:59 * ihrachys is amuzed by his own comment at https://review.openstack.org/#/c/230696/1//COMMIT_MSG
22:18:02 <njohnston> If the gate fits, you must commit!
22:18:10 <kevinbenton> njohnston: ha!
22:18:28 <armax> what counts is that we got this in time
22:18:29 <kevinbenton> ihrachys: whoops, forgot to add the docimpact
22:18:39 <kevinbenton> :)
22:18:44 <armax> so kevinbenton you’re guilty and charged if you don’t follow up now!
22:19:05 <kevinbenton> armax: but release notes didn't exist. i think there is something that says you can't be charged for something that wasn't illegal when you did it!
22:19:06 <armax> you did add the it
22:19:09 <ihrachys> kevinbenton: not that one; you actually added docimpact
22:19:30 * kevinbenton will write a release note for this
22:19:33 <armax> am I talking jibberish?
22:19:39 * dougwig wonders when the yak shaving about reno merging will stop.
22:19:47 * carl_baldwin notes this still sounds like a trial but it shouldn't.
22:20:29 <armax> it clearly isn’t :)
22:20:57 <armax> I seek your opinion on the matter
22:21:16 <armax> and I wonder if you think that closing a potential loophole is worth pursuing
22:21:35 * njohnston departs
22:22:10 <carl_baldwin> I think that providing better communication and avoiding dropping the ball is good.
22:23:02 <armax> we need to figure out a sane way to do that, but that’s the intention
22:23:04 <dougwig> knowing kevin, he's probably writing the reno patch right now rather than talking about it.
22:23:42 <armax> most likely
22:24:00 <amuller> armax: did you have a process in mind?
22:24:36 <armax> I haven’t formulated one yet, but I think it would go down like this
22:25:36 <armax> we’ll have period checks/updates (on a per-milestone basis?) on what’s been deprecated/removed in any given cycle
22:26:02 <armax> I’ll put together a script to pull this stuff from gerrit and launchpad
22:26:13 <amuller> just using the launchpad API, assuming we have the deprecation tag in place
22:26:16 <amuller> ?
22:26:19 <armax> amuller: right
22:26:33 <amuller> then send an email to openstack-operators with the list?
22:26:51 <armax> yes, and watching the bug reports for feedback too
22:26:56 <armax> in case it shows up there
22:27:08 <amuller> I think that's more than reasonable, it's a pretty light weight process
22:27:42 <armax> ihrachys: so let’s figure out how to integrate this in your patch https://review.openstack.org/#/c/291288/
22:27:55 * ihrachys makes notes
22:27:55 <armax> let’s all praise ihrachys for starting this by the way
22:28:06 <armax> ihrachys: or I can simply follow up
22:28:26 <ihrachys> armax: you are absolutely welcome to do while I will maybe sleep
22:28:31 <amuller> ihrachys++
22:28:41 <armax> am I, aren’t I?
22:28:46 <armax> :)
22:29:18 <armax> I did expand on deputy duties already here: https://review.openstack.org/#/c/291361/
22:30:15 <armax> ihrachys: people don’t seem to like you
22:30:19 <armax> only amuller praised
22:30:23 <armax> ihrachys: sorry
22:30:39 <ihrachys> armax: that's ok, they are just too shy of my magnificence
22:30:45 <armax> ihrachys: it might be it
22:31:00 <carl_baldwin> attaboy ihrachys
22:31:03 <armax> ok, one more thing I had in mind before I let you get back to your busy lives
22:31:40 <armax> does anyone of you have an update on rossella_s’ script to pull up a release dashboard?
22:32:04 <ihrachys> I remember she was working on getting it hosted and triggered in infra
22:32:04 <armax> last time I recall infra gave us a mechanism to publish the dashboard periodically
22:32:14 <armax> ihrachys: can I get a status update?
22:32:17 <armax> rossella_s: ^
22:32:35 <ihrachys> #link https://review.openstack.org/#/c/284281/
22:32:50 <armax> cool, I’ll dig into it
22:32:53 <armax> thanks ihrachys
22:33:00 <ihrachys> I don't have details, but I guess we wait for that to merge
22:33:07 <armax> understood
22:33:15 <ihrachys> your weight in infra circles may help to get it in ;)
22:33:25 <armax> then we can make changes to the script that generates the dashboard and everything works automagically?
22:33:36 <dougwig> i've still got it hosted in the meantime.
22:33:46 <armax> thanks dougwig
22:34:42 <ihrachys> armax: probably. honestly, we better ask Rossella ;) I can get to her with that when she comes here to Brno on Mon :)
22:35:32 <armax> ihrachys: sure
22:35:46 <armax> ok, I have nothing else on my plate
22:36:23 <dougwig> that dashboard in infra will be nice going forward.
22:36:33 <armax> yup
22:36:52 <armax> remember that next week we’ll cut RC1
22:37:02 <armax> so if there’s anything critical to be released
22:37:09 <dougwig> after that, COKE1, then PEPSI1
22:37:21 * dougwig goes back into his hole.
22:37:25 <armax> please scrub your gerrit/launchpad dashboards
22:37:41 <armax> to make sure we have nothing that falls through the cracks
22:37:48 <armax> dougwig: come back
22:38:06 <armax> anything else?
22:38:30 <carl_baldwin> Not from me.  Anxious to go scrub.
22:38:53 <armax> one
22:39:00 <armax> two
22:39:10 <armax> three
22:39:13 <armax> bye then
22:39:17 <armax> #endmeeting