21:02:55 <ttx> #startmeeting project
21:02:56 <openstack> Meeting started Tue Sep  2 21:02:55 2014 UTC and is due to finish in 60 minutes.  The chair is ttx. Information about MeetBot at http://wiki.debian.org/MeetBot.
21:02:57 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
21:03:00 <openstack> The meeting name has been set to 'project'
21:03:02 <ttx> Our agenda for today:
21:03:09 <ttx> #link http://wiki.openstack.org/Meetings/ProjectMeeting
21:03:16 <ttx> #topic News from the 1:1 sync points
21:03:22 <ttx> Here is the log:
21:03:31 <ttx> #link http://eavesdrop.openstack.org/meetings/ptl_sync/2014/ptl_sync.2014-09-02-08.00.html
21:03:38 <ttx> We missed Cinder and Heat, will cover those later in this meeting
21:03:46 <ttx> Swift released 2.1.0 on Monday and was excused
21:03:58 <ttx> The general idea is that today is the last day to approve stuff for juno
21:04:06 <ttx> Tomorrow we'll defer stuff that's not approved yet
21:04:13 <ttx> and then we'll wait for the gate to catch up and retry stuff
21:04:20 <ttx> In time for tagging J3 on Thursday
21:04:48 <ttx> so that means today is really last review day
21:05:02 <ttx> #topic Other program news
21:05:09 <ttx> Any other program with a quick announcement ?
21:05:20 <mtreinish> ttx: nothing from me this week
21:05:41 <ttx> jeblair: ?
21:06:09 <jeblair> ttx: nak
21:06:11 <ttx> ok, skipping to design summit discussion while we have everyone
21:06:15 <ttx> #topic Design Summit session scheduling
21:06:26 <ttx> I'm considering dropping the "session suggestion" website in favor of self-organization within teams (like etherpads and ML discussions)
21:06:42 <ttx> The idea being we'll likely have a lot less slots for scheduled sessions, so rather than denying a lot more (which is negative), we should encourage people to join schedule discussions (which is positive)
21:06:55 <ttx> We'd only use the session suggestion website for "other" projects and for the final scheduling step
21:06:56 <mestery> ++ to that idea!
21:07:04 <ttx> Would anybody regret the open "suggestion" thing ?
21:07:09 <mestery> I think this is a very productive way for teams to work together on proposing relevant sessions.
21:07:10 <jeblair> ttx: wfm
21:07:19 <david-lyle> +1
21:07:24 <eglynn_> ttx: why are we likely have a lot less slots for scheduled sessions this time round?
21:07:32 <ttx> I think we need to make sure we reach out to operators so that they participate in that per-program topic scheduling activity
21:07:40 <mtreinish> ttx: +1
21:07:55 <ttx> eglynn_: because we are likely to use Friday for contributors gatherings
21:08:04 <ttx> that's one day left
21:08:06 <ttx> less*
21:08:09 <SlickNik> ttx: I like the idea self-organization idea as well. So +1 from me.
21:08:13 <eglynn_> a-ha, k
21:08:28 <ttx> So that it works, you'll have to be very inclusive
21:08:36 <eglynn_> I'm open to open scheduling, for the Juno summit we did it a collaborative exercise among the ceilometer core team
21:08:43 <ttx> but overall I think it should yield better results
21:09:01 <eglynn_> "very inclusive" == "include a wider group than the core team"?
21:09:11 <ttx> yes, and reach to operators ion their list
21:09:30 <ttx> make sure you have a list of topics and all the data you need to choose
21:09:45 <ttx> we'll likely have scheduled slots and the "gathering" thing
21:09:49 <dhellmann> we've started an etherpad for oslo: https://etherpad.openstack.org/p/kilo-oslo-summit-topics
21:10:02 <ttx> the idea is to use scheduled slots for the stuff you want extra input in
21:10:09 <ttx> since it will appear on the schedule
21:10:27 <eglynn_> do we have an expectation that the proposer of a session is intending to drive the actual implementation of related BP(s) in Kilo?
21:10:30 <ttx> while team-specific issues can be covered on ML,; spec or the Friday thing
21:10:50 <eglynn_> as opposed to ... "this is something I'd love to see somebody else implement"
21:10:51 <SergeyLukjanov> fwiw +1 on self org
21:11:07 <ttx> eglynn_: no general expectation, but feel free to have a local one
21:11:08 <anteaya> ttx not sure where to ask so I'll just interrupt, where do you want any third party discussions?
21:11:14 <dolphm> eglynn_: i don't see much room for "session proposers" anymore. it's the community recognizing a need together.
21:11:18 <notmyname> ttx: I don't follow that last part. so each program still needs to come up with their list of "sessions"?
21:11:20 <dhellmann> eglynn_: my etherpad template separates the "driver" from "interested" parties
21:11:32 <notmyname> ttx: and then schedule them?
21:11:50 <notmyname> ttx: or are programs assigned a block and they manage whatever in it?
21:11:54 <ttx> notmyname: yes, you'll still have a number of scheduled slots, in addition to the informal thing on Friday
21:12:29 <ttx> notmyname: on the Wednesday the conference is still running, so alignement has /some/ value
21:12:49 <eglynn_> dhellmann: OK so the proposer doesn't have to be the actual driver, but surely there must be an identified driver at some point? (preferably in advance of the design session occurring)
21:13:07 <ttx> if really the 40-min step is a problem for you we could discuss non-alignment on the Tursday (since the conference will be over)
21:13:11 <notmyname> ttx: but instead of the webapp sessions suggestion, we have to get everyone together on an etherpad to choose the things to be talked about?
21:13:13 <dhellmann> eglynn_: right, but if the proposer isn't the driver and no one picks it up, we probably won't talk about it
21:13:34 <dhellmann> eglynn_: that still gives the community a way to ask us to work on something, though
21:13:41 <eglynn_> dhellmann: yep, that's fair enough
21:13:43 <ttx> notmyname: yes, and which ones should have scheduled slots, and which ones should be kept for the Friday
21:14:04 <dolphm> the logical conclusion to this is using gerrit to manage sessions. just sayin'
21:14:07 <ttx> I still have to check feasibility of the Friday thing wrt. the location
21:14:15 <ttx> I'm visiting there on Monday
21:14:22 <ttx> so we should have a final plan next week
21:14:38 <ttx> i'll keep you all posted
21:14:40 <ttx> ok, moving on
21:14:49 <notmyname> I guess I don't see the advantage of scheduling on an etherpad vs suggestion and scheduling via the webapp
21:15:14 <mestery> notmyname: To that point, I agree, the webapp has some advantage there.
21:15:24 <ttx> it's a way to choose more collectively, and to avoid crazy proposals
21:15:42 <ttx> I think eterpad is easier to mix and merge and discuss
21:15:54 <ttx> than the webapp with all its limitations
21:16:05 <notmyname> how are they mutually exclusive? suggest things on the webapp and then, if necessary, use an etherpad during a team meeting
21:16:25 <ttx> it's just a bit more painful to merge sessions etc
21:16:32 <dolphm> ++
21:16:49 <ttx> also there is less frustration about your proposal being "rejected"
21:16:53 <stevebaker> we transferred everything to an etherpad anyway
21:16:57 <ttx> should cut down my hate mail in half
21:16:58 <eglynn_> notmyname: yep, that kinda how we did for the ceilo track in Juno ... except with s/etherpad/googledoc/
21:17:10 <SlickNik> stevebaker / ttx: That's what we ended up doing as well
21:17:40 <anteaya> ttx +
21:17:40 <SlickNik> (transferring to etherpad)
21:17:42 <ttx> notmyname: if you want to use the webapp as your etherpad, we can do that for you
21:18:02 <ttx> the site will run for "other projects" track anyway
21:18:25 <notmyname> ttx: no, I'm not trying to do something different. I'm trying to figure out what problem is being solved. I never had too many issues with the existing tool
21:18:28 <ttx> but i figured removing the parallel CFP completely would just be clearer
21:19:03 <ttx> notmyname: so issues: parallel CFp confusing the hell out of people, hate mail when people see their session rejected
21:19:18 <ttx> vs. collaborative schedule editing in a friendly atmosphere.
21:19:20 <dhellmann> notmyname: we are trying to make the summit planning a discussion among the core team rather than a free-for-all
21:19:23 <notmyname> what parallel CFP? between conference and summit? or between differetn programs?
21:19:32 <ttx> notmyname: between conference and summit.
21:19:35 <notmyname> ok
21:19:53 <eglynn_> ttx: what approx timeline is envisaged for this design session proposal gathering?
21:20:03 <ttx> notmyname: also additional benefit of not ahving to parse crazy proposals from non-contributors
21:20:13 <ttx> mid Sept-mid-Oct
21:20:16 <dolphm> ttx: and one line proposals
21:20:20 <eglynn_> k, thanks
21:20:33 <ttx> ok, moving on
21:20:38 <ttx> #topic Early Kilo deferrals
21:20:47 <ttx> I'd like to look at stuff that just can't make it at this point and which should be deferred at this point
21:21:16 <ttx> If you have stuff that is missing so many reviews it can't possibly make it now, you should defer it to kilo by marking in Deferred on your j3 plan
21:21:48 <dhellmann> I went ahead and moved a few things to a kilo series already. Should I move those back?
21:21:53 <stevebaker> ttx: what is the FFE process?
21:22:02 <eglynn_> can't possibily make it even with an FFE?
21:22:04 <SlickNik> ttx: how should we handle BPs that we anticipate getting an FFE for?
21:22:04 <ttx> dhellman_: no, kilo is fine
21:22:07 * mestery has been moving things out of Juno but not into Kilo yet.
21:22:30 <ttx> SlickNik: you could move them to RC1 and we'll review them friday
21:22:55 <SlickNik> ttx: okay, sounds good.
21:22:58 <ttx> stevebaker: Starting friday we'll start ML threads for every exception
21:23:07 <stevebaker> ttx: ok, thanks
21:23:11 <ttx> and you can push the candidates to the RC1 milestone
21:24:18 <ttx> i'll talk to you all tomorrow to cut down the list to in-flight stuff
21:24:43 <ttx> Any question on that before we sync specifically on heat and Cinder ?
21:25:07 <ttx> mestery: any progress today?
21:25:31 <mestery> ttx: Some progress yes, still looking at needing a few FFEs post Thursday, but trying to minimize.
21:25:54 * mestery has been working the review board, LP, the phone, and email all day trying to wrangle things together.
21:26:27 <ttx> david-lyle: how about you ? your list is pretty large too
21:26:41 <david-lyle> cutting it down now
21:26:55 <david-lyle> a couple are close to inflight as well
21:27:07 <ttx> ok, we'll sync tomorrow on progress
21:27:14 <ttx> keep the gate busy tonight
21:27:19 <ttx> #topic Heat status
21:27:23 <ttx> stevebaker: around?
21:27:30 <stevebaker> y
21:27:59 <ttx> you still have 11 in progress
21:28:17 <ttx> stevebaker: which ones are you likely to require FFE for ?
21:28:34 <ttx> the 3 remaining "high" ones ?
21:28:54 <stevebaker> maybe some of the mediums too
21:29:22 <stevebaker> some just need reviews and may just miss the deadline
21:29:31 <ttx> any chance you could get a few more approved before they require exceptions?.
21:29:53 <ttx> exceptions create disruption, so we need to minimize them
21:30:28 <stevebaker> yes, will drum up reviews today. there are quite a few low-risk changes too, which may help their FFE case
21:30:53 <stevebaker> I don't see any bugs which must be in j-3
21:30:55 <ttx> stevebaker: when can I talk to you tomorrow to finalize the list (keep only in-flight stuff in) ?
21:31:05 <ttx> 19:00 UTC ?
21:31:05 <stevebaker> ttx: this time tomorrow?
21:31:21 <ttx> that's a bit late for me
21:31:59 <stevebaker> ttx: I'll ping you when I'm concious enough to form words
21:31:59 <ttx> stevebaker: how early can you make it?
21:32:22 <ttx> hm, ok I'll try to keep an eye on IRC
21:32:28 <stevebaker> ~1900UTC
21:32:31 <ttx> jgriffith: around?
21:32:41 <jgriffith> ttx: present
21:32:44 <ttx> stevebaker: i'll be back online around 19:00 utc tomorrow
21:32:52 <ttx> #topic Cinder status
21:32:53 <stevebaker> ttx: ok
21:33:08 <ttx> #link https://launchpad.net/cinder/+milestone/juno-3
21:33:19 <ttx> jgriffith: 6 still up
21:33:30 <jgriffith> ttx: so I believe I'll be dumping the following:
21:33:37 <jgriffith> ttx: secure NFS
21:33:48 <jgriffith> ttx: Cisco device FC zone lifecycle
21:33:59 <jgriffith> ttx: the others are "close"
21:34:05 <ttx> ok
21:34:07 <jgriffith> ttx: and would be willing to do exceptions for
21:34:22 <jgriffith> ttx: the smb stuff is locked up with dependencies on other issues
21:34:28 <ttx> i'll be in touch tomorrow so that we defer/FFE everything that's not in-flight by then
21:34:34 <jgriffith> ttx: ok
21:34:50 <ttx> so approve as much as you can today
21:35:15 <ttx> jgriffith: will be pinging you in your morning :)
21:35:20 <jgriffith> ttx: :)
21:36:46 <ttx> #topic Open discussion
21:36:50 <ttx> Anything else, anyone ?
21:37:20 <notmyname> ttx: I still dont' get the summit scheduling
21:37:28 <notmyname> ttx: but we can address that on the ML if necessary
21:38:21 <ttx> notmyname: oops, replied on -dev to you
21:38:36 <notmyname> is there one big collaborative list of proposed sessions for all programs?
21:39:01 <notmyname> or is it still a per-program list?
21:40:56 <ttx> notmyname: at this point, per-program lists
21:41:08 <ttx> some on etherpads, some on whatever
21:41:19 <ttx> most often on etherpads
21:41:26 <ttx> announced on -dev threads recently
21:41:43 <notmyname> ok
21:42:08 <ttx> ok, let's close this one
21:42:15 <stevebaker> notmyname: its up to you whether you have strict 40 minute slots, or larger chunks with topic themes
21:42:23 <eglynn_> BTW wouldn't the timing be likely to push the summit scheduling into the new PTLs' terms?
21:42:34 <eglynn_> ... not that it's unlikely a lot of existing PTLs will retain their positions
21:42:41 <eglynn_> ... just I guess we shouldn't be making any assumptions about the election outcomes, right?
21:42:42 <stevebaker> eglynn_: it always has been the new PTL
21:42:46 <notmyname> eglynn_: new PTLs always do the summit sessions
21:42:51 <notmyname> "new"
21:42:52 <ttx> eglynn_: yes, final stuff is definitely in the future PTLs hands
21:43:23 <eglynn_> stevebaker, notmyname, ttx: yep, so we shouldn't be deciding now how they might want to play this, or?
21:43:40 <ttx> eglynn_: unfortunately some things need to be decided before
21:43:49 <ttx> like the format
21:43:56 <mestery> ttx: ++
21:43:57 <ttx> which is what we discuss now
21:43:58 <stevebaker> eglynn_: I'm sure some pre-planning could be done though
21:44:02 <eglynn_> fair nuffski
21:44:12 <ttx> also, brainstorming themes can start now
21:44:28 <ttx> I doubt the new PTL would object to open brainstorming
21:44:32 <ttx> "new"
21:44:45 <eglynn_> ttx: yep, that's a fair point
21:45:09 <eglynn_> open-ness == "motherhood and apple pie" :)
21:45:28 <ttx> ok, really closing now
21:45:31 <ttx> thanks everyone
21:45:35 <ttx> #endmeeting