| *** markwash__ has quit IRC | 00:58 | |
| *** markwash__ has joined #openstack-relmgr-office | 01:00 | |
| *** markwash__ has quit IRC | 01:46 | |
| *** markwash__ has joined #openstack-relmgr-office | 01:48 | |
| *** markwash__ has quit IRC | 01:48 | |
| *** markwash__ has joined #openstack-relmgr-office | 01:49 | |
| *** david-lyle has joined #openstack-relmgr-office | 02:16 | |
| *** markwash__ has quit IRC | 03:44 | |
| *** markwash__ has joined #openstack-relmgr-office | 03:48 | |
| *** markwash__ has quit IRC | 03:53 | |
| *** markwash__ has joined #openstack-relmgr-office | 03:59 | |
| *** markwash__ has quit IRC | 05:10 | |
| *** david-lyle has quit IRC | 05:55 | |
| *** morganfainberg has quit IRC | 05:58 | |
| *** morganfainberg has joined #openstack-relmgr-office | 06:13 | |
| *** flaper87|afk is now known as flaper87 | 07:19 | |
| *** mestery has quit IRC | 07:30 | |
| *** mestery has joined #openstack-relmgr-office | 07:30 | |
| *** morganfainberg has quit IRC | 07:35 | |
| *** morganfainberg has joined #openstack-relmgr-office | 07:39 | |
| *** zz_johnthetubagu is now known as johnthetubaguy | 08:27 | |
| *** markwash__ has joined #openstack-relmgr-office | 08:59 | |
| *** markwash__ has quit IRC | 09:03 | |
| dhellmann | ttx: I've started the releases. I'm going to have to make a bunch of requirements updates as we go, so dependencies are correct. That's going to make this take a while. | 11:06 |
|---|---|---|
| dhellmann | ttx: How do you feel about me self-approving the changes to global requirements, after tests pass? | 11:07 |
| sdague | dhellmann: can you bunch all the requirements changes into a single patch? | 11:12 |
| sdague | because the lands trigger patch submissions to all the projects | 11:13 |
| dhellmann | sdague: I need to update the dependencies of the libraries as I go | 11:13 |
| sdague | so there is giant secondary load created | 11:13 |
| dhellmann | I think I need 6 patches | 11:13 |
| sdague | dhellmann: there is no way for the tests to pass without that? | 11:13 |
| dhellmann | to cover the 10 libraries being released today | 11:13 |
| dhellmann | I need to update the requirements in oslo.messaging after I release oslo.config, for example | 11:14 |
| sdague | because that means we're talking about 6 x 10 x 15 nodes consumed to do this | 11:14 |
| sdague | so basically, the entire zuul capacity will be consumed | 11:14 |
| dhellmann | it's not all going to happen at once | 11:14 |
| sdague | in secondary back scatter | 11:14 |
| sdague | dhellmann: right, so the question is is there no way that oslo.messaging can pass once there is a new oslo.config | 11:16 |
| dhellmann | sdague: we could turn off the thing that automatically updates the other projects | 11:16 |
| dhellmann | sdague: don't we still have a job that checks the requirements changes in a project? | 11:16 |
| dhellmann | I don't think I can change a dependency to something not found in global-requirements, but maybe we changed that? | 11:17 |
| sdague | no, that's probably true | 11:17 |
| sdague | so I guess the root issue is there are a bunch of libraries which really have lock step dependencies with other libraries, but are versioned independently? | 11:18 |
| sdague | ok, I guess it's not doable to do this with less work | 11:19 |
| dhellmann | I don't think it's lockstep, but it is weird to release a final oslo.messaging that depends on an alpha of oslo.config | 11:19 |
| sdague | dhellmann: I guess, though realistically, that's the truth, right? | 11:19 |
| dhellmann | what's the truth? | 11:19 |
| sdague | that you could work with the alpha | 11:19 |
| dhellmann | yeah, I'm tagging the same commits | 11:20 |
| sdague | so, in that case, is there any reason to bump it prior to the release? because it is semantically the same | 11:21 |
| dhellmann | I would rather not leave the dependencies just because we think it's going to consume a bunch of tests triggered by a bot we could turn off | 11:22 |
| sdague | well, then coord with infra folks about turning it off before the proposes | 11:22 |
| sdague | this comes back to things like the trove client question the other day | 11:23 |
| dhellmann | I wasn't involved in that one | 11:23 |
| sdague | where they wanted to bump min version because they release a library version | 11:23 |
| sdague | but that's not why we bump g-r | 11:23 |
| sdague | we bump it because we can't work with the minimum listed | 11:23 |
| sdague | anyway... I'll let you get back to it, but I think we need a session about libraries at summit because the fact that this process requires 6 g-r updates 2 weeks in FF... I think exposes that there are issues with it, and we should figure out if there are better ways | 11:25 |
| dhellmann | I don't really see the problem, so maybe you should run that session. | 11:28 |
| sdague | hmmm... you aren't concerned that there are a lot of extra moving releases parts 1 week before people are trying to get RC releases out? | 11:30 |
| dhellmann | I'm one of those people. | 11:30 |
| dhellmann | I do not consider oslo "other" or separate or whatever from the rest of the release. | 11:30 |
| sdague | so... maybe when we were talking about 5 git trees to integrate | 11:31 |
| sdague | but we're talking about 20+ that are all trying to RC the same week that all impact each other | 11:31 |
| sdague | that's not a recipe for success | 11:31 |
| dhellmann | I agree that we need to address the scale problem. I don't have a good answer this morning. | 11:37 |
| sdague | yep, that's fair, that's why I said summit :) | 11:37 |
| sdague | anyway, the gate is pretty idle now, I'd plow forward with the first rounds regardless | 11:37 |
| sdague | the capacity is there | 11:37 |
| dhellmann | yeah, the first check job is passing so I'm going to approve the patch and get it into the merge queue before I go to the dentist | 11:38 |
| sdague | sounds good | 11:38 |
| sdague | where is it? | 11:38 |
| sdague | I can +A | 11:38 |
| dhellmann | https://review.openstack.org/#/c/122388/1,publish | 11:38 |
| dhellmann | oh, hmm | 11:38 |
| dhellmann | you know, I probably can't do all of these updates anyway | 11:39 |
| sdague | ? | 11:39 |
| dhellmann | if I'm going back to tag the last alpha, and we've had changes since then that we are not going to release, there's no way for me to get the update in without releasing the changes we don't want | 11:39 |
| dhellmann | we had 1-2 cases where things merged and we said it's too late for the release | 11:39 |
| dhellmann | so I should probably just leave the requirements as they are :-/ | 11:40 |
| sdague | ok, right, because we're actually pulling and testing against git in the normal case? | 11:40 |
| dhellmann | I can go ahead and tag everything and update the global list in one (more) patch | 11:40 |
| dhellmann | the integration test is, yeah, but these were new features not fixes | 11:40 |
| dhellmann | we had a late merge for something that was approved and then jenkins/zuul did something weird and didn't actually land the patch and we didn't realize it until late | 11:41 |
| sdague | right, I guess the danger is if projects start using those new features prior to their release | 11:41 |
| dhellmann | yeah, nothing should be using it, it would have required a new feature added to another project this week | 11:41 |
| sdague | you'd be surprised what's still landing ... :) | 11:41 |
| dhellmann | heh | 11:41 |
| dhellmann | well, there can't be unit tests for it, I know that much :-) | 11:42 |
| sdague | yep | 11:42 |
| *** dhellmann is now known as dhellmann_ | 11:52 | |
| *** dhellmann_ is now known as dhellmann | 13:19 | |
| *** david-lyle has joined #openstack-relmgr-office | 14:20 | |
| Kiall | hey ttx - ready when you are.. | 14:22 |
| *** david-lyle has quit IRC | 14:27 | |
| ttx | Kiall: hey | 14:32 |
| ttx | #startmeeting incub_sync | 14:32 |
| openstack | Meeting started Thu Sep 18 14:32:20 2014 UTC and is due to finish in 60 minutes. The chair is ttx. Information about MeetBot at http://wiki.debian.org/MeetBot. | 14:32 |
| openstack | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 14:32 |
| Kiall | Now.. The Q is, where do we put the time.sleep(1) .. | 14:32 |
| openstack | The meeting name has been set to 'incub_sync' | 14:32 |
| ttx | #topic Designate | 14:32 |
| Kiall | oops - wrong room | 14:32 |
| Kiall | ;) | 14:32 |
| ttx | https://launchpad.net/designate/+milestone/juno-rc1 | 14:32 |
| ttx | Still have one blueprint open -- how is zone-migration-between-tenants doing? | 14:33 |
| Kiall | Good/Bad - Graham hasn't completed the rebase. If we can't have that in by mid-week next week, I'm going to push it to kilo. | 14:33 |
| Kiall | The good is re - it's just in need of a rebase, but everyone has been busy with other tasks! | 14:34 |
| Kiall | You suggested last week we might cut a rc1 once they merge, a number of new bugs have been found and are in progress - they're severe enough to warrant holding out. | 14:34 |
| ttx | your call. If you were integrated, that would be too late, and too much of a distraction from bugfixing | 14:34 |
| *** david-lyle has joined #openstack-relmgr-office | 14:35 | |
| Kiall | ttx: for integrated - what's the hard deadline? | 14:35 |
| ttx | end of this week for the last tragglers. We deferred most of the others earlier this week | 14:35 |
| ttx | stragglers* | 14:35 |
| ttx | as for RC1, the idea is to cut something you think will be your final release, not just another milestone | 14:36 |
| ttx | so if you have a set of targeted bugs, better wait for them to be fixed | 14:36 |
| Kiall | Okay - I think we can stick to that (well, closer to that!) .. If it's not merged Monday, we'll push it. I'd like to get it in :) | 14:36 |
| ttx | rathet than cut a RC1 and open a RC2 window hours after | 14:36 |
| Kiall | ttx: agreed | 14:36 |
| ttx | Kiall: there is less risk involved with incubated projects, so feel free to push the deadline farther away | 14:37 |
| ttx | basically, if you don't have a viable RC1 by October 16, you wouldn't make my life miserable. Only yours :) | 14:37 |
| Kiall | Hah - Yea, that won't happen - I won't allow it ;) | 14:38 |
| ttx | Is the bug list on the https://launchpad.net/designate/+milestone/juno-rc1 representative of all your release blockers ? | 14:38 |
| ttx | i.e. those 4 bugs left are the only ones you really need to fix before release ? | 14:38 |
| Kiall | Yes, although I haven't triaged 1370621 Bing9 PTR Zone error : multiple RRs of singleton type yet | 14:38 |
| ttx | ok, you seem to have a good handle on things. Questions ? | 14:39 |
| Kiall | Nothing new this week :) | 14:40 |
| Kiall | (I'll have plenty once we open kilo, but they can all wait!) | 14:42 |
| ttx | Kiall: sorry, got caught up in something. Thanks! | 15:32 |
| Kiall | ttx: no problem :) | 15:32 |
| *** jraim has joined #openstack-relmgr-office | 15:46 | |
| ttx | jraim: o/ | 15:48 |
| ttx | #topic Barbican | 15:48 |
| jraim | ttx: here | 15:48 |
| jraim | redrobot: you there? | 15:48 |
| redrobot | o/ | 15:48 |
| ttx | I see two bluepritns still open | 15:49 |
| ttx | add-ssl-ca-support | 15:49 |
| ttx | api-orders-add-more-types | 15:49 |
| redrobot | api-orders-add-more-types is just waiting on one more CR | 15:49 |
| jraim | Correct. We landed one and the api-orders one is in final review | 15:49 |
| ttx | it would be good to complete or defer them at this point, to turn focus to bugfixing | 15:49 |
| redrobot | I think we should be landing it in the next couple of days | 15:49 |
| ttx | ok | 15:49 |
| redrobot | add-ssl-ca-support probably needs to slip to Kilo | 15:50 |
| redrobot | I have low confidence we can complete it by RC1 | 15:50 |
| ttx | your call really, since you bear the consequences of failing to meet the release date | 15:50 |
| ttx | RC1 is not really a date on the calendar | 15:50 |
| jraim | We're okay with slipping it if we have too...it seems like it won't make it at this point | 15:51 |
| ttx | the idea is that when you have something you think would make a good release, we cut a branch and pusg a rc1 tag | 15:51 |
| ttx | ideally it all happens befor ethe final release date so that you can release at the same time | 15:51 |
| ttx | (at release date we just take the latest release candidate and promote it | 15:52 |
| ttx | ) | 15:52 |
| jraim | makes sense | 15:52 |
| ttx | in my experience, it's good to get rid of all feature work before the end of this week -- so that you can turn undivided attention to bugfixing | 15:52 |
| jraim | So we'll land the one that's almost done, but it looks like we'll push the SSL one to K | 15:52 |
| ttx | sound slike a good call | 15:53 |
| jraim | We did want to ask about design sessions | 15:53 |
| jraim | Did we come to a resolution on how those are getting planned? | 15:53 |
| ttx | yeah | 15:53 |
| ttx | so you'll have a number of sessions on the Tuesday, something like half a day | 15:54 |
| ttx | that's sessions that appear on the schedule | 15:54 |
| ttx | you'll also have a designated pod (table) to continue the discussions throughout the rest of the week | 15:54 |
| jraim | Okay, so we need to plan the tues sessions, but the rest is ad hoc | 15:55 |
| ttx | right | 15:55 |
| ttx | so it's good to use the scheduled sessions for stuff you want to communicate out of the project, things you want operator feedback on, etc | 15:55 |
| ttx | and keep project-internal discussion at the pod | 15:55 |
| jraim | Right. That sounds good. We have an etherpad we've been using to get those rolling | 15:56 |
| jraim | Last question I think - when are we doing PTL nominations for Kilo? | 15:56 |
| ttx | jraim: you might want to reference your etherpad at https://wiki.openstack.org/wiki/Summit/Planning | 15:56 |
| ttx | starting tomorrow | 15:56 |
| jraim | Will do | 15:56 |
| ttx | (PTL nominatins start tomorrow) | 15:56 |
| jraim | Perfect | 15:56 |
| ttx | jraim: any other question? | 15:57 |
| jraim | nope, we're good | 15:57 |
| ttx | ok, talk to you next week then | 15:57 |
| ttx | flaper87: around? | 15:57 |
| flaper87 | ttx: yes, sir | 15:58 |
| ttx | https://launchpad.net/zaqar/+milestone/juno-rc1 looks good | 15:58 |
| ttx | no FFE in progress | 15:58 |
| flaper87 | yup, as for the 2 bugs. 1 fix is in the gate and the other one coming soon | 15:58 |
| ttx | feel free to target more bugs to rc1, as we will tag (and open kilo) once the list is empty | 15:58 |
| flaper87 | roger, I'm planing to go through the list of bugs tomorrow and see if there's anything that makes sense fixing in rc1 | 15:59 |
| ttx | FWIW I scheduled the next session with the TC the week after next -- I figured we could use a one-week break | 16:00 |
| ttx | and start a few constructive ML threads | 16:00 |
| flaper87 | right | 16:00 |
| ttx | the urgency of the decision being part of the problem we had, I think | 16:01 |
| flaper87 | I sent one yday but I havne't seen any reply besides eglynn's | 16:01 |
| ttx | flaper87: any question on the RC1 / release ? | 16:01 |
| flaper87 | ttx: nope | 16:01 |
| ttx | flaper87: I'm a bit late on -dev, been traveling lately | 16:01 |
| flaper87 | no worries, I just hope it to help with the discussion | 16:01 |
| ttx | ok great, talk to you later | 16:01 |
| ttx | devananda: you around ? | 16:01 |
| devananda | ttx: o/ | 16:04 |
| ttx | #topic Ironic | 16:05 |
| ttx | devananda: so as far as Juno is concerned, you're still incubated | 16:05 |
| devananda | ttx: yep | 16:05 |
| ttx | https://launchpad.net/ironic/+milestone/juno-rc1 | 16:05 |
| ttx | All FFEs done | 16:05 |
| ttx | and a small list of RC bugs | 16:05 |
| ttx | Feels like you could hit a RC1 sooner rather than later | 16:06 |
| ttx | or do you need to add more to the list ? | 16:06 |
| devananda | we were going to collectively assess the bug sitaution today and see what, if anything, is really a blocker | 16:06 |
| devananda | we're using targeting as a way to track what we want in right now, not necessarily release blockers | 16:07 |
| devananda | and as I do not see any open Critical bugs, I would agree that we probably could tag now | 16:07 |
| devananda | but if it's not an impediment, I'd rather tag next week to give a little time to close some of the High bugs | 16:08 |
| devananda | also, i'm goign to tag our final client release today | 16:08 |
| ttx | devananda: ok, at this point you can add a lot, and starting next week, we'll start to cut down if necessary | 16:08 |
| ttx | the "good enough" approach to release management | 16:08 |
| devananda | sounds good | 16:08 |
| ttx | devananda: ok, just let me know if you have any question | 16:09 |
| ttx | otherwise we'll close shop | 16:09 |
| devananda | when can we start votign on nova :) | 16:09 |
| devananda | though that's probably a question for the qa team, not you | 16:09 |
| devananda | ttx: no more questions related to the release | 16:10 |
| ttx | devananda: alrighty then | 16:10 |
| ttx | #endmeeting | 16:10 |
| openstack | Meeting ended Thu Sep 18 16:10:41 2014 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 16:10 |
| openstack | Minutes: http://eavesdrop.openstack.org/meetings/incub_sync/2014/incub_sync.2014-09-18-14.32.html | 16:10 |
| openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/incub_sync/2014/incub_sync.2014-09-18-14.32.txt | 16:10 |
| openstack | Log: http://eavesdrop.openstack.org/meetings/incub_sync/2014/incub_sync.2014-09-18-14.32.log.html | 16:10 |
| dhellmann | ttx: if you have a sec, I think we're ready to branch oslo-incubator | 16:11 |
| dhellmann | commit eabe46630d53bd430361c9d9253864e473c567e1 | 16:12 |
| ttx | ok, did we say proposed/juno or stable/juno ? I know we hesitated... | 16:12 |
| dhellmann | let's go right to stable/juno | 16:12 |
| ttx | ok.. HEAD ? | 16:12 |
| dhellmann | yeah, that SHA should be the same as head | 16:12 |
| ttx | hmm, do you want to tag 2014.2 on it ? | 16:13 |
| dhellmann | if that's what we usually do, sure | 16:13 |
| dhellmann | it does look like that's what we ususally | 16:13 |
| dhellmann | do | 16:13 |
| ttx | we usually place a tag, yes, even if that doesn't mean a lot for the incubator | 16:13 |
| ttx | stable usually starts from it | 16:13 |
| dhellmann | yeah, I see some rc1 tags in there, too, but I don't think we need that this time around | 16:13 |
| dhellmann | should I tag, or do you want to do that all together? | 16:14 |
| ttx | yes, let's skip rc1 | 16:14 |
| ttx | dhellmann: no, please tag | 16:14 |
| dhellmann | oh, I wasn't sure if that release script worked for the incubator or not so I haven't run that yet either | 16:14 |
| ttx | I'll cut the branch from the tagged commit | 16:14 |
| ttx | which should all be HEAD. | 16:14 |
| dhellmann | ttx: I can't push tags to oslo-incubator? :-| | 16:16 |
| ttx | hm. | 16:17 |
| ttx | we should probably fix that. In the mean time let me do it for you | 16:17 |
| dhellmann | ttx: https://review.openstack.org/122459 should fix the acls | 16:18 |
| ttx | hmm, I'll just use the regular script to do all this | 16:19 |
| dhellmann | ok | 16:19 |
| ttx | dhellmann: can all oslo-incubator FixCommitted bugs be FixReleased ? | 16:19 |
| dhellmann | yes, if they are committed in the incubator we consider that released | 16:20 |
| ttx | hhhrm. so it's a bit more complicated than I thought | 16:21 |
| dhellmann | ttx: how so? | 16:21 |
| ttx | the scripts implement it the way we did it last cycles | 16:22 |
| ttx | so we tag a RC1 and push all FixCommitted stuff to it | 16:22 |
| ttx | then we promote rc1 to final | 16:22 |
| ttx | and move all juno-* bugs and blueprints to the "final" page | 16:22 |
| ttx | (2014.2 milestone page) | 16:23 |
| dhellmann | can we do those things back-to-back? | 16:23 |
| ttx | basically mimicking the integrated release modus operandi | 16:23 |
| dhellmann | right | 16:23 |
| ttx | maybe we can just tag the final, create the branch | 16:23 |
| ttx | and fix LP later | 16:23 |
| ttx | I'll have to run soon | 16:24 |
| dhellmann | I don't mind doing lp by hand | 16:24 |
| dhellmann | yeah, me, too :-) | 16:24 |
| ttx | ok tagging | 16:24 |
| ttx | tag in | 16:25 |
| ttx | and stable branch created | 16:26 |
| ttx | dhellmann: we'll fix up LP tomorrow :) | 16:26 |
| dhellmann | ttx: sounds good | 16:26 |
| dhellmann | ttx: I'm going to unleash the team to delete graduated modules tomorrow at our meeting | 16:26 |
| dhellmann | I'll send email to the -dev list after we talk about it in the meeting, explaining that incubator fixes should go right to stable/juno if the module has been deleted from master | 16:27 |
| dhellmann | ttx: thanks for your help today, ttyl | 16:28 |
| ttx | np | 16:33 |
| *** markwash__ has joined #openstack-relmgr-office | 17:09 | |
| *** johnthetubaguy is now known as zz_johnthetubagu | 17:16 | |
| *** dhellmann is now known as dhellmann_ | 17:22 | |
| SergeyLukjanov | ttx, sahara needs bump of client version - https://review.openstack.org/#/c/121942/ | 18:08 |
| *** zaneb has quit IRC | 18:34 | |
| *** zaneb has joined #openstack-relmgr-office | 18:36 | |
| *** dhellmann_ is now known as dhellmann | 19:01 | |
| *** dhellmann is now known as dhellmann_ | 19:01 | |
| *** dhellmann_ has quit IRC | 19:02 | |
| *** dhellmann has joined #openstack-relmgr-office | 19:02 | |
| *** markwash__ has quit IRC | 19:16 | |
| *** markwash__ has joined #openstack-relmgr-office | 19:27 | |
| *** skhamkar has joined #openstack-relmgr-office | 20:03 | |
| *** skhamkar has quit IRC | 20:03 | |
| *** david-lyle has quit IRC | 20:10 | |
| *** david-lyle has joined #openstack-relmgr-office | 20:47 | |
| *** fesp has joined #openstack-relmgr-office | 21:26 | |
| *** flaper87 has quit IRC | 21:27 | |
| *** fesp is now known as flaper87 | 21:27 | |
| *** sdague has quit IRC | 21:28 | |
| *** sdague has joined #openstack-relmgr-office | 21:31 | |
| *** flaper87 is now known as flaper87|afk | 22:18 | |
| *** david-lyle has quit IRC | 23:55 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!