12:31:52 #startmeeting incub_sync 12:31:52 Meeting started Thu Sep 25 12:31:52 2014 UTC and is due to finish in 60 minutes. The chair is ttx. Information about MeetBot at http://wiki.debian.org/MeetBot. 12:31:53 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 12:31:56 The meeting name has been set to 'incub_sync' 12:32:00 #topic Designate 12:32:15 https://launchpad.net/designate/+milestone/juno-rc1 12:32:17 So - We talked last week about pushing that last BP out, it's not merged, so we pushed it to Kilo. 12:32:29 2 RC bugs left before we can cut RC1 12:32:45 Just let me know when you're done, and we'll tag and open kilo 12:33:03 then if some new release-critical thing is detected we'll open a RC2 window to fix it 12:33:04 Yes, both have reviews up, one has a mistake or two, the other needs review. 12:33:32 Since neither are "it's totally broke" bugs - should we cut an rc1 and expect to do an rc2, or better to wait? 12:33:35 You seem in good shape 12:33:48 no, we should wait 12:34:08 RCs need to be release candidates 12:34:21 i.e. at that point in time, you're fine with it being released as final 12:34:34 Okay - Hopefully we can get them both merged today. I'll takeover Endre's and get the other cores to re-review both today. 12:34:43 (Endre is on holidays now..) 12:34:54 you can discover new things... and respin the RC... but there is no point in doing a RC1 when you already have bugs piled up for a RC2 12:35:07 Sure, makes sense. 12:35:16 so better wait a few days to be sure, rather than open a RC2 window just after RC1 iscooked 12:35:34 The rationale is.. people only test RCs if they believe it may be the final 12:35:46 if a RC2 is queued, then they wait 12:35:58 Releasing is a psychology exercise 12:36:23 Hah - OK - Makes sense.. I'll ping you once both have landed then? 12:36:30 Kiall: yes -- other questions ? 12:36:42 Yep - But all can wait till we open kilo ;) 12:37:02 Actually.. 1 12:37:04 Looking at design summit space -- nothing final, but current estimate: 12:37:22 Oo - I didn't know you were involved in that! Then I have questions ;) 12:37:25 You would get 3 scheduled slots on the Tuesday, then a pod for the rest of the week 12:37:32 (dedicated pod) 12:37:57 Okay - That sounds reasonable, I understand space in Paris is somewhat more limited than usual. 12:38:39 The scheduled slots, at what point do we finalize the topics etc? 12:38:42 (incubated projects in ATL got about the same fwiw) 12:38:54 Kiall: at least one week before summit starts 12:39:04 so people can start organizing and we can check conflicts 12:39:19 It's the job of the Kilo PTL, but you might be that person 12:39:19 Yep - I was expecting it to be less than ATL, since space is tighter and there's more projects! 12:39:50 Yep - I've nominated myself, and asked the others to nominate themselves if they wish. 12:39:59 Kiall: alright then, talk to you later ? 12:40:08 Yep - Enjoy your day :) 13:21:28 ttx: FYI both patches have 2x +2's .. Waiting on the check queue to +1, then I expect they will be approved shortly after.. 13:52:47 SergeyLukjanov, zaneb: ping re: RC1 buglist when you get a chance 14:55:18 ttx: Any idea on when you plan to cut RC1? Trying to see which bugs have a chance at this point. 14:55:49 When you're ready and confident it could make a release candidate 14:56:01 The goal is to hit that before the end of the month ideally 14:56:12 to leave 2 weeks for further testing 14:56:19 and potential respins 14:56:55 mestery: so as we get closer to the end of the month, we become more picky on what is considered a release blocker 14:57:34 ttx: Ack, I've been pairing things down, I think we're down to 10 bugs in RC1 now, with maybe 3-4 of those in the queue and the rest being respun today 14:57:36 and then we only reopen a new RC window if a bug is serious enough (like: it's a regression) 14:57:36 So we're close 14:57:42 Got it 14:58:12 the goal being, to trick testers into thinking it might well be the final, so they better test it 15:08:08 ttx: have an RC1 countdown chart running? 15:08:24 dolphm: http://old-wiki.openstack.org/rc/ 15:09:28 ttx: did i miss something with glance? no PTL candidacies for kilo + no RC blockers? 15:10:33 dolphm: the team around it wasn't very active in Juno. So it's not very surprising they are not broken hopelessly at this point 15:11:03 we still hope to find candidates before the end of the day though :) 15:12:23 ttx: o/ 15:13:05 zaneb: hey! Feel like the curve of RC issues is not going in the right direction those last days for Heat: http://old-wiki.openstack.org/rc/ 15:13:26 it might be time to split between real release blcokers and nice-to-haves-that-could -be-fixed-post-release 15:13:47 yep, that is my task for today 15:14:17 that and triaging the 'new' bugs to make sure none of those are actually release blockers 15:15:19 right, ok 15:15:41 i want that curve oriented in a way that makes it look like it may hit the floor before the end of the month 15:15:57 hint: it never really does :) 15:16:06 lol 15:16:15 Although I guess Cinder's curve is a good approximation 15:17:05 most of the others have a week or two of "RC1 what?", then a surge as bugs get triaged... then a few steep drops when I ping people about keeping only blockers 15:17:33 then they brutally hit the floor one or two days after the symbolic deadline I set 15:23:15 david-lyle: got a minute? 15:23:25 yes 15:23:53 david-lyle: same remark, horizon RC bugs not going in the right direction over the last day(s) 15:24:02 are they all release blockers? 15:24:50 or could you further split blockers from nice-to-haves ? 15:24:56 several last minute previously untargeted Sahara issues cause the bump in our curve 15:25:02 oh, right 15:25:15 I guess you can hit RC1 slightly later than the others though 15:25:23 like before the end of next week 15:25:58 SergeyLukjanov: around? 15:26:05 yes, I hope so. A few of the in progress are +A and making there way through the gate now 15:26:12 *their 15:26:14 ttx, yup 15:26:26 ttx, for sahara https://launchpad.net/sahara/+milestone/juno-rc1 is up to date 15:26:38 SergeyLukjanov: same remark, a bit concerned that your RC bug curve is generally not oriented in the right direction 15:26:50 oh. 0 ? 15:26:59 ttx, yup 15:27:01 some would say that's cheating now 15:27:20 SergeyLukjanov: how about we let it bake for one day and cut tomorrow when you get up ? 15:27:33 I'll run a few checks 15:27:35 ttx, +1 15:27:53 ttx, we're testing it now too 15:28:19 ttx, all other open issues are all about docs 15:29:02 alright, good 15:30:27 david-lyle, thanks for accepting our last minute patches, and sorry about this... 15:31:19 SergeyLukjanov: no worries, if we have them on our radar sooner in the future, there should be less scrambling 15:31:31 but we'll get them in 15:31:42 SergeyLukjanov: I cut horizon some slack precisely due to that 15:31:57 they are often hit by last-minute things in upstream projects 15:32:10 so the are naturally "ready" a bit later 15:32:16 jraim: around? 15:32:28 ttx: yes 15:32:31 #topic Barbican 15:33:04 https://launchpad.net/barbican/+milestone/juno-rc1 15:33:32 still seem like a lot to do before you can do a RC1 15:34:05 Are those 2 features still in progress ? 15:34:11 redrobot is working on it. we only have a little bit left from what I understand 15:34:21 hi ttx and jraim! Yes, I just bumped one of the BPs to Kilo-1. The other one is just waiting on a single patch, so we should have that merged in the next day or two. 15:34:44 OK, then the goal is to come up with a list of release blockers 15:34:48 after that we'll be in bugfix mode. Not sure if we'll want an RC1 now, and then an RC2 with bugfixes. 15:34:58 and keep non-release blockers out of the list, so that we know when we can tag a RC1 15:35:15 no, ideally RCs need to truly be potential release candidates 15:35:18 so better wait 15:35:28 ok 15:35:38 no big deal if you miss the "end of month" target I set for intgrated release 15:36:36 so yes, get those features out of the picture so that you can concentrate on identifying release blockers and getting them fixed 15:36:59 then when you believe you have a winer, we tag it RC1, which should result in added scrutiny from external testers 15:37:08 and potentially regressions being found 15:37:15 which would then warrant a RC respin 15:37:23 jraim, redrobot: makes sense ? 15:37:59 ttx yes, got it. Once we're comfortable with RC1, should I just ping you directly? 15:37:59 (the perk being when we tag RC1 we also unfreeze master) 15:38:07 redrobot: yeah 15:38:14 ok, will do. 15:38:38 jraim, redrobot: questions ? 15:38:56 I'm good. How about you jraim? 15:39:35 sorry - drive by meeting. Getting caught up 15:39:49 this makes sense to me 15:40:49 ok then! talk to you guys later 15:40:57 thanks! 15:41:01 flaper87: arond? 15:41:07 around, even? 15:41:17 ttx: o/ 15:41:20 sorry 15:41:23 #topic Zaqar 15:41:31 https://launchpad.net/zaqar/+milestone/juno-rc1 15:41:37 4 RC bugs left 15:41:41 which all seem RC to me 15:41:44 zaqar is in a pretty bad shape now, there are some bugs that I'd like to get in but our gate is blocked 15:41:49 so nothing to add really 15:42:15 patches for those bugs are up for review 15:42:25 we're waiting for the patch that fixes our gate to land 15:42:28 to get this done 15:42:56 ok 15:43:14 ping me when you get closer to rc1 and I'll run the scripts for you 15:43:19 flaper87: questions? 15:43:42 ttx: nope, all clear 15:46:04 flaper87: alright, thx 15:46:10 devananda: ready when you are 16:31:23 #endmeeting