08:04:11 #startmeeting ptl_sync 08:04:12 Meeting started Tue Mar 31 08:04:11 2015 UTC and is due to finish in 60 minutes. The chair is ttx. Information about MeetBot at http://wiki.debian.org/MeetBot. 08:04:13 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 08:04:15 The meeting name has been set to 'ptl_sync' 08:04:18 #topic Heat 08:04:32 #link https://launchpad.net/heat/+milestone/kilo-rc1 08:04:32 hi ttx 08:04:43 asalkeld: o/ 08:05:17 so let's see where thoise two FFEs stand 08:05:37 dam that one bp was approved, but failed on merge 08:05:44 convergence-push-data was supposed to be completed before end of last week 08:05:44 https://review.openstack.org/#/c/161306/ 08:05:58 * ttx looks 08:05:59 yeah, needs rebase i think 08:06:23 ttx it's not super important - if you need to cut a release we can push it 08:06:57 needs rebase ? Or just recheck ? 08:07:09 ttx: i just noticed, i'll check up on it 08:07:20 ttx i just pushed the mistral resources out 08:07:24 If it's in pipe, it's fine 08:07:31 (the convergence stuff) 08:07:34 ok 08:07:50 so most of those bugs are close 08:07:51 if more work is needed (than just rechecks) then we would push it out 08:08:10 i am fine with that 08:08:27 ttx: when do you want to do the rc1? 08:08:27 so please have a look after the sync 08:08:36 (will do) 08:08:42 as soon as your critical bug list is empty 08:09:03 Could be at the end of the week, could be next week 08:09:11 ok 08:09:50 Looks like you parsed the bug list, so your RC1 list should be pretty complete 08:10:23 yeah not a big deal, most look ok 08:10:25 #info convergence-push-data is in the pipe, so FFE extended. If more work needed, will be deferred 08:10:37 i'll chase up the really important ones 08:10:46 #info 6 bugs left 08:11:54 Anything else on your side ? Red flags ? 08:12:11 no, all looks ok to me 08:12:18 We'll discuss PTL election season and the design summit scheduling at the cross-poject meeting 08:12:19 (note 5 bugs now) 08:12:23 in case you can join 08:12:33 ok 08:12:50 ttx: note i'll take a week off and get stevebaker to cover 08:12:57 i'll email you the exact time 08:13:03 asalkeld: basically, RC1 is cut when you're satisfied enough with the known state of the code that you would not be ashamed if we called it "the release" 08:13:20 ok ttx 08:13:32 i'd say even now it's fine 08:13:35 then you discover new embarassing bugs and you reboot the process for a RC2 :) 08:13:46 yeah 08:13:56 too early and you miss the late bug reports 08:14:07 good point 08:14:27 so you don't want to cut too early (that's a recipe for an early RC2 round :) 08:14:44 so maybe the end of the week? 08:14:54 but yeah, end of this week/start of next week is probably fine. Early is not a bad thing 08:15:22 cool, i am easy with those 2 options 08:15:36 ttx: there is project meeting this week? 08:15:40 asalkeld: starting with RC2 you fix in master and backport to the release branch, so it's more work fixing issues 08:16:02 asalkeld: yes, see above 08:16:08 cool 08:16:14 "We'll discuss PTL election season and the design summit scheduling at the cross-project meeting" 08:16:27 (i have to wake up at 5:45am to make it) 08:16:35 just making sure 08:17:09 you can catch up with the log 08:17:20 yeah, bad form to miss it 08:17:35 I have to stay awake at 11pm to run it :) 08:17:42 happy to be there, but not fun waking up and it's not on 08:18:05 sure 08:18:11 johnthetubaguy: around? 08:18:15 ok, that's all i have 08:18:21 thanks ttx 08:18:22 ttx: good morning 08:18:31 asalkeld: looks like you're in good shape. Just get me that last feature patch in 08:18:35 #topic Nova 08:18:56 https://launchpad.net/nova/+milestone/kilo-rc1 08:19:12 johnthetubaguy: all done with FFEs ? 08:19:28 yes, I pushed out the online migration stuff on friday monring 08:19:32 #info No more FFEs 08:20:12 How much final is the list of RC1 bugs ? 08:20:17 I haven't looked closely at the bug list, honestly, in the casting the net out wide part of the process (is my current excuse) 08:20:54 I guess a few of those can get bumped though 08:21:00 I know a few nova people are staying on top of bugs 08:21:12 and I could add the kilo-rc-potential tag for the big ones 08:21:38 yes, sdague is doing his excellent work there, so I don't want to stop that happening at this point 08:21:58 I am hoping on thursday in the nova meeting we can sync up and see how we are looking 08:22:03 johnthetubaguy: Did someone go through the "Critical/High" list to spot release-critical material yet ? 08:22:46 (no idea if the current list is the result of a triaging work or just a random list) 08:22:54 ttx: I am not sure anyone has done it systematically yet, mostly as the list is so small 08:23:08 I think its the result of random triage work, if that makes sense 08:23:30 OK, so this week we need to come up with a clearer list of RC1 blockers 08:23:34 so that we know how far we are 08:23:42 +1 08:24:10 Then make sure all targets have assignees 08:24:24 https://bugs.launchpad.net/nova/+bug/1431291 is unassigned right now 08:24:27 Launchpad bug 1431291 in OpenStack Compute (nova) "Scheduler Failures are no longer logged with enough detail for a site admin to do problem determination" [High,Confirmed] 08:24:50 Sylvain was workign on it but removed himself 08:24:55 I thought sdague was going to do that, I must reach out and see where its at 08:24:56 ah... 08:25:20 #info 11 bugs on list, but list is likely incomplete 08:25:59 johnthetubaguy: ok, any red flag I should know about ? 08:26:09 I think you spotted the big one 08:26:27 the rc1 bug list not really being known at this point 08:26:50 Once it's done, you just need to watch incoming new bugs, which makes the job easier 08:27:03 but the initial scrub is a lot of work :) 08:27:35 yeah, we have failed to keep a long term bug czar, and its showing a bit I guess 08:28:00 johnthetubaguy: ok, if you don't have anythig else, I'll be talking to you later 08:28:03 we went for a trivial patch list, which has had some success, but we need to deal with the harder list soon 08:28:10 ttx: thanks, thats all for now 08:28:17 cheers 08:28:19 have a good day 12:00:12 eglynn: o/ 12:00:30 ttx: hey 12:00:33 #topic Ceilometer 12:00:35 #link https://launchpad.net/ceilometer/+milestone/kilo-rc1 12:00:45 looking good so far 12:00:49 It is bug-free! 12:01:07 let's not jump on the tag yet though 12:01:17 eglynn: did you go through reported bugs and see if there would be anything else blocking ? 12:01:18 in case something arises in the distro level testing of kilo3 12:01:38 right, I wouldn't tag before EOW at the minimum 12:01:47 yep, nothing absolutely critical as yet 12:01:48 so that distro testing catches up 12:01:48 cool 12:01:58 yep, that sounds reasonable 12:02:15 But good to see you're in such a good shape 12:02:27 cool 12:02:52 #info No FFEs, buglist empty, candidate for early tagging (end of week / start of next) 12:03:22 eglynn: We'll talk elections and design summit scheduling at the Cross-project meeting today 12:03:28 That is all I have for you. 12:03:33 ttx: coolness 12:03:41 thanks for your time 12:03:44 no warning / red flag on your side ? 12:03:56 nope 12:03:59 alright then 12:04:05 SergeyLukjanov: ready when you are 12:10:16 ttx, hey, I'm here 12:10:22 #topic Sahara 12:10:31 #link https://launchpad.net/sahara/+milestone/kilo-rc1 12:10:33 #link http://launchpad.net/sahara/+milestone/kilo-rc1 12:10:55 so, we have 2/3 blueprints merged and the last one is stuck on the one base CR 12:10:57 We are left with default-templates 12:11:34 yeah, IMO if it'll be not merged till the end of week - defer to liberty 12:11:48 regarding issues - mostly all of them are docs 12:12:07 btw is it ok to merge docs / tests after rc1 or better to avoid it? 12:12:12 So it is stuck on https://review.openstack.org/#/c/163649/ ? 12:12:25 ttx, yup 12:12:39 #info default-templates blocked on https://review.openstack.org/#/c/163649/ 12:13:03 SergeyLukjanov: how about we defer it to liberty if it's still not approved by Thursday ? 12:13:17 ttx, ack 12:13:20 ttx, I'm ok with it 12:13:23 That gives one extra day to pass gate in case of problems 12:13:42 #info default-templates FFE extended to Thursday (for patch approval) and Friday (for patch merging) 12:14:10 On the bugs side... 12:14:31 btw is it ok to merge docs after rc1 or better to avoid it? 12:14:32 Is it the blockers list, or do you still have work to do to parse Launchpad bugs ? 12:14:40 better avoid it 12:14:48 RC1 is in theory releasable 12:15:04 I've cleaned up the lp bugs, only important stuff is in the list 12:15:05 So if docs are missing and are release-critical, we need to wait until they land to tag 12:15:12 ok 12:15:15 and kilo-rc-potential is only on three bugs 12:15:22 #info 7 bugs on the RC list 12:15:27 I'm now evaluating their real importance 12:15:37 ttx, ack 12:15:57 OK, any red flag, or issue I should be aware of ? 12:16:24 ttx, nope 12:16:37 We'll discuss PTL elections and design summit scheduling at the cross-project meeting tonight 12:16:42 ttx, it looks like everything is working ok atm 12:16:49 Though it's late for you, so you can read the logs after 12:16:49 ttx, okay, I'll be their 12:17:13 OK, talk to you later then 12:17:56 ttx, thank you! 12:18:04 There won't be an Oslo update today for those following at home 12:18:08 as Doug is travelling 12:18:25 next up, mestery at 13:45 utc 12:59:27 ttx: so, I just noticed the following patch - https://review.openstack.org/169289 - which means that there has been 0 full stack testing on the glance artifacts service up until this point 12:59:37 does that still have an FFE? 12:59:58 sdague: was about to kill ittoday in my sync with Nikhil 13:00:09 ok, good to know 13:01:22 sdague: whil you're around -- anything we need to do in requirements now that we are in HardDepFreeze ? Apart from ignorning requests ? 13:01:39 Time to get the requirements sync merged I guess 13:03:36 honestly, I think just making sure projects merge stuff 13:03:51 rigth, that was my conclusion too, just doublechecking 13:03:56 as far as I know we just need to reject except for bug fixes 13:05:22 https://review.openstack.org/#/c/168187 - o_O ? 13:05:37 you'd think folks would know better than pushing bumps now 13:06:18 ttx: https://review.openstack.org/#/c/169132 does need to land, someone typoed a thing because of a bad font 13:07:00 auth-token-use-client is probably liberty material 13:07:23 ok approving now 13:07:53 done 13:09:08 https://review.openstack.org/#/c/166664/ is also probably safe, it's a remove 13:09:11 of a test tool 13:12:44 agreed 13:45:24 mestery: around? 13:45:29 ttx: o/ 13:45:48 #topic Neutron 13:45:52 #link https://launchpad.net/neutron/+milestone/kilo-rc1 13:46:02 I see 2 FFEs still up 13:46:16 Yes 13:46:19 Let me explain :) 13:46:25 We said "subnet-allocation -- should land before next 1:1, reconsider if not" 13:46:45 ttx: A problem came up in that we need to implement subnet as an extension, so it's being reworked now 13:46:49 I expect that to land by Friday 13:46:55 "multiple-ipv6-prefixes -- related to ipv6-router, needs to fully land by next Tuesday" 13:46:57 at the latest 13:47:13 That one is likely to be marked as implemented, I need to sync with the folks working on it 13:47:27 They had an action to fill in the workpage on LP but didn't yet 13:47:27 So, subnet is the last one 13:48:12 I think we can safely land subnet pools by this week yet 13:48:17 OK, if that is the only one, I guess we can give it a few more days 13:48:44 Yes, that's it. 13:48:48 The team worked hard to get the rest in! 13:48:58 the earlier the better, we won't be able to stretch that past this week 13:48:59 Honestly, I'm super proud of all the work around testing and stability we've done. 13:49:03 Ack 13:49:28 #info subnet-allocation slightly delayed, needs to fully land before Friday 13:49:39 Looking at the bugs side now 13:49:53 #info 14 targeted RC1 bugs 13:50:03 Yes, today I will start removing the bottom of those 13:50:04 One of them is missing an assignee 13:50:19 Yes, the one without an assignee is because the person on it said they could no longer allocate time for it 13:50:20 Did you go through the neutron bug list to build this ? 13:50:27 I'm trying to ifind someone for it 13:50:30 or is there some bug scrubbing to do ? 13:50:41 I have to do some slight bug scrubbing, but for the most part this is it. 13:50:48 Some new things have been appearing as they always do :) 13:50:48 to make sure nothing flies below radar 13:51:21 ok, keep it updated then 13:51:30 Is the plan to try and tag early next week? 13:51:56 I'd say "next week" is good enough 13:52:01 Ack 13:52:25 early risers are expected end of this week, normal people newt week, late people the week after :) 13:52:32 lol 13:52:37 * mestery targets being normal 13:52:51 yes, norpmal is good 13:52:58 normal is good too 13:53:01 lol 13:53:12 what else... 13:53:27 we'll be discussing pTl election and design summit scheduling at the cross-project meeting today 13:53:45 Yes, I saw the email! 13:54:01 Next week at the Neutron meeting I plan to open up the design summit melee for folks :) 13:54:04 any red flag or problem potentially impacting release, that wI should be aware of ? 13:54:14 Nothing at this point 13:54:27 Alright, I'll be talking to you soon. Hurry to next meeting :) 13:54:32 lol :) 13:54:39 Thanks sir! HAve a great day and talk soon. 14:19:42 nikhil_k: ready when you are 14:20:20 ttx: hi 14:20:23 #topic Glance 14:20:31 #link https://launchpad.net/glance/+milestone/kilo-rc1 14:20:40 So... our 2 FFEs live 14:20:56 Last week we said we'd defer both if they failed to make it by now 14:21:26 So unless they are already in the pipe... not sure waiting more would actually help 14:21:48 ttx: hmm, double checking if CIS got some traction 14:21:52 Artifacts is stuck 14:22:13 I could grant 1-2 more days to CIS if it's close 14:22:31 ttx: seems like, this one is pending https://review.openstack.org/#/c/161621/ 14:22:36 artifacts looks further away, and also as I wrote, not sure of the benefit to land it in-release rather than early in liberty 14:22:36 and some documentation patches 14:22:57 ttx: Yeah, I mentioned the same thing to the developers working on it 14:23:14 and made them aware about your stance on the same 14:23:33 Looking into CIS 14:23:51 so, whatever you decide on Artifacts should work. I'm tentative too after reading the testing messge earlier 14:24:39 If CIS is only missing https://review.openstack.org/#/c/161621/, I'd grant it until Thursday to fully merge 14:24:53 Artifacts I'd axe at this point 14:24:56 ttx: hmm, there seem to be some outside of Glance. Please see this too https://review.openstack.org/#/q/topic:bp/catalog-index-service,n,z (if you haven't already) 14:25:47 So the issue with the client fix is that it will trigger a requirements update, and we are at HardDepFreeze 14:25:54 ttx: ok, sure about Artifacts. Can you please educate me some on the different announcements that need to be made in the public? Like specs to L, BP to L and ...? 14:26:18 You can already open specs to L 14:26:32 master branch will switch to L as soon as we tag RC1 14:27:10 ok, sounds good. I will send a note that the rest of the development can happen in L and mark the patches with -2 then 14:27:32 sdague: an opinion on https://review.openstack.org/#/c/146074/ wrt. new client release and the cortresponding requirements mess ? 14:27:42 and indicate this info to the necessary people (about RC1, specs etc) 14:27:50 Looks like that FFE would trigger a new client release 14:28:24 which we'd had to give a HDF exception to 14:29:22 and then which other projects would have to sync requirements for 14:29:31 eh 14:29:33 so that one impacts beyond Glance 14:29:37 ok 14:29:51 That's messy for sure 14:30:06 maybe better to consider it a L feature too, after all 14:30:20 ttx: ok, I think we can create a new spec for client and see it in L 14:30:29 Will send out the info 14:30:41 sounds good. Split the client part, land the server part 14:31:15 cool 14:31:25 #info catalog-index-service gets until Thursday to merge the last serverside bits. Client work should be split and done in Liberty 14:31:26 and the rest of the devstack ones? 14:31:47 That can land without requiring an exception. Tests are good 14:32:00 ttx: cool. one more https://review.openstack.org/157209 14:32:07 Thought that isn't really WIP 14:32:28 ttx: can we get that one in? 14:32:33 Test addition is not considered a feature 14:32:43 ok, sounds good 14:32:44 so yes, it can land until RC1 14:33:15 Thanks for letting me know, I would request removal of the bp tag from the commit message and ask for a bug instead 14:33:16 nikhil_k: Did you have some time to triage the bugs and find release blockers ? 14:33:28 ttx: yes, please let's do that 14:33:34 sorry couldn't get to it earlier 14:34:00 Would be good to work on it soon, so that we know how far we are from RC1 14:34:20 Maybe something you could delegate 14:34:47 #info only 10 bugs in RC buglist but that one is still incomplete 14:34:56 ttx: oh, I read that you wanted to do it now. Sorry context switched from the use of language my team does .. 14:35:00 I need to jump to my next customer 14:35:06 otherwise we could have ;) 14:35:19 ttx: sure :) 14:35:34 nikhil_k: Let's talk again on Thursday, hopefully FFE will be done and buglist built 14:35:52 yeah, catch ya then 14:35:55 Thanks! 14:35:56 cheers 14:36:00 thingee: around? 14:36:11 ttx: hi 14:36:15 #topic Cinder 14:36:22 Sorry for running late 14:36:29 #link https://launchpad.net/cinder/+milestone/kilo-rc1 14:36:30 no problem 14:36:50 So.. what date did you finally give for driver readdition ? 14:37:06 april 6th http://lists.openstack.org/pipermail/openstack-dev/2015-March/059990.html 14:37:36 Looks good. Correspnds to the end of the "normal" RC1s (the week after that is kept for stragglers) 14:38:17 #info Some drivers may be reintroduced, deadline Apr 6 14:38:23 Looking at the bug list 14:38:31 #info 7 targeted RC bugs 14:38:34 I've been happy with the progress drivers are making for being readded. 14:38:48 Did you triage most bugs to find those 7 ? 14:39:07 ttx: I think - https://review.openstack.org/#/c/146074/ is too late 14:39:20 I am actively triaging yes. 14:40:05 sdague: right, nikhil_k and I decided to defer it 14:40:19 thingee: ok, good, looks like you have stuff under control 14:40:32 Any red flag, besides the late driver readdition ? 14:42:20 nope. we have some nice bad bugs but patches are posted. 14:42:27 glad they're being found now. 14:42:30 ok, great! ttyl 14:42:37 david-lyle: around? 14:42:41 thanks 14:51:10 david-lyle might have missed the European DST memo 14:52:43 ttx: o/ 14:52:49 #topic Horizon 14:52:58 #link https://launchpad.net/horizon/+milestone/kilo-rc1 14:53:11 I see two FFEs still in there, unfortunately assigned to the same person 14:53:19 Last week we said: 14:53:44 federated-identity -- late feature catch-up, ideally would merge before Tuesday next week 14:53:51 angularize-identity-tables -- needs to be finalized before next week 14:54:02 How far are those from landing ? 14:54:21 angularize-identity-tables seems to have plenty of patches left 14:54:36 identity tables is farther 14:54:55 Federated identity seems to hinge on https://review.openstack.org/#/c/151842/ 14:55:31 in which case we could give it a couple more days to go through the CI pipe 14:55:54 I think Federated Identity can land 14:55:55 How about we defer angularize-identity-tables and give federated-identity until Friday ? 14:56:02 does that make sense ? 14:56:20 Yes, that seems appropriate given the timeframe 14:56:33 OK, I'll let you defer the table stuff 14:56:37 will do 14:56:53 #info federated-identity is really close, FFE extended until Friday 14:57:13 #info 11 targeted bugs on the RC buglist 14:57:29 Is tahht all the known blockers, or the triage work is still a todo ? 14:57:41 that* 14:58:03 some of those will clean out with removing the tables bp 14:58:29 that's all I am aware of 14:58:52 ok, good 14:58:58 Looks like we should be on time here too 14:59:11 Yes, I think so 14:59:12 Any red flag I should know of ? 14:59:25 other than the gate fiasco, no 14:59:31 heh, that happens 14:59:38 alright, have a good week then 14:59:44 thanks, you too 15:00:11 ttx: did want to ask about horizon decomposition 15:00:30 but you may not have enough time right now 15:00:47 Hmm, could you ping me tomorrow ? About to jump on a call right now 15:00:58 yes, sure 15:01:01 and today is meeting-crazy day 15:01:06 ok, great 15:01:14 ack 15:01:18 will ping tomorrow 15:09:44 ttx: o/ 15:14:47 devananda: moved our sync one hour later 15:14:54 devananda: on a call right now 15:15:00 oh. darn time changes. I thought it was now ... thanks 15:21:37 ttx: heya - can you settle a Q for me? At what point do we branch stable/kilo and move master -> icehouse? I seem to remember some changes recently on this.. 15:37:59 Kiall: when we tag RC1 we cut the kilo release branch from it and let master evolve in liberty 15:38:25 devananda: I'm of the call now, we can sync now if you want 15:38:29 off* 15:45:29 ttx: perfect, for some reason I thought it had changed to cut stable/kilo after kilo final was released 15:46:42 Kiall: it's also true 15:47:02 We cut proposed/kilo for release candidates, and switch to stable/kilo once it's stable/released 15:47:32 hah - i thought proposed/foo was dropped? 15:47:32 * Kiall needs to re-read the branching wiki page 15:48:01 Kiall: no.. "milestone-proposed" was dropped. 15:48:11 proposed/foo is still alive 15:48:39 https://wiki.openstack.org/wiki/ReleaseTeam/How_To_Release#proposed.2F.24SERIES_branch_cut_.28switch_master_to_next_version.29 15:48:52 morganfainberg: ready when you are 15:49:14 * morganfainberg wakes up. 15:49:18 Hi :) 15:49:30 * ttx violently shakes morganfainberg 15:49:45 Lol. 15:49:46 #topic Keystone 15:49:52 * ttx dcc sends coffee 15:49:58 Yessssssss 15:50:11 #link https://launchpad.net/keystone/+milestone/kilo-rc1 15:50:18 Coffee will happen right after this sync up 15:50:19 I see one FFE still standing there 15:50:37 Yes. There is one outstanding review and a new bug / issue to address. 15:50:46 interestingly we didn't mention that one last week 15:50:50 That will be discussed today w/ henrynash 15:51:00 We did, domain-sql? 15:51:34 It was on the list of cleanup/final patches for a feature to land. 15:51:36 I failed to #info it correctly 15:51:40 Ahh 15:51:43