Friday, 2015-12-04

*** sileht has quit IRC00:15
*** sigmavirus24_awa is now known as sigmavirus2400:26
*** sileht has joined #openstack-release00:27
*** devananda has quit IRC00:28
*** amotoki has joined #openstack-release00:28
*** amotoki has quit IRC00:44
*** sigmavirus24 is now known as sigmavirus24_awa00:53
*** devananda has joined #openstack-release00:54
*** doug-fish has quit IRC01:08
dims_lifeless : would you have some time to look at the change?
lifelessdims_: not today sorry, its gone crazy on me and that change needs my full attention.I'll make time in the weekend / monday01:16
dims_no worries. thanks lifeless01:17
lifelessdims_: (as you know, its gnarly)01:17
dims_i hear ya.01:17
*** mriedem has joined #openstack-release01:29
openstackgerritFei Long Wang proposed openstack/releases: Release Zaqar 2.0.0b1
*** kzaitsev_mb has quit IRC02:12
*** amotoki has joined #openstack-release02:13
*** kzaitsev_mb has joined #openstack-release02:13
*** kzaitsev_mb has quit IRC02:18
*** kzaitsev_mb has joined #openstack-release02:34
*** kzaitsev_mb has quit IRC03:13
*** stevemar has quit IRC04:09
*** stevemar_znc has joined #openstack-release04:12
openstackgerritAngus Lees proposed openstack/releases: Release oslo.privsep 0.1.0
*** mriedem has quit IRC04:27
openstackgerritAngus Lees proposed openstack/releases: Force basepython=python2.7
*** amotoki_ has joined #openstack-release05:09
*** amotoki_ has quit IRC05:12
*** amotoki has quit IRC05:13
*** amotoki has joined #openstack-release05:17
*** armax has quit IRC05:48
skraynevttx: was done. thx for leaving comment to review ;)06:24
*** amotoki has joined #openstack-release06:30
* ttx yawns08:02
*** dims_ has quit IRC09:05
*** dims has joined #openstack-release09:11
*** dims has quit IRC09:23
-openstackstatus- NOTICE: Tox tests are broken at the moment. From openstack-infra we are working to fix them. Please don't approve changes until we notify that tox tests work again.09:31
*** dims has joined #openstack-release09:34
*** dtantsur|afk is now known as dtantsur09:35
*** kzaitsev_mb has joined #openstack-release10:00
*** bauzas is now known as bauwser10:05
*** kzaitsev_mb has quit IRC10:14
*** openstackgerrit has quit IRC10:32
*** openstackgerrit has joined #openstack-release10:32
ttx#success The new mitaka release schedule page is up!10:49
openstackstatusttx: Added success to Success page10:49
openstackgerritMerged openstack/releases: Release oslo.privsep 0.1.0
*** kzaitsev_mb has joined #openstack-release11:52
johnthetubaguyttx: great stuff with that page, I will try add a review for the nova items, etc11:59
*** dtantsur is now known as dtantsur|brb12:09
*** dtantsur|brb is now known as dtantsur12:46
*** kzaitsev_mb has quit IRC13:03
dtantsurhi folks! does the new procedure mean that the release team no longer manages (e.g. closes) milestone launchpad pages?13:25
dtantsurI remember when I used the release scripts myself, they closed the milestone and uploaded the tarball to launchpad13:26
*** david-lyle has quit IRC13:43
*** kzaitsev_mb has joined #openstack-release13:46
dhellmannSergeyLukjanov : thanks for confirming13:47
dhellmanndtantsur : yes, that's right. We're moving away from using launchpad for tracking completed work. That's why reno integration was a requirement before tagging this milestone.13:48
dhellmannttx: good morning13:48
dhellmannttx: nice work on that release schedule page, it looks good13:56
ttxyeah, the challenge was to make it not significantly more ugly than the wiki page, and then make it more rich13:57
dhellmannyeah, rst is good for docs but less good for ensuring legibility of complex data like that13:57
ttxdhellmann: I added a few points to the meeting agenda13:58
*** mriedem has joined #openstack-release13:59
dhellmannok, I saw a couple of them13:59
dtantsurdhellmann, thanks! does it mean we can stop creating milestones in LP completely, if we feel like?14:00
*** gordc has joined #openstack-release14:09
openstackgerritMerged openstack/releases: Release Zaqar 2.0.0b1
*** dansmith is now known as superdan14:53
ttxdhellmann: see f4c4dd2 in the post column at
ttxI'll delay the announcement until that is completed, since we link to the page and it's currently missing the zaqar line15:03
*** kzaitsev_mb has quit IRC15:05
*** kzaitsev_mb has joined #openstack-release15:11
ttxok, ready now15:17
ttxand sent15:18
*** sigmavirus24_awa is now known as sigmavirus2415:24
dhellmanndtantsur : yes, though you may want to consider using them for planning. it's up to you, though.15:27
dhellmannttx: thanks!15:27
dtantsurunderstood, thanks15:28
*** mriedem has quit IRC15:30
ttx#success mitaka-1 complete!15:31
openstackstatusttx: Added success to Success page15:31
dimsttx : "clouseau" - hahaha +115:33
*** david-lyle has joined #openstack-release15:42
dhellmannttx: I forgot to mention in the meeting that I'll be out for the next 2 fridays. Do you want to chair those meetings with dims and jokke_?15:42
ttxdhellmann: sure!15:43
ttxthough obviously nothing will get done while you're away.15:43
dimsttx: fyi, i am in Russia/Ukraine TZ the next week and half. So i may miss one meeting.15:43
mordredso, if I wanted to get on the reno bandwagon for occ, do I need to run any of that by y'all? or just start doing it?15:45
mordredalso - I have a feature request that I'd be more than happy to do the coding on if it would be an acceptable thing to y'all15:46
mordredwhich is that I think I may also want to use reno in shade - but I'd kind of just like to integrate the output of it with ... wait. nevermind15:47
mordredI think I just realized how to do what I wanted while saying things outloud15:47
dhellmannmordred : you can just add reno to your project, no need to coordinate, though I'd be happy to review patches if you like15:49
dhellmannand if you have a feature request, let me know -- reno is far from "done" so we may already be thinking about the same thing15:50
mordreddhellmann: well, I was about to say "I don't want to have a second sphinx location for shade" - but I do not believe I need one now that I read the docs :)15:51
dhellmannmordred : we split them out so the publish URL could be /releasenotes instead of /developer but that's up to you, yes15:52
dhellmannand for other libs we split them so the deployer-focused release notes were not in the developer docs15:52
mordredyah - makes sense for openstack for sure15:53
mordredand I think following the pattern for occ makes total sense15:53
dhellmannyeah, I'm just explaining the logic so you understand how we decided -- what you do is up to you15:53
dhellmannttx: do you have a minute to look over my email to unmanaged projects about recording their milestones?
ttxdhellmann: sure16:00
dhellmannI left out reno this time, and just focused on the version change16:00
ttxlet's do that before executing that monitor-frying xrandr script16:00
* dhellmann googles xrandr16:00
ttxlooks good16:01
dhellmannoh, I sometimes get nostalgic for X and then something like this comes up and I remember why I'm glad I don't use it any more16:01
ttxdhellmann: you can happily ignore it if you don't need it16:01
ttxemail looks good, just changed it so that people don't assume release:unmanaged is a thing16:02
ttxand now boom16:02
dhellmannyeah, I realized as I was pasting it into my mailer that I would need to change that16:03
*** armax has joined #openstack-release16:18
openstackgerritDoug Hellmann proposed openstack/releases: add instructions for setting URL to release notes
*** david-lyle has quit IRC16:48
*** david-lyle has joined #openstack-release16:49
mordreddhellmann: any way to get the release notes for things that are not released yet to be labeled "unreleased changes" instead of "1.3.0-1"? or is there a different approach I should be using?16:50
dhellmannmordred: we would have to change reno to rename those versions, but we can do that16:54
-openstackstatus- NOTICE: The earlier JJB bug which disrupted tox-based job configurations has been reverted and applied; jobs seem to be running successfully for the past two hours.16:54
*** dtantsur is now known as dtantsur|afk16:54
openstackgerritDoug Hellmann proposed openstack-infra/release-tools: elaborate on how to check out releases repo to run releases by hand
openstackgerritDoug Hellmann proposed openstack-infra/release-tools: elaborate on requirements handling in the scripts
openstackgerritDoug Hellmann proposed openstack-infra/release-tools: add instructions for using
openstackgerritDoug Hellmann proposed openstack-infra/release-tools: remove
openstackgerritDoug Hellmann proposed openstack-infra/release-tools: cosmetic changes to command examples
mordreddhellmann: I can probably make a patch - do you think that shoudl be configurable behavior?17:03
dhellmannmordred : no, I don't think we need an option for that17:04
mordreddhellmann: there is a comment that says it looks at pbr or git describe - but the code only looks at git describe - should I also make a patch to mkae the code match the comment? or should the comment go away?17:05
openstackgerritLingxian Kong proposed openstack/releases: Release mistral-dashboard 2.0.0 beta 1
dhellmannmordred : the comment is wrong17:06
dhellmannif you would clean that up, too, I'd appreciate it17:06
mordreddhellmann: kk17:06
mordredsure nuff!17:07
*** kfarr has joined #openstack-release17:09
*** david-lyle has quit IRC17:13
mordreddhellmann: btw- your version walking code is better than what we use in pbr to generate changelogs17:18
mordreddhellmann: I think we should steal it17:18
mordreddhellmann: becuase pbr changelogs erroniously put entries with tags based on merge commit - but commits that were made before one tag then merged later will show up in the wrong spot17:19
dhellmannmordred : it's broken :-/
openstackLaunchpad bug 1522153 in reno "Merge commits confuse the version detection" [Undecided,New]17:21
dhellmannso I think both sets of logic are broken in different ways17:21
mordredit's a lovely tree walking problem17:23
dhellmannI was planning to look at that next week, if I have time17:23
dhellmannyeah, it looks like there are some options we can give to git log to change the order, so I was going to experiment with them. I need to come up with a test case demonstrating the brokenness first17:24
dhellmannand if that fails, mtreinish's suggestion of brute-force querying by tag separately may be the best alternative17:25
*** kzaitsev_ws has quit IRC17:28
*** kzaitsev_ws has joined #openstack-release17:29
*** stevemar_znc is now known as stevemar17:47
*** kfarr has quit IRC18:00
mtreinishdhellmann: heh, it's ugly, but it would be pretty simple I think :)18:11
*** kzaitsev_mb has quit IRC18:21
dhellmannmtreinish : true18:53
mtreinishdhellmann: I tried to doodle on it a bit the other day, and the only hangup I found was projects that have non-numeric tags or change versioning18:55
mtreinishbecause tags don't have dates associated with them just a sha118:55
dhellmannmtreinish : yeah, you almost have to pull the tags in order first, then go back and pull the notes between the tags18:55
dhellmannmtreinish : that's why I was looking at some of the options like --topo-order18:57
mtreinishdhellmann: oh, that looks like it should do the right thing18:58
mtreinishI figured there would be a git option to do it, just didn't know what it was18:58
dhellmannyeah, I need to put together a good test case to demonstrate how it's failing now, and then I can work out how to fix it18:59
dhellmannI'm worried it might introduce a different ordering bug, showing commits too early instead of too late or not at all18:59
mtreinishwell the bug I hit was the commit showing up too early, but yeah I can understand that concern19:00
dhellmannso was it showing up with a version number too low, or too high?19:01
dhellmannI was using early in terms of the git log output stream, which would give something a higher version number than it should have19:02
mtreinishdhellmann: it's showing up on 1.0.2 but it was part of 1.1.019:04
dhellmannok, that's what I thought19:05
*** david-lyle has joined #openstack-release19:06
*** david-ly_ has joined #openstack-release19:06
fungidhellmann: i'm working on the juno-eol repos list now before i begin pulling triggers... is the list of repos which belong to official deliverables tagged release:managed and release:has-stable-branches which have branches named stable/juno19:08
fungidhellmann: can i assume those 40 are in the safe set?19:08
mtreinishfungi: can you also do devstack and grenade while you're in a tagging mood :)19:08
fungidhellmann: additionally are the ones with release:has-stable-branches but not release:managed and i know there are at least some we'll add19:08
fungimtreinish: ^ you read my mind!19:09
mtreinishfungi: heh19:09
fungidevstack, grenade, oslo-incubator and requirements at a minimum19:09
dhellmannfungi : that description matches my expectations, but it'd be good to confirm with ttx since he handled the juno release and he and I haven't discussed this19:09
fungittx: ^ when you're bored enjoying your friday evening and get a hankering to do more exciting openstack things19:10
mtreinishfungi: oh, this reminds me I never pushed the tempest tag to indicate the end of juno support (which we stopped as soon as the last tag was pushed)19:10
* mtreinish goes to do that19:10
fungigood deal19:11
dhellmannfungi : you may want to talk to the astara team before deleting branches in their repo, since they're new and may have different expectations -- just to avoid a surprise19:12
fungidhellmann: right, that second list is stuff i wouldn't delete without explicit confirmation since they're not release:managed19:12
dhellmannright, ok19:12
fungioh, for completeness i'll also make an outlier list of repos with stable/juno branches but no release:has-stable-branches tag in governance (there will be a lot, official and unofficial, but makes a good crosscheck of our tags)19:13
dhellmannare you querying gerrit for that list directly or are you running something against all of the repos locally?19:13
fungilots of local experimentation19:13
fungia combination of git commands over a complete clone of all our hosted repos, and manipulating a pyyaml load of the governance projects.yaml file19:14
fungiand the some combinatorial comparisons in shell loops19:14
fungiin other words, terrible ugliness19:14
fungiunfortunately, like with our election process, the eol process has to be revisited every cycle because of changes to how we model and track things in governance19:15
* dhellmann nods and hands fungi a beer19:15
fungiso attempting to automate it would probably not be a worthwhile use of my time19:16
fungithanks, beer indeed19:16
fungidhellmann: 33 official repos have a stable/juno branch but lack the release:has-stable-branches governance tag
fungii don't see anything there which jumps out at me as missed19:28
fungimostly puppet/chef19:28
dhellmannfungi : ttx and I have planned to audit the repos with branches but not the tag as part of spinning up the new stable team.19:31
fungiand finally, 17 unofficial repos with stable/juno branches which we shouldn't need to care about at all
dhellmannfungi : the kolla and mistral-extra repos are product repos, I think19:32
fungimakes sense19:32
fungianyway, the current plan is the 40 release:managed release:has-stable-branch repos plus the other four known qa/oslo/cp repos i mentioned earlier19:33
fungimtreinish already okayed devstack and grenade19:34
fungiand i'll pretend that as infra ptl i can justify the requirements repo myself19:35
fungidims: any opposition to me getting rid og the stable/juno branch on oslo-incubator when i do the others even though it's not release:managed?19:35
fungier, getting rid of19:36
fungithere won't be any repos left taking syncs of that branch from incubator anyway, so presumably it's safe19:36
*** david-lyle has quit IRC19:39
*** david-ly_ is now known as david-lyle19:39
dimsfungi: +1 from me19:41
dhellmannfungi : that plan sounds reasonable19:43
*** gordc has quit IRC20:13
*** kzaitsev_mb has joined #openstack-release20:45
*** cp16net has quit IRC21:14
*** cp16net has joined #openstack-release21:16
openstackgerritDevananda van der Veen proposed openstack/releases: Release Ironic 4.3.0
*** kzaitsev_mb has quit IRC21:18
openstackgerritDevananda van der Veen proposed openstack/releases: Release Ironic 4.2.2 (stable/liberty)
devanandadhellmann: ^ my first time proposing releases since the switch, since jroll is out this week.21:24
dhellmanndevananda : I'll take a look now, but probably won't process them until monday21:25
devanandaI'm unclear on how reno is going to work for 4.3.0 ... seems like there needs to be something esle21:25
devanandadhellmann: no worries. the security fixes are out there21:25
dhellmanndevananda : oh, right, forgot this was a security thing. if you need me to release them today I can21:25
dhellmannfor reno, it'll show up on Since you're cycle-with-intermediary you might want to rename that page master.html or something similar21:27
devanandadhellmann: 4.2.2 might be more important of a release, as the vuln affects stable/liberty21:27
dhellmannok, sure. for that one the notes will show up on
devanandaI think you mean :)21:28
dhellmannoops, yes21:28
devanandaand yes, it shows up there. except without any other liberty notes21:28
dhellmannthat explains why those version numbers were so different :-)21:28
dhellmannyou could back-port your liberty notes from the wiki if you want21:29
devanandahm, right. I htink someone was working on that21:30
*** kzaitsev_mb has joined #openstack-release21:30
dhellmannI've been thinking that in addition to the smart detection, reno might need to let us specify the version for a given note file21:31
dhellmannthat way we can add notes about old releases like this21:31
dhellmannI'll process the 4.2.2 release right now21:32
devanandaright now, we'd have to convert them into a reno-like format21:32
devanandathanks much. I think we can land the release notes change to stable/liberty without needing a new release, right?21:32
dhellmannyou could also just paste them into that file as restructured text, under a new version heading21:32
dhellmannyou can, yes21:32
devananda ?21:33
dhellmannin fact, pasting them in as restructured text is probably safer21:33
dhellmannyeah, there you go :-)21:33
* devananda pushes that one in21:34
dhellmannfungi, ttx: this conversation with devananda reminded me that one of the things we will need to remember to change about the stable branch EOL process is to collect all of the release notes for the branch in a static file that we check in once (in master) to replace the dynamically generated notes21:35
*** kzaitsev_mb has quit IRC21:36
openstackgerritMerged openstack/releases: Release Ironic 4.2.2 (stable/liberty)
openstackgerritJohn Dickinson proposed openstack/releases: python-swiftclient 2.6.1 release
dhellmanndevananda : the release is tagged. In the past I've let the PTL/liaison handle the announcements for server projects because the current announcement script is geared towards libraries not servers (deployers probably don't need the full git changelog)21:39
dhellmanndevananda : I'm also running the batch update script against all of your current fix-committed bugs to mark them fix-released21:40
dhellmannit looks like there are quite a few of them21:40
devanandadhellmann: yes, tyvm21:40
dhellmanndevananda : ok, all done21:41
devanandaI'll leave the 4.3.0 announcement to jroll on Monday, too21:41
dhellmanndevananda : yeah, I only processed 4.2.2 so far. I'll do 4.3.0 now21:42
dhellmanndevananda : for 4.3.0, do you  not want to call that 5.0.0 since it's the first mitaka cycle release?21:43
dhellmannyou don't have to, but that's what most of the projects have started doing21:43
devanandaI think jroll wants to follow a more semver-like numbering scheme21:46
devanandanot date or cycle based21:47
dhellmannok. I don't think we have a firm policy on that, so I'll leave it open until next week and we can discuss it21:47
devanandaworks for me21:47
dhellmannI'm on the fence about it myself21:48
devanandadhellmann: in theory, it could mean we have a 4.4 in the spring, and a 4.5 next fall21:48
dhellmannsure, and that may be fine21:48
*** sigmavirus24 is now known as sigmavirus24_awa21:49
*** sigmavirus24_awa is now known as sigmavirus2421:49
*** kzaitsev_mb has joined #openstack-release22:09
*** kzaitsev_mb has quit IRC22:19
openstackgerritKyle Mestery proposed openstack/releases: Neutron 7.0.1
*** david-ly_ has joined #openstack-release22:29
*** bauwser is now known as bauzas22:30
*** david-ly_ is now known as david-lyle_22:31
*** david-lyle has quit IRC22:32
*** kzaitsev_mb has joined #openstack-release22:36
devanandadhellmann: will artifacts like will be automatically created now?22:57
*** dims has quit IRC23:10
dhellmanndevananda : no, we're not using launchpad for tracking releases any more. That information is accessible via now23:11
*** kzaitsev_mb has quit IRC23:32
*** openstack has joined #openstack-release23:36
*** sigmavirus24 is now known as sigmavirus24_awa23:36
*** dims has joined #openstack-release23:47
devanandadhellmann: ok, still shows 4.2.1-7 -- what updates that to 4.2.2 ?23:54

Generated by 2.14.0 by Marius Gedminas - find it at!