| *** Qiming has quit IRC | 00:02 | |
| *** shwetaap has joined #openstack-stable | 00:32 | |
| *** Qiming has joined #openstack-stable | 01:07 | |
| *** dims has quit IRC | 01:26 | |
| *** shwetaap has quit IRC | 01:43 | |
| *** nikhil has joined #openstack-stable | 02:25 | |
| *** niedbalski has quit IRC | 04:18 | |
| *** niedbalski has joined #openstack-stable | 04:20 | |
| *** shwetaap has joined #openstack-stable | 04:30 | |
| *** rcernin has joined #openstack-stable | 05:47 | |
| *** rcernin has quit IRC | 05:57 | |
| -openstackstatus- NOTICE: gerrit is going to be restarted due to bad performance | 07:24 | |
| *** ChanServ changes topic to "gerrit is going to be restarted due to bad performance" | 07:24 | |
| *** armax has joined #openstack-stable | 07:25 | |
| *** ChanServ changes topic to "reviews: http://tiny.cc/os-stable-dash | issues: https://etherpad.openstack.org/p/stable-tracker" | 07:27 | |
| *** shwetaap has quit IRC | 07:39 | |
| *** shwetaap has joined #openstack-stable | 07:39 | |
| *** shwetaap has quit IRC | 07:44 | |
| *** unlaudable has joined #openstack-stable | 07:46 | |
| *** pcaruana has joined #openstack-stable | 08:25 | |
| *** shwetaap has joined #openstack-stable | 08:40 | |
| *** shwetaap has quit IRC | 08:44 | |
| *** shwetaap has joined #openstack-stable | 09:40 | |
| *** shwetaap has quit IRC | 09:44 | |
| *** derekh has joined #openstack-stable | 09:44 | |
| *** apevec has joined #openstack-stable | 09:46 | |
| *** apevec has quit IRC | 09:46 | |
| *** apevec has joined #openstack-stable | 09:46 | |
| *** dtantsur|afk is now known as dtantsur | 09:50 | |
| *** apevec has left #openstack-stable | 09:58 | |
| *** ihrachys has joined #openstack-stable | 10:00 | |
| *** e0ne has joined #openstack-stable | 10:15 | |
| *** Qiming has quit IRC | 10:16 | |
| *** shwetaap has joined #openstack-stable | 10:40 | |
| *** dims_ has joined #openstack-stable | 10:41 | |
| *** shwetaap has quit IRC | 10:45 | |
| *** Qiming has joined #openstack-stable | 11:05 | |
| *** e0ne has quit IRC | 11:06 | |
| *** e0ne has joined #openstack-stable | 11:06 | |
| *** dims_ has quit IRC | 11:09 | |
| *** dims_ has joined #openstack-stable | 11:22 | |
| *** shwetaap has joined #openstack-stable | 11:41 | |
| *** shwetaap has quit IRC | 11:45 | |
| *** daemontool has joined #openstack-stable | 11:54 | |
| daemontool | hi | 11:54 |
|---|---|---|
| daemontool | can anyone take a look please at https://jenkins02.openstack.org/job/gate-freezer-api-requirements/3/console | 11:55 |
| daemontool | in particular | 11:55 |
| daemontool | 2016-03-07 11:32:06.269 | File "/usr/local/jenkins/slave_scripts/project-requirements-change.py", line 260, in <module> | 11:55 |
| daemontool | 2016-03-07 11:32:06.270 | main() | 11:55 |
| daemontool | 2016-03-07 11:32:06.270 | File "/usr/local/jenkins/slave_scripts/project-requirements-change.py", line 231, in main | 11:55 |
| daemontool | 2016-03-07 11:32:06.271 | if req.extras: | 11:55 |
| daemontool | 2016-03-07 11:32:06.272 | AttributeError: 'Requirement' object has no attribute 'extras' | 11:55 |
| daemontool | bknudson, ^^ | 11:56 |
| *** slashme has joined #openstack-stable | 11:57 | |
| daemontool | please disregard the previous lines, the issue was resolved at our end, thanks | 12:17 |
| daemontool | bknudson, ^^ | 12:17 |
| *** shwetaap has joined #openstack-stable | 12:41 | |
| *** shwetaap has quit IRC | 12:45 | |
| *** takedakn has joined #openstack-stable | 12:53 | |
| *** nikhil has quit IRC | 12:58 | |
| *** Kiall has quit IRC | 13:03 | |
| *** Kiall has joined #openstack-stable | 13:03 | |
| *** Kiall has quit IRC | 13:03 | |
| *** Kiall has joined #openstack-stable | 13:03 | |
| *** takedakn has quit IRC | 13:18 | |
| *** dims_ has quit IRC | 13:25 | |
| *** shwetaap1 has joined #openstack-stable | 13:32 | |
| bknudson | daemontool: when we hit this problem on keystonemiddleware stable/kilo it was because the requirements job didn't update the requirements files properly, so check on that. | 13:43 |
| *** shwetaap1 has quit IRC | 13:45 | |
| daemontool | bknudson, ty | 13:45 |
| *** e0ne has quit IRC | 13:58 | |
| *** e0ne has joined #openstack-stable | 14:00 | |
| *** eharney has joined #openstack-stable | 14:20 | |
| *** ihrachys has quit IRC | 14:40 | |
| *** shwetaap has joined #openstack-stable | 14:42 | |
| *** mriedem has joined #openstack-stable | 14:42 | |
| *** dims has joined #openstack-stable | 14:54 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 14:56 | |
| *** ihrachys has joined #openstack-stable | 14:56 | |
| *** unlaudable has quit IRC | 15:32 | |
| *** Qiming has quit IRC | 15:36 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 15:57 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 15:57 | |
| *** xhku has joined #openstack-stable | 15:58 | |
| *** xhku_ has joined #openstack-stable | 16:05 | |
| *** nikhil has joined #openstack-stable | 16:06 | |
| *** xhku has quit IRC | 16:06 | |
| *** xhku_ is now known as xhku | 16:21 | |
| *** xhku has quit IRC | 16:30 | |
| *** xhku has joined #openstack-stable | 16:30 | |
| *** xhku has quit IRC | 16:31 | |
| *** xhku has joined #openstack-stable | 16:31 | |
| *** david_lyle has quit IRC | 16:46 | |
| *** david_lyle has joined #openstack-stable | 16:46 | |
| *** e0ne has quit IRC | 16:49 | |
| *** armax has quit IRC | 16:56 | |
| *** eharney has quit IRC | 17:12 | |
| *** ihrachys has quit IRC | 17:25 | |
| *** eharney has joined #openstack-stable | 17:26 | |
| *** armax has joined #openstack-stable | 17:40 | |
| *** eharney_ has joined #openstack-stable | 17:51 | |
| *** eharney has quit IRC | 17:55 | |
| *** eharney_ is now known as eharney | 18:00 | |
| *** doug-fish has quit IRC | 18:07 | |
| *** doug-fish has joined #openstack-stable | 18:08 | |
| *** dtantsur is now known as dtantsur|afk | 18:11 | |
| *** doug-fish has quit IRC | 18:13 | |
| *** ihrachys has joined #openstack-stable | 18:14 | |
| *** xhku has quit IRC | 18:20 | |
| *** doug-fish has joined #openstack-stable | 18:27 | |
| *** thingee has joined #openstack-stable | 18:32 | |
| thingee | mriedem: any decision with stable team's summit room allocation? | 18:32 |
| mriedem | thingee: i emailed ttx, dhellmann, mtreinish and fungi about that this morning | 18:33 |
| mriedem | trying to see if we're going to do a joint session | 18:34 |
| mriedem | otherwise i wasn't really planning on doing a stable team specific session | 18:34 |
| fungi | i'm in the process of replying now | 18:34 |
| thingee | mriedem: cool kids just do something informal in the dev lounge anyways | 18:35 |
| mriedem | i thought cool kids smoked in the bathroom? | 18:35 |
| mriedem | or is lung cancer no longer cool? | 18:35 |
| thingee | vape now I'm afraid | 18:35 |
| *** sigmavirus24 is now known as sigmavirus24_awa | 18:37 | |
| *** ihrachys has quit IRC | 18:39 | |
| mriedem | fungi: can you give me a rough idea of what the infra impacts are for stable branch maintenance? beyond just having more jobs running. i know that project-config gets messy with filters and such in zuul/layout.yaml. are there other things? | 18:54 |
| fungi | mriedem: it increases requirements for platform diversity. while we're not using centos 6 or ubuntu precise in any stable branches now, we're probably going to end up testing mitaka on ubuntu xenial, so the longer we support kilo, liberty and mitaka the longer it will be before we can stop maintaining trusty workers | 18:57 |
| fungi | er, i mean testing newton on xenial | 18:57 |
| *** sigmavirus24_awa is now known as sigmavirus24 | 18:57 | |
| mriedem | yeah, so at one time we had precise and trusty workers | 18:58 |
| mriedem | seems we've been mostly in solely trusty for awhile | 18:58 |
| fungi | right, xenial releases around the mitaka release | 18:58 |
| mriedem | i think precise went out with icehouse? and py26 went out with juno | 18:58 |
| fungi | yep | 18:58 |
| mriedem | well, we'll still be running trusty for stable/liberty | 18:59 |
| fungi | and it's possible we'd end up growing a dependence on centos 7 for testing certain features not in ubuntu sometime between now and when centos 8 is out | 18:59 |
| mriedem | so i'd think it shouldn't be any different to keep stable/kilo at least around for as long as stable/liberty, even if things get weird if they were to both EOL at the same time... | 18:59 |
| mriedem | fungi: yeah, but that's newton+ right? | 18:59 |
| fungi | mriedem: right, i wasn't responding on kilo specifically but on policy in general | 18:59 |
| mriedem | i'm not sure what those features are | 18:59 |
| mriedem | probably doesn't matter | 19:00 |
| mriedem | for me i mean | 19:00 |
| fungi | things like test matrixing also get far more complex, regular expressions skipping/adding releases get longer, number of additional periodic jobs increases, fallback behaviors in devstack-gate and zuul-cloner grow more conditionals... | 19:01 |
| mriedem | yes, so there is technical debt in the tooling | 19:01 |
| *** rockyg has joined #openstack-stable | 19:02 | |
| mriedem | i think we'd only run periodic stable jobs for projects that have the stable:follows-policy tag | 19:02 |
| fungi | so more surface area to grow problems on | 19:02 |
| mriedem | and the periodic jobs are just running unit tests (as far as i'm aware) | 19:02 |
| mriedem | well, fwiw, no projects have the stable:follows-policy tag yet :) | 19:02 |
| fungi | grenade gets especially worse to deal with when we have issues which need backporting to all supported stable branches (as you've experienced plenty first hand) | 19:03 |
| mriedem | i'm assuming you're referring to the problems with wedges due to pip resolver? | 19:03 |
| *** doug-fis_ has joined #openstack-stable | 19:03 | |
| fungi | so in general it's not a significant infra concern, rather that i hope there's an upper bound on the number of branches we'd be supporting in parallel | 19:03 |
| mriedem | ok, that's good to know | 19:04 |
| mriedem | i think i'm really just wanting to extend kilo another cycle, | 19:04 |
| mriedem | so that makes the oldest branch 2 years old when it goes EOL | 19:04 |
| *** doug-fi__ has joined #openstack-stable | 19:04 | |
| fungi | mriedem: really any issue that needs a fix in kilo so that the fix can be landed in liberty, so that upgrades from liberty to mitaka can be tested, so that the fix can be applied to mitaka, so that upgrades from mitaka to newton can be tested, so that the fix can be applied in newton | 19:04 |
| mriedem | there are implications for lifeless' dependency spec, but that's been mostly idle | 19:04 |
| mriedem | fungi: yeah | 19:05 |
| mriedem | i honestly can't remember one of those since we axed juno | 19:05 |
| fungi | the more stable branches you add, the longer that chain gets | 19:05 |
| *** doug-f___ has joined #openstack-stable | 19:05 | |
| fungi | anyway, i'm not opposed to the longer kilo support suggested, which is why i said 2-3 stable branches | 19:06 |
| *** doug-fish has quit IRC | 19:06 | |
| *** doug-fish has joined #openstack-stable | 19:06 | |
| fungi | just make sure you give yourself an out in case nobody steps up to keep kilo working in a couple months when it falls apart for some reason we're not capable of predicting right now | 19:06 |
| mriedem | so maybe in general, each branch has 18 months (as today), with an option for 24 months, unless exceptional circumstances bar a thing from going 24 months | 19:07 |
| mriedem | yeah that ^ you read my mind | 19:07 |
| mriedem | also, once kilo is eol, that's the last mass release at the same time stable branch, so liberty becomes easier to eol b/c we can do it on a per-project basis | 19:07 |
| *** doug-fis_ has quit IRC | 19:08 | |
| rockyg | ++ | 19:08 |
| fungi | my objections from the vmt perspective have more to do with the fact that we need an ever increasing number of backports provided as attachments to private bugs and manually tested and reviewed there, possibly extending embargoes further | 19:08 |
| mriedem | if a change would break a project's oldest stable branch, it would break the stable policy for that project, so their only option would be to EOL the branch early | 19:08 |
| *** doug-fi__ has quit IRC | 19:09 | |
| mriedem | fungi: yeah...although, how hard are stable branch VMT reviews once there is agreement on the master branch fix? i know it's case by case depending on the fix. | 19:09 |
| mriedem | we still support going back to grizzly internally, so i realize the backports become monsters | 19:09 |
| *** doug-f___ has quit IRC | 19:10 | |
| fungi | frequently quite hard. pretty often we have to make the call to not treat something as a vulnerability because it can't be fixed safely in a backward-compatible way in all "supported" stable releases | 19:10 |
| fungi | and the more supported releases we consider, the greater the risk that not all of them will be capable of being fixes as trivially as the most recent stable release | 19:11 |
| fungi | er, being fixed | 19:11 |
| mriedem | fungi: so that might be a trigger for EOL | 19:11 |
| mriedem | for the project in question anyway | 19:12 |
| fungi | premature eol is also not great for downstream users though, which is why making some attempt at predicting when you expect to cease being able to support a release is preferred | 19:12 |
| mriedem | i don't know about that too much, i expect most distros that are depending on stable branches have longer support contracts than what is upstream | 19:13 |
| fungi | "oh, look, everyone running kilo is vulnerable to this particular nasty thing but fixing it there would mean a non-backward-compatible behavior change. time to hang them out to dry and tell them to scramble on an unplanned upgrade to liberty" | 19:13 |
| mriedem | like ibm, red hat at least | 19:13 |
| fungi | right, "use a distro with some long-term support guarantees and don't roll your own deployment from git" | 19:14 |
| fungi | i suppose as long as we make sure the distros and the people doing installs from stable branches of git know that we could up and decide to eol them at any point without warning, then that's something the vmt can work with | 19:15 |
| mriedem | but to be fair, wasn't stable originally mostly supposed to be distro people? | 19:15 |
| fungi | yeop | 19:15 |
| fungi | er, yep | 19:15 |
| fungi | however, distro people have historically had little interest in keeping testing working upstream for stable branches, which means they don't really have anywhere to collaborate once it breaks | 19:16 |
| mriedem | yeah, i actually think your VMT example and the stable:follows-policy tag as the basis go together here | 19:16 |
| mriedem | if you have to make a backward incompatible change to a stable branch to fix a bug, that would break the stable policy, and end of life the branch | 19:16 |
| mriedem | for that project | 19:17 |
| fungi | which is why the various discussions about cross-distro collaboration in the past have always seemed hollow to me | 19:17 |
| mriedem | sure, and i'm not really advocating for them either | 19:17 |
| mriedem | some distro could maintain their stable branches in github if they wanted right? | 19:17 |
| mriedem | like if ibm had a fork of openstack/nova and still maintained juno and icehouse in it | 19:18 |
| rockyg | But, at least with the projects more independent, there is a chance that you could roll forward with the just the eol'ed project and leave the rest of the install in place. | 19:18 |
| rockyg | Only works for a couple of the projects, but....better than none. | 19:18 |
| fungi | yep, it would be outside the openstack community infrastructure, and so would be considered unofficial from openstack's perspective, but they can certainly host whatever they want wherever they want if they choose to define "collaboration" as "not really tested, use at your own risk, kthxbai" | 19:19 |
| mriedem | fungi: yeah | 19:19 |
| fungi | that is below our quality assurance threshold as a community, so we can't really endorse it | 19:19 |
| mriedem | so, i guess we probably have enough meat on this bone that i should consider a session to talk through a proposal like this... | 19:19 |
| fungi | i'll attend it if you slot it | 19:20 |
| mriedem | fishbowl style as ttx suggested | 19:20 |
| mriedem | ok | 19:20 |
| fungi | and try to make sure we have good vmt/infra representation present | 19:20 |
| rockyg | Please! I'll attend. If I don't have other conflicts. But I'll make the Huawei release guy attend, at least. | 19:20 |
| fungi | since i don't really want to be seen as assuming the opinions of others on those teams | 19:20 |
| rockyg | Sorry to barge in on this discussion, but I really am happy to see it. | 19:22 |
| mriedem | fungi: yeah, for sure. also, i know you're a busy guy so i appreciate your time to talk through this. | 19:23 |
| fungi | mriedem: it's an important topic. especially if it means we reach a conclusion which makes me a less busy guy in the long run (or at least not even more busy) | 19:24 |
| mriedem | thingee: so i'm going to be looking for a fishbowl slot | 19:26 |
| mriedem | thingee: preferrably opposite a qa/vmt/infra slot if possible | 19:27 |
| mriedem | well, opposite as not at the same time | 19:27 |
| *** doug-fish has quit IRC | 19:38 | |
| *** rockyg has quit IRC | 19:50 | |
| *** doug-fish has joined #openstack-stable | 19:53 | |
| *** ihrachys has joined #openstack-stable | 19:55 | |
| *** rockyg has joined #openstack-stable | 20:04 | |
| *** doug-fish has quit IRC | 20:05 | |
| *** doug-fish has joined #openstack-stable | 20:06 | |
| *** rockyg has quit IRC | 20:09 | |
| *** doug-fis_ has joined #openstack-stable | 20:10 | |
| *** doug-fish has quit IRC | 20:11 | |
| *** david-lyle_ has joined #openstack-stable | 20:12 | |
| *** david_lyle has quit IRC | 20:13 | |
| *** david-lyle_ is now known as david-lyle | 20:26 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 20:58 | |
| *** stevemar has quit IRC | 20:59 | |
| *** shwetaap1 has joined #openstack-stable | 20:59 | |
| *** shwetaap has quit IRC | 21:01 | |
| *** stevemar has joined #openstack-stable | 21:01 | |
| *** dims has quit IRC | 21:07 | |
| *** e0ne has joined #openstack-stable | 21:10 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 21:16 | |
| *** e0ne has quit IRC | 21:34 | |
| *** derekh has quit IRC | 21:36 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 21:41 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 21:42 | |
| *** e0ne has joined #openstack-stable | 21:43 | |
| *** dims has joined #openstack-stable | 21:58 | |
| *** dims has quit IRC | 22:01 | |
| *** david-lyle has quit IRC | 22:02 | |
| *** david-lyle has joined #openstack-stable | 22:02 | |
| *** gnuoy_ has joined #openstack-stable | 22:07 | |
| *** e0ne has quit IRC | 22:11 | |
| *** gnuoy_ has quit IRC | 22:12 | |
| *** doug-fis_ has quit IRC | 22:20 | |
| *** doug-fish has joined #openstack-stable | 22:20 | |
| *** doug-fish has quit IRC | 22:23 | |
| *** doug-fish has joined #openstack-stable | 22:24 | |
| tonyb | fungi: A tangent to that discussion .... Now that we have Gerrit 2.11 deployed can we look at a "private" gerrit for VMT use? | 22:24 |
| *** doug-fish has quit IRC | 22:26 | |
| *** doug-fish has joined #openstack-stable | 22:26 | |
| fungi | tonyb: yes, we can. zaro had a solution mostly designed for 2.8 but we held off until we could get production upgraded so that we didn't have two to upgrade | 22:26 |
| fungi | some additional infra bandwidth on that would be helpful | 22:27 |
| tonyb | fungi: Clearly not infra but willing to help where I can. | 22:29 |
| fungi | tonyb: we're all infra, there's just only so many of us who can approve patches or get shell access to the servers to troubleshoot oddities | 22:29 |
| fungi | that said, i'm continuing to grow the core reviewers and root sysadmins, so hopefully we'll find sufficient support among the ranks to finish that up | 22:30 |
| tonyb | fungi: Sure. Poorly choice of words on my behalf. | 22:30 |
| *** jamielennox|away is now known as jamielennox | 22:38 | |
| *** ihrachys has quit IRC | 22:43 | |
| thingee | mriedem: ack | 22:43 |
| *** mriedem is now known as mriedem_away | 22:44 | |
| *** doug-fish has quit IRC | 22:53 | |
| *** doug-fish has joined #openstack-stable | 23:01 | |
| *** eharney has quit IRC | 23:05 | |
| *** mtreinish has quit IRC | 23:13 | |
| *** mtreinish has joined #openstack-stable | 23:14 | |
| *** shwetaap1 has quit IRC | 23:29 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 23:30 | |
| *** Qiming has joined #openstack-stable | 23:34 | |
| *** doug-fish has quit IRC | 23:37 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!