| *** michchap has joined #openstack-infra | 00:00 | |
| sarob | strange oxygen has no problems, but maven fails | 00:02 |
|---|---|---|
| sarob | def valid section id | 00:03 |
| clarkb | or selections based on id, the included document must have a DOCTYPE declaration that correctly points to the DocBook DTD. It is the DTD that declares that id attributes are of the ID type (the name id is not sufficient). If the file does not have the DOCTYPE or if the DTD cannot be opened, then such references will not resolve. <- from some documentation returned by google | 00:04 |
| clarkb | sarob: I don't see a DOCTYPE in the file you are including | 00:05 |
| sarob | okay, so which doctype declar? | 00:07 |
| clarkb | I have no idea. | 00:07 |
| clarkb | annegentle: ^ thoughts? | 00:07 |
| sarob | hey there | 00:08 |
| clarkb | sarob: git grep DOCTYPE in the manuals repo returns a lot of possibilities | 00:09 |
| sarob | okay im there | 00:10 |
| clarkb | also http://www.sagehill.net/docbookxsl/ModularDoc.html | 00:10 |
| clarkb | is where I got that line above | 00:10 |
| *** pcrews has quit IRC | 00:11 | |
| sarob | right, thanks for your help | 00:12 |
| *** blamar has quit IRC | 00:13 | |
| clarkb | ttx: mikal: what about writing rootwrap in something like lua | 00:14 |
| clarkb | it should be ridiculously fast and just about as easy to code for as python | 00:14 |
| *** juice has quit IRC | 00:14 | |
| clarkb | some of the regex stuff might get hairy though as lua doesn't support that out of the box though | 00:14 |
| *** juice has joined #openstack-infra | 00:15 | |
| *** juice has quit IRC | 00:16 | |
| *** nati_ueno_2 has joined #openstack-infra | 00:16 | |
| *** juice has joined #openstack-infra | 00:17 | |
| *** juice has quit IRC | 00:17 | |
| *** juice has joined #openstack-infra | 00:18 | |
| *** juice has joined #openstack-infra | 00:19 | |
| *** pcrews has joined #openstack-infra | 00:23 | |
| *** sarob has quit IRC | 00:29 | |
| *** sarob has joined #openstack-infra | 00:30 | |
| *** dkliban has quit IRC | 00:30 | |
| *** nati_uen_ has joined #openstack-infra | 00:33 | |
| *** Ryan_Lane has quit IRC | 00:33 | |
| *** DinaBelova has joined #openstack-infra | 00:33 | |
| *** cody-somerville has joined #openstack-infra | 00:34 | |
| *** weshay has quit IRC | 00:36 | |
| *** nati_ueno has quit IRC | 00:36 | |
| *** DinaBelova has quit IRC | 00:38 | |
| *** UtahDave has quit IRC | 00:45 | |
| *** w_ has joined #openstack-infra | 00:46 | |
| *** UtahDave has joined #openstack-infra | 00:46 | |
| *** harlowja has quit IRC | 00:47 | |
| *** olaph has quit IRC | 00:49 | |
| *** nati_uen_ has quit IRC | 00:57 | |
| *** dkliban has joined #openstack-infra | 00:58 | |
| *** bclifford has joined #openstack-infra | 01:18 | |
| *** bclifford has left #openstack-infra | 01:24 | |
| *** DinaBelova has joined #openstack-infra | 01:34 | |
| *** yaguang has joined #openstack-infra | 01:34 | |
| *** nati_ueno_2 has quit IRC | 01:35 | |
| *** DinaBelova has quit IRC | 01:37 | |
| lifeless | jeblair: ok so I would like to move this patch through without stalling; am ready to discuss more whenever you are | 01:39 |
| *** sarob has quit IRC | 01:40 | |
| *** yaguang has quit IRC | 01:45 | |
| *** zehicle_at_dell has quit IRC | 01:49 | |
| *** sarob has joined #openstack-infra | 01:50 | |
| *** anteaya has quit IRC | 01:56 | |
| *** yaguang has joined #openstack-infra | 01:57 | |
| *** markmcclain has quit IRC | 02:14 | |
| *** CaptTofu has joined #openstack-infra | 02:17 | |
| *** yaguang has quit IRC | 02:26 | |
| *** dguitarbite has joined #openstack-infra | 02:28 | |
| *** sdague has quit IRC | 02:34 | |
| *** DinaBelova has joined #openstack-infra | 02:34 | |
| *** yaguang has joined #openstack-infra | 02:39 | |
| *** DinaBelova has quit IRC | 02:39 | |
| *** melwitt has quit IRC | 02:40 | |
| *** sdague has joined #openstack-infra | 02:44 | |
| *** HenryG has joined #openstack-infra | 02:45 | |
| *** sgviking has joined #openstack-infra | 02:47 | |
| *** jhesketh__ has quit IRC | 02:51 | |
| *** jhesketh__ has joined #openstack-infra | 02:54 | |
| *** UtahDave has quit IRC | 03:19 | |
| *** afazekas_zz is now known as afazekas | 03:29 | |
| *** DinaBelova has joined #openstack-infra | 03:35 | |
| *** nati_ueno has joined #openstack-infra | 03:37 | |
| *** DinaBelova has quit IRC | 03:40 | |
| *** nati_ueno has quit IRC | 03:41 | |
| openstackgerrit | John Dewey proposed a change to openstack-infra/config: Enabled gate jobs for openstack-metering cookbook https://review.openstack.org/38730 | 03:51 |
| *** nati_ueno has joined #openstack-infra | 03:57 | |
| *** DinaBelova has joined #openstack-infra | 04:21 | |
| *** sarob has quit IRC | 04:23 | |
| *** nati_ueno has quit IRC | 04:26 | |
| *** SergeyLukjanov has joined #openstack-infra | 04:27 | |
| *** DinaBelova has quit IRC | 04:28 | |
| *** nati_ueno has joined #openstack-infra | 04:42 | |
| *** harlowja has joined #openstack-infra | 04:46 | |
| mikal | How do I stop https://review.openstack.org/#/c/37251/ from merging? | 04:54 |
| mikal | I approved it, and then realized that the migration number was already taken | 04:54 |
| clarkb | mikal: removing your +A was sufficient | 04:55 |
| clarkb | mikal: if someone else had +A'd it you could -2 code review it as an alternative | 04:55 |
| mikal | Oh good | 04:55 |
| * fungi is not entirely not here but nonetheless intrigued by the idea of a rootwrap rewrite in lua. as a language it doesn't get much free software love | 04:55 | |
| mikal | It seems I got lucky and it failed py27 unit tests | 04:55 |
| mordred | mikal: -2 it | 04:56 |
| * mordred sees he was scrolled back just a bit | 04:56 | |
| mikal | Heh | 04:56 |
| clarkb | fungi: and its bloody fast | 04:56 |
| fungi | it IS! | 04:57 |
| fungi | as a number of game development houses will attest | 04:57 |
| clarkb | mordred: how did the talk go? | 04:57 |
| fungi | he was on point. trademarks and all ;) | 04:57 |
| clarkb | fungi: we used it at garmin for testing of embedded software which I always found slightly odd considering that there are no integer types in lua | 04:58 |
| mordred | clarkb: it went well! only a mild amount of trolling | 04:58 |
| clarkb | did anyone go to pleia2's talk? how did that go? | 04:58 |
| mordred | mikal: MarkAtwood had a rootwrap suggestion tonight at dinner | 04:58 |
| fungi | lua is great as an interpreter within an interpreter, agreed | 04:58 |
| fungi | float all the ints | 04:59 |
| mordred | mikal: which was that we have a separate project, let's call it 'oslo.security' or 'oslo.root' or 'oslo.sheep' | 04:59 |
| fungi | pleia2 did amazingly well | 04:59 |
| mordred | mikal: and that will be comprised of a small set of very scruitinzed shell scripts | 04:59 |
| mordred | mikal: which will get installed somewhere special | 04:59 |
| jeblair | initial docs-draft rsync done, started a second one | 04:59 |
| mordred | mikal: and then sudo rules would be installed that would enable use of those scripts | 05:00 |
| fungi | i went to pleia2's talk and she provided an accurate overview of how we use code review for sysadmin work | 05:00 |
| fungi | seemed very well received | 05:00 |
| mordred | mikal: and we could put in gating checks similar to the requirements check to ensure that nobody attempts to do sudo things that reference anything other than the scripts in oslo.goats | 05:00 |
| fungi | goat herding tests | 05:01 |
| mikal | mordred: that's not a bad idea | 05:03 |
| mikal | mordred: its better / safer than my generation idea I suspect | 05:03 |
| clarkb | mikal: or we could just do lua | 05:03 |
| mikal | clarkb: You'd have to talk to the TC about that | 05:03 |
| mikal | clarkb: Those guys are dicks | 05:03 |
| jeblair | mikal: we can ask them when we're asking about bash. | 05:05 |
| jeblair | ;) | 05:05 |
| mikal | Heh | 05:05 |
| clarkb | lua has the added benefit of running on windows whereas bash less so | 05:05 |
| mikal | Ewww windows | 05:05 |
| clarkb | however the benefit of rootwrap on windows is probably nil | 05:05 |
| jeblair | clarkb: i'm not sure you're helping. :) | 05:05 |
| mikal | Also, is cygwin still a thing? | 05:05 |
| clarkb | see what I did there | 05:06 |
| clarkb | no one caught the lua joke | 05:06 |
| jeblair | clarkb: i got your lisp joke | 05:06 |
| *** nayward has joined #openstack-infra | 05:08 | |
| clarkb | mikal: I guess its still a thing if it was ever a thing | 05:09 |
| clarkb | mikal: I only ever used it for its X server which was useful for x forwarding thunderbird to windows when I was too lazy to set up a mail client on windows | 05:10 |
| mordred | (rootwrap (clarkb (lua (goat t)))) | 05:10 |
| * fungi caught said joke somewhat late after being afk | 05:11 | |
| fungi | mordred: careful, you'll use up all your parens | 05:12 |
| *** nati_ueno has quit IRC | 05:12 | |
| *** nayward has quit IRC | 05:13 | |
| clarkb | is the osconing over or does it continue into tomorrow? | 05:14 |
| jeblair | clarkb: first half of tomorrow. | 05:15 |
| *** afazekas_ has joined #openstack-infra | 05:15 | |
| * clarkb puts https://review.openstack.org/#/c/36886/ on everyone elses radar | 05:17 | |
| clarkb | and goes to bed | 05:17 |
| *** amotoki has joined #openstack-infra | 05:20 | |
| *** mriedem has quit IRC | 05:26 | |
| *** sgviking has left #openstack-infra | 05:27 | |
| *** nayward has joined #openstack-infra | 05:28 | |
| *** DinaBelova has joined #openstack-infra | 05:29 | |
| * fungi fires his radar back up for a bit | 05:30 | |
| *** nayward has quit IRC | 05:32 | |
| *** DinaBelova has quit IRC | 05:33 | |
| mordred | fungi: if you have radar: https://review.openstack.org/#/c/38391/ and https://review.openstack.org/#/c/35730/ https://review.openstack.org/#/c/38328/ are all things that other folks need | 05:35 |
| mordred | fungi: and I'd like to land and then cut a new release | 05:35 |
| *** odyssey4me has joined #openstack-infra | 05:35 | |
| *** DinaBelova has joined #openstack-infra | 05:39 | |
| *** koolhead17 has joined #openstack-infra | 05:43 | |
| *** DinaBelova has quit IRC | 05:44 | |
| *** Ryan_Lane has joined #openstack-infra | 05:46 | |
| openstackgerrit | Sergey Lukjanov proposed a change to openstack-infra/config: Exec pip freeze in gate-requirements-install https://review.openstack.org/38746 | 05:49 |
| openstackgerrit | Sergey Lukjanov proposed a change to openstack-infra/config: Add dev-requirements.txt to gate-requirements-install https://review.openstack.org/38749 | 05:51 |
| *** zehicle_at_dell has joined #openstack-infra | 05:51 | |
| *** vipul is now known as vipul-away | 05:54 | |
| *** nati_ueno has joined #openstack-infra | 05:55 | |
| *** odyi has quit IRC | 05:56 | |
| *** vipul-away is now known as vipul | 05:56 | |
| *** odyi has joined #openstack-infra | 05:56 | |
| *** SergeyLukjanov has quit IRC | 05:57 | |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/zuul: Add option to validate function names https://review.openstack.org/38753 | 06:04 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/zuul: Add option to validate job names https://review.openstack.org/38753 | 06:08 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/config: Validate that zuul-referenced jobs exist https://review.openstack.org/38755 | 06:09 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/config: Validate that zuul-referenced jobs exist https://review.openstack.org/38755 | 06:16 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/zuul: Add option to validate job names https://review.openstack.org/38753 | 06:17 |
| jeblair | that's been bugging me for a while; i think it will be nice to have. bed now. | 06:18 |
| *** amotoki has quit IRC | 07:02 | |
| *** nayward has joined #openstack-infra | 07:06 | |
| *** yaguang has quit IRC | 07:09 | |
| *** nayward has quit IRC | 07:16 | |
| openstackgerrit | Erwan Gallen proposed a change to openstack-infra/config: Add french list openstack-fr https://review.openstack.org/38761 | 07:19 |
| *** yaguang has joined #openstack-infra | 07:22 | |
| *** SergeyLukjanov has joined #openstack-infra | 07:26 | |
| *** DinaBelova has joined #openstack-infra | 07:26 | |
| *** koolhead17 has quit IRC | 07:27 | |
| *** plomakin has quit IRC | 07:30 | |
| *** plomakin has joined #openstack-infra | 07:32 | |
| *** Ryan_Lane has quit IRC | 07:33 | |
| *** Ryan_Lane has joined #openstack-infra | 07:35 | |
| *** yaguang has quit IRC | 07:48 | |
| *** dguitarbite has quit IRC | 07:51 | |
| *** dguitarbite has joined #openstack-infra | 07:53 | |
| *** nati_ueno has quit IRC | 07:54 | |
| *** yaguang has joined #openstack-infra | 08:01 | |
| *** jpich has joined #openstack-infra | 08:02 | |
| *** hashar has joined #openstack-infra | 08:10 | |
| *** hashar has quit IRC | 08:10 | |
| *** hashar has joined #openstack-infra | 08:11 | |
| *** Ryan_Lane has quit IRC | 08:13 | |
| *** SergeyLukjanov has quit IRC | 08:18 | |
| *** derekh has joined #openstack-infra | 08:22 | |
| *** yaguang has quit IRC | 08:31 | |
| *** DinaBelova has quit IRC | 08:35 | |
| *** alexpilotti has joined #openstack-infra | 08:39 | |
| *** yaguang has joined #openstack-infra | 08:44 | |
| *** mikal has quit IRC | 08:46 | |
| *** mikal has joined #openstack-infra | 08:48 | |
| *** SergeyLukjanov has joined #openstack-infra | 08:55 | |
| *** DinaBelova has joined #openstack-infra | 08:56 | |
| *** afazekas has quit IRC | 09:04 | |
| *** afazekas_ is now known as afazekas | 09:08 | |
| *** yaguang_ has joined #openstack-infra | 09:12 | |
| *** nayward has joined #openstack-infra | 09:14 | |
| *** yaguang has quit IRC | 09:15 | |
| *** turul_ has joined #openstack-infra | 09:18 | |
| *** DinaBelova has quit IRC | 09:26 | |
| *** DinaBelova has joined #openstack-infra | 09:27 | |
| *** DinaBelova has quit IRC | 09:27 | |
| *** DinaBelova has joined #openstack-infra | 09:29 | |
| *** GheRivero has quit IRC | 09:36 | |
| *** GheRivero has joined #openstack-infra | 09:37 | |
| *** GheRivero has quit IRC | 09:37 | |
| *** GheRivero has joined #openstack-infra | 09:38 | |
| *** fbo is now known as fbo_away | 09:49 | |
| *** DinaBelova has quit IRC | 10:02 | |
| *** dkehn has quit IRC | 10:03 | |
| *** dkehn has joined #openstack-infra | 10:03 | |
| openstackgerrit | Sean Dague proposed a change to openstack-infra/config: enable htmlify on main log vhost https://review.openstack.org/38785 | 10:46 |
| *** yaguang_ has quit IRC | 10:59 | |
| *** DinaBelova has joined #openstack-infra | 11:02 | |
| *** jhesketh__ has quit IRC | 11:04 | |
| *** DinaBelova has quit IRC | 11:07 | |
| *** afazekas is now known as afazekas_zz | 11:07 | |
| *** BobBall_ is now known as BobBall | 11:11 | |
| *** DinaBelova has joined #openstack-infra | 11:12 | |
| *** dprince has joined #openstack-infra | 11:16 | |
| *** DinaBelova has quit IRC | 11:17 | |
| *** hashar has quit IRC | 11:20 | |
| *** zehicle_at_dell has quit IRC | 11:24 | |
| BobBall | dprince: Why, in reviewday, are features ranked as higher priority than bug fixes? | 11:26 |
| *** DinaBelova has joined #openstack-infra | 11:29 | |
| *** DinaBelova has quit IRC | 11:30 | |
| dprince | BobBall: I think I initially used Thierry's ranking system. | 11:49 |
| dprince | BobBall: I'd mostly be fine with making priority bug fixes rank at the top I think. | 11:49 |
| BobBall | I think it's crazy that medium features get higher up the list than high bugfix.. | 11:50 |
| BobBall | and low feature above medium bugfix too :) | 11:50 |
| dprince | BobBall: The one *big* ranking feature that reviewday is missing is the ability to factor in the age of the review as part of the equation. | 11:50 |
| *** sdake_ has quit IRC | 11:50 | |
| BobBall | it encourages buggy code | 11:50 |
| BobBall | that should be easy enough to add? | 11:50 |
| BobBall | 1 point per day? ;) | 11:50 |
| *** cppcabrera has joined #openstack-infra | 11:50 | |
| dprince | BobBall: we could start there. Sure. | 11:51 |
| BobBall | I wonder if it's worth a discussion on -infra or even -dev as to what the rankings for features vs bugfixes should be | 11:51 |
| *** mriedem has joined #openstack-infra | 11:52 | |
| dprince | BobBall: If I recall it wasn't easy. Gerrit doesn't provide that information in the query we are making. Or at least at the time it didn't. Could be the newer version we have now contains the date info we need. The other way to get it would be to use Git would we be both heavy handed and slow. | 11:52 |
| BobBall | ahhhh okay :/ Shame!! | 11:52 |
| dprince | BobBall: Feel free to double check me though. I think it was late 2011 when I initially wrote it... | 11:53 |
| BobBall | checking now | 11:54 |
| BobBall | we have createdOn and lastUpdated | 11:54 |
| BobBall | that'd be good enough | 11:54 |
| BobBall | so createdOn should be the date that patchset was uploaded | 11:54 |
| *** w_ is now known as olaph | 11:58 | |
| *** krtaylor has joined #openstack-infra | 11:59 | |
| *** fbo_away is now known as fbo | 12:05 | |
| jpich | Hey. I'm having some trouble debugging why this grizzly review is failing: http://logs.openstack.org/48/38648/1/check/gate-horizon-pep8/3932/console.html . The distribution for pytz can't be found even though lots of versions are available, and pytz isn't capped or pinned in the requirements file | 12:12 |
| *** alexpilotti has quit IRC | 12:17 | |
| cppcabrera | jpich: That might be related to this article I found this morning: http://blog.vrplumber.com/b/2013/07/25/pippytz-fails-14-and-2013b/ | 12:20 |
| *** hashar has joined #openstack-infra | 12:20 | |
| cppcabrera | If I understand correctly, pytz made an update recently and their version number doesn't play nice with pip 1.4 | 12:20 |
| cppcabrera | I think the solution is to pin to version == 2013b | 12:21 |
| jpich | cppcabrera: Wow that is weird. Thanks for the link. | 12:22 |
| jpich | I can't reproduce in a local grizzly env after destroying the venv and pip cache though (nor is it happening on master (version set to >=2010h, installs 2013b)) | 12:25 |
| hashar | pabelanger: hi :-] Do you happen to know how Debian packages are exported in Ubuntu? python-pbr is in Debian testing since June 6th but not yet in Ubuntu | 12:26 |
| hashar | nm it is not always automatic, got to fill a bug to request the sync ( https://wiki.ubuntu.com/UbuntuDevelopment/NewPackages ) | 12:28 |
| *** cppcabrera has left #openstack-infra | 12:28 | |
| *** rfolco has joined #openstack-infra | 12:31 | |
| *** dkranz has joined #openstack-infra | 12:33 | |
| jpich | As, it's just like ccpcabrera said indeed. https://github.com/openstack/horizon/commit/5abe67709c61ae705526ee41ef13795216ec1eec Backport time | 12:34 |
| pabelanger | hashar, Ya, I you can manually sync if FF is not closed | 12:35 |
| hashar | pabelanger: apparently python-pbr is already in saucy https://launchpad.net/ubuntu/+source/python-pbr | 12:36 |
| hashar | pabelanger: though it does not show up on http://packages.ubuntu.com/python-pbr | 12:36 |
| *** cody-somerville has quit IRC | 12:37 | |
| pabelanger | hashar, maybe a build error? or packages out of sync | 12:38 |
| pabelanger | not too sure | 12:38 |
| hashar | I managed to download it from the saucy repository :-] | 12:39 |
| hashar | so I can backport it in Precise (which we are using) | 12:39 |
| pabelanger | cool | 12:39 |
| hashar | of course the dependencies are not satisfied under Precise :] | 12:41 |
| *** pabelanger has quit IRC | 12:43 | |
| *** xchu has joined #openstack-infra | 12:47 | |
| *** dguitarbite has quit IRC | 12:51 | |
| *** krtaylor has quit IRC | 12:54 | |
| *** weshay has joined #openstack-infra | 12:55 | |
| *** SergeyLukjanov has quit IRC | 12:57 | |
| *** afazekas_zz has quit IRC | 13:03 | |
| *** anteaya has joined #openstack-infra | 13:06 | |
| *** pabelanger has joined #openstack-infra | 13:07 | |
| *** DinaBelova has joined #openstack-infra | 13:18 | |
| openstackgerrit | Paul Belanger proposed a change to openstack-infra/config: Invert modules.conf logic https://review.openstack.org/38801 | 13:20 |
| *** SergeyLukjanov has joined #openstack-infra | 13:20 | |
| *** lbragstad has joined #openstack-infra | 13:22 | |
| *** changbl has quit IRC | 13:24 | |
| *** avtar has joined #openstack-infra | 13:27 | |
| *** turul_ is now known as afazekas | 13:28 | |
| *** dprince has quit IRC | 13:29 | |
| *** rnirmal has joined #openstack-infra | 13:31 | |
| *** prad has joined #openstack-infra | 13:41 | |
| *** DinaBelova has quit IRC | 13:41 | |
| *** russellb is now known as rustlebee | 13:42 | |
| *** blamar has joined #openstack-infra | 13:50 | |
| *** pentameter has joined #openstack-infra | 13:57 | |
| *** odyssey4me has quit IRC | 13:57 | |
| *** dprince has joined #openstack-infra | 13:59 | |
| *** ndipanov has joined #openstack-infra | 14:00 | |
| *** anteaya is now known as ant-tree-ya | 14:05 | |
| *** mriedem has quit IRC | 14:07 | |
| *** burt has joined #openstack-infra | 14:08 | |
| *** DinaBelova has joined #openstack-infra | 14:12 | |
| *** DinaBelova has quit IRC | 14:12 | |
| *** DinaBelova has joined #openstack-infra | 14:12 | |
| *** DinaBelova has quit IRC | 14:15 | |
| sdague | clarkb: is is possible to get a copy of what the logs vhost file looks like when all is said and done? | 14:18 |
| mordred | dhellmann_: around? the pbr gate is fixed - I was wondering if I could get a few reviews from you this morning so that we can get a new release cut for you | 14:18 |
| jeblair | #status notice Performing maintenance on docs-draft site, unstable docs jobs expected for the next few minutes; use "recheck no bug" | 14:18 |
| sdague | because locally for me I can have ReadmeFile and the wsgi stuff together | 14:18 |
| annegentle | jeblair: all docs or ci docs? | 14:18 |
| jeblair | annegentle: all gating docs jobs (dev docs) | 14:19 |
| annegentle | jeblair: ok thanks I'll let #openstack-doc know too | 14:19 |
| mordred | jpich: we updated the requirements file for the issue - you need to do a >=2010h with it | 14:19 |
| *** openstackstatus has joined #openstack-infra | 14:19 | |
| jeblair | #status notice Performing maintenance on docs-draft site, unstable docs jobs expected for the next few minutes; use "recheck no bug" | 14:19 |
| openstackstatus | NOTICE: Performing maintenance on docs-draft site, unstable docs jobs expected for the next few minutes; use "recheck no bug" | 14:19 |
| mordred | jpich: because of 1.4 not installing pre-release thigns | 14:19 |
| *** krtaylor has joined #openstack-infra | 14:19 | |
| jeblair | (i am going to wait 5 more mins to start to let a few gate jobs finish) | 14:20 |
| ttx | dprince: I think my original ranking placed bugs tagged "regression" first, he ncritical bugs, then essential features targeted to the next milestone, then High/Med features, then High/Med bugs etc | 14:20 |
| jpich | mordred: Thanks! I saw that later and backported your patch to Grizzly, cheers | 14:21 |
| mordred | jpich: sweet! | 14:21 |
| ttx | jeblair: we can place breakfast before 8:15, i'm awake | 14:21 |
| BobBall | ttx: The concern I have is that medium features are "higher priority" than medium bugs | 14:21 |
| * mordred supposes that perhaps he should look through the project for folks using unpinned pytz... | 14:21 | |
| BobBall | ttx: "low feature" is higher priority than "medium bug"... | 14:21 |
| *** markmcclain has joined #openstack-infra | 14:22 | |
| ttx | BobBall: so the trick is it should depend on where you are in the cycle | 14:22 |
| *** changbl has joined #openstack-infra | 14:22 | |
| ttx | BobBall: because there are so many medium bugs | 14:22 |
| BobBall | ttx: see https://github.com/openstack-infra/reviewday/blob/master/reviewday/mergeprop.py#L22 for the current ranking | 14:22 |
| ttx | but yes, you could make a case than a med bugfix review is more important than a low feature impl | 14:23 |
| ttx | BobBall: that ranking dates from back when Low features were "important" | 14:23 |
| jeblair | ttx: sure, let me do the docs-draft thing and then let's head down. | 14:23 |
| ttx | Now that "Low" means "i don't really care either way"... | 14:23 |
| ttx | BobBall: I guess it should be downgraded | 14:24 |
| ttx | jeblair: just ping me when you are ready | 14:24 |
| *** yaguang has joined #openstack-infra | 14:24 | |
| ttx | I'd still argue that Med feature > Med bug, as long as you're in the impl stages (-> h3) | 14:24 |
| *** mriedem has joined #openstack-infra | 14:25 | |
| BobBall | Is there a way to programatically identify where we are in the cycle? i.e. during H-1 we might not care about bugfixes but post H-3 we only care about bugfixes | 14:25 |
| ttx | BobBall: theoretically you should not care about features if you're past H3 | 14:25 |
| jeblair | starting docs draft work now | 14:25 |
| dprince | BobBall: the release dates our roughly fixed on a 6 month cycle so you could use those dates | 14:26 |
| BobBall | indeed - but anything will still show up in reviewday as more important than bugfixes | 14:26 |
| dprince | BobBall: I'm not sure I like it automagically changing priorities based on dates though. | 14:26 |
| BobBall | yeah, makes me nervous too | 14:26 |
| dprince | BobBall: I think I could go for just making bugs a higher priority. | 14:26 |
| ttx | BobBall: you /could/ query milestone release dates and figure it out, since the naming rules are pretty consistent | 14:27 |
| dprince | ttx: I'm wary of putting in any more LP hits. We've had performance issues with some of the queries already. | 14:27 |
| ttx | dprince: heh, no kidding | 14:27 |
| BobBall | I don't so much care about those - the milestone release dates can be got once per run - not for each review | 14:28 |
| *** avtar has quit IRC | 14:28 | |
| mordred | jeblair, clarkb: I +2'd https://review.openstack.org/#/c/36886/ but did not APRV because I think someone talked about sending out a mailing list message | 14:29 |
| *** vijendar has joined #openstack-infra | 14:30 | |
| *** _TheDodd_ has joined #openstack-infra | 14:30 | |
| ttx | BobBall: even better they could even use the file that describes them in infra, let me fetch that for you | 14:31 |
| ttx | https://github.com/openstack-infra/config/blob/master/modules/releasestatus/files/integrated.yaml | 14:32 |
| ttx | this is what is used to generate the progress gauge at http://status.openstack.org/release/ | 14:33 |
| BobBall | but that would still need us to have priorities that change based on the current date | 14:33 |
| *** dhellmann_ is now known as dhellmann | 14:34 | |
| *** avtar has joined #openstack-infra | 14:35 | |
| BobBall | Perhaps the real problem I have with the priorities is that there are so many that are considered "high" priority and only 7 that are "low"... | 14:35 |
| BobBall | whereas the opposite is probably true of bugs - people tend to down-play bugs and up-play features | 14:35 |
| jeblair | docs-draft filesystems swapped, final rsync and find running. there's probably a further 20-30 minute window where docs jobs have a small chance of being unstable while the permissions are updated | 14:36 |
| annegentle | jeblair: okie | 14:36 |
| BobBall | ttx: Are you on-board with the reviewday ranking to give a bigger score for things that have been around longer? to pull them to the top | 14:40 |
| ttx | BobBall: sure, aging was in the original scoring | 14:40 |
| BobBall | oh - anything else in there that I could add while I'm here? | 14:41 |
| openstackgerrit | A change was merged to openstack-infra/gitdm: Add mattt@defunct.ca to Rackspace group https://review.openstack.org/37036 | 14:41 |
| *** datsun180b has joined #openstack-infra | 14:46 | |
| annegentle | ttx: wondering if I need to prep for Documentation Program at next week's TC? re: http://lists.openstack.org/pipermail/openstack-dev/2013-July/012491.html | 14:46 |
| BobBall | ttx, dprince: What should the per-day ranking be? 1 point per day doesn't affect the ranking much / at all between groups due to the large gaps between them (e.g. a medium bugfix would have to be waiting for review for 4 months before being ranked above a medium feature) | 14:48 |
| *** mrodden has quit IRC | 14:48 | |
| annegentle | jeblair: couple of notes on https://review.openstack.org/#/c/38694/, the CLI gate job isn't necessary and the CLI guide is being removed, so I removed it. Just letting you know, would like that merged soon if possible (probably a travel day today for some) | 14:48 |
| annegentle | jeblair: if I'm unclueful about that gate job, let me know | 14:49 |
| dprince | ttx: your old *junk* scripts did this right? I can't recall what you had in there | 14:49 |
| *** cp16net is now known as cp16net|away | 14:50 | |
| *** cp16net|away is now known as cp16net | 14:50 | |
| dprince | BobBall: maybe put in a multiplier so that anything there for a really long time gets bumped up significantly. | 14:50 |
| BobBall | a multiplier is fine - but should it be 2? 3? 5? ;) Personally I'd set it high because we need to get old reviews in before new reviews IMO | 14:51 |
| BobBall | 5 x days-since-last-patchset could be good | 14:52 |
| dprince | BobBall: not sure. I'd like to see it pan out I guess against the other values. | 14:53 |
| dprince | BobBall: I'd probably just plug it in and see if it seems logical and go with it. | 14:53 |
| BobBall | I'll go for 3. That way a medium bugfix would have to be around 10 days longer than a low feature to be ranked higher | 14:56 |
| *** mrodden has joined #openstack-infra | 14:59 | |
| *** ndipanov has quit IRC | 15:01 | |
| *** rpodolyaka has joined #openstack-infra | 15:02 | |
| *** krtaylor has quit IRC | 15:02 | |
| jeblair | rsync done; find is running to fix any stray permission errors, but there shouldn't be any | 15:15 |
| *** ndipanov has joined #openstack-infra | 15:16 | |
| clarkb | sdague: yes I can grab the vhost file for you when I switch to a computer with keys | 15:17 |
| *** ^demon has joined #openstack-infra | 15:18 | |
| *** ^demon has joined #openstack-infra | 15:18 | |
| jeblair | oh i can do that | 15:18 |
| clarkb | mordred: yup I think that type of change needs an announcement. otherwise it will be new goodness that doesn't get used | 15:18 |
| jeblair | sdague, clarkb: logs-dev? | 15:18 |
| jeblair | sdague: http://paste.openstack.org/show/41993/ | 15:19 |
| clarkb | jeblair: I think that is what sdague wanted | 15:19 |
| clarkb | not sure if all is said and done has been reached, but that should be a start | 15:19 |
| sdague | jeblair: yep | 15:20 |
| sdague | hmph | 15:20 |
| sdague | well, I'll have to replicate the trey locally and figure out why this doesn't work. Seems really weird | 15:21 |
| *** crank has quit IRC | 15:24 | |
| *** rockstar` is now known as rockstar | 15:28 | |
| *** ndipanov has quit IRC | 15:30 | |
| *** vijendar has quit IRC | 15:31 | |
| *** fbo is now known as fbo_away | 15:31 | |
| *** fbo_away is now known as fbo | 15:31 | |
| *** fbo is now known as fbo_away | 15:31 | |
| BobBall | dprince: I think that days-old should only count if we've not got -1/-2 score - i.e. if we're waiting on a new patch then don't bump it up - what do you think? | 15:31 |
| *** krtaylor has joined #openstack-infra | 15:33 | |
| *** Alexei_987 has joined #openstack-infra | 15:33 | |
| Alexei_987 | Hello everyone. I wonder could we move /tmp on jenkins centos to tmpfs | 15:34 |
| Alexei_987 | this will allow us to greatly speed up db tests in ceilometer | 15:34 |
| dprince | BobBall: would that mean a single -1 would deflate it? | 15:35 |
| dprince | BobBall: -2 is probably fine as those would come from core. | 15:35 |
| BobBall | only back to the non-inflated score. Would you rather see a penalty for -ve reviews? | 15:35 |
| BobBall | okay - that'd work | 15:35 |
| *** vijendar has joined #openstack-infra | 15:43 | |
| *** ndipanov has joined #openstack-infra | 15:43 | |
| clarkb | Alexei_987: I can't think of an immediate reason for not doing that assuming we can have puppet do it for us. Can you submit a bug against openstack-ci for that? | 15:45 |
| Alexei_987 | clarkb: Ok I'll do it | 15:46 |
| *** rpodolyaka has left #openstack-infra | 15:47 | |
| *** comstud is now known as bearhands | 15:50 | |
| jeblair | clarkb: /tmp used to get very large. I don't know its current habits; we should make sure we understand them first. | 15:52 |
| jeblair | (but it would be cool if we could use tmpfs) | 15:52 |
| * jeblair -> oscon | 15:53 | |
| *** DinaBelova has joined #openstack-infra | 15:56 | |
| *** DinaBelova has quit IRC | 15:56 | |
| sdague | Alexei_987: are the db tests in sqlite? | 15:56 |
| *** DinaBelova has joined #openstack-infra | 15:56 | |
| sdague | if so there is a different solution for making those fast | 15:57 |
| Alexei_987 | sdague: for sqllite we use inmemory db | 15:57 |
| Alexei_987 | sdague: for mongodb we use /tmp | 15:57 |
| sdague | oh, and mongodb doesn't have an in memory driver? | 15:57 |
| Alexei_987 | no.. it creates a small db file when you create a database | 15:57 |
| Alexei_987 | it's not intended to be used completely in memory | 15:58 |
| sdague | the other option might be to do it not on /tmp but some other tmpfs place | 15:58 |
| mgagne | Can't the location of this folder be controlled? | 15:58 |
| sdague | to not impact other things | 15:58 |
| Alexei_987 | yes.. we use mktemp -d to determine path for the database | 15:59 |
| Alexei_987 | instead we could use some other known path | 15:59 |
| *** SergeyLukjanov has quit IRC | 15:59 | |
| *** vogxn has joined #openstack-infra | 16:00 | |
| *** DinaBelova has quit IRC | 16:01 | |
| *** datsun180b has quit IRC | 16:05 | |
| *** cp16net is now known as cp16net|away | 16:06 | |
| *** westmaas is now known as westmau5 | 16:06 | |
| *** nayward has quit IRC | 16:07 | |
| *** crank has joined #openstack-infra | 16:10 | |
| *** hashar has quit IRC | 16:12 | |
| *** UtahDave has joined #openstack-infra | 16:14 | |
| *** bclifford has joined #openstack-infra | 16:15 | |
| *** changbl has quit IRC | 16:16 | |
| *** xchu has quit IRC | 16:18 | |
| *** nati_ueno has joined #openstack-infra | 16:18 | |
| *** sarob has joined #openstack-infra | 16:21 | |
| *** Alexei_987 has quit IRC | 16:21 | |
| *** sarob has quit IRC | 16:22 | |
| *** sarob has joined #openstack-infra | 16:23 | |
| *** sarob has quit IRC | 16:24 | |
| *** sarob_ has joined #openstack-infra | 16:24 | |
| *** nati_ueno has quit IRC | 16:25 | |
| *** BobBall is now known as BobBall_Away | 16:28 | |
| openstackgerrit | Bob Ball proposed a change to openstack-infra/reviewday: Added time-based score for length of time latest patch has been waiting https://review.openstack.org/38836 | 16:29 |
| sarob_ | i think jenkins has died, lots of jobs queued, graph shows no jenkins nodes available | 16:31 |
| *** sarob_ is now known as sarob | 16:31 | |
| *** ndipanov has quit IRC | 16:33 | |
| openstackgerrit | Bob Ball proposed a change to openstack-infra/reviewday: Added time-based score for patches https://review.openstack.org/38836 | 16:34 |
| *** _TheDodd_ has quit IRC | 16:36 | |
| *** michchap has quit IRC | 16:37 | |
| *** nati_ueno has joined #openstack-infra | 16:38 | |
| *** nati_ueno has quit IRC | 16:42 | |
| *** cp16net|away is now known as cp16net | 16:42 | |
| *** bnemec is now known as beanemec | 16:43 | |
| jeblair | sarob: i believe it's under heavy load, but i don't think it's completely broken | 16:45 |
| jeblair | sarob: the graph is of idle nodes; there are lots of nodes that are running jobs | 16:45 |
| *** koolhead17 has joined #openstack-infra | 16:46 | |
| clarkb | jeblair: yeah I have checked the launch jobs and they appear to be adding nodes | 16:46 |
| *** datsun180b has joined #openstack-infra | 16:46 | |
| clarkb | just not fast enough | 16:46 |
| sarob | roger that, false alarm | 16:47 |
| jeblair | clarkb: i've seen some spurious jenkins errors when adding nodes, which isn't helping | 16:47 |
| *** nati_ueno has joined #openstack-infra | 16:47 | |
| sarob | next time ill chk jenkins active jobs directly | 16:47 |
| jeblair | so i think when we daemonify this system, we'll probably want to have some check/retry/verify logic around manipulating jenkins nodes | 16:48 |
| *** jpich has quit IRC | 16:48 | |
| *** avtar has quit IRC | 16:50 | |
| openstackgerrit | Russell Bryant proposed a change to openstack-infra/reviewstats: Fix oslo.json to be valid syntax https://review.openstack.org/38839 | 16:51 |
| *** nati_ueno has quit IRC | 16:51 | |
| *** krtaylor has quit IRC | 16:52 | |
| *** gyee has joined #openstack-infra | 16:52 | |
| beanemec | Hey, just got this error in one of my changes: https://jenkins.openstack.org/job/gate-tempest-devstack-vm-postgres-full/35676/console | 16:57 |
| beanemec | Should I just reverify no bug when it finishes? | 16:57 |
| *** cp16net is now known as cp16net|away | 16:58 | |
| openstackgerrit | A change was merged to openstack-infra/reviewstats: Fix oslo.json to be valid syntax https://review.openstack.org/38839 | 16:58 |
| clarkb | beanemec: that will work | 17:01 |
| *** pcrews has quit IRC | 17:01 | |
| beanemec | clarkb: Okay, thanks. | 17:01 |
| jeblair | docs-draft find is done | 17:02 |
| ant-tree-ya | sarob if that was your email to the list, would you like to update it? | 17:03 |
| jeblair | old docs-draft lvm volume removed | 17:03 |
| *** _TheDodd_ has joined #openstack-infra | 17:04 | |
| jeblair | i think that completes the docs-draft work | 17:04 |
| *** mjfork has quit IRC | 17:04 | |
| jeblair | i'm in a talk about "Automating Community Contributions to Puppet with Github, Heroku, Trello and Travis" | 17:05 |
| jeblair | i am very glad that i work on a project where "community members" and "community contributions" aren't second-class. | 17:06 |
| ant-tree-ya | + | 17:06 |
| mordred | oh - hey sarob ! | 17:07 |
| * mordred didn't connect your irc nick with your name before - but it now seems obvious | 17:07 | |
| mordred | sarob: you know the http://status.openstack.org/zuul/ page, yeah? | 17:07 |
| clarkb | mordred: hey I spun up logstash-worker6.openstack.org and pip pkg_resources is derping which breaks puppet | 17:08 |
| clarkb | mordred: any chance you want to ssh into that host and share a screen session with me so that I can watch you fix it? | 17:08 |
| mordred | sure! | 17:08 |
| clarkb | mordred: also this was broken after hte first puppet run | 17:08 |
| clarkb | and during the puppet run | 17:08 |
| mordred | I'm excitied | 17:09 |
| mordred | mordred@camelot:~$ ssh logstash-worker6.openstack.org | 17:09 |
| mordred | ssh: Could not resolve hostname logstash-worker6.openstack.org: No such file or directory | 17:09 |
| clarkb | mordred: I have a giant monitor so I will screen -x once you have screen going | 17:09 |
| clarkb | mordred: oh yeah I didn't do dns yety | 17:09 |
| clarkb | mordred 166.78.162.15 | 17:09 |
| mordred | clarkb: k. I'm in as root in screen | 17:10 |
| clarkb | mordred: I am attached go for it | 17:10 |
| clarkb | mordred: now try pip --version | 17:10 |
| clarkb | I think that is the underlying problem | 17:10 |
| jeblair | ooh me too. :) | 17:10 |
| *** derekh has quit IRC | 17:11 | |
| *** reed has joined #openstack-infra | 17:11 | |
| dstufft | sounds like setuptools got uninstalled | 17:11 |
| mordred | dammit. what's the screen thing for doing scrollback? | 17:11 |
| jeblair | mordred: i do C-A esc | 17:12 |
| mordred | cool. thanks | 17:12 |
| jeblair | which is copy mode, and then navigate there, and esc gets out of it | 17:12 |
| clarkb | :( something is hosed | 17:13 |
| mordred | ok. so - the getaddrinfo doesn't seem great | 17:13 |
| mordred | there's no proper puppet.conf there | 17:13 |
| clarkb | oh interesting | 17:13 |
| mordred | did the initial puppet apply not run? | 17:13 |
| clarkb | mordred: it did run. you can attach to my screen on the puppetmaster to see. window 1 | 17:14 |
| clarkb | oh wait | 17:14 |
| clarkb | I know what happened | 17:14 |
| clarkb | in my sillyness I uninstalled puppet and reinstalled. must've purged the config | 17:14 |
| mordred | nod | 17:14 |
| clarkb | however it was the same pip error before and after that | 17:14 |
| mordred | zul: is there any chance we can get a backport of current setuptools in ubuntu cloud archive? | 17:15 |
| clarkb | maybe same symtptom for two different derps | 17:15 |
| mordred | clarkb: no - this might be a thing we need to sort out properly | 17:15 |
| clarkb | ok | 17:15 |
| mordred | in that I mean that we might be triggering the distribute upgrade bug | 17:15 |
| mordred | and have not properly communicated to puppet how to do the right thing | 17:15 |
| clarkb | I am sure we are | 17:16 |
| mordred | which is "for the love of god, upgrade setuptools by itself before ever using pip to install anything" | 17:16 |
| dstufft | lol python packaging | 17:16 |
| clarkb | mordred: I think we may want to put the setuptools upgrade in the pip manifests and force setuptools to upgrade first | 17:17 |
| clarkb | mordred: rather than whenever puppet decides to attempt it | 17:17 |
| mordred | oh - I see what you mean | 17:17 |
| mordred | yes. I think that is an EXCELLENT idea | 17:17 |
| mordred | do you want to do that - or want me to? | 17:17 |
| clarkb | mordred: I can do it if you want to underp that host for me :) | 17:17 |
| *** nati_ueno has joined #openstack-infra | 17:17 | |
| *** nati_uen_ has joined #openstack-infra | 17:17 | |
| clarkb | actually will that fix the problem? we install pip with apt. that will install older setuptools | 17:18 |
| clarkb | or is only a problem when going old to new and not when new is already present? | 17:19 |
| jeblair | mordred: puppet addresses the github pull request workflow/state problem by sucking all pr info into trello | 17:19 |
| reed | hi all | 17:19 |
| clarkb | reed: hello | 17:19 |
| jeblair | reed: hey | 17:19 |
| clarkb | mordred: maybe it will be easier to delete that host and try again without the purge derpage | 17:19 |
| reed | back in SF, winter here | 17:19 |
| clarkb | mordred: 166.78.162.153 is a sister node with the same error but I didn't do anything to the puppet setup on that host | 17:20 |
| clarkb | mordred: if you want to derp around there | 17:20 |
| jeblair | reed: i got roped into doing an impromptu talk later today. | 17:20 |
| reed | jeblair, nice :) | 17:20 |
| reed | where? | 17:20 |
| clarkb | mordred: hiera | 17:21 |
| jeblair | reed: oscon | 17:21 |
| *** cp16net|away is now known as cp16net | 17:21 | |
| reed | jeblair, regular sessions or at the booth? | 17:21 |
| jeblair | reed: regular session, d135@11:50 | 17:21 |
| reed | looks like the plan to move the General list off of launchpad is going rapidly thanks to rockstar | 17:21 |
| mordred | clarkb: let's delete and re-spin the other node as a way to test your pip module change | 17:21 |
| jeblair | reed: mark atwood, fungi, pleia2 and i will be there | 17:22 |
| reed | jeblair, sweet... sorry to miss it, I'd love to learn more about RE | 17:22 |
| lbragstad | clarkb: Taking another look at the Oslo failure I had from the other day with threading. I've ran the tests a few times but it seems to be failing out in different spots each time (checking the last test run in the /tmp dir). This is running locally and rebased on the latest Oslo changes as of this morning. Any suggestions for debugging tests under these circumstances? | 17:22 |
| clarkb | mordred: we reverted the setuptools change. I remember now and am paranoid | 17:23 |
| *** vogxn has quit IRC | 17:23 | |
| * reed announces that we're going *very* fast with the migration of the mailing list | 17:24 | |
| rockstar | reed, the plan is to have the migration complete by 7pm my time at the very latest. | 17:24 |
| rockstar | It has to be fast, like a band-aid. :) | 17:24 |
| *** vogxn has joined #openstack-infra | 17:24 | |
| reed | I have one question for you all: do we want openstack@lists.openstack.org or openstack-general@lists.openstack.org? (assuming we can still change that) | 17:24 |
| mordred | clarkb: ok. machine no-longer derped | 17:25 |
| clarkb | mordred: looks like you got it to run. what did you have to do? | 17:25 |
| reed | I kinda like the -general one more because it has the name of the list in the address | 17:25 |
| mordred | clarkb: I did that | 17:25 |
| clarkb | mordred: I see | 17:25 |
| rockstar | reed, ooh. That's a good one. Probably should have asked that before I started migrating. | 17:25 |
| mordred | clarkb: after doing "apt-get install hiera-puppet" | 17:25 |
| clarkb | mordred: can you check what version(s) of setuptools are installed? | 17:25 |
| clarkb | mordred: is the apt package installed? | 17:26 |
| mordred | yes | 17:26 |
| reed | rockstar, it's alright.. if it's too late to use another name we can live with it | 17:26 |
| clarkb | mordred: so for this to work right, we want to apt version to be installed then the pip version? | 17:26 |
| rockstar | reed, well, it's probably not too late, other than that I've announced it as openstack@ | 17:26 |
| mordred | clarkb: yes | 17:26 |
| mordred | apt-get install python-pip python-setuptools ; pip install -U setuptools | 17:27 |
| mordred | also, I'm going to work on an apt repo for us | 17:27 |
| mordred | because this is all unworkable | 17:27 |
| reed | I don't think it's going to impact much since I doubt many people have already changed their addressbooks | 17:27 |
| *** nati_ueno has quit IRC | 17:27 | |
| clarkb | mordred: will this break d-g again? | 17:28 |
| mordred | clarkb: it should not break d-g again - although it should not have broken d-g before | 17:29 |
| rockstar | reed, I'm happy to change it, actually. | 17:31 |
| reed | mordred, jeblair, clarkb: what do you think? openstack@lists or openstack-general@lists? | 17:31 |
| ttx | openstack@ | 17:31 |
| ttx | or openstack-users@ | 17:32 |
| mordred | openstack@ | 17:32 |
| mordred | or openstack-users@ | 17:32 |
| ttx | not -general | 17:32 |
| * jeblair agrees with ttx and mordred | 17:32 | |
| ttx | openstack-users is more accurate but then we als owant it to be the default list | 17:32 |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/config: Ensure latest setuptools after installing pip. https://review.openstack.org/38846 | 17:32 |
| clarkb | mordred: ^ I think maybe that is what we want | 17:33 |
| ttx | and old list was named openstack@ | 17:33 |
| * mordred look | 17:33 | |
| openstackgerrit | Paul Belanger proposed a change to openstack-infra/jenkins-job-builder: Switch to flake8+hacking https://review.openstack.org/38013 | 17:33 |
| * ttx sometimes wishes all airports would be like pdx | 17:33 | |
| ttx | went thru security using regular lines in ~3min. | 17:34 |
| mordred | ttx: all airports are like that ... for me | 17:34 |
| ttx | and then free wifi. | 17:34 |
| ttx | mordred: except sfo and other non-TSA-precheck-enables airports. | 17:34 |
| ttx | also tsa precheck is not exactly an option for me. | 17:35 |
| clarkb | I have enabled the puppet daemon on logstash-worker6. clear to create DNS records for it? | 17:35 |
| clarkb | mordred: ^ | 17:35 |
| mordred | clarkb: yup. logstash-worker6 should be fine | 17:35 |
| reed | where does 'users' come from? | 17:36 |
| mordred | clarkb: I want to poke at the other machien real quick | 17:36 |
| reed | we've always called it the 'General' list, not 'users' list | 17:36 |
| ttx | reed: traditional split beween -dev and -users ML on most projects. | 17:36 |
| reed | ttx, we have our own traditions :) | 17:36 |
| ttx | reed: I'm fine with openstack@ | 17:36 |
| ttx | that's our traditoin | 17:36 |
| reed | alrigth then | 17:36 |
| ttx | not -general | 17:36 |
| ttx | or -colonel | 17:36 |
| reed | easier | 17:37 |
| clarkb | mordred: go for it | 17:37 |
| clarkb | mordred: I am half of the opinion you/we should just fix the other machine and keep it | 17:37 |
| clarkb | mordred: then use some test machines to test my change | 17:37 |
| reed | i find it annoyingly redundant but "oh well" | 17:37 |
| reed | rockstar, and openstack@lists.openstack.org it is :) | 17:37 |
| rockstar | Yay. | 17:38 |
| * ttx starts to get invite requests for the openstack team i nLP :( | 17:38 | |
| ttx | rockstar: what should we do with those ? | 17:38 |
| jeblair | everything-else@lists.openstack.org | 17:38 |
| rockstar | ttx, it's a temporary problem. | 17:38 |
| ttx | rockstar: do you plan to put it back to open membership once the transition is completed ? | 17:39 |
| ttx | ok | 17:39 |
| clarkb | mordred: that way we can iterate on the fix and I can get my new nodes up :) | 17:39 |
| * clarkb works on fixing logstash-worker8 | 17:39 | |
| rockstar | ttx, there's no reason to have the launchpad group after the mailing list is gone, right? | 17:39 |
| ttx | rockstar: the only thing left we use it for is, awkwardly enough, LP polls | 17:40 |
| ttx | but yeah, maybe simpler to just get rid of it | 17:40 |
| ttx | and build some polling capability into the foundation membership system | 17:41 |
| rockstar | ttx, unfortunately, we can't "get rid of it", per se. Launchpad is weird that way. | 17:41 |
| mordred | clarkb: on 7, I fetched your change, apt-get removed python-pkg-resources | 17:41 |
| mordred | clarkb: and am trying running puppet apply | 17:41 |
| ttx | rockstar: keeping it around doesn't really hurt anything | 17:41 |
| clarkb | mordred: ok | 17:41 |
| rockstar | ttx, right, except confuse users. | 17:41 |
| rockstar | ttx, but truthfully, with a community this big, there are bound to be lots of confused users | 17:42 |
| mordred | clarkb: that seems to have worked | 17:42 |
| *** Ryan_Lane has joined #openstack-infra | 17:42 | |
| *** Ryan_Lane has quit IRC | 17:42 | |
| *** Ryan_Lane has joined #openstack-infra | 17:42 | |
| * rockstar remember handling Launchpad source import requests for svn+ssh://please-send-me-some-free-ubantu-cds | 17:42 | |
| fungi | skimming scrollback, but openstack@ has the beneficial property of acting as a magnet to attract messages from people who don't know where to post them, which in almost all cases means that's the list they should go to | 17:42 |
| ttx | rockstar: good times | 17:43 |
| ttx | so close and yet so far away | 17:43 |
| *** krtaylor has joined #openstack-infra | 17:43 | |
| mordred | rockstar: svn+ssh://please-send-me-some-free-ubantu-cds sounds like a great repo. I would like some of your product | 17:44 |
| rockstar | mordred, lawlz | 17:44 |
| clarkb | mordred: worker8 was brought to life by removing pkg resources installing it and pip by hand then upgrading setuptools and running puppet | 17:45 |
| clarkb | I have created dns records for 8 | 17:45 |
| jeblair | rockstar: please subscribe me to your newsletter :) oh wait that's what you're doing. | 17:45 |
| clarkb | mordred: should I start the puppet daemon on 7 and create dns records there now? | 17:45 |
| rockstar | Trolls, the lot of you. | 17:45 |
| clarkb | mordred: I am going to do that | 17:46 |
| mordred | clarkb: yes please | 17:46 |
| clarkb | all done | 17:47 |
| clarkb | mordred: I think we should hold off on merging my puppet change though | 17:47 |
| mordred | clarkb: ok | 17:47 |
| clarkb | mordred: I don't think we will need to spin up any more hosts today or over the weekend | 17:47 |
| mordred | clarkb: I believe that is true | 17:47 |
| clarkb | mordred: and I am slightly paranoid about breaking d-g while you guys are conferencing and traveling | 17:48 |
| mordred | clarkb: yah. | 17:48 |
| clarkb | I am going to kill the screen on 6 if no one needs it | 17:48 |
| *** dprince has quit IRC | 17:49 | |
| jeblair | clarkb: please don't break d-g right before i show it. :) | 17:50 |
| fungi | heh, yeah that would be doubleplusungood | 17:51 |
| *** pabelanger has quit IRC | 17:51 | |
| clarkb | jeblair: well now that you mention you depend on it I bet that if I break it I could get help fixing it :P | 17:51 |
| clarkb | jeblair: don't worry I want a quiet weekend | 17:51 |
| jeblair | clarkb, fungi: searching for "com.jcraft.jsch.JSchException: verify: false", which is a moderately frequent cause of unstable jobs atm... | 17:53 |
| jeblair | clarkb, fungi: leads me to believe the scp plugin should use a newer version of jsch | 17:53 |
| jeblair | i'll prioritize fixing up the scp issues so i can upgrade that and cut a release | 17:53 |
| clarkb | o | 17:54 |
| clarkb | *ok | 17:54 |
| jeblair | but also, i'll just try building a local copy with the new version for us | 17:54 |
| *** pcrews has joined #openstack-infra | 17:57 | |
| fungi | i agree those random failures trying upload result logs are frequent enough i get questions. i thought maybe it was intermittent network breakage, but maybe not in that case | 17:58 |
| *** ladquin has quit IRC | 18:00 | |
| *** bclifford has quit IRC | 18:02 | |
| openstackgerrit | A change was merged to openstack-infra/config: Add three more logstash workers nodes. https://review.openstack.org/38389 | 18:02 |
| *** bclifford has joined #openstack-infra | 18:02 | |
| *** yaguang has quit IRC | 18:05 | |
| jeblair | restarting jenkins-dev with the new scp plugin | 18:07 |
| *** vipul is now known as vipul-away | 18:09 | |
| vishy | fungi: ping | 18:09 |
| fungi | hi vishy! are you here at oscon too? | 18:10 |
| *** sdake_ has joined #openstack-infra | 18:10 | |
| *** sdake_ has joined #openstack-infra | 18:10 | |
| vishy | fungi: no. :( | 18:10 |
| fungi | aww, too bad. pinging me about the stable/folsom regression in nova? | 18:10 |
| vishy | fungi: I'm hoping this will unblock the stable/folsom gate: https://review.openstack.org/#/c/38854/ | 18:10 |
| vishy | yup | 18:10 |
| * fungi looks | 18:10 | |
| jeblair | fungi, clarkb: scp plugin on -dev appears to still work. we can put it in prod when convenient. | 18:14 |
| jeblair | fungi, clarkb: i'll go ahead and upload it to prod, but not restart | 18:15 |
| *** melwitt has joined #openstack-infra | 18:15 | |
| jeblair | done; should take effect with next restart | 18:16 |
| zul | mordred: possibly | 18:17 |
| *** pabelanger has joined #openstack-infra | 18:18 | |
| clarkb | jeblair: ok | 18:18 |
| *** _TheDodd_ has quit IRC | 18:18 | |
| *** cp16net is now known as longlostnickname | 18:18 | |
| openstackgerrit | Matthew Treinish proposed a change to openstack-infra/config: Add gerritbot for openstack-glance channel https://review.openstack.org/38855 | 18:19 |
| fungi | jeblair: awesome | 18:21 |
| *** CaptTofu has quit IRC | 18:22 | |
| *** ladquin has joined #openstack-infra | 18:24 | |
| vishy | fungi: so should we just force it in then? | 18:25 |
| fungi | vishy: the devstack change or the nova one? i think we want to manually merge the devstack change and then recheck the nova one | 18:27 |
| vishy | the devstack | 18:27 |
| vishy | exactly | 18:27 |
| sdague | fungi: you got super submit powers? | 18:27 |
| fungi | sdague: i can grant them to myself, but i feel a little uneasy doing that since i'm not on devstack-core | 18:28 |
| sdague | fungi you have my blessing | 18:28 |
| sdague | I'm in the office for probably the next 3 hours, so I'd say sooner rather than later | 18:29 |
| jeblair | fungi: +1 | 18:29 |
| *** dprince has joined #openstack-infra | 18:29 | |
| fungi | also, it's debatable whether that's a great idea to do while 75% of the infra core devs are osconing, but since it's only hitting folsom and folsom is already broken, i guess it's safe enough | 18:29 |
| fungi | can't make it any more broken than broken | 18:29 |
| sdague | clarkb doesn't need to sleep tonight | 18:29 |
| sdague | :) | 18:29 |
| clarkb | wait what? | 18:30 |
| jeblair | fungi: if it somehow breaks something other than folsom we'll forcibly revert it. easy. | 18:30 |
| clarkb | this is my weekend off, I am going to watch a baseball game or something | 18:30 |
| * fungi nods | 18:30 | |
| clarkb | oh if folsom needs unborking carry on | 18:30 |
| clarkb | new logstash worker nodes are working | 18:30 |
| fungi | clarkb: talking about manually submitting https://review.openstack.org/36698 to devstack stable/folsom since gate tests on it are not going to succeed (because of a nova regression which needs to merge too) | 18:32 |
| *** ladquin has quit IRC | 18:32 | |
| jeblair | fungi: why are we still talking about it? :) | 18:33 |
| clarkb | its fine with me | 18:33 |
| *** sarob has quit IRC | 18:33 | |
| clarkb | I am going to approve the meetbot change after lunch if nothing else borks | 18:34 |
| fungi | was just answering clarkb's question | 18:34 |
| *** bclifford has quit IRC | 18:37 | |
| *** bclifford has joined #openstack-infra | 18:37 | |
| *** vipul-away is now known as vipul | 18:42 | |
| *** mjfork has joined #openstack-infra | 18:43 | |
| *** mriedem has quit IRC | 18:44 | |
| reed | jeblair, fungi: when you have time, remember to pull the list of ATC from the date of the last time we pulled it to <today> | 18:45 |
| *** sarob has joined #openstack-infra | 18:46 | |
| jeblair | fungi: don't forget you have a talk. :) | 18:46 |
| *** ladquin has joined #openstack-infra | 18:47 | |
| *** vogxn has quit IRC | 18:48 | |
| fungi | eek, the mornig break is shorter than i realized. omw | 18:48 |
| *** mriedem has joined #openstack-infra | 18:49 | |
| clarkb | fungi is talking too? on what subject? | 18:52 |
| *** UtahDave has quit IRC | 18:55 | |
| *** vipul is now known as vipul-away | 19:05 | |
| *** vipul-away is now known as vipul | 19:07 | |
| *** bclifford has quit IRC | 19:18 | |
| *** bclifford has joined #openstack-infra | 19:18 | |
| *** mjfork has quit IRC | 19:20 | |
| *** rpodolyaka1 has joined #openstack-infra | 19:21 | |
| rpodolyaka1 | mordred: ping | 19:23 |
| *** harlowja has quit IRC | 19:23 | |
| *** bclifford has quit IRC | 19:24 | |
| *** afazekas is now known as afazekas_zz | 19:24 | |
| sdague | clarkb: any chance in releasing more devstack nodes? | 19:26 |
| sdague | http://status.openstack.org/zuul/ | 19:26 |
| sdague | we're kind of tapped out and backing up :) | 19:26 |
| clarkb | sdague it wont help we cant create them fast enough | 19:32 |
| sdague | ok | 19:32 |
| clarkb | we hit d-g with a sufficiently large thundering herd | 19:33 |
| sdague | well we might want to up the overall allocations leading up to H3 | 19:33 |
| sdague | as I expect this is only going to get worse | 19:33 |
| sdague | until we can get converted to testr to bring the through times down | 19:33 |
| clarkb | I think we need a different solution. if you look at the launch jobs logs we are continuously spinning up nodes | 19:34 |
| clarkb | a bigger pool will help until you empty it then you have this situation | 19:34 |
| *** SergeyLukjanov has joined #openstack-infra | 19:35 | |
| clarkb | a second jenkins master may help if jenkins is the bottleneck | 19:35 |
| clarkb | we definitely need to do something | 19:36 |
| sdague | yep | 19:36 |
| clarkb | maybe go back to pep8 first | 19:36 |
| clarkb | I need to do more digging. can do thst after lunch | 19:36 |
| fungi | clarkb: jeblair and i provided behind-the-podium ground cover for mc mark atwood | 19:43 |
| fungi | more or less the ci overview talk | 19:43 |
| fungi | was a last-minute entry into the schedule | 19:44 |
| *** hashar has joined #openstack-infra | 19:45 | |
| clarkb | cool | 19:45 |
| clarkb | fungi if you guys could give sdague's concerns above some thought that would be great | 19:46 |
| reed | ttx, http://gizmodo.com/new-nexus-7-now-on-amazon-update-923138791 | 19:48 |
| openstackgerrit | Yegor Minin proposed a change to openstack-infra/gerritlib: Add option to query commit messages https://review.openstack.org/38865 | 19:48 |
| reed | lunch time | 19:49 |
| *** mrodden has quit IRC | 19:54 | |
| *** sdake has quit IRC | 19:54 | |
| *** DinaBelova has joined #openstack-infra | 19:58 | |
| *** DinaBelova has quit IRC | 20:03 | |
| *** rainya has joined #openstack-infra | 20:05 | |
| *** rpodolyaka1 has left #openstack-infra | 20:06 | |
| *** sarob_ has joined #openstack-infra | 20:09 | |
| *** mrodden has joined #openstack-infra | 20:11 | |
| *** sarob has quit IRC | 20:13 | |
| *** sarob_ has quit IRC | 20:14 | |
| *** bclifford has joined #openstack-infra | 20:19 | |
| *** nayward has joined #openstack-infra | 20:21 | |
| *** dprince has quit IRC | 20:24 | |
| mordred | jeblair: what do you think about making some of the non-voting jobs asymetrical | 20:26 |
| mordred | ? | 20:26 |
| *** DinaBelova has joined #openstack-infra | 20:26 | |
| mordred | specifically, the swift folks were just asking about why they're always having to wait on non-votingjobs that they don't care about | 20:26 |
| mordred | which made me think - perhaps the non-voting-because-job-is-broken jobs should only be triggered on projects where the output is likely to help someone make the job better | 20:27 |
| mordred | like, the non-voting neutron jobs should perhaps only trigger on neutron and devstack | 20:27 |
| mordred | it might also help with H3 node starvation | 20:27 |
| *** hashar has quit IRC | 20:27 | |
| mordred | and then if someone actually thinks they have the job fixed and we're non-voting it to make sure it is, we can go back to symetrical non-voting | 20:28 |
| *** nayward has quit IRC | 20:29 | |
| *** pabelanger has quit IRC | 20:29 | |
| ttx | reed: a few hours too late I guess | 20:30 |
| jd__ | mordred: you could also trigger it on demand via a comment or something | 20:30 |
| *** ^demon is now known as ^demon|lunch | 20:34 | |
| clarkb | mordred: the problem with treating non voting jobs that way is they essentially become noops | 20:36 |
| clarkb | mordred: no one cares about them they never become voting and they are ignored ala the bitrot jobs | 20:36 |
| mordred | clarkb: the feedback is that they are nooops in situations | 20:36 |
| mordred | like, the swift team does not care about the fact that neutron doesn't work | 20:36 |
| mordred | and I doubt additional runs triggered by the swift team will help anything | 20:36 |
| *** DinaBelova has quit IRC | 20:37 | |
| *** hashar has joined #openstack-infra | 20:37 | |
| ttx | wow. Google recognized I posted a beer picture and tagged my g+ post appropriately. | 20:38 |
| ttx | https://plus.google.com/u/0/113997404392832194109/posts/AN4zAbpbxCW | 20:38 |
| clarkb | mordred: they should care | 20:38 |
| clarkb | mordred: and not just because it makes their testing slower | 20:38 |
| clarkb | mordred: but because this is a large cooperative project | 20:38 |
| *** dkliban has quit IRC | 20:39 | |
| mordred | possibly | 20:40 |
| mordred | on the other hand | 20:40 |
| mordred | the neutron guys don't even know why neutron is not working | 20:40 |
| clarkb | (personally I am all for running fewer d-g jobs because it will make my life easier, jenkins is in a deep pile of poo right now) | 20:40 |
| clarkb | but I think that the instant we stop running something such that it is in your face we should just stop running it ocmpletely | 20:40 |
| mordred | I don't think we should stop running it | 20:40 |
| mordred | I think we should run the CRAP out of it for the relevant folks | 20:41 |
| mordred | but I'm guessing that the neutron team does not see swift test results | 20:41 |
| mordred | and I'm guessing tha tthe swift team will probably not fix the neutron bug that no one knows how to fix | 20:41 |
| mordred | I could totally be wrong though, of course | 20:41 |
| clarkb | mordred: I think you are right, but I think that this becomes a big carrot for everyone involved | 20:42 |
| *** pentameter has quit IRC | 20:42 | |
| clarkb | either we (and that is a collective openstack we) make $thing work or we don't do it at all | 20:42 |
| jeblair | i am going to spend my time making there be enough nodes. | 20:42 |
| clarkb | jeblair: ++ | 20:42 |
| jeblair | because the problem us not that we are 1 node short. | 20:42 |
| jeblair | is not. | 20:42 |
| mordred | yah | 20:43 |
| clarkb | jeblair: after thinking about it I can try bumping the pool size to see if asking the launch jobs to do more work each time helps, but with the timeout I don't expect it to | 20:43 |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/devstack-gate: Build a devstack-gate image with diskimage-builder https://review.openstack.org/38871 | 20:44 |
| *** dkliban has joined #openstack-infra | 20:45 | |
| clarkb | mordred: one worry of my is bitrot in the layout (we are already seeing that where neutron itself does not run the neutron full job) | 20:45 |
| clarkb | s/my/mine/ | 20:45 |
| mordred | clarkb: that is an excellent point | 20:45 |
| clarkb | granted these are all problems that can be worked through, but there doesn't seem to be any interest so why remove the little carrot that we have | 20:46 |
| Alex_Gaynor | Alternate policy: don't allow non-voting jobs. If you want a thing to run it must be voting, and you cannot remove a running job on a temporary basis. | 20:46 |
| mordred | Alex_Gaynor: that makes developing/vetting jobs hard | 20:46 |
| Alex_Gaynor | mordred: In the time I've been observing OS CI, I don't think I've seen a job move from non-voting to voting. Maybe non-voting jobs should require a formal declaration by a sub-project of purpose, length of time, and evaluation strategy? | 20:47 |
| clarkb | jeblair: https://jenkins.openstack.org/job/devstack-launch-vms-hpcloud-az2/43250/console we are seeing the occasional jenkins failure to add a slve | 20:48 |
| mordred | Alex_Gaynor: that's not a bad idea | 20:48 |
| clarkb | I am sure that isn't helping with our ability to add nodes | 20:48 |
| Alex_Gaynor | mordred: the evaluation strategy and requrested length of time are the key bits | 20:49 |
| *** harlowja has joined #openstack-infra | 20:50 | |
| clarkb | I'm going to bump the pool sizes to see if that helps anything. | 20:52 |
| *** rfolco has quit IRC | 20:53 | |
| clarkb | and done. bumped to 20 * 3 | 20:54 |
| *** lbragstad has quit IRC | 20:54 | |
| sdague | there are a bunch of doc fails in the gate | 20:54 |
| clarkb | sdague: don't say that | 20:54 |
| * clarkb looks | 20:54 | |
| clarkb | sdague: I think that is the issue jeblair hopes will be fixed with the new version of the scp plugin that can be installed if we restart jenkins | 20:56 |
| clarkb | https://jenkins.openstack.org/job/gate-nova-docs/28697/console https://jenkins.openstack.org/job/gate-nova-docs/28698/console for those following along | 20:57 |
| clarkb | also I just noticed that those docs aren't precompressed like the build logs. we might want to change that | 20:57 |
| *** burt has quit IRC | 20:57 | |
| mordred | clarkb: ++ | 20:58 |
| clarkb | I don't see any new filesystem errors in static's dmesg so it isn't that problem (also jeblair should've fixed that with the new fs) | 20:58 |
| *** pabelanger has joined #openstack-infra | 20:59 | |
| clarkb | I am going to push that meetbot change through now | 21:00 |
| openstackgerrit | A change was merged to openstack-infra/meetbot: Adding ability for a non-chair to end a meeting after time https://review.openstack.org/36825 | 21:00 |
| *** alaski_ is now known as alaski | 21:02 | |
| *** bclifford has quit IRC | 21:03 | |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/devstack-gate: Build a devstack-gate image with diskimage-builder https://review.openstack.org/38871 | 21:10 |
| *** openstack has joined #openstack-infra | 21:14 | |
| openstackgerrit | A change was merged to openstack-infra/config: Update yaml file to add new OpenStack End User Guide https://review.openstack.org/38694 | 21:15 |
| *** jjmb has joined #openstack-infra | 21:15 | |
| jeblair | clarkb: i think you are right. we need a 2nd jenkins then can make the slave spinup more efficient when we have more capacity. | 21:15 |
| clarkb | it looks like the adding machine to jenkins bit is taking ~30 seconds or more per slave | 21:16 |
| clarkb | https://jenkins.openstack.org/job/devstack-launch-vms-hpcloud-az1/43255/console | 21:16 |
| *** HenryG has quit IRC | 21:16 | |
| *** woodspa has quit IRC | 21:16 | |
| *** woodspa has joined #openstack-infra | 21:17 | |
| clarkb | I am willing to bet we are encountering more global locks? | 21:17 |
| *** bclifford has joined #openstack-infra | 21:17 | |
| jeblair | yeah so we need another master to scale past that | 21:18 |
| *** HenryG has joined #openstack-infra | 21:18 | |
| clarkb | I almost think a rate limiter in zuul could potentially help too | 21:19 |
| fungi | seems lock-contention-like since load average is low and cpu isn't pegged on either jenkins or devstack-launch | 21:20 |
| clarkb | jenkins doesn't do so good when you throw as much stuff at it as possible but if you let it go by at a reasonable rate it is fine | 21:20 |
| annegentle | clarkb: jeblair: thanks coremudgeons! | 21:21 |
| clarkb | fungi: ya | 21:21 |
| *** weshay has quit IRC | 21:24 | |
| dhellmann | jeblair: you like entry points, right? https://github.com/dhellmann/entry_point_inspector | 21:28 |
| *** lbragstad has joined #openstack-infra | 21:28 | |
| *** SergeyLukjanov has quit IRC | 21:30 | |
| mordred | dhellmann: neat | 21:32 |
| dhellmann | esp. useful for debugging import errors in plugins | 21:33 |
| *** lbragstad is now known as ldbragst1 | 21:33 | |
| *** vipul is now known as vipul-away | 21:34 | |
| *** vipul-away is now known as vipul | 21:34 | |
| *** ldbragst1 has left #openstack-infra | 21:35 | |
| *** DinaBelova has joined #openstack-infra | 21:38 | |
| *** jjmb has quit IRC | 21:39 | |
| *** odyi has quit IRC | 21:40 | |
| *** DinaBelova has quit IRC | 21:42 | |
| *** ^demon|lunch is now known as ^demon | 21:42 | |
| dstufft | dhellmann: heh, I think I'll finally be able to get PyPI to use sha256 instead of md5 | 21:45 |
| dhellmann | :-) | 21:45 |
| dstufft | dhellmann: which was either the first or the second thing I tried to do like 2 years ago, before I got pissed and made crate.io | 21:45 |
| dhellmann | is it any easier to contribute to pypi these days? last time I looked it was...messy | 21:46 |
| dhellmann | it's been ages, though | 21:46 |
| *** DinaBelova has joined #openstack-infra | 21:48 | |
| *** hashar has left #openstack-infra | 21:49 | |
| *** dhellmann is now known as dhellmann_ | 21:49 | |
| dstufft | dhellmann_: right now, not really. However my big push besides random sec improvements is trying to move PyPI from a home grown framework of horribleness spread across multiple multi thousand line files to django | 21:51 |
| dstufft | dhellmann_: Aka PyPI 2.0 Aka Warehouse | 21:51 |
| *** dhellmann_ is now known as dhellmann | 21:52 | |
| dhellmann | dstufft: that sounds very cool, and I look forward to seeing it show up on stackforge (or at least github) :-) | 21:52 |
| *** DinaBelova has quit IRC | 21:53 | |
| dstufft | dhellmann: dstufft/warehouse | 21:54 |
| dstufft | will eventually move to PyPA but right now it's not very contributor friendly since I'm attempting to do this half phased migration so things have to be done really carefully | 21:54 |
| *** Ryan_Lane has quit IRC | 21:55 | |
| dhellmann | dstufft: sure | 21:55 |
| *** Ryan_Lane has joined #openstack-infra | 21:55 | |
| * dhellmann "watches" warehouse | 21:55 | |
| dhellmann | hey, the weekend started without me | 21:56 |
| dhellmann | see you guys next week :-) | 21:56 |
| *** dhellmann is now known as dhellmann_ | 21:56 | |
| *** Ryan_Lane has quit IRC | 21:59 | |
| *** pcrews has quit IRC | 21:59 | |
| *** Ryan_Lane has joined #openstack-infra | 21:59 | |
| *** UtahDave has joined #openstack-infra | 21:59 | |
| *** Ryan_Lane has quit IRC | 22:01 | |
| openstackgerrit | Steve Baker proposed a change to openstack-infra/config: Add tempest check and gate for python-heatclient. https://review.openstack.org/38884 | 22:01 |
| *** Ryan_Lane has joined #openstack-infra | 22:01 | |
| *** Ryan_Lane has quit IRC | 22:03 | |
| *** Ryan_Lane has joined #openstack-infra | 22:03 | |
| *** vijendar has quit IRC | 22:04 | |
| *** rnirmal has quit IRC | 22:04 | |
| *** vipul is now known as vipul-away | 22:16 | |
| *** prad has quit IRC | 22:22 | |
| *** nati_ueno has joined #openstack-infra | 22:23 | |
| *** bknudson has joined #openstack-infra | 22:24 | |
| bknudson | pytz is the pits. | 22:24 |
| jeblair | clarkb: modules/openstack_project/templates/logstash/agent.conf.erb is the pip input section there used? | 22:30 |
| jeblair | s/pip/pipe | 22:30 |
| clarkb | jeblair: no, that file isn't used anymore I should clean that up | 22:31 |
| clarkb | jeblair: indexer.conf.erb does all of the work now | 22:31 |
| jeblair | clarkb: ok, that's the conclusion i was oming to. good. | 22:31 |
| reed | this week's newsletter is taking a lot longer to assemble than usual | 22:31 |
| clarkb | and the gearman workers speak tcp into that guy | 22:31 |
| clarkb | jeblair: unix sockets might make more sense there... | 22:31 |
| *** harlowja has quit IRC | 22:35 | |
| reed | rockstar, how are things proceeding? | 22:35 |
| rockstar | reed, getting ready to finish things up. | 22:35 |
| *** datsun180b has quit IRC | 22:35 | |
| reed | great | 22:35 |
| rockstar | Took a bit to get the pickle from the losa | 22:35 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/config: Add new jenkins masters. https://review.openstack.org/38885 | 22:35 |
| rockstar | Probably should have scheduled that ahead of time. | 22:35 |
| reed | I imagined that would have been the bigger unknown | 22:36 |
| jeblair | clarkb: ^ for next week | 22:36 |
| * jeblair airports | 22:37 | |
| *** blamar has quit IRC | 22:37 | |
| clarkb | jeblair: I am excited. already reviewing :) have a good flight | 22:38 |
| * Alex_Gaynor dreams of the day that the tempest devstack runs are parallel | 22:41 | |
| *** mgagne has quit IRC | 22:41 | |
| mikal | Are requirements files case sensitive? | 22:42 |
| clarkb | mikal: yes and no, but you should treat them as if they were | 22:42 |
| mikal | Ok, thanks | 22:42 |
| clarkb | mikal: some things (not pip) do treat the names case sensiteively | 22:42 |
| clarkb | Alex_Gaynor: some of them are :) | 22:42 |
| Alex_Gaynor | clarkb: not the gating ones :( | 22:43 |
| clarkb | Alex_Gaynor: they just don't vote and only run in the check queue :/ | 22:43 |
| clarkb | ya | 22:43 |
| Alex_Gaynor | clarkb: If I want to make the testr one the gating one who do I mail the suitcase of cash to, you or mordred ? | 22:43 |
| clarkb | Alex_Gaynor: definitely mordred because he is not afraid of jeblair :) | 22:44 |
| Alex_Gaynor | mordred: DO you have a mailing address? | 22:44 |
| *** mriedem has quit IRC | 22:46 | |
| *** DinaBelova has joined #openstack-infra | 22:48 | |
| jog0 | anyone else seeing a pyhton keystone client conflict in horizon? | 22:50 |
| clarkb | jog0: I haven't seen that fail again in the gate/check queues | 22:51 |
| *** hartsocks has joined #openstack-infra | 22:51 | |
| clarkb | there are a lot of failures in there today but I haven't seen keystone causing problems | 22:51 |
| reed | damn it | 22:51 |
| clarkb | jog0: is this where django openstack auth wants a really new version but other things pin below that? | 22:51 |
| clarkb | jog0: if so I would try updating all the things as this problem should've gotten sorted | 22:51 |
| Alex_Gaynor | There's a nova failure, the fix is in the queue, btu like at #23 in gating | 22:52 |
| clarkb | Alex_Gaynor: is that to fix the python26 flakyness? | 22:52 |
| reed | many duplicate accounts on the members database, make it very hard to use it easily/reliably to track affiliations :( | 22:52 |
| jog0 | clarkb: this is in my env | 22:52 |
| Alex_Gaynor | clarkb: no, it's in tempest, is there more than one :/ | 22:52 |
| Alex_Gaynor | clarkb: https://bugs.launchpad.net/tempest/+bug/1205344 | 22:52 |
| uvirtbot | Launchpad bug 1205344 in tempest "mkfs error in test_stamp_pattern" [Critical,In progress] | 22:52 |
| clarkb | Alex_Gaynor: there are many sadly | 22:52 |
| clarkb | Alex_Gaynor: nova has a python26 non deterministic thing that dansmith and jog0 were looking at iirc | 22:53 |
| jog0 | clarkb: yeah horizon==2013.2.a18.g07b1343 | 22:53 |
| jog0 | clarkb: AFAIK thats a something else haha | 22:53 |
| clarkb | Alex_Gaynor: oh that was one of them | 22:53 |
| *** DinaBelova has quit IRC | 22:53 | |
| jog0 | I think comstud was looking at the mkfs issue | 22:53 |
| clarkb | Alex_Gaynor: so good at least one should be fixed at some point :( | 22:53 |
| jog0 | clarkb: IMHO we should revert the offending patch, until a fix is ready but thats jsut me | 22:54 |
| clarkb | jog0: re horizon? | 22:55 |
| clarkb | thats what we did when it was broken, but I thought it got sorted out | 22:55 |
| jog0 | sorry split conversation. | 22:55 |
| jog0 | clarkb: I pulled that from the logs | 22:55 |
| jog0 | clarkb: hmm looks like something fishy with my env | 22:56 |
| clarkb | meetbot restarted earlier today after it got updated with the timeout patch | 23:02 |
| *** bknudson has quit IRC | 23:02 | |
| clarkb | I have started a test meeting in #openstack-meeting-alt. If someone want to try a #endmeeting in there in about 57 minutes that would be excellent | 23:02 |
| clarkb | I should've done this earlier but got distracted | 23:02 |
| fungi | i'll try to remember to test that out | 23:04 |
| fungi | and report back | 23:04 |
| clarkb | ty | 23:04 |
| jog0 | clarkb: ceilometer/requirements.txt:python-keystoneclient>=0.2,<0.3 | 23:08 |
| jog0 | don't think that was my issue but ... | 23:08 |
| clarkb | yeah sdague has been pestering them to update it so thatwe can shift the lower bound | 23:08 |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/devstack-gate: Proof of concept of kexec/takeovernode https://review.openstack.org/38887 | 23:09 |
| mordred | clarkb: ^^ | 23:09 |
| jog0 | and horizon/requirements.txt:python-keystoneclient>=0.3.0 | 23:09 |
| mordred | clarkb: UNTESTED and MOST LIKELY INCOMPLETE | 23:09 |
| clarkb | mordred: :) | 23:09 |
| clarkb | jog0: horizon should not have done that yet | 23:09 |
| jog0 | keystone/requirements.txt:python-keystoneclient>=0.3.0 | 23:09 |
| clarkb | did we jump the gun on bumping the lower bound? | 23:10 |
| mordred | the problem is that ceilometer needs to unpin theirs - horizon literally does not operate with < 0.3 now | 23:10 |
| clarkb | mordred: yes they do sdague has been bugging them for that | 23:10 |
| clarkb | mordred: but | 23:10 |
| jog0 | clarkb: http://paste.openstack.org/show/42065/ | 23:10 |
| clarkb | but the other projects should play nice with everyone else and wait for the ability to bump the lower bound | 23:10 |
| jog0 | that is after I ran a git pull on everythin | 23:10 |
| clarkb | jog0: so, I think you may want to just disable ceilometer until they bump their reqs ifyou can get away with it | 23:11 |
| reed | rockstar, I'll wait for the migration to be over before sending the weekly newsletter, so we test it immediately :) | 23:11 |
| rockstar | reed, awesome. I'll also send a post mortem as well. | 23:11 |
| reed | I'd hope it's a post-vitam instead :) | 23:12 |
| mordred | mikal: what clarkb said - please please please do not put a requirement in requirements.txt that does not match the package name on pypi | 23:12 |
| reed | I have the edit for the wiki page also ready | 23:12 |
| jog0 | funny thing is i don't think I was using ceilometer | 23:13 |
| jog0 | this may be a pip version issue | 23:17 |
| jog0 | or evil python stuff like that | 23:17 |
| rockstar | reed, I'm starting to suspect this pickle is not good. | 23:18 |
| rockstar | :/ | 23:18 |
| reed | different than the other one? | 23:18 |
| rockstar | reed, yeah. | 23:19 |
| rockstar | It was generated by another losa than the first one was. | 23:19 |
| reed | uhm ... | 23:19 |
| reed | what makes you think that this one is borked? | 23:20 |
| clarkb | Alex_Gaynor: pytz isntall failures due to their weird versioning is another issue | 23:22 |
| clarkb | Alex_Gaynor: I think that may only affect the stable branches though | 23:22 |
| mordred | it should be fixed in openstack/requirements for trunk. some projects may not have landed patches updating their local requirments files | 23:22 |
| clarkb | mordred: cinder is definitely affected | 23:22 |
| mordred | sigh | 23:23 |
| * mordred has only been working on getting this stuff fixed for like a month now | 23:23 | |
| * mordred cries | 23:23 | |
| clarkb | :( | 23:23 |
| mordred | WOW. the job stats graphs are CRAZY | 23:25 |
| *** UtahDave has quit IRC | 23:25 | |
| reed | mordred, where? | 23:26 |
| mordred | reed: bottom of http://status.openstack.org/zuul/ | 23:26 |
| mordred | reed: it's been a day | 23:26 |
| reed | 125 gerrit events per hour? | 23:27 |
| rockstar | reed, I can't actually even open this one with pickle. :/ | 23:27 |
| rockstar | Mailman just tells me there's no list by that name, when I know there is. It must be trapping the actual error and saying "No list" | 23:27 |
| reed | 20 changes per hour? wow | 23:27 |
| reed | rockstar, damn... you're following the same process as you did with the other one, right? | 23:28 |
| rockstar | reed, yeah. | 23:28 |
| reed | is there any chance to get a new pickle in time for this project to be salvaged today or tomorrow? | 23:29 |
| rockstar | The losa that did it this time wasn't a losa when I was at Canonical, and I'm not sure he sterilized the pickle correctly. | 23:29 |
| rockstar | Possibly. I can probably give him a call. | 23:29 |
| reed | it's worth to give it a shot | 23:29 |
| reed | and if not possible let's abort and set a new schedule | 23:29 |
| reed | maybe this LOSA can get the instructions from the other LOSA | 23:29 |
| rockstar | That's what he did. | 23:30 |
| rockstar | reed, yeah, I'll have to get it done Monday, methinks. | 23:30 |
| rockstar | Er, if I can't get it figured out soon. | 23:30 |
| reed | makes sense | 23:30 |
| reed | let's put a hard deadline on this and release ourselves from ejoying the weekend :) | 23:31 |
| *** nati_ueno has quit IRC | 23:33 | |
| jeblair | mordred, clarkb: http://tinyurl.com/q4en3ao | 23:38 |
| jeblair | mordred, clarkb: perhaps we should update the status page to use that graph | 23:38 |
| mordred | jeblair: I like that graph | 23:38 |
| jeblair | (forgot to remove ready from the title) | 23:39 |
| mordred | jeblair: also - I was just trying to make another graph and I couldn't figure out how ... | 23:39 |
| jeblair | mordred: it's because graphite is so easy | 23:39 |
| jeblair | that's why i have trouble with it | 23:39 |
| mordred | jeblair: which is "total machine minutes of tests run per hour" | 23:39 |
| mordred | jeblair: which I think should be able to be derived since we do collect test run times | 23:39 |
| mordred | jeblair: but then graphite is too easy | 23:40 |
| *** sdake has joined #openstack-infra | 23:40 | |
| jeblair | looking at the graph i pasted, it looks like we were runing about 115 devstack tests concurrently for a while. peaking to 150 nodes probably when some of them were aborting and restarting. | 23:41 |
| jeblair | mordred: i will think about your graph | 23:41 |
| mordred | jeblair: that is a bananas graph | 23:41 |
| rockstar | reed, it just so happens that I have the number for this particular losa, and have called him. Should have a new pickle within the hour. | 23:41 |
| reed | sounds good then | 23:42 |
| jog0 | http://logs.openstack.org/51/38851/1/check/gate-nova-python27/30187/console.html | 23:42 |
| jog0 | ^ looks like another dep blowup | 23:42 |
| uvirtbot | jog0: Error: "looks" is not a valid command. | 23:42 |
| mordred | jog0: looking | 23:42 |
| jog0 | mordred: thanks | 23:43 |
| * mordred stabs | 23:43 | |
| mordred | wow | 23:43 |
| mordred | that's amazing | 23:43 |
| jog0 | the words stab and amazing shouldn't be used next to eachother | 23:43 |
| mordred | jog0: that's gonna be hard to fix. did babel just upgrade? | 23:43 |
| jog0 | mordred: yes https://pypi.python.org/pypi/Babel | 23:44 |
| mordred | jog0: but we're pulling 0.9.6 | 23:45 |
| mordred | oh - we're pulling Babel >= 0.9.6 | 23:45 |
| mordred | is how that started blowing u | 23:45 |
| mordred | up | 23:45 |
| jog0 | https://pypi.python.org/packages/source/B/Babel/ | 23:46 |
| jog0 | so just uploaded a few hours ago | 23:46 |
| jog0 | mordred: why don't we stage packages before pulling them into our pypi mirror? | 23:46 |
| *** zaro0508 has joined #openstack-infra | 23:46 | |
| *** sdake_ has quit IRC | 23:47 | |
| jog0 | as in anything that isn't related to an explicit change to requirements repo we run in devstack a few times before pushing up | 23:47 |
| *** gyee has quit IRC | 23:47 | |
| jog0 | or do you do that and this snuck by the goalie | 23:48 |
| *** DinaBelova has joined #openstack-infra | 23:49 | |
| jeblair | jog0: the current thinking on that is that it is better to allow it to break because it's a real problem that needs dev attention. | 23:50 |
| jeblair | (otherwise it would be broken for everyone except us) | 23:51 |
| jog0 | jeblair: ahh but that potentially breaks trunk. what about staging and auto alert when it fails. and if no one fixes it it merges anyway | 23:51 |
| jog0 | so there is a window to fix before trunk is broken | 23:51 |
| *** DinaBelova has quit IRC | 23:53 | |
| * reed takes some time for expense report... yippie | 23:54 | |
| jeblair | jog0: trunk is broken as soon as a dependency does something that breaks it, regardless of what's in our mirror. | 23:55 |
| jog0 | jeblair: good point | 23:56 |
| jog0 | ignore everthing I said in that case | 23:56 |
| jeblair | mordred: i think that graph will break graphite because the data aggregation takes too long | 23:57 |
| jog0 | so the babel thing confuses me, it says it cannot find pytz but I see it in our mirrors | 23:58 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!