Thursday, 2014-04-10

*** markmcclain has joined #openstack-infra00:02
adam_ghttps://review.openstack.org/#/c/85227/1  could also use a quick +A if anyone is feeling generous!00:02
openstackgerritDavid Pursehouse proposed a change to openstack-infra/jenkins-job-builder: Allow to specify multiple branches per project in Gerrit trigger  https://review.openstack.org/8564800:05
*** dizquierdo has quit IRC00:08
*** wchrisj has joined #openstack-infra00:11
*** thuc has joined #openstack-infra00:13
*** thuc has quit IRC00:14
*** thuc has joined #openstack-infra00:14
*** markwash has quit IRC00:15
*** msabramo has joined #openstack-infra00:15
pabelangerfungi: Ya, it seems I'm getting high timeouts when loading from a snapshot on rax, which is why my deletes are ignored.00:15
*** thuc_ has joined #openstack-infra00:15
*** lcostantino has quit IRC00:18
*** thuc has quit IRC00:19
*** thuc_ has quit IRC00:19
*** mbacchi has quit IRC00:20
*** markmcclain has quit IRC00:20
*** thuc has joined #openstack-infra00:22
*** thuc has quit IRC00:23
*** thuc_ has joined #openstack-infra00:23
*** thuc_ has quit IRC00:24
*** dcramer_ has quit IRC00:24
*** wenlock has quit IRC00:25
*** matsuhashi has joined #openstack-infra00:26
*** pdmars has quit IRC00:31
*** tomhe has quit IRC00:31
*** adalbas has quit IRC00:31
*** mrodden has quit IRC00:31
*** dcramer_ has joined #openstack-infra00:36
*** adalbas has joined #openstack-infra00:44
*** yamahata has quit IRC00:46
*** Sukhdev has quit IRC00:47
sdaguehttps://review.openstack.org/#/c/86439/ - we should merge this as soon as possible, as we let a break slip through because it wasn't on more projects00:54
*** sabari has joined #openstack-infra00:58
clarkbsdague: done00:58
* clarkb walks home00:58
sdagueclarkb: thanks00:59
openstackgerritA change was merged to openstack-infra/config: add grenade-dsvm-neutron to integrated gate  https://review.openstack.org/8643901:01
sdagueclarkb: gah, markmcclain's info about this passing doesn't look right01:01
openstackgerritSean Dague proposed a change to openstack-infra/config: make grenade-dsvm-neutron non voting  https://review.openstack.org/8648101:03
clarkbk /me reviews01:03
*** yamahata has joined #openstack-infra01:03
sdagueyeh, that just turns off voting01:04
sdaguehopefully they can fix it tomorrow then. Sorry I didn't double check that one.01:04
clarkbits ok. modern tech allows me to review from sidewalk01:04
sdague:)01:04
*** saschpe has quit IRC01:04
sdagueclark boylan, in the 21st and a half century!01:05
*** kevinbenton has joined #openstack-infra01:05
clarkbdone01:05
*** saschpe has joined #openstack-infra01:06
*** CaptTofu has quit IRC01:07
*** CaptTofu has joined #openstack-infra01:07
chmoueljeblair: here's is a errbot plugin for gerrit https://github.com/chmouel/err-gerrit-changes/blob/master/gb.py but i could not make it works with the gerritlib ssh loop and used gerrit changes 2.801:07
chmouelnot sure if via gerritlib is doable01:08
*** wchrisj has quit IRC01:08
*** yaguang has joined #openstack-infra01:10
openstackgerritMichael Krotscheck proposed a change to openstack-infra/storyboard: [WIP] Collate story metadata for status  https://review.openstack.org/8645201:11
*** CaptTofu has quit IRC01:11
openstackgerritA change was merged to openstack-infra/config: make grenade-dsvm-neutron non voting  https://review.openstack.org/8648101:12
*** dims has joined #openstack-infra01:19
*** wendar_ has joined #openstack-infra01:19
*** miarmak_ has joined #openstack-infra01:19
*** GheRivero_ has joined #openstack-infra01:19
*** SpamapS_ has joined #openstack-infra01:21
*** StevenK_ has joined #openstack-infra01:21
*** proffalk1n has joined #openstack-infra01:22
*** gaelL has quit IRC01:22
*** jeckersb` has joined #openstack-infra01:22
*** juice has quit IRC01:22
*** vipul has quit IRC01:22
*** SergeyLukjanov has quit IRC01:22
*** spiffxp has quit IRC01:22
*** ryanpetrello has quit IRC01:22
*** SpamapS has quit IRC01:22
*** jd__ has quit IRC01:22
*** proffalken has quit IRC01:22
*** StevenK has quit IRC01:22
*** jeckersb has quit IRC01:23
*** jesusaur has joined #openstack-infra01:23
*** mattoliverau has quit IRC01:23
*** antonym has quit IRC01:23
*** mordred has quit IRC01:23
*** jesusaurus has quit IRC01:23
*** timrc has quit IRC01:23
*** akscram has quit IRC01:23
*** jamielennox has quit IRC01:23
*** jesusaur is now known as jesusaurus01:23
*** bknudson has quit IRC01:23
*** wendar has quit IRC01:23
*** timrc has joined #openstack-infra01:23
*** miarmak has quit IRC01:23
*** GheRivero has quit IRC01:23
*** sileht has quit IRC01:24
*** vipuls has joined #openstack-infra01:24
*** mordred has joined #openstack-infra01:24
*** ryanpetrello has joined #openstack-infra01:24
*** spiffxp has joined #openstack-infra01:24
*** juice has joined #openstack-infra01:24
*** SergeyLukjanov has joined #openstack-infra01:24
*** StevenK_ is now known as StevenK01:24
*** jd__ has joined #openstack-infra01:24
*** mattoliverau has joined #openstack-infra01:24
*** gaelL has joined #openstack-infra01:24
*** sileht has joined #openstack-infra01:24
*** jamielennox has joined #openstack-infra01:24
*** akscram has joined #openstack-infra01:25
*** antonym has joined #openstack-infra01:25
*** bknudson has joined #openstack-infra01:25
*** yfried has quit IRC01:25
*** yfried has joined #openstack-infra01:26
*** Ryan_Lane has quit IRC01:28
*** Ryan_Lane has joined #openstack-infra01:29
*** Ryan_Lane has quit IRC01:33
*** thuc has joined #openstack-infra01:34
*** thuc has quit IRC01:39
*** CaptTofu has joined #openstack-infra01:41
*** ianw has quit IRC01:41
*** sabari has quit IRC01:42
*** ianw has joined #openstack-infra01:42
*** dcramer_ has quit IRC01:44
anteayachmouel: can I just get a link to a gif for the comment on line 39: # Crap code :p01:45
*** zehicle_at_dell has quit IRC01:46
*** nosnos has joined #openstack-infra01:48
*** SumitNaiksatam has joined #openstack-infra01:51
*** saschpe has quit IRC02:00
*** dcramer_ has joined #openstack-infra02:00
jeblairchmouel: could you just start a thread that consumes events from gerritlib and calls self.send() with them?  that's basically what the current gerritbot does02:03
*** saschpe has joined #openstack-infra02:03
jeblairchmouel: i think there are 2 main things we'd want -- to use stream-events so that it's more responsive and we don't hammer the server, and to configure it with a yaml file through code review like we do now02:04
jeblairchmouel: thanks for getting that started02:05
*** nati_ueno has quit IRC02:06
*** changbl has joined #openstack-infra02:07
*** khyati_ has quit IRC02:08
*** signed8bit has quit IRC02:10
*** amcrn has quit IRC02:17
*** rloo has quit IRC02:18
openstackgerritDavanum Srinivas (dims) proposed a change to openstack-infra/devstack-gate: Temporary HACK : Enable UCA & Build latest libvirt  https://review.openstack.org/8605702:18
*** CaptTofu has quit IRC02:20
*** zehicle_at_dell has joined #openstack-infra02:34
*** wenlock has joined #openstack-infra02:34
*** thomasem_ has quit IRC02:35
*** thomasem has joined #openstack-infra02:35
*** sabari has joined #openstack-infra02:37
*** dcramer_ has quit IRC02:38
*** harlowja is now known as harlowja_away02:39
*** harlowja_away is now known as harlowja02:39
*** zz_gondoi is now known as gondoi02:39
*** thomasem has quit IRC02:40
*** UtahDave has quit IRC02:42
*** david-lyle has joined #openstack-infra02:56
*** wchrisj has joined #openstack-infra02:58
*** wchrisj has quit IRC03:02
openstackgerritZhongyue Luo proposed a change to openstack-infra/config: Fixes git repository brower line number misalignment  https://review.openstack.org/8649503:05
*** cody-somerville has quit IRC03:07
*** jgrimm has joined #openstack-infra03:14
*** khyati_ has joined #openstack-infra03:17
*** matsuhashi has quit IRC03:22
*** nati_ueno has joined #openstack-infra03:28
*** matsuhashi has joined #openstack-infra03:32
*** CaptTofu has joined #openstack-infra03:33
*** CaptTofu has quit IRC03:38
*** wchrisj has joined #openstack-infra03:42
*** BadCub01 has quit IRC03:43
openstackgerritJoshua Hesketh proposed a change to openstack-infra/zuul: Allow zuul-server to run a merger process  https://review.openstack.org/8649803:45
*** wchrisj has quit IRC03:46
*** wchrisj has joined #openstack-infra03:51
*** aconrad has joined #openstack-infra03:56
*** asettle is now known as alex-away03:57
openstackgerritA change was merged to openstack-infra/jenkins-job-builder: Make overrideDefaultExcludes line into 3 lines  https://review.openstack.org/8643404:05
*** wchrisj has quit IRC04:06
*** nosnos has quit IRC04:09
*** chandan_kumar_ has joined #openstack-infra04:14
openstackgerritJeremy Stanley proposed a change to openstack-infra/config: Add script for counting commits  https://review.openstack.org/8650004:18
*** thuc has joined #openstack-infra04:18
*** sabari has quit IRC04:18
*** thuc_ has joined #openstack-infra04:20
StevenKfungi: The massive for loop at the start is difficult to parse04:21
*** chandan_kumar_ has quit IRC04:21
*** zhiyan_ is now known as zhiyan04:22
*** thuc has quit IRC04:23
*** yfried has quit IRC04:23
openstackgerritJoshua Hesketh proposed a change to openstack-infra/zuul: Allow zuul-server to run a merger process  https://review.openstack.org/8649804:24
*** wchrisj has joined #openstack-infra04:29
fungiStevenK: that's a quick and dirty work in progress. foundation marketing people were looking for some stats and i wanted to stick the algorithm i used somewhere so i wouldn't lose it04:29
StevenKfungi: Heh, makes sense04:30
StevenKHuh. I would have thought puppet would have pushed out my prepare_tripleo.sh change by now04:30
fungii'll add comments and stuff to it tomorrow if the method turns out to be sound04:31
*** coolsvap has joined #openstack-infra04:31
fungiStevenK: the nodepool prepare scripts are used when updating images, not when booting individual nodes04:32
StevenKAww04:32
StevenKfungi: Does that happen every night for the tripleo cloud too?04:32
fungithat change will take effect the next time new images are created. i'll see when that's happening04:33
*** gondoi is now known as zz_gondoi04:35
*** chandan_kumar_ has joined #openstack-infra04:36
*** jhesketh_ has quit IRC04:37
*** krotscheck has left #openstack-infra04:42
fungiStevenK: apparently the images in that cloud have failed to update for the past day or two04:44
fungii've started a couple shell loops actively working to keep trying to build replacement images04:45
StevenKfungi: How have they failed?04:45
fungino idea. i'm in bed on a very constrained device, and about to fall asleep (nearly 1am here)04:46
*** gyee has quit IRC04:47
*** nosnos has joined #openstack-infra04:47
fungilooks like more floating ip association failures04:47
fungiBadRequest: Error. Unable to associate floating ip (HTTP 400) (Request-ID: req-81d6d03a-4356-4327-ba33-2fffc35980e4)04:48
fungii think previously it was something to do with the template instance coming up with interfaces switched to the wrong neutron nets,,, if memory serves, lifeless filed a nova or neutron bug on it04:50
fungilooks like it tried enough times to get past that bug so you might have updated images soon04:53
StevenKfungi: Love your work. Now go to sleep? :-)04:53
*** gema has quit IRC04:54
*** thuc_ has quit IRC04:55
*** thuc has joined #openstack-infra04:56
*** wchrisj has quit IRC04:57
*** sdake has quit IRC04:57
* fungi does04:59
*** thuc has quit IRC05:00
*** jhesketh has joined #openstack-infra05:07
*** jgrimm has quit IRC05:08
*** russell_h has quit IRC05:09
*** harlowja is now known as harlowja_away05:10
*** russell_h has joined #openstack-infra05:10
*** russell_h has quit IRC05:10
*** russell_h has joined #openstack-infra05:10
openstackgerritJoshua Hesketh proposed a change to openstack-infra/zuul: Implement a simple mysql reporter.  https://review.openstack.org/6588505:17
*** msabramo has quit IRC05:17
*** msabramo has joined #openstack-infra05:18
*** ilives has joined #openstack-infra05:19
*** CaptTofu has joined #openstack-infra05:21
*** gema has joined #openstack-infra05:23
lifelessfungi: arosen1 fixed it in nova... but the fix broke other ways of requeseting similar things.05:25
*** kevinbenton has quit IRC05:25
*** marktraceur has quit IRC05:25
lifelessfungi: we're running the initial patch of arosen1 's that landed05:25
*** ilives has quit IRC05:25
*** ilives has joined #openstack-infra05:25
*** CaptTofu has quit IRC05:25
*** marktraceur has joined #openstack-infra05:26
*** aconrad has quit IRC05:28
*** kevinbenton has joined #openstack-infra05:30
*** chandan_kumar_ has quit IRC05:31
*** jhesketh has quit IRC05:36
*** jhesketh has joined #openstack-infra05:36
*** denis_makogon has joined #openstack-infra05:37
openstackgerritJoshua Hesketh proposed a change to openstack-infra/zuul: Implement a simple mysql reporter.  https://review.openstack.org/6588505:40
*** _nadya_ has joined #openstack-infra05:42
*** ihrachyshka has joined #openstack-infra05:45
openstackgerritJoshua Hesketh proposed a change to openstack-infra/zuul: Implement a simple mysql reporter.  https://review.openstack.org/6588505:46
*** ilives has quit IRC05:47
*** ilives has joined #openstack-infra05:47
*** ilives has quit IRC05:49
*** melwitt has quit IRC05:50
*** skraynev_afk is now known as skraynev05:50
openstackgerritJoshua Hesketh proposed a change to openstack-infra/zuul: Implement a simple mysql reporter.  https://review.openstack.org/6588505:51
*** ilives has joined #openstack-infra05:53
*** unicell has joined #openstack-infra05:55
*** _nadya_ has quit IRC06:03
*** jlibosva has joined #openstack-infra06:05
*** kevinbenton has quit IRC06:05
*** thuc has joined #openstack-infra06:06
zigolifeless: When I try to build testrepository in my jenkins, I get: AttributeError: 'module' object has no attribute 'test_slowest'06:08
zigoHowever, it does work well in my laptop, or on my sid box.06:08
*** jhesketh_ has joined #openstack-infra06:08
zigoWhat am I missing?06:08
*** ilives has quit IRC06:08
zigolifeless: Full traceback is here: http://paste.debian.net/92868/ (with the commands I run on top of it)06:09
*** wenlock has quit IRC06:10
*** ihrachyshka has quit IRC06:10
zigoOnce installed, it does work well though... it's just testrepository package own unit tests which are failing over (even if I know they should work).06:10
*** thuc has quit IRC06:11
lifelessnot sure06:13
*** yfried has joined #openstack-infra06:15
*** sdake has joined #openstack-infra06:15
*** kevinbenton has joined #openstack-infra06:16
*** markwash has joined #openstack-infra06:17
*** SpamapS_ is now known as SpamapS06:20
*** SpamapS has joined #openstack-infra06:20
*** jamielennox is now known as jamielennox|away06:24
*** _nadya_ has joined #openstack-infra06:25
*** xchu_ has joined #openstack-infra06:26
*** nati_ueno has quit IRC06:27
*** khyati_ has quit IRC06:37
*** chandan_kumar has joined #openstack-infra06:38
*** cody-somerville has joined #openstack-infra06:40
*** sdake has quit IRC06:41
openstackgerritRicardo Carrillo Cruz proposed a change to openstack-infra/jeepyb: Add change abandoned logic to jeepyb  https://review.openstack.org/8638206:48
*** ilives has joined #openstack-infra06:55
*** _nadya_ has quit IRC06:55
*** afazekas has joined #openstack-infra06:56
*** turul_ has joined #openstack-infra06:56
*** turul_ has quit IRC06:56
*** dims has quit IRC06:58
*** cody-somerville has quit IRC06:58
*** hogepodge has quit IRC07:03
*** ilives has quit IRC07:06
*** ilives has joined #openstack-infra07:07
*** CaptTofu has joined #openstack-infra07:09
*** dims has joined #openstack-infra07:10
*** cody-somerville has joined #openstack-infra07:10
*** flaper87|afk is now known as flaper8707:11
*** CaptTofu has quit IRC07:14
*** Clabbe has quit IRC07:16
*** hashar has joined #openstack-infra07:17
*** hogepodge has joined #openstack-infra07:18
*** afazekas has quit IRC07:20
zigolifeless: Not even a clue or a direction where I could investigate?07:20
zigoBTW, we now have a python3-testrepository package if you didn't notice! :)07:20
zigoIt was required by PBR 0.807:21
*** proffalk1n has quit IRC07:21
*** oomichi has joined #openstack-infra07:21
*** proffalken has joined #openstack-infra07:22
*** oomichi is now known as ken1ohmichi07:22
*** proffalken has quit IRC07:22
*** proffalken has joined #openstack-infra07:24
*** jgallard has joined #openstack-infra07:26
*** zigo has quit IRC07:29
mattoliveraunight all07:31
*** tnurlygayanov has quit IRC07:32
*** tnurlygayanov has joined #openstack-infra07:33
*** ildikov has quit IRC07:34
*** zigo has joined #openstack-infra07:35
*** msabramo has quit IRC07:35
*** jcoufal has joined #openstack-infra07:35
*** ihrachyshka has joined #openstack-infra07:37
*** julim_WFH has quit IRC07:38
*** arborism has joined #openstack-infra07:40
*** arborism is now known as amcrn07:40
*** xchu_ has quit IRC07:40
*** e0ne has joined #openstack-infra07:40
*** matrohon has joined #openstack-infra07:41
*** julim has joined #openstack-infra07:41
*** amotoki has joined #openstack-infra07:43
*** afazekas has joined #openstack-infra07:44
*** ildikov has joined #openstack-infra07:46
*** ihrachyshka has quit IRC07:48
*** ihrachyshka has joined #openstack-infra07:48
*** reed has joined #openstack-infra07:49
*** jp_at_hp has joined #openstack-infra07:51
*** e0ne has quit IRC07:51
*** ihrachyshka has quit IRC07:52
*** ihrachyshka has joined #openstack-infra07:53
*** e0ne_ has joined #openstack-infra07:53
*** Grishkin has joined #openstack-infra07:54
*** ildikov_ has joined #openstack-infra08:01
*** e0ne has joined #openstack-infra08:02
openstackgerritMike Scherbakov proposed a change to openstack-infra/reviewstats: Added fuel-core team  https://review.openstack.org/8651908:03
*** ildikov has quit IRC08:04
*** e0ne_ has quit IRC08:05
*** Grishkin has quit IRC08:08
*** che-arne has quit IRC08:08
*** skolekonov has joined #openstack-infra08:09
*** andreaf has joined #openstack-infra08:12
*** rallegre has joined #openstack-infra08:14
*** jpich has joined #openstack-infra08:17
openstackgerritNikita Konovalov proposed a change to openstack-infra/storyboard: Hard delete for auth stuff  https://review.openstack.org/8627108:20
*** matsuhashi has quit IRC08:21
*** ildikov has joined #openstack-infra08:21
*** zigo has quit IRC08:21
*** matsuhashi has joined #openstack-infra08:21
*** zigo has joined #openstack-infra08:23
*** ildikov_ has quit IRC08:23
*** matsuhashi has quit IRC08:25
*** derekh has joined #openstack-infra08:26
*** markmc has joined #openstack-infra08:27
*** bogdando has quit IRC08:29
*** bogdando has joined #openstack-infra08:30
*** denis_makogon has quit IRC08:31
*** jooools has joined #openstack-infra08:32
*** ruhe has quit IRC08:32
*** NikitaKonovalov has quit IRC08:32
*** SergeyLukjanov has quit IRC08:32
*** DinaBelova has quit IRC08:32
*** DinaBelova has joined #openstack-infra08:33
*** NikitaKonovalov has joined #openstack-infra08:34
*** ruhe has joined #openstack-infra08:35
*** SergeyLukjanov has joined #openstack-infra08:35
openstackgerritNikita Konovalov proposed a change to openstack-infra/storyboard: Hard delete for auth stuff  https://review.openstack.org/8627108:37
*** matsuhashi has joined #openstack-infra08:37
*** yjiang has joined #openstack-infra08:54
openstackgerritZhenguo Niu proposed a change to openstack-infra/config: Add django_openstack_auth to the integrated gate  https://review.openstack.org/8652808:57
*** CaptTofu has joined #openstack-infra08:58
*** rallegre has quit IRC09:00
openstackgerritMike Scherbakov proposed a change to openstack-infra/reviewstats: Added fuel-core team  https://review.openstack.org/8651909:00
*** CaptTofu has quit IRC09:03
*** morganfainberg is now known as morganfainberg_Z09:03
*** dizquierdo has joined #openstack-infra09:05
reedttx, if I wanted to get the programs.yaml for Icehouse only, is there a tag in git for it?09:06
reedor an older version before the Juno resolutions for new incubation/graduation?09:06
ttxreed: you could extract an older version yes. But ideally we would record history in the file. Anne started to address that @ https://review.openstack.org/#/c/81859/09:08
reedttx,  thanks, that new format seems messy09:09
ttxyes, still being worked on09:09
openstackgerritNikita Konovalov proposed a change to openstack-infra/storyboard: Hard delte for tasks  https://review.openstack.org/8653109:09
ttxdifficult to use a file both as a reference for current and as history09:09
ttxbut we really need some reference for that09:09
reedi'm adding myself to the review09:10
ttxusing tags won't be convenient either09:10
reedare you considering the possibility that projects can be incubated for more than one cycle?09:10
ttxthey can and they are09:10
reedttx, why not using tags?09:10
ttxtags make it difficult to see "when a project was incubated" and "when a project was integrated" which is the question everyone asks09:11
ttxyou basically have to bisect on tags to answer09:11
* ttx checks for a version that would be "icehouse"09:12
*** mihgen has joined #openstack-infra09:13
ttxreed: the closest would be http://git.openstack.org/cgit/openstack/governance/tree/reference/programs.yaml?id=8895283847717bdce000b89dda908a70432491ce09:14
ttxreed: although gantt should probably be removed from that09:14
ttxit's a fork of nova so keeping it in results in counting nova commits twice09:14
ttxand then it was abandoned.09:14
ttxtricky huh09:15
ttxmaybe we should have stable branches for the governance :)09:15
*** david-lyle has quit IRC09:16
*** ilives has quit IRC09:18
reedindeed :)09:18
reedwhat is gantt anyway? so many projects :)09:19
*** ilives has joined #openstack-infra09:19
*** andreaf has quit IRC09:22
reedttx, the list of repositories currently being used for activity board stats http://paste.openstack.org/show/75452/09:24
* ttx looks09:25
*** glenng1 has quit IRC09:26
*** glenng has joined #openstack-infra09:26
*** mrda is now known as mrda_away09:27
reedi need to print things to cross check them :(09:28
ttxreed: doesn't look completely wrong. It's missing a few projects, like oslo.rootwrap. it's also picky with what -infra projects it considers09:29
reedI can have them add them09:29
ttxfun to see openstack-planet in there but not the governance repo09:29
ttxso it's just a bit outdated I think09:29
*** matsuhashi has quit IRC09:30
ttxfwiw the list at programs.yaml is regularly compared with the actual repo list we have09:30
reedyep, indeed09:30
* ttx has a test-projects-yaml.py to compare them09:31
reedttx, the dashboard gives 6683 commits for the integrated projects only up to yesterday for icehouse09:32
reedsimilar number for stackalytics09:32
ttxbeware stackalytics counts sahara as integrated09:32
reedoh09:32
reedalready?09:33
ttxreed: well, it looks at programs.yaml09:33
ttxand the time infirmation isn't there yet :)09:33
reedbut maybe they have another way to track the time?09:33
*** matsuhashi has joined #openstack-infra09:34
*** rallegre has joined #openstack-infra09:34
ttxI asked Sergey, and apparently they have only one list at a time, not one list per cycle09:34
ttxon the bright side, hat means the list is the canonical list09:34
ttxBUT that means the project splits between incubated/integrated/other might not be relevant for the cycle under consideration09:35
ttxas soon as we add timing info, they will make use of it09:35
reedI think a stable branch should be used as reference09:37
reed not a tag but a git branch?09:38
ttxmaybe. trick is the rest of the governance repo doesn't really care about cycles09:38
ttxso the stable branch only makes sense for programs.yaml09:38
ttxbut then maybe it's worth it09:39
*** zhiyan is now known as zhiyan_09:39
reedbranch only the file? :)09:40
reednew feature for git09:40
ttxanyway, for our immediate need, crafting relevant programs.yaml file is probably the easiest09:41
reedyep09:41
ttxlet me publish that somewhere. For havana and Icehouse09:42
*** reed has quit IRC09:45
openstackgerritPavlo Shchelokovskyy proposed a change to openstack/requirements: troveclient 1.0.4 released  https://review.openstack.org/8654209:46
*** ociuhandu has quit IRC09:47
*** hashar has quit IRC09:54
*** ominakov has joined #openstack-infra09:56
*** reed has joined #openstack-infra10:07
*** miarmak_ has quit IRC10:10
*** miarmak has joined #openstack-infra10:10
*** sahumada has quit IRC10:12
*** sahumada has joined #openstack-infra10:14
*** ilives has quit IRC10:16
*** ilives has joined #openstack-infra10:17
reedttx, am I right that some of the infra and oslo repositories are forks?10:18
ttxreed: some of oslo repos are partial forks from oslo-incubator (with cleaned up commits and history)10:19
ttxreed: infra... any example ?10:19
reedgerrit?10:19
reedon oslo, none of those libs are forks from pypi?10:19
reedon infra, lodgeit?10:20
*** matsuhashi has quit IRC10:20
ttxyes, gerrit repo is definitely a fork -- wonder if we've been inviting all gerrit devs to summits lately :)10:20
*** aconrad has joined #openstack-infra10:21
ttxoslo, no. Pypi is not a git repo10:21
ttxwe publish there.10:21
ttxmy head hurts10:21
reedttx, I know that :) I meant to say if on oslo we keep forks of libs that are also published on pypi10:21
reedrecent discussions on the -dev list made me believe we have such things10:22
ttxreed: we.. shouldn't ?10:22
ttxoslo libs are oslo libs.10:22
reedi'm not saying we should or not, that's not for me to even get close to such discussion :)10:22
ttxwe have adopted a number of projects but those usually live in stackforge10:22
ttxand not under oslo10:23
reedi need to understand if by counting commits under the oslo repositories we count commits done by people not working on openstack10:23
reedlike we would if we counted the commits to gerrit on the infra repository10:23
ttxreed: I don't think we do. But the gererit fork being listed under the infra program is definitely an issue imho10:23
ttxoslo libraries are original and openstack. Not forks.10:24
reedcool, thanks for clarifying that10:24
*** matsuhashi has joined #openstack-infra10:25
reedttx, on infra, meetbot I think it's a fork too10:26
*** tnurlygayanov has quit IRC10:26
ttxreed: looks like it yes10:27
* ttx lunches10:27
reedI need to go to get something to eat for my dad10:28
*** CaptTofu has joined #openstack-infra10:33
*** kmartin has quit IRC10:33
*** rallegre has quit IRC10:33
*** tnurlygayanov has joined #openstack-infra10:35
*** CaptTofu has quit IRC10:36
*** CaptTofu has joined #openstack-infra10:37
*** mrda_away is now known as mrda10:40
*** zhiyan_ is now known as zhiyan10:41
*** alexpilotti has joined #openstack-infra10:43
*** reed has quit IRC10:45
*** jgallard has quit IRC10:49
*** andreaf has joined #openstack-infra10:50
*** miqui has quit IRC10:51
*** e0ne has quit IRC10:52
*** e0ne has joined #openstack-infra10:53
*** e0ne has quit IRC10:57
*** nosnos has quit IRC10:59
*** nosnos has joined #openstack-infra10:59
*** nosnos has quit IRC11:03
*** yjiang has quit IRC11:09
openstackgerritNikita Konovalov proposed a change to openstack-infra/storyboard: Logging configuration improvement  https://review.openstack.org/8655811:11
markmcttx, summit.openstack.org bug report: "Incorrect project name: oslo.messaging" :)11:12
markmcttx, in the blueprints field11:12
*** coolsvap is now known as coolsvap|afk11:12
*** reed has joined #openstack-infra11:13
sdaguettx when you get back, this requirements shuffle looks ready - https://review.openstack.org/#/c/86171/11:14
*** matsuhashi has quit IRC11:14
sdaguethat should let us decouple so that hacking can control it's pyflakes/pep8 version directly, which is where we want to be headed11:15
*** mburned is now known as mburned_out11:18
*** ociuhandu has joined #openstack-infra11:22
*** matsuhashi has joined #openstack-infra11:22
*** ociuhandu has quit IRC11:23
*** mkerrin has quit IRC11:25
*** YorikSar has joined #openstack-infra11:25
YorikSarHello, everyone. Can someone help me with devstack-gate instructions?11:27
YorikSarI'm trying to follow README there.11:27
YorikSarI've got Puppet::Parser::AST::Resource failed with error ArgumentError: Could not find declared class openstack_project::single_use_slave at line 2 on node testserver.openstacklocal11:27
YorikSarWhat could go wrong?11:29
rcarrillocruzlooks like you are missing some module11:29
YorikSarBut it's there, it's in /root/config/modules/openstack_project/manifests/single_use_slave.pp11:31
*** matsuhashi has quit IRC11:31
*** lcostantino has joined #openstack-infra11:31
rcarrillocruzcheck if that folder is being passed to puppet with the --modulepath flag11:31
YorikSarIt does11:32
*** mkerrin has joined #openstack-infra11:32
rcarrillocruzyou may have it, but puppet may now know about it11:32
YorikSarI've copied the command to run from README.rst. There's this call:11:33
*** matsuhas_ has joined #openstack-infra11:33
YorikSarpuppet apply --modulepath=/root/config/modules:/etc/puppet/modules -e "class { openstack_project::single_use_slave: install_users => false, ssh_key => \"$( cat .ssh/authorized_keys )\" }"11:33
*** Ryan_Lane has joined #openstack-infra11:36
*** CaptTofu has quit IRC11:38
ttxsdague: +2ed11:39
ttxmarkmc: will look into it11:39
*** Ryan_Lane has quit IRC11:40
*** dims has quit IRC11:42
*** mbacchi has joined #openstack-infra11:43
*** reed has quit IRC11:43
markmcttx, cool, thanks11:44
*** mrda is now known as mrda_away11:44
*** rfolco has joined #openstack-infra11:44
*** e0ne has joined #openstack-infra11:46
*** OmPrakashPandey has joined #openstack-infra11:47
*** e0ne has quit IRC11:48
*** e0ne has joined #openstack-infra11:49
*** e0ne has quit IRC11:53
*** yaguang has quit IRC11:54
*** matsuhas_ has quit IRC11:57
*** rallegre has joined #openstack-infra11:59
*** e0ne has joined #openstack-infra12:00
*** adalbas has quit IRC12:01
*** mburned_out is now known as mburned12:01
*** Clabbe has joined #openstack-infra12:02
*** miqui has joined #openstack-infra12:04
SergeyLukjanovttx, sdague, https://review.openstack.org/#/c/86171/ lgtm, are we waiting for someone else to review?12:04
sdagueSergeyLukjanov: I don't think so12:05
sdagueclarkb might have useful feedback, but he's pretty buried with cred reissue12:06
sdagueI think in the worst case, the mirror-requirements doesn't work like I think it does12:06
sdaguebut it shouldn't hurt anything else12:06
openstackgerritRicardo Carrillo Cruz proposed a change to openstack-infra/jeepyb: Remove --is-draft flag parsing  https://review.openstack.org/8657012:06
*** yfried_ has joined #openstack-infra12:08
*** yfried has quit IRC12:09
SergeyLukjanovsdague, re mirror, all removed dependecies are transient througth the hacking12:09
*** jgallard has joined #openstack-infra12:09
SergeyLukjanovsdague, but I have a small concern re projects.txt12:09
*** weshay has joined #openstack-infra12:09
*** ociuhandu has joined #openstack-infra12:09
*** yfried__ has joined #openstack-infra12:10
SergeyLukjanovsdague, hacking is listed in this file, so, we'll sync removal of pep8/flake8 to hacking12:10
*** pdmars has joined #openstack-infra12:10
sdagueSergeyLukjanov: right, once we have this, jogo will remove hacking from projects.txt12:11
*** vponomaryov has quit IRC12:11
SergeyLukjanovsdague, okay, so, I'm ok with it, approving12:13
*** yfried_ has quit IRC12:13
*** vponomaryov has joined #openstack-infra12:14
*** hashar has joined #openstack-infra12:14
*** aconrad has quit IRC12:15
*** CaptTofu has joined #openstack-infra12:15
openstackgerritNikita Konovalov proposed a change to openstack-infra/storyboard-webclient: Navigate to story details after create complete  https://review.openstack.org/8657112:16
*** yfried_ has joined #openstack-infra12:19
*** yfried__ has quit IRC12:21
*** yfried_ has quit IRC12:24
*** yfried_ has joined #openstack-infra12:25
*** dims has joined #openstack-infra12:35
*** yfried__ has joined #openstack-infra12:36
*** Ryan_Lane has joined #openstack-infra12:37
*** yfried_ has quit IRC12:39
*** dkranz has joined #openstack-infra12:39
*** Ryan_Lane has quit IRC12:41
*** msabramo has joined #openstack-infra12:42
*** CaptTofu has quit IRC12:42
msabramohashar: when you get a chance (thanks for all your reviews!): https://review.openstack.org/8558012:44
hasharmsabramo: hi :-)12:45
msabramohashar: oh and https://review.openstack.org/84778 :-)12:45
hasharI do get email for any activity happening on JJB12:45
*** rlandy has joined #openstack-infra12:45
msabramoheading to PyCon today - you going?12:45
hashardid too much review this week though, will throttle till monday12:45
hasharunless it happens in my city I am not going :]12:45
msabramohashar: ok, understood12:46
hasharI think James E. Blair is attending12:46
hasharyou should definitely try to get in touch with him12:46
msabramoyeah was going to drop in on the Openstack session on Saturday12:46
msabramowhat city are you in?12:47
hasharNantes, France12:47
hasharon the south of Britanny, the last big city along the Loire Valley12:47
msabramoah, nice12:47
hasharit probably has weather similar to Seattle12:48
msabramorainy?12:48
hasharand although its is the 5th or 6th biggest city in France, it is only 550k people including the urban area12:48
hasharit is not that rainy, we have sun everyday12:48
hashar(between waves of clouds)12:48
msabramoah ok, Seattle is famous for rain12:49
sdaguefungi, you awake yet?12:49
hasharclimate listed at http://en.wikipedia.org/wiki/Nantes#Geography12:49
msabramohashar: really appreciate your comments on my pull requests; you deserve a break! :-)12:49
hasharI which we had some interface to compare climates between different cities around the wolrd12:49
sdagueI want to bounce ideas needed by - https://review.openstack.org/#/c/86577/1/specs/branchless-tempest.rst12:49
*** rallegre has quit IRC12:49
msabramoI would go to PyCon if it were in Nantes!12:50
sdagueespecially the devstack gate pieces12:50
hasharmsabramo: I am having a bad time right now wearing sunglasses on my terrace and trying to debug a Gearman bug I got with Zuul/Jenkins :]12:50
fungisdague: what's up?12:50
*** dprince has joined #openstack-infra12:50
msabramohashar: at least Montreal has tons of French speakers… :-)12:51
sdaguein order for use to do branchless tempest we're going to need much more granular feature selection in devstack gate12:51
sdagueso I was thinking we should take that piece out of functions.sh and make it a python program that reads a yaml file12:52
msabramothough I understand from my French friend/colleague, that they have a weird accent/dialect12:52
*** glenng has quit IRC12:52
*** rallegre has joined #openstack-infra12:52
hasharmsabramo: they have a government agency that find out french equivalent for english names. So instead of "email" they use the very nice "courriel"  (short for "Courrier electronique" or "Electronic Mail").12:53
msabramohashar: oh yes, I think I heard that12:53
hasharmsabramo: and their slang is slightly different.12:53
*** yamahata has quit IRC12:54
hasharmsabramo: ok reviewing your browse and test patches to jjb12:55
*** che-arne has joined #openstack-infra12:56
msabramohashar: thanks, I'll have time on the plane to implement your suggestions :-)12:57
*** thuc has joined #openstack-infra12:57
fungisdague: having some more clarity around devstack setup for a given test branch would be a plus. i expect extracting that and making it a little more intelligent could help in that regard. still a little iffy on switching bits of devstack-gate from shell to python though (the last time we did that, we moved them to a new project named nodepool)12:57
*** thuc_ has joined #openstack-infra12:58
sdaguethis would be a call out just for the feature grid12:58
fungireed: ttx: we don't count commits from upstream gerrit in the atc generation script... that uses a query against our own gerrit looking for approval votes (it doesn't analyze the git repos directly)12:58
ttxfungi: I was assuming something of the like, yes :)12:59
sdaguefungi: because I think the logic for doing that in shell is going to get nuts12:59
fungisame for meetbot, for the same reasons12:59
sdaguebecause we're going to need supported extensions for each service, by each config, by each branch13:00
fungisdague: sounds like a well-defined scope, we'd just probably want to make sure it didn't become a gateway to more heavily hybridizing d-g into multiple languages for no good reason13:00
hasharmsabramo: I am afraid you will have to sleep on the plane, got too many pings from coworkers right now :-/13:01
*** ihrachyshka has quit IRC13:02
fungiYorikSar: the readme in devstack-gate has bitrotted slightly and doesn't take a recent puppet refactoring into account. i've got something i'm using which takes advantage of our nodepool image prep scripts instead... i'll hurry and post a review for that and give you the link13:02
*** thuc has quit IRC13:02
*** ken1ohmichi has quit IRC13:02
sdaguefungi: yeh, the slicing out is going to get a little messy, but I think it's doable. Basically extract this logic: https://github.com/openstack-infra/devstack-gate/blob/master/devstack-vm-gate.sh#L36-L24213:03
sdagueanyway, should chat with jeblair on it as well13:04
sdaguehe at pycon this week?13:04
*** ameade has joined #openstack-infra13:04
*** mfer has joined #openstack-infra13:04
*** nkinder has quit IRC13:05
fungihe is, but has been around before/after/between sessions13:06
*** thomasem has joined #openstack-infra13:09
*** mriedem has joined #openstack-infra13:09
*** thuc has joined #openstack-infra13:09
*** eharney has joined #openstack-infra13:10
*** jeckersb` is now known as jeckersb13:10
sdaguecool13:11
sdaguesadly not there this year, next year I plan to though13:11
*** thuc_ has quit IRC13:12
fungii think his main reason for attending was that the python project is talking about adopting some of our infrastructure solutions (or at least workflows) and he wanted to be around to provide ready answers when questions arose13:12
sdaguecool13:14
fungiwhich i believe was going on more at the language summit than the conference proper, but not sure13:14
*** reed has joined #openstack-infra13:15
*** ogelbukh has quit IRC13:15
*** rallegre has quit IRC13:15
*** ihrachyshka has joined #openstack-infra13:16
*** ogelbukh has joined #openstack-infra13:16
*** jgallard has quit IRC13:18
*** rallegre has joined #openstack-infra13:22
jeblairtoday is about followup from yesterday  :)13:26
fungijeblair: today is always a combination of followup from yesterday and preparation for tomorrow... it's the ratio which makes each day distinct and memorable ;)13:27
SergeyLukjanov:)13:27
SergeyLukjanovjeblair, how is pycon?13:28
*** adalbas has joined #openstack-infra13:31
*** zz_gondoi is now known as gondoi13:31
*** saschpe has quit IRC13:32
*** yamahata has joined #openstack-infra13:33
jeblairSergeyLukjanov: good, i've been having some great discussions with people13:33
*** signed8bit has joined #openstack-infra13:34
*** wchrisj has joined #openstack-infra13:34
*** signed8bit has joined #openstack-infra13:34
SergeyLukjanovjeblair, cool, /me would like to participate it someday13:35
*** msabramo has quit IRC13:35
SergeyLukjanovjeblair, is it located in one place each time?13:35
SergeyLukjanovor there is a chance to see it in Moscow :)13:35
openstackgerritA change was merged to openstack/requirements: add mirror requirements list  https://review.openstack.org/8617113:37
*** saschpe has joined #openstack-infra13:37
rcarrillocruzhey guys, anyone willing to look at https://bugs.launchpad.net/openstack-ci/+bug/1305801 ? I filed it myself, I think the --is-draft is not used at all...13:40
uvirtbotLaunchpad bug 1305801 in openstack-ci "update_bug.py from jeepyb parses for --is-draft option that is never passed by Gerrit" [Undecided,In progress]13:40
*** zehicle_at_dell has quit IRC13:40
*** wchrisj has quit IRC13:41
rcarrillocruzand now that i check...it's not just update_bug, but all jeepyb/cmd scripts...13:41
*** thuc has quit IRC13:42
fungircarrillocruz: possibly. those hooks are mostly based off the hook api docs for gerrit 2.4, and keep in mind that particular script is called by more than one type of hook13:42
*** thuc has joined #openstack-infra13:42
fungiand different gerrit hooks pass different options into them13:42
rcarrillocruzi linked a change that removes it...13:42
rcarrillocruzper the gerrit docs, they are not passed at all13:43
rcarrillocruzand checked it on my machine as well13:43
fungicool--thanks!13:43
rcarrillocruzshould I remove all --is-draft from all cmds?13:43
rcarrillocruzor open different bugs for each?13:43
rcarrillocruzi opened it against update_bug.py and just noticed it happens to other scripts13:44
*** wchrisj has joined #openstack-infra13:44
*** dkliban has joined #openstack-infra13:44
fungircarrillocruz: it might help to identify first why it was there... unfortunately git blame for that is likely to go back before it was split out of openstack-infra/config into its own project13:44
fungircarrillocruz: but perhaps a little research in the gerrit 2.1-2.3 api docs could highlight a time when it was used, and explain why we had it there13:44
rcarrillocruzworking on https://bugs.launchpad.net/openstack-ci/+bug/1078231 when i noticed that main method was parsing it...13:45
uvirtbotLaunchpad bug 1078231 in openstack-ci "Draft shouldn't update Launchpad before being Published" [Low,Confirmed]13:45
rcarrillocruzwill do thx13:45
fungircarrillocruz: as long as we have a clear reason like "this was used by gerrit 2.2 but dropped in 2.3" or something, then i have no qualms with removing it13:45
*** wchrisj has quit IRC13:46
*** esker has joined #openstack-infra13:47
*** thuc has quit IRC13:47
rcarrillocruzlet's see what i can find13:47
*** saju_m has joined #openstack-infra13:50
*** aconrad has joined #openstack-infra13:51
openstackgerritAntoine Musso proposed a change to openstack-infra/gear: Use Job repr instead of handle in Client debug log  https://review.openstack.org/8659613:53
*** CaptTofu has joined #openstack-infra13:54
openstackgerritSergey Lukjanov proposed a change to openstack-infra/devstack-gate: Prepare for Juno dev  https://review.openstack.org/8307513:54
openstackgerritSergey Lukjanov proposed a change to openstack-infra/devstack-gate: Enable sahara as default in gate  https://review.openstack.org/8307613:55
SergeyLukjanovsdague, ^^13:55
YorikSarfungi: No, it's cool. I've used basic Ubuntu cloud image and after "sudo -s" had ~ pointing to /home/ubuntu instead of /root.13:57
YorikSarfungi: After I recreated VM and did everything in /root, it worked fine.13:57
*** nkinder has joined #openstack-infra13:58
sdagueSergeyLukjanov: still need to wait for the branches :)13:58
*** msabramo has joined #openstack-infra13:58
fungiYorikSar: okay, great. nevertheless, keep an eye on that readme since i expect it to improve soon by borrowing more from our nodepool image prep scripts, and hopefully in the not-too-distant future it might even point to downloadable nightly images13:58
msabramojeblair: maybe see you at PyCon13:59
fungiYorikSar: (once we get glance integration going in nodepool)13:59
*** skolekonov has quit IRC13:59
msabramojeblair: can hack on python-jenkins? https://code.launchpad.net/~msabramo/python-jenkins/python3/+merge/21466413:59
*** jlibosva has quit IRC14:00
YorikSarfungi: And I'm thinking about wrapping it into a script that would provide some shiny CLI in terms "please run that job on my private cloud"14:00
hasharmsabramo: I think i will end up asking to get python-jenkins to be migrated to Gerrit if original author dont mind :]14:01
*** jaypipes has joined #openstack-infra14:01
*** dkliban has quit IRC14:01
*** jlibosva has joined #openstack-infra14:01
msabramohashar: oh I would love it if python-jenkins were on GitHub instead of launchpad14:01
msabramo+1 +1 +114:02
hasharthat is merely because I am too lazy to learn bzr :-(14:02
msabramohashar: yep, me too14:02
*** pdmars has quit IRC14:02
*** pdmars_ has joined #openstack-infra14:02
hasharalso I think jeblair is attending pycon, if so is review queues is growing faster than he can proceed it hehe14:03
*** rallegre has quit IRC14:04
*** markmcclain has joined #openstack-infra14:04
SergeyLukjanovsdague, I know, /me just remembered to update them :)14:04
*** vhoward has left #openstack-infra14:04
jamespagemsabramo, hashar: I'd +1 that as well - even better if we can get gerrit lined up for reviews as well :-)14:07
hashar+ openstack CI system :-]14:07
hasharthe little gem is being used by Jenkins Job Builder and NodePool (a system to automatically create slaves)14:07
*** e0ne has quit IRC14:07
hasharmailing the openstack-infra list14:08
*** e0ne has joined #openstack-infra14:08
openstackgerritFlorian Preinstorfer proposed a change to openstack-infra/jenkins-job-builder: Add support for the Jenkins Doxygen publisher plugin.  https://review.openstack.org/8660414:08
msabramoI have migrated from Google Code to GitHub; never from Launchpad14:08
*** jgallard has joined #openstack-infra14:09
fungiall of early openstack migrated from lp+bzr to git repositories14:09
fungii believe'14:10
fungi(i was not around for that... before my time)14:10
*** sandywalsh has quit IRC14:10
msabramohttp://blog.jozilla.net/2012/02/16/moving-my-code-from-launchpad-to-github/14:10
*** e0ne has quit IRC14:11
*** e0ne_ has joined #openstack-infra14:11
*** rallegre has joined #openstack-infra14:11
*** freyes has joined #openstack-infra14:13
YorikSarfungi: It was just GitHub at first.14:13
*** msabramo1 has joined #openstack-infra14:13
*** prad has joined #openstack-infra14:13
YorikSarAnd then all that Gerrit and stuff came.14:14
fungisure, but those were still git repositories14:14
jeblairjamespage: would you like to move python-jenkins to openstack's gerrit?  if so, i'd be happy to do so.  :)14:14
jeblairYorikSar: nope, we moved from launchpad to gerrit14:14
jamespagejeblair, I'd be happy for that to happen yes14:14
hashar\O/14:14
* hashar reached the nirvana of opensource awesomeness14:15
YorikSarjeblair: Yes, I mixed them up.14:15
jeblairjamespage: ok, will do, probably next week since i'm at pycon14:15
*** msabramo has quit IRC14:15
hasharjeblair: should I poke openstack-infra about the migration of python-jenkins ?14:15
hasharjeblair: I was about to send an email to poke everyone about it14:15
jeblairhashar: i'm just planning to do it :)14:17
* hashar discards email14:17
msabramo1:-)14:17
hasharI think Node Pool is extending python-jenkins with the ability to add/remove slaves14:18
hasharwould be nice to have that part "upstreamed" back in python-jenkins14:18
jeblairjamespage: any chance you are javacruft on pypi?  if so, can you add 'openstackci' as an owner or maintainer to  https://pypi.python.org/pypi/python-jenkins/0.2.1 ?14:18
msabramo1I added 100% test coverage and Python 3 compat to python-jenkins14:18
jamespagejeblair, I am14:18
jeblairjamespage: cool, if you do that, we can use the release automation we have14:19
jamespagejeblair, done14:19
jeblairhashar: i have been a bad person and haven't done that;  i did merge one change and reviewed zaro's change though14:20
hasharjeblair: if we get python-jenkins under Gerrit that would probably easier to handle :]14:20
*** yamahata__ has joined #openstack-infra14:21
jeblairhashar: i'd expect so :)14:21
*** jpeeler has joined #openstack-infra14:21
*** jpeeler has quit IRC14:21
*** jpeeler has joined #openstack-infra14:21
*** sdake has joined #openstack-infra14:22
jeblairmsabramo1: would you mind if we hold your changes until after the move?14:22
msabramo1jeblair: Would I have to resubmit them on Gerritt?14:22
jeblairor should we try to merge all the pending reqs in lp?14:23
msabramo1s/Gerritt/Gerrit/14:23
msabramo1seems like it might be easier to merge pending PRs if there are not too many14:23
msabramo1that's what I did when I migrated pymssql from Google Code to GitHub because migrating the code is pretty straightforward but migrating issues and pull requests is not14:23
*** jlibosva has quit IRC14:24
jeblairthere's some automation around it, but it's not straightforward either way; the main thing is that we're likely to have more review bandwidth after the move14:26
*** yamahata__ has quit IRC14:26
*** yamahata has quit IRC14:26
jeblairhashar: let's see what we can do in the next few days14:26
*** wchrisj has joined #openstack-infra14:27
hasharsure14:27
hasharenjoy pycon jeblair :]14:27
*** jgrimm has joined #openstack-infra14:29
*** jgrimm has quit IRC14:29
*** jgrimm has joined #openstack-infra14:30
jhesketh_jeblair, clarkb: I'm about to head to bed and realise you guys are really busy (pycon et all) but if you have a chance to look at the zuul changes in review that'd be awesome :-)14:31
jeblairjhesketh_: i'll try, but realistically i won't be reviewing much until next week14:32
jhesketh_that's cool, no worries14:32
jeblairjhesketh_: is there something particularly urgent i should put at the top of the list?14:32
jhesketh_not particularly14:32
jhesketh_jeblair: in other news, I have turbo-hipster running basic jobs (like pep8) from JJB14:33
jeblairjhesketh_: !!!!! :)14:33
jhesketh_it's pretty rudimentary, I'm not really sure where we want to head with it overall14:33
jhesketh_which is something I proposed for discussion at the summit14:34
slaglehi, this override for the acls on the stable tripleo branches has merged: https://review.openstack.org/#/c/85485/, but i don't see the change in gerrit yet14:34
slagleshould i see it by now?14:34
slaglei don't see the acls updated in gerrit, i should say14:34
jeblairslagle: because the new project machinery is broken, acl updates only take effect when we run the script manually which is on fridays14:35
jeblairslagle: mordred is working on a fix to that14:36
slaglejeblair: ok, thanks14:36
msabramo1jeblair: https://github.com/jeblair/python-jenkins (empty) :-)14:36
jeblairmsabramo1: wip14:37
hasharjhesketh_: well done on jjb/pep8/turbo hipster :]14:37
hasharjhesketh_: so sorry I did not have anytime to help on that front :-(14:37
msabramo1jeblair: You just created that? It says last updated 6 months ago14:37
jhesketh_hashar: cheers, and no worries :-)14:38
jeblairmsabramo1: yes, github has a weird conception of time apparently14:38
*** Ryan_Lane has joined #openstack-infra14:39
jhesketh_I'm not sure what the roadmap to switching away from Jenkins will be.. I suspect there'll be lots of teething and bugs14:39
jhesketh_and one thing I think we need to discuss is if we really think it's something worth doing14:39
jhesketh_I'm cautious of just duplicating Jenkins and ending off no better14:39
*** wenlock has joined #openstack-infra14:39
jhesketh_especially with all this extra work going into nodepool and python-jenkins14:39
msabramo1what is turbo hipster?14:41
openstackgerritA change was merged to openstack-infra/config: Increase min-ready for f20 nodes to 5  https://review.openstack.org/8643214:41
jhesketh_msabramo1: It's a job-worker that speaks to zuul14:42
pabelangerfungi, clarkb: Do you have any thoughts about: https://bugs.launchpad.net/zuul/+bug/1302756.  Zuul reports lost jobs from gearman when it is restarted14:42
uvirtbotLaunchpad bug 1302756 in zuul "Gearman data does not persist across restarts" [Undecided,New]14:42
*** Ryan_Lane has quit IRC14:43
pabelangerI mentioned it a few weeks ago, but cannot remember who suggested a possible fix.14:43
fungipabelanger: right, the proposed "simple" fix is for zuul to just assume all jobs exist and rely on external validation of the correctness of job names in your layout.yaml14:44
*** che-arne has quit IRC14:44
*** aconrad has quit IRC14:44
fungipabelanger: otherwise zuul has to wait for your jenkins master(s) to register those jobs, which it will only do when you have slaves assigned to it which are capable of running those jobs14:44
jeblairjhesketh_: sure, we can talk about it, but we _barely_ use jenkins at this point, and there's very little to duplicate.  jenkins has an unsustainable scalabality and security model.  there are _lots_ of compelling reasons for us to drop it.14:45
jeblairjhesketh_: that doesn't invalidate the work going on in python-jenkins or elsewhere; what makes sense for us doesn't necessarily make sense for everyone14:45
jeblairi expect lots of people will continue to use jenkins with zuul14:46
pabelangerfungi, Correct, that's the catch-22 I am in.  Keep my slaves un-registered as much as possible because of the financial cost.  So, you are not opposed to some sort of config flag to possible toggle a 'simple' fix?14:46
jhesketh_sure14:46
jhesketh_as a side, it's a pity that turbo-hipster wasn't more mature before the influx of 3rd parties to openstack ci land14:47
*** chandan_kumar has quit IRC14:47
pabelangerfungi, I guess the smarter solutions, would have zuul bump nodepool somehow, to launch and register said job14:47
*** ildikov_ has joined #openstack-infra14:47
jhesketh_but yes, I do think there is good reason to work towards a replacement, I'm just cautious of where to focus as it's important not to make matters worse14:47
*** ilives has quit IRC14:47
jeblairjhesketh_: but yes, there will be teething and bugs so i think we'll want to do it the way we do most big changes if we can -- phased with both running as we spin up and wind down14:47
*** khyati has joined #openstack-infra14:48
jhesketh_yep, that's what I envisioned14:48
jhesketh_anywho, it's nearly 1am so I might get some sleep and bash it into submission some more tomorrow (ack, today)14:48
*** ilives has joined #openstack-infra14:48
fungipabelanger: yeah, i believe jeblair suggested it was good fodder for a config option, and one which we'd likely enable ourselves since we already have jobs which check that the job names in zuul's layout exist in our jjb configs14:48
jhesketh_have fun at pycon!14:48
jeblairpabelanger: yup14:49
*** OmPrakashPandey has quit IRC14:49
jeblairjhesketh_: thanks!  good night!14:49
* jhesketh_ away!14:49
*** jhesketh has quit IRC14:49
pabelangerjeblair, fungi: Okay, let me see if I can come up with something for it. THanks14:49
msabramo1jeblair, hashar: https://github.com/msabramo/python-jenkins14:50
*** ildikov has quit IRC14:50
*** markwash has quit IRC14:50
jeblairpabelanger: to clarify, i support the config option idea ; i don't think the oob poke is the way to go (the enqueue is supposed to do that, so it fundamentally would have the same problem)14:51
jeblairmsabramo1: well you beat me too it while i was talking away :) thanks14:52
*** afazekas has quit IRC14:52
pabelangerjeblair, Yup, checking out the config options first14:52
*** thuc has joined #openstack-infra14:53
msabramo1jeblair, hashar: https://github.com/msabramo/python-jenkins/pull/114:53
*** thedodd has joined #openstack-infra14:53
msabramo1but I assume you'll want to create a more official repo, but maybe you can clone from mine to make it easier14:54
openstackgerritA change was merged to openstack-infra/config: Run docs jobs on qa-specs commits  https://review.openstack.org/8543114:54
hasharmsabramo1: does that import foreign branches as well ?:D14:57
*** marun has joined #openstack-infra14:58
*** rallegre has quit IRC14:58
*** thuc has quit IRC14:58
*** dkliban has joined #openstack-infra14:59
hasharmsabramo1: I can see my bzr changes being merged that is nice14:59
*** rallegre has joined #openstack-infra14:59
msabramo1hashar: you want me to import some of your branches?15:00
*** rallegre has quit IRC15:00
hasharmsabramo1: they are single commits that got merged in. So I dont think that is needed15:00
*** thedodd has quit IRC15:00
msabramo1hashar: oh, stuff that was already merged15:01
hasharthe pending pull requests in laucnhapd can probably be copied to Gerrit and reviewed there15:01
*** thedodd has joined #openstack-infra15:01
*** rallegre has joined #openstack-infra15:01
msabramo1hashar: I don't know Gerrit or Launchpad well; I'll trust you on that one15:01
*** thuc has joined #openstack-infra15:03
*** thuc_ has joined #openstack-infra15:04
*** saju_m has quit IRC15:05
*** yfried__ has quit IRC15:07
*** thuc has quit IRC15:08
*** mwagner_lap has quit IRC15:10
*** nati_ueno has joined #openstack-infra15:10
*** skraynev is now known as skraynev_afk15:11
*** aconrad has joined #openstack-infra15:12
*** msabramo1 has quit IRC15:14
*** saschpe has quit IRC15:14
*** cody-somerville has quit IRC15:15
*** kmartin has joined #openstack-infra15:15
*** ihrachyshka has quit IRC15:15
*** coolsvap|afk is now known as coolsvap15:16
*** thuc_ has quit IRC15:18
*** thuc has joined #openstack-infra15:18
openstackgerritJames E. Blair proposed a change to openstack-infra/config: Import python-jenkins project to stackforge  https://review.openstack.org/8662715:21
jeblairi would adopt it in openstack-infra, but official projects of openstack programs with something other than an apache license is unexplored territory15:22
*** eyerediskin has joined #openstack-infra15:22
*** thuc has quit IRC15:23
fungii think that's a sane position15:23
openstackgerritMatt Riedemann proposed a change to openstack-infra/elastic-recheck: Add query for heat auth bug 1285833  https://review.openstack.org/8609315:23
uvirtbotLaunchpad bug 1285833 in heat "Keystone client racing on certificate lookups causing 401 Unauthorized on API calls" [Critical,Confirmed] https://launchpad.net/bugs/128583315:23
*** saschpe has joined #openstack-infra15:23
fungialso, as we use jenkins less still, i forsee a time in the not too distant future where the community of devs and reviewers for it don't overlap with openstack infra much15:23
*** markmcclain has quit IRC15:23
jeblairbut then again... http://git.openstack.org/cgit/openstack-infra/meetbot/tree/README.txt15:23
jeblair(that well predates the current situation)15:24
fungithat's a fork, right? is gerrit apache-licensed?15:24
jeblairfungi: yes on both15:24
fungihttp://git.openstack.org/cgit/openstack-infra/gerrit/tree/COPYING15:24
fungii guess so!15:24
fungithough what a strangely gpl-esque name for the license copy15:25
*** jcoufal has quit IRC15:25
*** nati_ueno has quit IRC15:26
eyerediskinfungi: can you pls take a look? https://review.openstack.org/#/c/86331/15:28
*** che-arne has joined #openstack-infra15:28
eyerediskinfungi: im not sure if should work like that15:28
eyerediskinfungi: devstack-checkout and then tox builders15:28
*** cody-somerville has joined #openstack-infra15:31
fungieyerediskin: no idea... have you tried running devstack-gate/devstack-vm-gate-wrap.sh with no configuration at all and then feeding those other parameters into run-tox.sh?15:32
*** pdmars_ has quit IRC15:33
fungieyerediskin: i have a feeling calling devstack-vm-gate-wrap.sh is going to do more than you want, but maybe the envvar/config default values for it are sane enough that it will work15:33
*** pdmars has joined #openstack-infra15:34
openstackgerritA change was merged to openstack-infra/elastic-recheck: Add query for heat auth bug 1285833  https://review.openstack.org/8609315:34
uvirtbotLaunchpad bug 1285833 in heat "Keystone client racing on certificate lookups causing 401 Unauthorized on API calls" [Critical,Confirmed] https://launchpad.net/bugs/128583315:34
sdaguefungi: ok, so apparently my hack in requirements didn't seem to do the thing I wanted15:36
sdaguebased on everything I read about the mirror building - https://review.openstack.org/#/c/86171/ seemed like it would pull those additional requirements into the mirror15:36
sdague(unless it's still rebuilding and new pep8 just isn't there yet)15:36
*** msabramo has joined #openstack-infra15:36
dimssdague,  thanks. Anyone else with devstack karma? please nudge this tiny fix - https://review.openstack.org/#/c/86107/ - will help me test all gates jobs with libvirt 1.2.2+/precise15:37
*** wchrisj has quit IRC15:37
sdaguedims: if dtroyer doesn't +A by the time I get back from lunch I'll hit it15:37
sdagueit's pretty straight forward15:38
dimssdague, thanks again15:38
*** imcsk8_lap has quit IRC15:38
*** pdmars has quit IRC15:38
*** imcsk8_lap has joined #openstack-infra15:38
* zul jeblair ping15:40
*** rallegre has quit IRC15:40
fungisdague: that commit message says "The current mirror process processes things that look like requirements files from global-requirements..." but i don't see how that is reflected at http://git.openstack.org/cgit/openstack-infra/pypi-mirror/tree/pypi_mirror/cmd/run_mirror.py#n25515:40
fungisdague: can you explain a little as to how you were expecting it to work?15:40
*** wchrisj has joined #openstack-infra15:41
*** basha_ has joined #openstack-infra15:42
*** dkliban has quit IRC15:42
mordredsdague: hey! - so, I had drinks with jogo last night and he was telling me about what you were working on with the mirror - have a second to chat about it?15:43
mordredI'm not convinced it's needed15:43
mordredbut I'd like to either convince you of that or get you to convince me I'm wrong :)15:43
*** msabramo has quit IRC15:43
*** rallegre has joined #openstack-infra15:44
mordredspecifically, while in theory I support the idea of reworking the mirror to be for resilience and not enforcement, it is not necessary to do that split to affect the upgrade of pep8 safely for hacking15:45
*** salv-orlando has left #openstack-infra15:45
*** dkliban has joined #openstack-infra15:45
mordredalso, pypi-mirror does a lot of work already to ensure that transitive deps are captured - so we don't need to list them explicitly15:46
dimsdtroyer, thanks!15:46
mordredthe only time we do is when there is a transitive dep of a transitive setup_requires dep because of the specific mechanics about how those get grabbed breaks15:46
*** yfried__ has joined #openstack-infra15:47
mordredthat, in particular, is no longer needed with wheels, as the consumers of our mirror will cease needing transitive deps that are only setup_requires deps15:47
*** basha_ has quit IRC15:50
openstackgerritMatt Riedemann proposed a change to openstack-infra/elastic-recheck: Add query for heat timeout bug 1306029  https://review.openstack.org/8663615:51
*** yfried__ has quit IRC15:51
uvirtbotLaunchpad bug 1306029 in heat "tempest.api.orchestration.stacks.test_stacks times out during tearDownClass" [Undecided,New] https://launchpad.net/bugs/130602915:51
zaromorning15:54
phschwartzzaro: morning15:55
*** comstud has quit IRC15:55
*** comstud has joined #openstack-infra15:55
*** krotscheck has joined #openstack-infra15:56
jeblairzul: pong (sorry, crappy hotel ethernet cable fell out :)15:56
*** e0ne_ has quit IRC15:56
*** e0ne has joined #openstack-infra15:57
*** asselin_ has quit IRC15:58
*** ihrachyshka has joined #openstack-infra15:58
*** thuc_ has joined #openstack-infra15:58
*** russellb has quit IRC15:59
*** russellb has joined #openstack-infra15:59
*** marun has quit IRC16:00
*** marun has joined #openstack-infra16:01
fungijeblair: what was the glanceclient workaround rackspace needed?16:01
fungii was spending a minute trying out python-openstackclient and realized that image list relies on glance working16:02
*** aglarendil has quit IRC16:02
jeblairfungi: it needs the glance endpoint and version specified manually16:02
*** yfried__ has joined #openstack-infra16:02
*** wchrisj has quit IRC16:03
fungiaha16:04
fungijeblair: thanks!16:04
jeblairfungi: i can dig up the specifics if you need them16:04
fungijeblair: no need, i think i can figure it out from there16:05
*** bookwar has quit IRC16:05
* mordred has it...16:05
mordred02:01:59         jeblair | glance -d --os-image-url=https://iad.images.api.rackspacecloud.com --os-image-api-version=2 image-list16:06
mordredfungi: ^^16:06
fungimordred: thanks!16:06
*** markmc has quit IRC16:06
*** jcoufal has joined #openstack-infra16:07
zigomordred: Hi there!16:08
*** atiwari has joined #openstack-infra16:08
*** yfried__ has quit IRC16:08
*** bookwar has joined #openstack-infra16:08
*** wchrisj has joined #openstack-infra16:08
mordredhey zigo16:08
tteggelis there a way, in a script, to get a definitive list of cores for a given project?16:09
tteggelgerrit ls-members not supported :(16:09
zigomordred: I was wondering if there was a way we could go ahead with moving the Debian packaging to review.openstack.org16:09
*** SumitNaiksatam has quit IRC16:09
zigomordred: I got contributors knocking at the door, and it's hard to ask to go through the old process of sending patches to a mailing list. :(16:10
mordredzigo: I'm a big fan of that idea - where's the packaging now?16:11
zigomordred: Hosted on alioth.debian.org, with a simple Git that FusionForge provides.16:12
mordredzigo: cool - I mean you gota  link?16:12
fungiyay, rackspace having a separate api endpoint for glance is basically unsupportable in openstackclient since it uses one common os_url and seems not to let you pass through api-specific endpoints16:13
fungidtroyer: ^ any suggested workarounds for that besides complain at the provider or extend the client? is there already a hidden feature i'm not finding?16:14
zigomordred: https://alioth.debian.org/projects/openstack/ You can have a look there, but there's not much to see. Otherwise, it's the usual git://anonscm.debian.org/git/openstack/<debian-package-name.git>16:15
mordredI thnk we shoudl complain to the provider - the url should be registered with keystone16:15
zigomordred: Or for example http://anonscm.debian.org/gitweb/?p=openstack/nova.git16:15
mordredzigo: awesome - i'm about to run out the door - let me look at that real quick and then come back to you with a proposed plan16:15
*** hogepodge has quit IRC16:15
*** aglarendil has joined #openstack-infra16:15
zigomordred: Ok. Though I'm about to ... go to sleep! :)16:16
dtroyerfungi: that is a weakness for sure…if you use token auth you can give it the endpoint directly.  and that's bad for all the right reasons, but the only way at present.16:16
*** hashar has quit IRC16:16
fungidtroyer: that's what i assumed--thanks for confirming. papering over it just gives providers an excuse to continute doing things like that16:16
dtroyerbut their sdk handles it perfectly!  go figure16:17
fungiyep... *grumble-grumble*16:18
*** mrodden has joined #openstack-infra16:18
mordredfungi, dtroyer: I've pinged troy over there to see what the story is16:18
tteggelnevermind: found reviewstats/projects/... which will do fine16:22
kevinbentonthere seems to be a connectivity problem with jobs in the gate16:23
*** afazekas has joined #openstack-infra16:23
kevinbentonhttps://jenkins05.openstack.org/job/gate-neutron-pep8/1108/console16:24
kevinbentonhttps://jenkins01.openstack.org/job/gate-nova-pep8/21771/console16:24
kevinbentondying on requirements installs16:24
*** SumitNaiksatam has joined #openstack-infra16:25
fungikevinbenton: we've seen lots of what seem to be network and disk issues in hpcloud az2. we're actively working to shift more of our workload out of the old hpcloud availability zones and into their next-gen regions16:26
*** e0ne has quit IRC16:26
*** david-lyle has joined #openstack-infra16:26
*** e0ne has joined #openstack-infra16:27
fungikevinbenton: that one does seem to be an instance in az2 which got an icmp/net-unreach packet back when connecting to our pypi mirror16:27
fungibut only after having already connected to it a bunch prior to that. definitely not a good sign16:27
YorikSarCan we somehow update Python on our Jenkins slaves? I'm running into a bug that doesn't seem to be easily worked around and this bug has been fixed in Python 2.7.4 while there is 2.7.3 on Jenkins.16:28
fungiYorikSar: ubuntu 12.04 lts would need the fix backported16:28
kevinbentonfungi: ah. is there a way to see the association of jenkins slave to provider?16:28
fungikevinbenton: it's encoded into the hostname, the "Building remotely on..." line at the top of the log16:29
kevinbentonfungi: oh i see it. thx16:29
YorikSarfungi: Can you point me at how can I try to push the patch there? I've never messed with Ubuntu packaging before.16:30
fungijamespage: zul: ^ ?16:30
*** thuc has joined #openstack-infra16:30
*** mwagner_lap has joined #openstack-infra16:30
fungijamespage: zul: YorikSar is inquiring about a python bugfix backport from 2.7.4 to the packages on precise16:30
*** aconrad has quit IRC16:30
kevinbentonfungi: is there a bug i can reference for this issue for the reverify statement?16:31
*** e0ne has quit IRC16:31
jamespagefungi, YorikSar: I expect that could be fixed; is there/can you raise a bug against python in Ubuntu on launchpad please16:31
YorikSarfungi, jamespage, zul: It's about http://bugs.python.org/issue605616:31
*** jpich has quit IRC16:31
jamespageYorikSar, ubuntu-bug XXX is the right was16:31
jamespageway16:32
fungikevinbenton: there may be several... the way these failures manifest is not easily classified by the errors in logs since they can impact a variety of different parts of any job16:32
kevinbentonfungi: is there a generic infra connectivity bug that you would prefer i reference?16:33
fungikevinbenton: i'm hunting16:33
*** msabramo has joined #openstack-infra16:33
YorikSarjamespage: Should I file it here? https://bugs.launchpad.net/bugs/bugtrackers/python-roundup16:33
*** OmPrakashPandey has joined #openstack-infra16:33
fungiwe have a bunch open since openstack-ci turns into a dumping ground for any job failure a dev can't figure out16:33
*** thuc_ has quit IRC16:34
jamespageYorikSar, if you can use ubuntu-bug on the server where you have the issue it gets tagged correctly and put in the right place in launchpad16:34
*** gyee has joined #openstack-infra16:34
fungikevinbenton: keeping on top of them gets challenging sometimes16:34
jamespageYorikSar, ubuntu-bug python2.716:34
kevinbentonfungi: yeah, i can imagine16:34
jamespageshould do it I think16:34
YorikSarjamespage: Ah.. Ok, I'm trying to get gate-like VM locally. I'll file the bug from it once it's ready then.16:35
*** yamahata has joined #openstack-infra16:35
fungikevinbenton: maybe bug 1272746 or bug 127599416:35
uvirtbotLaunchpad bug 1272746 in openstack-ci "Error <urlopen error [Errno 101] Network is unreachable> while getting http://pypi.openstack.org/openstack/..." [Undecided,Confirmed] https://launchpad.net/bugs/127274616:35
uvirtbotLaunchpad bug 1275994 in openstack-ci "Network is unreachable errno string reported in various ways killing pip and git operations gate/check jobs" [Medium,Expired] https://launchpad.net/bugs/127599416:35
kevinbentonfungi: ok, thanks16:36
fungiheh, looks like i reported that first one ;)16:36
*** david_lyle_ has joined #openstack-infra16:37
fungiand yep, the log i linked had "Building remotely on devstack-precise-hpcloud-az2-..."16:37
*** dklyle has joined #openstack-infra16:38
kevinbentonfungi: i’ll reference that one then :-)16:38
msabramohashar, jeblair: I know you guys want Gerrit and Jenkins and not Travis CI, but at least this illustrates that the tests are passing - https://travis-ci.org/msabramo/python-jenkins/builds/2270376716:38
*** david-lyle has quit IRC16:40
*** Ryan_Lane has joined #openstack-infra16:40
*** david-lyle has joined #openstack-infra16:40
*** markmcclain has joined #openstack-infra16:41
*** david_lyle_ has quit IRC16:41
*** david_lyle_ has joined #openstack-infra16:41
*** dklyle has quit IRC16:43
*** alexpilotti has quit IRC16:43
*** alexpilotti has joined #openstack-infra16:44
msabramow00t! jenkins/__init__     221      0   100%16:44
*** Ryan_Lane has quit IRC16:45
msabramo100% coverage makes me smile16:45
*** david-lyle has quit IRC16:45
*** fbo is now known as fbo_away16:46
*** lcostantino has quit IRC16:47
*** andreaf has quit IRC16:48
*** saju_m has joined #openstack-infra16:48
*** hogepodge has joined #openstack-infra16:53
*** markwash has joined #openstack-infra16:54
*** Ryan_Lane has joined #openstack-infra16:54
*** markmcclain has quit IRC16:55
*** derekh has quit IRC16:57
*** david_lyle_ has quit IRC16:58
*** melwitt has joined #openstack-infra16:59
*** aconrad has joined #openstack-infra17:06
*** wchrisj has quit IRC17:06
*** harlowja_away is now known as harlowja17:07
*** Sukhdev has joined #openstack-infra17:08
openstackgerritA change was merged to openstack-infra/reviewstats: Added fuel-core team  https://review.openstack.org/8651917:09
*** mrodden has quit IRC17:12
*** yongli has quit IRC17:13
*** aconrad has quit IRC17:13
*** yongli has joined #openstack-infra17:14
clarkbmorning17:15
*** eyerediskin has left #openstack-infra17:15
clarkbapparently my early start yesterday ruined todays ability to get going17:15
*** nati_ueno has joined #openstack-infra17:17
mgagneclarkb: I feel your pain =(17:17
sdaguefungi: because apparently I'm dumb and never found that script17:17
*** aconrad has joined #openstack-infra17:18
*** _nadya_ has joined #openstack-infra17:18
fungiclarkb: that often happens to me too17:18
sdaguemordred: if you don't think it's needed, that's cool.17:18
sdagueI think I was maybe operating under bad information17:19
sdagueor bad understanding on my behalf17:19
*** rallegre has quit IRC17:19
*** mihgen has quit IRC17:20
sdaguefungi: is the pypi mirror borked?17:21
sdaguehttp://logs.openstack.org/62/86562/1/gate/gate-tempest-dsvm-neutron/721a02a/logs/devstacklog.txt.gz#_2014-04-10_16_24_26_97017:21
sdaguezuul just when xmass tree style17:21
fungisdague: see discussion with kevinbenton but i think hpcloud az2 is experiencing network issues17:22
openstackgerritMathieu Gagné proposed a change to openstack-infra/jenkins-job-builder: Fix H306 errors reported by flake8/hacking  https://review.openstack.org/8644517:22
openstackgerritMathieu Gagné proposed a change to openstack-infra/jenkins-job-builder: Fix H302 errors reported by flake8/hacking  https://review.openstack.org/8644717:22
openstackgerritMathieu Gagné proposed a change to openstack-infra/jenkins-job-builder: Fix H301 errors reported by flake8/hacking  https://review.openstack.org/8644617:22
openstackgerritMathieu Gagné proposed a change to openstack-infra/jenkins-job-builder: Fix H101 errors reported by flake8/hacking  https://review.openstack.org/8644817:22
openstackgerritMathieu Gagné proposed a change to openstack-infra/jenkins-job-builder: Fix H403 and H404 errors reported by flake8/hacking  https://review.openstack.org/8644917:22
openstackgerritMathieu Gagné proposed a change to openstack-infra/jenkins-job-builder: Fix H231, H233, H902 and H703 errors reported by flake8/hacking  https://review.openstack.org/8645017:22
openstackgerritMathieu Gagné proposed a change to openstack-infra/jenkins-job-builder: Migrate to flake8/hacking  https://review.openstack.org/8645117:22
kevinbentonsdague: it looks cool :-)17:22
fungisdague: oh, yours is actually hpcloud-az1 though17:23
dhellmannare we still planning to have our own openid server at some point?17:24
*** bradm has quit IRC17:25
dhellmannI can't edit the wiki because launchpad's login service is returning 500 errors17:25
fungidhellmann: http://git.openstack.org/cgit/openstack-infra/openstackid/tree/ https://git.openstack.org/cgit/openstack-infra/config/tree/modules/openstackid/manifests/17:25
*** dizquierdo has quit IRC17:25
clarkband openstackid-dev.o.o is athing17:25
*** _nadya_ has quit IRC17:25
fungidhellmann: work in progress but i gather it's nearing production17:25
dhellmannfungi: cool17:26
fungiworking now to see whether i can correlate locations having connectivity issues to pypi.o.o17:27
fungiso far 10 out ot 10 voting failures in the gate are all nodes in legacy hpcloud zones17:29
jeblairmgagne: we ignore hacking rules in infra projects.  there is no need to fix them.17:30
mgagnejeblair: "need" no?17:30
mgagnejeblair: sorry, mistyped17:30
mgagnejeblair: "no need" ?17:30
*** jgallard has quit IRC17:30
*** OmPrakashPandey has quit IRC17:30
fungiactually, here's a few similar failures in hpcloud region b too17:31
harlowjahmmm, is the sign-in for review.openstack.org not working for anyone else (or maybe its just me)17:31
fungii'm starting to wonder if there are troubles at a peering point somewhere, which is causing communication with rackspace outside of rackspace to run into problems17:32
jeblairright.  everyone is welcome to follow the recommendations of course and we often do.  but we dont enforce it and we have enough reviews as it is so i would rather not ask people to review those kinds of changes.17:32
funginothing as of yet mentioned on their status page though17:32
jeblairmgagne: usually we just depend on hacking to get the deps and ignore H rules in the config17:34
*** dkliban has quit IRC17:34
*** dkliban has joined #openstack-infra17:34
mgagnejeblair: so my understanding is that it's a matter of human resources/time more than H rules being "not good".17:34
*** nati_ueno has quit IRC17:35
fungijust went through the several dozen job failures currently showing in the gate and every one of them was in hpcloud, none in rackspace. whether this means a network problem in one provider or the other or something more widespread in the backbone between them, hard to say at the moment17:36
*** nati_ueno has joined #openstack-infra17:36
JayFfungi: are the job failures with network issues in Jenkins on Rackspace Cloud?17:38
jeblairmgagne: also many of us disagree with the rules.  we are a small enough community that doesnt nit pick each other on reviews.  i like that about us.  ha king solves a problem that nova has but we dont.  i dont like the arguments that come up around hacking rules.17:39
fungiJayF: see my comment just above your question17:39
JayFfungi: heh. I apparently read every bit of the scrollback except for the last line.17:39
JayFfungi: was going to offer to help, but glad it's outside of my sphere of influence (aka Rackspace Cloud isn't having network issues :D)17:40
fungiJayF: if it is, looks like it's probably at your border/peering17:40
fungiJayF: since i don't see any evidence of connectivity problems between servers/tenants/regions inside rackspace17:41
sdaguefungi: have we thought about putting a mirror inside each cloud provider?17:41
sdagueto help with things like this17:41
fungisdague: i believe it has been discussed, yes17:42
*** wchrisj has joined #openstack-infra17:42
*** sdake has quit IRC17:42
jeblairfungi: pull hp from np config?17:43
fungijeblair: we could. it'll chop our quota in half, but maybe that's still better than the alternative17:44
fungiif that doesn't totally stink, i'll get a review through for it asap17:44
jeblairwhats the failure rate?17:45
fungimost changes in the gate are failing17:45
fungiso yes, slow is better than no17:45
jeblairagreed17:46
fungijeblair: best to just set the max-servers to 0, or rip all hp.* out of the config for now?17:46
jeblairmax 0 i think17:46
*** amotoki has quit IRC17:47
openstackgerritJeremy Stanley proposed a change to openstack-infra/config: Temporarily disable hpcloud in nodepool  https://review.openstack.org/8666217:50
fungijeblair: clarkb: mordred: SergeyLukjanov: ^17:50
*** yfried__ has joined #openstack-infra17:50
* bnemec stops rechecking against the pypi.o.o bug17:50
SergeyLukjanovfungi, oh...17:51
jeblairfungi: lgtm. im phoning it in clarkb style.17:52
fungijeblair: good enough ;)17:53
fungiclarkb is indeed a trend-setter17:53
*** jcoufal has quit IRC17:53
mgagnejeblair: looking at infra projects, only jeepyb and nodepool have H checks disabled. To me, disabling H checks looks more of an exception than the rule. jeepyb/nodepool are mostly developed and used by the infra team, I could understand the choice to disable them for whatever reasons. However, it's not like JJB which came to have a life of his own. So I still don't fully understand your POV on the subject.17:53
*** UtahDave has joined #openstack-infra17:54
dhellmannharlowja: I'm having trouble with the launchpad sign-in for the wiki, too17:54
harlowjadhellmann ya, i eventually got it to work, maybe launchpad is being flakey17:54
dhellmannharlowja: yeah17:54
anteayaharlowja: dhellmann I get this when I try to sign into lp https://login.launchpad.net/9Gk2hm6HXc98D616/+decide17:55
anteayaso yeah, lp signin is having a sad17:55
dhellmannanteaya: I was seeing 500 errors or their equivalent earlier, but it just worked17:55
*** morganfainberg_Z is now known as morganfainberg17:56
jeblairmgagne: i think hacking brings more harm than good to small projects.  it othen has rules added to it that people dont agree with which cause huge debates.  k disagree strongly with several rules.17:56
anteayayeah, I just kept refreshing until lp let me in17:56
anteayanow gerrit lets me in as well17:56
jeblairmgagne: "i" disagree17:57
mgagnejeblair: are you referring to yourself as the letter k ? =)17:57
clarkbfungi: jeblair wfm17:58
clarkbsorry was afk for a couple minutes there17:58
fungimgagne: it's because he lives entirely inside of franz kafka's "the trial"17:58
anteayaredundant to say it in this channel, but it is the last day to vote for the ptl elections17:58
jeblairmgagne: hacking is for projects where people argue over nit picky things. we dont have that problem and if we start using the tool it will encourage people to behave that way.17:58
*** msabramo has quit IRC17:58
*** amcrn has quit IRC17:59
*** freyes has quit IRC17:59
*** amcrn has joined #openstack-infra18:00
jeblairmgagne: the better approach is just to review for good code, expect people to match the existing style for the most part and dont worry about the little things.18:00
*** amcrn has quit IRC18:00
*** khyati has quit IRC18:00
fungijeblair: clarkb: mordred: SergeyLukjanov: i'm forcing 86662 in because it was hoisted upon its own petard18:01
jeblairmgagnd: at least thats the kind of project id rather work on.  not worry about whether commit msgs need periods on the first line.18:01
jeblairmgagne: sorry im typing on a phone18:01
mgagnejeblair: What can I say. I love good looking code and feel hacking style fits my personal taste. But it looks to not be shared by everyone.18:01
openstackgerritA change was merged to openstack-infra/config: Temporarily disable hpcloud in nodepool  https://review.openstack.org/8666218:01
jeblairmgagne: im very sorry i didnt add something to jjb sooner about this.18:03
SergeyLukjanovfungi, ack18:04
fungiJayF: i meant to ping you sooner, but you expressed interest in pitching in on a keysigning activity at the summit... there's a list thread at http://lists.openstack.org/pipermail/openstack-dev/2014-March/031337.html if you want to weigh in18:05
fungiJayF: continued over the month change at http://lists.openstack.org/pipermail/openstack-dev/2014-April/031537.html18:05
*** ildikov_ has quit IRC18:06
anteayaso bug 1272746 is the # for recheck reverify for the poor patches with red dots?18:06
uvirtbotLaunchpad bug 1272746 in openstack-ci "Error <urlopen error [Errno 101] Network is unreachable> while getting http://pypi.openstack.org/openstack/..." [Undecided,Confirmed] https://launchpad.net/bugs/127274618:06
*** mrodden has joined #openstack-infra18:07
JayFfungi: I saw the thread; but the time for when to do it never seemed to materialize18:07
JayFfungi: given you/others have run it in the past, I'll just be at you all's disposal to help in whatever way you need18:07
*** ominakov has quit IRC18:08
*** dprince has quit IRC18:08
*** CaptTofu has quit IRC18:08
*** che-arne has quit IRC18:08
*** CaptTofu has joined #openstack-infra18:08
fungiJayF: well, i just happened to be the concerned citizen who coordinated the first one, but am happy for others to pick up the reigns if they want18:08
JayFI've never personally run a keysigning party18:08
JayFso if you've ever done it at all, for at least the first time around, I'd probably rather apprentice :)18:09
fungier, reins18:09
fungiJayF: that's fine too. i agree the thread sort of fizzled out with no real consensus18:10
JayFYeah, it seems like maybe for next summit, we should get some space carved out specifically for a keysigning party18:10
JayFsince we sign tagged releases, it's very important to the project to have a strong web of trust18:10
*** jlk has quit IRC18:15
*** jlk has joined #openstack-infra18:16
*** dkliban has quit IRC18:16
fungiJayF: agreed. ksps aren't the only way to get that done, but the summits are definitely a good opportunity to fill in the gaps (i always carry two forms of id and a stack of business cards with my key fingerprint and related contact info with me)18:17
JayFKey fingerprint on business cards18:17
JayFis a DAMN good idea18:17
*** hashar has joined #openstack-infra18:18
JayFand getting new business cards before the summit is also probably something I should do18:18
JayFso I might do this18:18
*** hogepodge has quit IRC18:18
*** hogepodge_ has joined #openstack-infra18:18
JayF<newtopic> for all these network-related build failures, should I keep spamming recheck no bug? Open a bug? Go for a walk then recheck?18:18
fungiJayF: recheck bug 127274618:18
uvirtbotLaunchpad bug 1272746 in openstack-ci "Error <urlopen error [Errno 101] Network is unreachable> while getting http://pypi.openstack.org/openstack/..." [Undecided,Confirmed] https://launchpad.net/bugs/127274618:18
fungiJayF: hopefully it's dying down now18:19
funginode counts in hpcloud are falling sharply18:20
fungihttps://community.hpcloud.com/status/ and https://status.rackspace.com/ are both fairly quiet about the situation18:22
*** _nadya_ has joined #openstack-infra18:23
*** zul has quit IRC18:25
JayFfungi: I've seen failures even trying to clone the git repo from the builder18:26
*** bookwar has quit IRC18:26
fungiJayF: also from a slave in hpcloud?18:26
*** amcrn has joined #openstack-infra18:26
JayFHow do I tell from console.html?18:27
fungiJayF: one of the first few lines of the console log mentions the hostname of the jenkins slave and we encode the provider and region/zone in that18:27
JayFah I see, the hostname18:27
*** sdake_ is now known as sdake18:28
JayFfungi: http://logs.openstack.org/75/86475/6/check/gate-ironic-python-agent-python27/74a752f/console.html hp cloud18:28
*** gema has quit IRC18:29
JayFfungi: plus all the network unreachables I've seen while digging for that failure have been all hp cloud18:29
*** mriedem1 has joined #openstack-infra18:29
JayFsame deal here -> http://logs.openstack.org/75/86475/6/check/gate-ironic-python-agent-pep8/af35ac2/console.html18:29
openstackgerritMichael Krotscheck proposed a change to openstack-infra/storyboard: Collate story metadata for status  https://review.openstack.org/8645218:29
fungiyep, that seems to match what i've correlated so far18:29
*** mrodden has quit IRC18:30
*** mriedem has quit IRC18:30
JayFIf there's anything I can do to help, lemme know. I don't know much about -infra in general but devops is my background so if there's a specific place I can dig, lemme know. Doubt there's much I can figure out without specific knowledge of the environment though.18:30
*** jergerber has joined #openstack-infra18:31
fungithanks, though i have doubts the problem is within the environment. looks like a misbehaving layer-3 device at some network boundary18:31
JayFyup, that's what I'd suspect as well. If you have issues similar to this at Rackspace, feel free to ping me. I work directly with folks who'd wanna know and be able to fix (but I doubt I'm alone in that here)18:32
openstackgerritMichael Krotscheck proposed a change to openstack-infra/storyboard: Collate story metadata for status  https://review.openstack.org/8645218:33
*** asselin has quit IRC18:33
*** asselin has joined #openstack-infra18:33
openstackgerritJames E. Blair proposed a change to openstack-infra/jenkins-job-builder: Add hacking to requirements  https://review.openstack.org/8667918:36
*** yolanda has quit IRC18:36
fungiJayF: do you happen to know why rackspace has a bunch of images which show up in glance image-list but not nova image-list for the same region?18:39
*** zul has joined #openstack-infra18:40
*** mrodden has joined #openstack-infra18:40
fungiJayF: for example, i see two debian images in nova image-list output, but glance image-list gives me 4418:40
*** marun has quit IRC18:41
*** Ryan_Lane1 has joined #openstack-infra18:41
*** thuc has quit IRC18:43
*** mdenny has quit IRC18:43
*** mdenny has joined #openstack-infra18:43
*** thuc has joined #openstack-infra18:43
mgagnefungi: could be they filter out "old" images in Nova based on metadata18:44
fungimgagne: good possibility. though some of these look like beta images or maybe images from other tenants (hard to tell without booting them)18:45
fungibut yes, probably metadata filtering in nova18:45
mgagnefungi: well, "old" images still exist, they aren't listed in Nova though18:45
*** thuc has quit IRC18:45
*** thuc has joined #openstack-infra18:46
*** gyee has quit IRC18:46
*** thuc has quit IRC18:46
*** Ryan_Lane1 has quit IRC18:47
*** thuc has joined #openstack-infra18:47
fungimakes sense18:47
*** thuc has quit IRC18:47
*** coolsvap is now known as coolsvap|afk18:48
*** thuc has joined #openstack-infra18:48
*** tjones has joined #openstack-infra18:52
*** _nadya_ has quit IRC18:53
*** gema has joined #openstack-infra18:54
*** zhiyan is now known as zhiyan_18:56
*** jcoufal has joined #openstack-infra18:59
*** yamahata has quit IRC19:04
*** sdake_ has joined #openstack-infra19:04
*** ominakov has joined #openstack-infra19:05
*** Sukhdev has quit IRC19:08
*** Ryan_Lane has quit IRC19:11
*** gema_ has joined #openstack-infra19:12
*** ominakov has quit IRC19:12
*** nati_uen_ has joined #openstack-infra19:13
openstackgerritPaul Belanger proposed a change to openstack-infra/zuul: Add option to toggle isJobRegistered check for gearman  https://review.openstack.org/8668519:13
*** ildikov has joined #openstack-infra19:13
*** gema has quit IRC19:14
openstackgerritPaul Belanger proposed a change to openstack-infra/zuul: Add option to toggle isJobRegistered check for gearman  https://review.openstack.org/8668519:14
*** nati_ueno has quit IRC19:14
*** ihrachyshka has quit IRC19:16
*** gema_ has quit IRC19:16
*** chandan_kumar has joined #openstack-infra19:19
*** chandan_kumar_ has joined #openstack-infra19:19
*** chandan_kumar_ has quit IRC19:19
*** dachary has quit IRC19:19
*** denis_makogon_ has joined #openstack-infra19:20
*** davidhadas has joined #openstack-infra19:20
jogomordred: https://review.openstack.org/#/c/86171/19:22
*** freyes has joined #openstack-infra19:22
*** ilives has quit IRC19:24
*** ociuhandu has quit IRC19:28
*** dachary has joined #openstack-infra19:31
*** bradm has joined #openstack-infra19:34
JayF.19:34
*** krtaylor has quit IRC19:36
*** nati_uen_ has quit IRC19:36
*** nati_ueno has joined #openstack-infra19:37
*** gondoi is now known as zz_gondoi19:38
*** khyati_ has joined #openstack-infra19:40
jogosdague: https://review.openstack.org/#/c/86171/ didn't seem to add the deps19:40
*** mriedem has joined #openstack-infra19:41
*** zns has quit IRC19:41
*** ihrachyshka has joined #openstack-infra19:42
*** cody-somerville has quit IRC19:42
*** salv-orlando has joined #openstack-infra19:43
*** mriedem1 has quit IRC19:43
*** mrodden has quit IRC19:46
*** mrodden has joined #openstack-infra19:47
openstackgerritPaul Belanger proposed a change to openstack-infra/zuul: Add option to toggle isJobRegistered check for gearman  https://review.openstack.org/8668519:47
openstackgerritPhilip Marc Schwartz proposed a change to openstack-infra/config: Added the abillity to use launch_node.py to connect to and create infrastructure instances in openstack setups that are not setup to use the ssl secured clients.  https://review.openstack.org/8669019:48
mordredjogo: yah - I read that this morning19:49
*** wayneeseguin has left #openstack-infra19:49
mordredsdague: I don't _think_ it's needed - but I very well could be wrong19:49
mordredsdague: there is also a case you're showing in that patch that, while not accomplishing what you were setting out to do - I don't believe is supported currently which we may want to think about19:50
*** wayneeseguin has joined #openstack-infra19:52
*** gema has joined #openstack-infra19:53
*** aconrad has quit IRC19:59
*** krtaylor has joined #openstack-infra20:03
openstackgerritJoe Gordon proposed a change to openstack-dev/hacking: Make test_doctest use entry points  https://review.openstack.org/8669220:03
openstackgerritJoe Gordon proposed a change to openstack-dev/hacking: Move comment checks into there own module  https://review.openstack.org/8669320:03
openstackgerritJoe Gordon proposed a change to openstack-dev/hacking: Move except checks into there own module  https://review.openstack.org/8669420:03
openstackgerritJoe Gordon proposed a change to openstack-dev/hacking: Only compile regex for hacking_except_format once.  https://review.openstack.org/8669520:03
*** dkranz has quit IRC20:03
jogohoping to get this sorted out so I can actually have these patches pass ^20:03
*** jp_at_hp1 has joined #openstack-infra20:03
*** markmcclain has joined #openstack-infra20:03
*** signed8bit has quit IRC20:04
*** jcoufal has quit IRC20:04
*** ihrachyshka has quit IRC20:04
*** nkinder has quit IRC20:05
*** jp_at_hp has quit IRC20:06
*** signed8bit has joined #openstack-infra20:07
*** sabari has joined #openstack-infra20:08
fungimordred: are you potentially aware of any network issues in hpcloud today? not sure if you saw earlier but i zeroed the nodepool max-servers on all hpcloud regions due to numerous jobs failing on connectivity errors (which had the effect of basically failing every change in the gate)20:10
fungiwe were seeing it from region b as well as the legacy azs20:10
fungii see https://community.hpcloud.com/status/incident/2536 which looks possibly similar, but was from yesterday20:13
fungiresolved ~29 hours ago and only lasted 30 minutes20:14
sdaguemordred: right, so I messed up my understanding of what got processed20:14
sdagueso what we're trying to do at the end of the day is let hacking control pep8 and flake8 versions20:15
*** e0ne has joined #openstack-infra20:15
sdagueand not let anyone else specify them20:15
*** NobodyCam is now known as NotNobodyCam20:15
sdaguebecause projects should consume those through hacking20:15
*** NotNobodyCam is now known as NobodyCam20:16
fungisdague: perhaps you want a hacking check which fails if flake8 or pep8 or pyflakes are in the {,test-}requirements files?20:16
sdaguefungi: well, actually, if we drop it from global requirements, we get that for free20:16
sdaguebecause it enforces not having unknown requirements20:16
fungisdague: but then you lose getting to enforce/sync global reqs on hacking20:16
sdaguebut why would we want that on hacking?20:17
fungii don't know for sure that you do, just pointing it out as a potential drawback20:17
sdaguehacking requirements today are:20:17
sdagued2to1>=0.2.10,<0.320:17
sdaguepbr>=0.5.10,<0.620:17
sdaguepep8==1.4.520:17
sdaguepyflakes==0.7.220:17
sdagueflake8==2.020:17
sdague3 of those we want it to be the only owner of20:17
sdaguethe other 2 seem overkill to be part of g-r sync20:18
fungiand especially, given that hacking is built to check things, having it check that projects using it don't use its dependencies as test-reqs would be a sane rule20:18
mtreinishsdague: do you still need d2to1 if you update the pbr req?20:19
*** dkranz has joined #openstack-infra20:19
sdaguemtreinish: probably not20:19
sdaguefungi: sure, that would be a fine addition. Probably not a priority20:20
*** mrodden has quit IRC20:21
*** mrodden1 has joined #openstack-infra20:21
fungisdague: especially if there are projects which want to test with flake8 but not hacking, yet still want to participate in requirements enforcement/sync20:21
fungitheir only option would become to depend on hacking and then disable it in config20:22
mtreinishsdague: your grenade-neutron add merged right?. So on https://review.openstack.org/#/c/84840/ should I add the grenade-neutron job to the new group?20:22
sdaguemtreinish: yes20:22
fungithough the size of that set of projects may, in reality, be null. no idea20:22
sdaguefungi: well for integrated projects, it's a null set20:23
sdagueand I think it's fine policy that folks have to access flake8 through hacking, given that you just need to add a small bit to tox to ignore all of the hacking rules. Especially as it gives you a hook to write local rules which more projects find useful.20:24
*** dkranz has quit IRC20:25
sdaguethe important issue is that pep8 isn't entirely stable. so for synchronizing reasons we need to have hacking version be what we flag on.20:25
openstackgerritMatthew Treinish proposed a change to openstack-infra/config: Only run jobs with neutron on neutron changes  https://review.openstack.org/8484020:25
mordredsdague: yeah - that was my concern with dropping pep8 and pyflakes from the g-r file - is that you'd have to drop hacking from mirror enforcement20:26
mordredbecause then you really _wouldn't_ be able to update the vesrion20:27
mordredif we do the other route, which is keep the mirror logic the current simple logic, then each upgrade is self-gating20:27
*** Steap has quit IRC20:27
sdagueright, but the issue that we have today is that for jogo to update pep8 in hacking, he has to propose it to the mirror20:27
mordredright20:28
mordredwhich is fine and will work as expected20:28
sdaguewhich then proposes changes to all the projects20:28
*** saju_m has quit IRC20:28
mordredright. which should be no-op patches since none of them shoudl be specifyin pep820:28
fungionly to the projects which declare it as a requirement though20:28
fungiright20:28
mordredif they aren't, they'll learn real quick because those patches won't work20:28
sdaguewhich may or may not work, because he can run the integration tests with the new pep8 until20:28
mordredso if req jobs propose a pep8 bump to anyone, they'll want to fix their requirements file20:29
*** saju_m has joined #openstack-infra20:29
mordredthen, at some point, we propose a hacking update to people20:29
*** gyee has joined #openstack-infra20:29
*** dprince has joined #openstack-infra20:29
mordredthe only issue in the system is if someone is direct depending on pep8 - which is a bug in that project20:29
sdagueright20:30
mordrednow - the thing you're pointing out is that we don't have a great generalized story for _blocking_ the listing of a req in a project20:30
sdagueright20:30
mordredwhich I think would be a nice thing - a requirement blacklist20:30
mordredbut I think we should do that in some way that isn't tied to mirror creation20:30
mordredmaybe20:30
fungi(which is something hacking could enforce)20:30
mordredor something20:30
mordredyeah20:30
sdaguewell, that's the question about adding another mirror-requirements piece20:30
mordredbecause nobody should be listing setuptools, distribute or d2to1 or M2Crypto20:31
fungii still see this as a problem for hacking which hacking can solve by checking for this when it's used20:31
sdaguebecause that's things we want in the mirror, but we don't want projects to be able to use directly20:31
mordredI think that expresses the wrong thing20:31
sdagueok20:31
mordredwhat we want is "things you shouldn't declare"20:31
*** markmcclain has quit IRC20:31
sdaguewell today that's anything that's not in global requirements20:31
mordredin the mirror or not in the mirror is handled by transitive deps20:31
*** markmcclain has joined #openstack-infra20:31
*** e0ne has quit IRC20:32
sdagueok20:32
*** e0ne has joined #openstack-infra20:32
sdaguewell I feel like we have an enforcement mechanism already in g-r, so it seems weird to create another one. However if that seems like the best way forward, I'm cool with it20:33
sdagueit still seems weird to manager hacking's requirements through g-r, because that's a multi day, multi hrs of CI, to proxy through a change that only hacking cares about20:35
*** e0ne has quit IRC20:36
sdaguerealistically 36 or 48 nodes consumed (given clean check, assuming longer than 24hrs to land) to bump pep8 in hacking, for instance. Which maybe is an over optimizing concern.20:38
*** dkranz has joined #openstack-infra20:38
fungihow often do you perform this particular task?20:40
sdaguenot all that often20:40
sdaguecouple times a cycle.20:40
sdagueif that20:40
openstackgerritClark Boylan proposed a change to openstack-infra/config: Update jenkins-dev ssh public key.  https://review.openstack.org/8670520:42
openstackgerritClark Boylan proposed a change to openstack-infra/config: Add second jenkins ssh key.  https://review.openstack.org/8670620:42
*** freyes has quit IRC20:42
*** Ryan_Lane has joined #openstack-infra20:43
sdaguewell, I'll leave the approach up to jogo on what he thinks is sane20:43
*** e0ne has joined #openstack-infra20:43
sdaguewe've at least gotten pyflakes out of all the project requirements files in this process20:43
sdagueso that was good, even if we back off this particular approach20:44
sdagueso I supposed I should propose the revert?20:46
*** Ryan_Lane has quit IRC20:47
openstackgerritSean Dague proposed a change to openstack/requirements: Revert "add mirror requirements list"  https://review.openstack.org/8670720:49
*** jamielennox|away is now known as jamielennox20:49
*** jooools has quit IRC20:51
*** harlowja is now known as harlowja_away20:56
*** reed has quit IRC20:57
*** reed has joined #openstack-infra20:59
*** mrda_away is now known as mrda21:01
*** mburned is now known as mburned_out21:01
*** mrodden1 is now known as mrodden21:01
*** thuc_ has joined #openstack-infra21:02
*** markmcclain has quit IRC21:03
*** thuc_ has quit IRC21:04
*** mrda is now known as manslaughter21:05
*** thuc_ has joined #openstack-infra21:05
*** thuc_ has quit IRC21:05
*** thuc has quit IRC21:05
*** rfolco has quit IRC21:06
*** mfer has quit IRC21:09
*** e0ne has quit IRC21:10
*** e0ne has joined #openstack-infra21:11
*** miqui has quit IRC21:13
*** ameade has quit IRC21:13
*** markmcclain has joined #openstack-infra21:13
dhellmanndid we ever decide as a project if we wanted to keep run_tests.sh? I have several patches related to making it work for oslo-incubator.21:14
*** markmcclain has quit IRC21:14
*** e0ne has quit IRC21:15
sdagueI think that's one of the great undecideds21:15
mtreinishdhellmann: I plan to keep it around in tempest at least short term. (there are actually 2 of them there) But I know nova pulled it out of their tree.21:15
dhellmannI'm inclined to just remove it, too21:16
dhellmannor make it just run tox :-)21:17
*** markmcclain has joined #openstack-infra21:17
sdagueyeh, honestly, I think readme to explain tox is probably more useful21:17
openstackgerritMichael Krotscheck proposed a change to openstack-infra/storyboard-webclient: Added task status display to story list view  https://review.openstack.org/8671221:17
*** esker has quit IRC21:19
*** CaptTofu has quit IRC21:20
phschwartzdhellmann: Here at Rax we requested that it stay a little longer as we use it to kick off tests on our prebuilt venvs. We have also been talking with the tox devs who are addressing the issue we opened with them about being able to call tox with a param stating to use a venv instead of building one for the tests.21:20
*** markmcclain has quit IRC21:20
*** markmcclain has joined #openstack-infra21:21
dhellmannphschwartz: ah, we use tox to make the venv and then package that :-)21:21
phschwartzAs soon as tox allows the using of a preexisting venv we will be all for removing run_test.sh21:21
*** markmcclain has quit IRC21:22
*** denis_makogon_ has quit IRC21:22
phschwartzdhellmann: we have been looking at doing that, but tox doesn't support pinning of single requirements outside of requirements.txt which we use since we still have melange lurking about (don't even say it)21:22
dhellmannphschwartz: tox --notest && .tox/env/bin/pip install something && tox21:23
*** dkranz has quit IRC21:23
phschwartzas we move from the need to pin deps as we come closer to upstream completely (just started deploying neutron), we will have less of that need.21:23
dhellmanna bit caveman, but works21:23
phschwartzhmm, we never thought of doing it that way.21:23
*** chandan_kumar has quit IRC21:23
dhellmannI'll be happy to share our script with you21:23
phschwartzThat might be a way around it, but does that second run of tox still use the venv built with the first tox command?21:24
dhellmannyep21:24
phschwartzif you could, it would be appreciated21:24
dstufftas long as tox doesn't think it needs to recreate the venv21:24
*** dprince has quit IRC21:24
dstufftwhich it shouldn't if you run those commands in succesion21:24
dhellmannright21:24
*** eharney has quit IRC21:25
tjonescan you run tox with pdb?  that is one reason i sill use runtests21:25
dstufftyou can pass stuff to the underlying commands21:26
dstufftassuming the tox.ini is set up to21:26
dstufftall my projects support tox -- --pdb to pass the --pdb flag to py.test21:26
tjonescool - i have a bug in nova to remove runtests but can't until we could use a different venv.  sounds like you have already solved the problem dhellmann21:27
sdaguedhellmann: is that something you could slip into pbr ?21:28
sdagueand give us a sane calling path through setup.py ?21:28
sdaguesounds like lots of projects would like it21:29
dstufftI'm not sure if the words setup.py and "sane path" exist in the same sentence21:30
dstufftcan exist*21:30
*** Ryan_Lane has joined #openstack-infra21:30
*** reed has quit IRC21:31
*** mwagner_lap has quit IRC21:31
*** mrodden1 has joined #openstack-infra21:32
dhellmannsdague: into pbr?21:32
openstackgerritA change was merged to openstack-infra/config: Update jenkins-dev ssh public key.  https://review.openstack.org/8670521:32
openstackgerritA change was merged to openstack-infra/config: Add second jenkins ssh key.  https://review.openstack.org/8670621:33
openstackgerritMichael Krotscheck proposed a change to openstack-infra/storyboard-webclient: Added story status label filter  https://review.openstack.org/8671821:33
*** mrodden has quit IRC21:34
dhellmanntjones: you can't just use any-old venv, it still needs to live in the place tox expects to find it21:35
*** nkinder has joined #openstack-infra21:35
tjonesdhellmann: could you use a sym link to point to any-old venv?21:35
*** thedodd has quit IRC21:36
dhellmanntjones: I'm not sure how tox would view that, I haven't tried it.21:36
dhellmannwhat's wrong with just having tox create the virtualenv?21:36
dhellmannyou can always modify it21:36
tjonesdhellmann: can't recall the name of the guy who had the usecase to ask him……21:36
tjoneslet me find it21:36
dhellmannI did have some trouble with tox creating stuff on an fs mounted into a vagrant vm, but I solved that by making the .tox directory a symlink to a filesystem inside the vm21:37
dhellmannso the parent of the venv dir is a symlink, and that works fine21:37
dhellmannor worked, I don't do that any more21:37
tjonesdhellmann: see #link https://review.openstack.org/#/c/76627/21:38
*** adalbas has quit IRC21:38
tjoneskevin mitchell's comment21:38
dhellmanntjones: yeah, that's what we do, but we let tox build the venv21:39
dhellmannhttp://paste.openstack.org/show/75508/21:39
phschwartztjones: yeah, that is what we were just talking about. We are going to mess with dhellmann's build script to see if we can do similar with tox directly and go forward with out needing run_tests.sh (kevin mitchell work on the same team at Rax)21:40
*** mrodden1 has quit IRC21:41
tjonesphschwartz: awesome!  it would be great to have 1 way to run the tests ;-)21:41
dhellmannphschwartz: in my case, I'm talking about the copy of run_tests.sh in oslo-incubator, and my guess is you don't run that one21:41
openstackgerritMichael Krotscheck proposed a change to openstack-infra/storyboard-webclient: Added story status to project story list.  https://review.openstack.org/8671921:42
phschwartztjones: I agree, but unless we can get tox fixed or dhellmann's suggestion proves to work for us, we would be stuck as we build a venv and then deploy that built venv to all of our production and test infrastructure.21:42
tjonesphschwartz: i started to look into tox but my head exploded21:43
phschwartzdhellmann: no, the oslo one would not block us. We mirror for local modifications for our infrastructure all the core components directly, but don't have a need for oslo. Basically things like nova, glance, etc, ... and the associated clients.21:43
phschwartzdhellmann: the change tjones pointed out is the one that we noticed it coming through with as it is removing the one in nova.21:44
*** hashar has quit IRC21:44
*** Ryan_Lane1 has joined #openstack-infra21:44
* dhellmann nods21:45
sdaguedhellmann: I thought it had some additional test hooks, maybe I'm mistaken21:45
mordredtjones, phschwartz: I support whatever things obviate run_tests.sh scripts - let me know if I can help on things21:45
*** harlowja_away is now known as harlowja21:45
dhellmannsdague: see the pastebin link above, the script relies on some customization scripts that live in our packaging repo but not in the app repo21:46
*** krtaylor has quit IRC21:46
dhellmannare projects syncing run_tests.sh from oslo, or does everyone have their own custom thing?21:47
*** dizquierdo has joined #openstack-infra21:47
*** tjones has quit IRC21:47
phschwartzdhellmann: to my knowledge they weren't syncing it from oslo. But I might have been completely mistaken21:47
dhellmannI would hate to delete the oslo version if everyone is expecting to keep it up to date that way21:47
*** mrodden has joined #openstack-infra21:48
*** Ryan_Lane1 has quit IRC21:48
*** tjones has joined #openstack-infra21:48
dhellmannlooks like install_venv is the same, but runtests is different21:49
dhellmannok, I'll leave it in the incubator until the other projects are ready to remove it21:49
*** yonglihe_ has joined #openstack-infra21:49
*** mriedem has quit IRC21:49
* mordred has been working on deleting run_tests.sh for like 3 years now21:50
*** yongli has quit IRC21:50
* mordred holds grudges apparently21:50
*** mrodden1 has joined #openstack-infra21:50
tjonesmordred: since the nova project has decided tox is the way to run the unit tests, i wanted to remove run_tests.sh and change the docs as it's confusing for newbies.  But then discovered that 2 use cases were not covered - using a diff venv and running the debugger.  sounds like there could be a solution though ;-)21:51
tjoneswhich is why i jumped in here -21:51
* mordred does a little jig21:51
tjoneslol21:51
clarkbso I am going to be bad and I am beginning to think we should dump tox21:51
clarkbthe posargs bug doesn't appear to be going anywhere21:52
tjonesclarkb: you are bad !21:52
sdagueclarkb: what would the alternative be?21:52
morganfainbergmordred, i know keystone is reworking run_tests to just wrap around tox w/ nice options.21:52
*** mrodden has quit IRC21:53
clarkbsdague: documentation. I think we can get away with documenting how to run tests. But maybe I am full of wishful thinking21:53
morganfainbergmordred, waiting for a feature for tox (and obv. upgrade of tox in ci) and we will be using run_tests as just a wrapper for conviencne21:53
clarkbhow hard is it to virtualenv venv && venv/bin/testr run --parallel ?21:53
morganfainbergcc dhellmann, ^ re: keystone and run_tests21:53
clarkb(I skipped the dep install process intentionally because that is where this falls apart I think)21:54
sdagueclarkb: well I think the current testr ui makes some things difficult, we create a lot of convenience targets21:54
*** wchrisj has quit IRC21:54
sdaguelook at tempest tox.ini, for instance21:54
dhellmannclarkb: the problem is you have to redo all of the settings present in the tox.ini in your wrapper script in order to have dependencies installed properly and the tests run properly -- at that point, you're building a thing to parse and reuse the tox.ini, so just use tox21:54
morganfainbergdhellmann, ++21:54
dhellmannmorganfainberg: what sorts of options are you adding to run_tests.sh?21:55
clarkbdhellmann: tox is terrible though21:55
clarkbdhellmann: and learing how tools work isn't21:55
morganfainbergdhellmann, failfast, the -- --subunit | subunit-2ot121:55
clarkbdhellmann: I don't think you have to redo all of the settings21:55
clarkbinstead you tell people how to use virtualenv and pip and testr21:56
morganfainbergdhellmann, etc. just a nice thin wrapper that is more straightforward to explain21:56
dhellmannmorganfainberg: makes sense21:56
sdaguedhellmann: though if tox has broken our use case for it with a bug no one is fixing, we're stuck on old tox forever, which is breaking plenty of people that don't realize they can't pip install tox21:56
morganfainbergdhellmann, it helps encourage running unit tests locally vs. waiting for check/gate to fail them21:56
dhellmannsdague: yeah, I'm about ready to just fork it or build something sane to replace it21:56
openstackgerritIan Wienand proposed a change to openstack-infra/config: Add note on ci.openstack.org source  https://review.openstack.org/8593321:56
clarkbdhellmann: oh good, making a botox isn't completely crazy :)21:57
sdagueheh21:57
morganfainbergdhellmann, especialyl with newer contributors (eventually people tend to move to tox directly, but even some seasoned develppers like run_tests since it wraps everything up nicely)21:57
dhellmannclarkb: have you already started?21:57
clarkbdhellmann: I have considered just taking my patch moving to git and seeing how that goes21:57
clarkbdhellmann: well if you consider my patch that will never merge as a fork maybe :P21:57
sdagueanyone at pycon to try to talk to devs face to face?21:57
morganfainbergsdague, what is broken in tox?21:57
mordredjeblair: ^^21:57
*** thomasem has quit IRC21:57
morganfainbergsdague, the {posargs} thing?21:57
*** thuc has joined #openstack-infra21:57
clarkbmorganfainberg: variable substitution21:57
mordredjeblair: any chance you've seen hpk?21:57
jogosdague mordred:  I think just going down this route first https://review.openstack.org/#/c/85496/ is better21:57
dhellmannclarkb: I thought a recent comment on that patch said they should do something like what you propose?21:57
clarkbmorganfainberg: all variable substitution is context sensitive now21:58
morganfainbergclarkb, wait... what?21:58
mordredjogo: ++21:58
clarkbdhellmann: no they want to switch to doing {posargs-string} and make only that substitution regular21:58
morganfainbergclarkb, seriously?21:58
jeblairmordred: yes, i saw him yesterday but did not have an opportunity to bring it up21:58
clarkbdhellmann: I want to make all substitutions regular so that they are well substitutions21:58
jeblairi can try to convince him; i'm not sure how successful that would be21:58
dhellmannclarkb: yeah21:58
clarkbneither option is backward compatible21:58
sdaguejeblair: well it would be worth a shot in person21:58
mordredjeblair: I believe in you21:59
jogosdague: I like the idea about removing pep8 etc from global reqs but  we need to sort out the mirror stuff first.21:59
clarkbso I think both are equally bad in hpk's opinion21:59
dhellmannjeblair: you have the power21:59
* mordred sends supportive thoughts to jeblair21:59
mordredjogo: no mirror stuff to sort out21:59
mordredreally21:59
jogoand don't think we *need* to tackle that right now21:59
jeblairi mean, if you don't buy "hey can we not break backwards compat" in the bug tracker, i don't know why doing so in person will help.  i mean, unless my hat is THAT awesome21:59
mordredoh - unless there is - but yes, I don't thin kwe need to sort it out yet21:59
dhellmannclarkb: how much of what tox does do we really need?21:59
jogomordred: making the mirror have any version of pep8?21:59
dhellmannjeblair: I thought holger was on vacation when this came up the first time21:59
morganfainbergjeblair, your hat is awesome. i am confident it is _that_ awesome21:59
mordreddhellmann: almost none of it21:59
sdaguejogo: actually, after the prior conversation, I'm fine with reverting what I did, and going don't the route of hacking enforcing that it's 3 deps aren't in the project test-requirements.txt22:00
sdaguegoing down22:00
sdaguedamn, I can't type this afternoon22:00
clarkbdhellmann: not a whole lot. The dependency installs and command running22:00
clarkbyou know what22:00
clarkbwe should just use make22:00
clarkbmake is available on all the OSes22:00
clarkband has targets22:00
mordreddhellmann: having a file in the repo that expresses what we want do is nice22:00
clarkband well works22:00
* dhellmann wonders if clarkb is trolling22:00
morganfainbergclarkb, scons or make would be well supported22:00
clarkbmordred: you are the auto* expert. Is make silly22:00
clarkbdhellmann: not trollin at all22:01
mordredI think make is silly22:01
jeblairclarkb: https://bitbucket.org/hpk42/tox/pull-request/85/fix-command-expansion-and-parsing22:01
morganfainbergclarkb, distro wise that is22:01
jeblairclarkb: is that what you want to happen?22:01
dhellmannmordred: yeah, I like that the config file is in the repo22:01
mordrednot because you're a bad person or hitler or anything22:01
clarkbjeblair: yes22:01
jogosdague:  that would work nicely22:01
mordredbut because most of our users will get confused by a Makefile22:01
clarkbmordred: huh really?22:01
mordredyah22:01
morganfainbergmordred, lets go really confusing, use m4 and then generate a scons config for it22:01
mordredmake is a thing that only C people and old unix greybeards understand anymore22:01
dstufftlol makefiles22:01
jeblairclarkb: okay, i will try to read and understand that and bring it up if i see him22:01
* mordred shoots morganfainberg in teh face22:02
clarkbjeblair: thanks22:02
dstufftmakefiles are brain pain :[22:02
clarkbseriously though we don't need complicated make22:02
clarkbdstufft: not for what we need22:02
* dhellmann strokes his greying beard22:02
jogosdague: want to drop your -1 https://review.openstack.org/#/c/85496/ so I can get closer to at least having my patches pass check queue22:02
mordredclarkb: I think, sadly enough, that a simple ini file makes sense to people22:02
sdagueyep22:02
jeblair++make22:02
morganfainbergdhellmann, i dunno, greying isn't the same as grey22:02
clarkbseriously this make file would be so simple22:02
mordredas soon as we put in makefiles - people will use them to do MOAR THINGS22:02
dhellmannmorganfainberg: don't be pedantic22:02
sdagueI actually like make22:02
*** dizquierdo has quit IRC22:02
* dhellmann was being poetic22:02
jeblairi mean, i cat 'tox.ini' all the time to figure out how to run zuul's tests22:02
clarkbpy27: pip_install22:02
sdagueI do fear mordred's concerns22:02
jeblairit's like python setup.py testr something something something22:02
clarkb\t.make/py27/testr run --parallel22:03
mordred'virtualenv foo ; foo/bin/pip install . ; foo/bin/tox run --parallel' :)22:03
dhellmannif we're going to build our own thing, it should only have the targets we want all projects to use, and only store config data in the project22:03
*** nijaba has quit IRC22:03
*** wchrisj has joined #openstack-infra22:04
clarkbif I write a make file for this no one will scream?22:04
jeblairanyway, i don't want my winging to detract from the convo too much22:04
clarkbin fact it solves the run_tests problem too22:04
jeblairi think tox is a huge win22:04
mordredclarkb: I will scream22:04
sdagueit's not even April 122:04
dhellmannthis is fun, but wine calls22:04
clarkbit is weird to me that a tool purpose built for this stuff that has existed forever and installed everywhere is a bad thing22:05
jeblairalso, tox seems very highly regarded in the python community22:05
mordredprimarily because I don't want two build systems in the tree - if we wanted to go makefiles- I would want to go ALL THE WAY and get rid of setuptools22:05
mordredand have make distcheck output a tarball with a generated static setup.py in it22:05
morganfainbergjeblair, i like tox.22:05
mordredbut I think that might be a little bit extra crazypants if we can just get a patch landed to tox22:05
morganfainbergi'm asking some folks here the holywar question... what would they use instead of tox.22:06
mordredoh god22:06
* mordred ducks22:06
clarkbmordred: have you used testrepositories make?22:06
morganfainbergclarkb, mordred, i'm happy to help if i can for landing a patch to tox22:06
clarkbmordred: *Makefile? it is actually pretty legit22:06
jeblairi think if we have to do one big project-wide update to use new tox args but can keep using tox stably after that, it's probably not the end of the world22:06
morganfainbergmordred, general consensus "don't use make... please don't22:06
morganfainberg"22:06
jeblairi won't tell hpk that though22:06
mordredmorganfainberg: that's what I'm saying22:06
clarkbmorganfainberg: the patch is written see https://bitbucket.org/hpk42/tox/pull-request/85/fix-command-expansion-and-parsing/diff22:06
morganfainbergclarkb, i saw.22:07
mordredclarkb: it's not that it's not good - it's the "we have 2000 devs, many of whom barely understand many facets of the toolchain" problem22:07
mordredit's the reason I don't think we should follow dstufft's advice about how to handle requirements, even though I recognize his point22:07
dstufftwhich advice was that22:08
mordreddstufft: no versions22:08
clarkbmordred: right but who doesn't understand make22:08
mordredclarkb:  many many many many many people22:08
clarkbmake is the first build tool everyone learns22:08
morganfainbergclarkb, sadly, a lot of python developer types22:08
mordredclarkb: you're forgetting corp devs who work in IDEs all the time22:08
mordredand python devs who have never done C22:08
dstufftI don't understand make above a basic level22:08
morganfainbergmordred, ++22:08
dstufftFWIW22:08
clarkbdstufft: you run `make target` and it runs that target for you + dependencies22:09
clarkbdstufft: it is really simple22:09
dstufftright I can make like22:09
dstufftbasic ones22:09
dstufftbut when it starts using % and shit22:09
dstufftI just go WELP22:09
mordredjust so everyone knows - while we're bikeshedding22:09
dstufftalso I dunno but maybe the clients care about windows22:09
jogomordred fungi: https://review.openstack.org/#/c/85496/22:09
mordredI DO have a bunch of automake rules that do all of the same VCS-based versioning that pbr does22:09
dstufftand make doesn't run there22:09
*** jp_at_hp1 has quit IRC22:10
morganfainbergmordred, should i be scared?22:10
mordredmorganfainberg: very22:10
mordredmorganfainberg: I know a LOT about automake22:10
morganfainbergmordred, i feel like being scared is warranted in this case22:10
sdaguewell it's beer-o-clock on the east coast, so bikes their sheds22:10
mattoliverauMorning all22:10
clarkbdstufft: http://gnuwin32.sourceforge.net/packages/make.htm22:10
clarkbdstufft: make runs everywhere22:10
morganfainbergmordred, i know a lot about scons (past job) - not sure i like it or make better than tox.22:10
tjoneshttp://en.wiktionary.org/wiki/bikeshedding22:10
clarkbgranted that is gnumake but close enough22:10
mordredmorganfainberg: http://bazaar.launchpad.net/~mordred/pandora-build/devel/files/head:/m4/22:10
mordredmake is MUCH better22:11
morganfainbergmordred, nice. that is scary22:11
mordredit's complete22:11
*** yamahata_ has quit IRC22:11
mordredand it does many things I desperately wish we had22:11
mordredbut I think the cost is too high for our project22:11
*** yamahata_ has joined #openstack-infra22:11
clarkbtjones: it feels like bikeshedding but I no longer believe tox is a viable tool. I would be so happy for upstream tox to prove me wrong22:11
clarkbwe will see what jeblair is able to do22:11
tjonesclarkb: i was just looking it up cause i had no idea what it was22:12
clarkbah22:12
tjonesclarkb: sorry - wasn' pointing at you ;-)22:12
*** atiwari has quit IRC22:13
jeblairokay, i'm pretty much the worst booth person ever; i really need to get into the expo hall and represent22:13
mordredjeblair: you're the BEST booth person ever22:14
tjonesmaybe i am being naive here - but does it matter that most people don't know the internals of make?  most people don't know the internals of run_tests or tox, but they can type ./run-tests.sh or tox -epy2722:14
clarkbtjones: exactly22:14
mordredtjones: that is an excellent point22:14
clarkbbut make is actually a proper language so it can do the things we want it to do (which is also slightly scary but hey)22:14
mordredtjones: otoh - if they're on windows or mac - and they don't have make - is there a universal and understandable instruction like "pip install tox" that gets them set up?22:15
clarkband run_tests should be a make file22:15
clarkbthere is no reason to redo that logic in bash22:15
clarkbmordred: why can't they have make on windows or mac?22:15
tjonesmac has make22:15
tjonesas long as you have xcode installed22:15
dstufftif you don't have xcode you don't have the ability to isntall c-exts using pip22:16
tjoneswindows….  dunno.  install cygwin and then you have make?22:16
dstufftso you won't get very far in python land dependencies w/o xcode22:16
tjonesdstufft:  right22:16
dstufftwindows is lolland and idk shit about there but probably if someone can figure out how to set up a compiler toolchain on Windows that works in Python they can download that thing clarkb linked22:17
*** dcramer_ has joined #openstack-infra22:18
*** saju_m has quit IRC22:20
morganfainbergtjones, mordred, if you don't have xcode installed you're not going to have git or able to install all the deps via pip22:20
jogosdague: working on making hacking faster, looks like there is a bunch of low hanging fruit22:20
morganfainbergdstufft, hah you beat me to it!22:20
sdaguejogo: great22:21
sdaguegiving me an option to disable the import checks would be great as well22:21
openstackgerritJoe Gordon proposed a change to openstack-dev/hacking: Speed up license checks  https://review.openstack.org/8672622:21
openstackgerritJoe Gordon proposed a change to openstack-dev/hacking: Only compile regex for hacking_except_format once  https://review.openstack.org/8669522:21
tjonesmorganfainberg: yeah xcode is already a dependency22:21
sdagueI really want to be able to run this in flymake, and the import tests aren't ever going to be able to handle that22:21
*** dkliban has joined #openstack-infra22:21
jogosdague: well once the global reqs is updated, you can help review them22:22
sdaguecool22:22
sdagueso I think we need the revert to land first22:22
jogosdague: turns out the license check was pretty silly btw22:22
sdagueheh22:22
jogohttps://review.openstack.org/8672622:22
*** dkranz has joined #openstack-infra22:22
sdagueI blame me22:22
sdaguethough I don't think I did that silliness22:23
sdaguebut I might have22:23
sdaguejogo: can you put your req change on top of https://review.openstack.org/#/c/86707/22:24
*** dims has quit IRC22:24
sdaguethat will get you back to the repo you expect22:24
sdaguethen I'll +222:24
fungi300 lines of scrollback in an hour. i feel like i don't know you people any more ;)22:24
sdaguefungi: I thought you were a grey beard at heart and would love the Make idea22:25
jogosdague: it was you ;)  32f820bcaf385f00fdce5d00281073838031db1822:27
jogoId3f9b6099301f998ce3185e9fd0ed6907e6e71f022:27
*** prad has quit IRC22:27
sdagueso I was right at first22:28
sdague:)22:28
*** thuc has quit IRC22:28
*** thuc has joined #openstack-infra22:29
openstackgerritJoe Gordon proposed a change to openstack/requirements: Bump pep8, pyflakes and flake8 versions  https://review.openstack.org/8549622:31
jogosdague: there ^22:31
sdague+222:31
fungialso, late chiming in, but jeblair's hat *is* that awesome. none of my hats are even half as awesome22:31
jogoand I'm out fora bit22:31
fungi(i compensate with hawaiian shirts)22:32
*** thuc has quit IRC22:33
*** thuc has joined #openstack-infra22:34
*** thuc_ has joined #openstack-infra22:35
*** thuc_ has quit IRC22:36
fungiclarkb: also, cc was the first build tool i learnt, if that counts ;)22:36
*** thuc_ has joined #openstack-infra22:36
* fungi hands sdague another vBeer22:37
*** signed8bit has quit IRC22:37
fungijeblair: you're on boot duty?!? :022:38
*** mriedem has joined #openstack-infra22:39
*** thuc has quit IRC22:39
* fungi is a bad openstack foundation person for not realizing we have a booth at pycon22:39
openstackgerritJamie Lennox proposed a change to openstack/requirements: Add Kerberos to global requirements  https://review.openstack.org/8672722:40
fungisdague: unsurprisingly, i don't find the make proposition that scary and sort of wonder what the fuss is all about. then again, debian packaging uses make as its language of choice, so i may be unduly biased22:42
*** signed8bit has joined #openstack-infra22:44
*** Ryan_Lane1 has joined #openstack-infra22:45
JayFthere are jobs that have been pending for 4h+ in zuul on python3.3 checks -> http://status.openstack.org/zuul/22:47
JayFand pypy22:47
JayFall the long pends seem to be on py33/pypy22:47
*** dcramer_ has quit IRC22:47
*** Ryan_Lane1 has quit IRC22:49
fungiJayF: probably starved on py3k-precise nodes with the temporary removal of hpcloud22:49
JayFI'm not seeing /any/ of the ones in queue running py33 checks at all22:49
JayFlike the top of queue is queued for py33, and I can't see anything successfully running py33 checks22:50
JayFthat's why I suspected the number of active py3k/pypy nodes might be more like '0' than >022:50
fungijogo: pardon my small mind, but i really don't understand how or why 85496 works. you're setting packages exactly equal to more than one version22:50
fungiJayF: yep, i'm checking to make sure we have some in nodepool ;)22:51
JayFfungi: is that something I can check in the future? Or does it require access?22:51
sdaguefungi: because pip is *fun*22:52
sdagueand it seems to or those in22:52
*** jgrimm has quit IRC22:52
fungiJayF: right now it requires one of the 4 people with ssh access into nodepool.openstack.org to run a local nodepool cli command. however it's something you might be able to check in the future if you contributed patches to expose that "somehow" *waves hands wildly*22:52
*** CaptTofu has joined #openstack-infra22:58
fungiJayF: it looks like we may have had enough py3k-precise nodes stuck in a building state in hpcloud that nodepool wasn't trying to add any in a working provider. i've set them to a delete state now, in hopes that will prompt some renewed building elsewhere22:58
*** dims has joined #openstack-infra22:58
JayFfungi: thanks. When I see stuff like this is complaining in here the right move? File a bug?22:59
fungiJayF: filing a bug is always acceptable, though in this case i'm not yet sure we have enough information to make it at all actionable22:59
*** jhesketh_ is now known as jhesketh23:00
jheskethMorning23:00
*** thomasem has joined #openstack-infra23:00
sdaguealso, I think infra folks might want to comment on the branchless tempest qa-spec - https://review.openstack.org/#/c/86577/1/specs/branchless-tempest.rst23:00
*** jpeeler has quit IRC23:00
sdaguebecause it's definitely overlapping between qa and infra23:00
sdagueso commentary would be appreciated to make sure cases are covered well23:01
*** wenlock has quit IRC23:01
sdagueI'm going to integrate all the feedback tomorrow and take to the list again to make sure we're all good with it23:01
*** mbacchi has quit IRC23:01
*** khyati_ has quit IRC23:02
*** yamahata has joined #openstack-infra23:03
mordredsdague: I love the idea23:03
*** thuc_ has quit IRC23:04
fungiper our original discussion in #-qa, assuming it at all reflects what we talked about, i'm wholly in favor. haven't had time to read though23:05
*** thuc has joined #openstack-infra23:05
*** zehicle_at_dell has joined #openstack-infra23:05
*** thuc has quit IRC23:09
*** asettle has joined #openstack-infra23:10
*** arosen1 has quit IRC23:11
*** asettle has quit IRC23:15
*** mburned_out is now known as mburned23:18
*** mikal is now known as stillbert23:19
*** tchaypo has quit IRC23:20
*** flaper87 is now known as flaper87|afk23:20
*** tchaypo has joined #openstack-infra23:21
*** CaptTofu has quit IRC23:23
*** tchaypo has quit IRC23:24
*** tchaypo has joined #openstack-infra23:25
*** tchaypo has quit IRC23:25
*** markwash has quit IRC23:26
*** tchaypo has joined #openstack-infra23:26
*** jergerber has quit IRC23:29
*** SumitNaiksatam has left #openstack-infra23:30
*** dkliban is now known as dkliban_afk23:31
*** jeckersb is now known as jeckersb_gone23:34
*** thuc has joined #openstack-infra23:35
adam_ganyone have a sec for a quick +A  to enable ironic scenario testing in its devstack job? https://review.openstack.org/#/c/85227/123:36
*** markmcclain has joined #openstack-infra23:40
*** oomichi has joined #openstack-infra23:40
*** markmcclain has quit IRC23:43
*** thuc has quit IRC23:44
*** markmcclain has joined #openstack-infra23:44
*** Ryan_Lane1 has joined #openstack-infra23:45
*** mattoliverau is now known as mat-the-woli23:47
*** markmcclain has quit IRC23:48
*** thuc has joined #openstack-infra23:48
*** thuc has quit IRC23:49
openstackgerritJames Polley proposed a change to openstack-infra/pypi-mirror: Add a pointer to the DIB pypi element's README  https://review.openstack.org/8674623:49
adam_gfungi, thanks!23:49
*** thuc has joined #openstack-infra23:50
*** thuc has quit IRC23:50
*** Ryan_Lane1 has quit IRC23:50
fungiadam_g: any time23:50
*** khyati_ has joined #openstack-infra23:50
*** thuc has joined #openstack-infra23:50
*** yamahata has quit IRC23:50
*** thuc has quit IRC23:51
*** thuc has joined #openstack-infra23:52
*** wchrisj has quit IRC23:56
pabelangerany chance for some nodepool reviews / approvals? https://review.openstack.org/#/c/83975/23:59

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!