Sunday, 2013-12-08

*** jcooley_ has joined #openstack-infra00:00
*** nati_ueno has joined #openstack-infra00:03
*** nati_ueno has quit IRC00:03
*** dripton has quit IRC00:07
*** nati_ueno has joined #openstack-infra00:10
*** jcooley_ has quit IRC00:12
*** jcooley_ has joined #openstack-infra00:13
*** jcooley_ has quit IRC00:17
*** herndon has joined #openstack-infra00:20
*** nati_ueno has quit IRC00:25
*** nicedice_ has quit IRC00:32
*** nicedice_ has joined #openstack-infra00:32
*** senk has joined #openstack-infra00:44
*** herndon has quit IRC00:45
*** senk has quit IRC00:45
zarois the 'gate-jenkins-job-builder-docs' broken?  all links are broken.00:45
*** cyeoh has quit IRC00:45
fungizaro: that might be me... have an example?00:46
*** dripton has joined #openstack-infra00:50
fungizaro: ahh, yeah i think it was my fault (the files are there, just compressed). fixing, but it'll take a few minutes to finish00:52
zarofungi: i noticed that for all jjb reviews.  it doesn't even look like the link points to the right directory.00:55
fungizaro: yeah, you're right, that directory isn't there. something replaced one of my temporary symlinks, but i'll have it hooked back up here shortly01:05
fungizaro: should be better now01:10
*** yamahata_ has joined #openstack-infra01:12
fungiand now the symlinks replaced by a directory again. i suspect a jenkins publisher is doing that01:14
fungii'll have to hardlink it instead01:15
fungioh, right, directories can't be hardlinks. duh01:16
fungiand yeah, the scp publisher obliterates any symlink in the destination path and replaces it with a new directory. how thoughtful01:27
*** mgagne has joined #openstack-infra01:33
*** paul-- has joined #openstack-infra01:36
*** michchap_ has quit IRC01:42
*** mgagne has quit IRC01:42
*** mgagne1 has joined #openstack-infra01:42
*** mgagne1 has joined #openstack-infra01:42
*** michchap has joined #openstack-infra01:43
*** nati_ueno has joined #openstack-infra01:43
*** moted has quit IRC01:45
*** nati_ueno has quit IRC01:48
*** tma996 has joined #openstack-infra02:06
*** che-arne has quit IRC02:10
*** loq_mac has joined #openstack-infra02:11
*** che-arne has joined #openstack-infra02:11
*** paul-- has quit IRC02:15
*** paul-- has joined #openstack-infra02:19
*** fbo is now known as fbo_away02:24
*** mgagne1 is now known as mgagne02:25
*** thedodd has quit IRC02:28
*** senk has joined #openstack-infra02:30
*** senk has quit IRC02:40
*** loq_mac has quit IRC02:40
*** paul-- has quit IRC02:45
*** senk has joined #openstack-infra02:46
*** bpokorny has joined #openstack-infra03:06
*** senk has quit IRC03:07
*** bpokorny1 has joined #openstack-infra03:15
*** bpokorny has quit IRC03:17
*** fifieldt has quit IRC03:38
*** zehicle_at_dell has joined #openstack-infra03:38
*** rcleere has joined #openstack-infra03:53
*** rcleere has quit IRC04:01
*** syerrapragada has quit IRC04:04
*** syerrapragada has joined #openstack-infra04:05
*** bpokorny1 has quit IRC04:13
*** bpokorny has joined #openstack-infra04:13
*** bpokorny has quit IRC04:14
*** boris-42 has quit IRC04:38
*** talluri has joined #openstack-infra04:43
*** luisg has quit IRC04:51
*** luisg has joined #openstack-infra04:52
*** herndon has joined #openstack-infra05:01
*** herndon has quit IRC05:02
*** jcooley_ has joined #openstack-infra05:03
*** talluri has quit IRC05:03
*** senk has joined #openstack-infra05:09
*** talluri has joined #openstack-infra05:11
*** sdake_ has joined #openstack-infra05:12
*** senk has quit IRC05:13
*** sdake_ has quit IRC05:17
*** jcooley_ has quit IRC05:17
*** A-G has joined #openstack-infra05:27
*** syerrapragada has quit IRC05:29
*** syerrapragada has joined #openstack-infra05:30
*** dcramer_ has joined #openstack-infra05:30
mordredfungi, clarkb, jeblair: meet A-G. he's starting tomorrow and I'm at conferences. abuse him well05:31
*** julim has joined #openstack-infra05:31
clarkbA-G: ohai05:31
fungihowdy!05:31
clarkbmordred: are you ever not at conferences?05:31
mordredclarkb: define ever05:32
*** senk has joined #openstack-infra05:32
clarkbmordred: long enough to do laundry at home05:32
mordredclarkb: A-G was just telling me fascinating things about heira in 3.x - apparently we can use it to deal with package define duplication sanely05:32
mordredclarkb: and it might make sense to add a public heira tree to our repo that doesn't contain passwords to be able to do some things, once we're on 3.x05:33
mordredit's 730 in the morning here though, so that mostly just hurts my head05:34
clarkbA-G: also you don't have to work on sunday05:35
*** julim has quit IRC05:35
*** senk has quit IRC05:36
*** dcramer_ has quit IRC05:37
mordredA-G: ignore clarkb. there must be working at all times. just redefine "work" as "fun" and then it's not a problem ;)05:37
mordredclarkb: I'm happy to be continuing the fine infra tradition of not being around for someone's first week of work though05:38
mordredI _may_ have been around for pleia2's first day?05:38
clarkbmaybe?05:40
lifelessA-G: welcome! Also you might like to put a real realname in your IRC client rather than ... - helps folk trying to map IRC names to email threads05:40
* clarkb gets away with being boring enough to have a realname as his nick05:41
mordredlifeless:  bah. where's the fun of that :)05:46
A-Gah, sorry, stepped away to have a word with a rabbit.  heira stuff is very cool, and yeah, going to rename momentarily.05:53
*** A-G has quit IRC05:54
*** AaronG has joined #openstack-infra05:54
AaronGThere, i'm like a real person.05:55
mordredoh - my. you've changed05:57
fungimy real name actually *is* fungi...05:58
*** afazekas has joined #openstack-infra05:58
fungiat least it's the realest name i can ever remember05:59
AaronGlast year i interviewed a person for a devops role, named Dev Ghai.05:59
AaronGunsurprisingly, he got the job.05:59
funginame and title in one06:00
*** sdake_ has joined #openstack-infra06:00
mordredo m g06:03
mordredthat should be someone's nick on casual nick friday06:04
*** SergeyLukjanov has joined #openstack-infra06:04
*** elasticio has joined #openstack-infra06:04
*** talluri has quit IRC06:17
*** sdake_ has quit IRC06:18
*** sdake_ has joined #openstack-infra06:28
*** sdake_ has quit IRC06:32
*** elasticio has quit IRC06:34
*** SergeyLukjanov has quit IRC06:37
*** rwsu has quit IRC06:40
*** rwsu has joined #openstack-infra06:47
*** jcooley_ has joined #openstack-infra06:59
*** jcooley_ has quit IRC07:09
*** jcooley_ has joined #openstack-infra07:10
*** talluri has joined #openstack-infra07:12
*** jcooley_ has quit IRC07:14
*** Protux has quit IRC07:39
*** rongze has joined #openstack-infra07:56
*** mgagne has quit IRC08:00
*** senk has joined #openstack-infra08:03
*** A-G has joined #openstack-infra08:12
*** A-G has quit IRC08:12
*** AaronG has quit IRC08:15
*** AaronG has joined #openstack-infra08:15
*** AaronGr has joined #openstack-infra08:16
*** AaronG has left #openstack-infra08:17
*** avishayb has joined #openstack-infra08:26
*** AaronGr is now known as AaronG_Zzz08:26
*** sdake_ has joined #openstack-infra08:34
*** francesco87 has joined #openstack-infra08:44
*** francesco87 has quit IRC08:45
*** senk has quit IRC08:48
*** elasticio has joined #openstack-infra08:50
*** avishayb has quit IRC08:50
*** sdake_ has quit IRC08:51
*** sdake has joined #openstack-infra08:51
*** senk has joined #openstack-infra09:13
*** guohliu has joined #openstack-infra09:14
*** senk has quit IRC09:17
*** senk has joined #openstack-infra09:18
*** avishayb has joined #openstack-infra09:27
*** tma996 has quit IRC09:41
*** guohliu has quit IRC09:58
*** tma996 has joined #openstack-infra10:04
*** Protux has joined #openstack-infra10:09
*** che-arne has quit IRC10:20
*** avishayb has quit IRC10:20
*** avishayb has joined #openstack-infra10:21
*** elasticio has quit IRC10:26
*** afrittoli has joined #openstack-infra10:27
*** andreaf has quit IRC10:29
*** tma996 has quit IRC10:39
*** avishayb has quit IRC10:40
*** fbo_away is now known as fbo10:41
*** tma996_ has joined #openstack-infra10:42
*** senk has quit IRC10:44
*** nati_ueno has joined #openstack-infra10:46
*** avishayb has joined #openstack-infra10:51
*** fbo is now known as fbo_away10:53
openstackgerritAntoine Musso proposed a change to openstack-infra/jenkins-job-builder: cloverphp publisher  https://review.openstack.org/5294011:02
openstackgerritAntoine Musso proposed a change to openstack-infra/jenkins-job-builder: cloverphp publisher  https://review.openstack.org/5294011:02
*** rongze has quit IRC11:11
*** rongze has joined #openstack-infra11:24
*** senk has joined #openstack-infra11:28
*** senk1 has joined #openstack-infra11:29
*** senk has quit IRC11:33
*** persia has joined #openstack-infra11:47
*** marun has joined #openstack-infra11:51
*** senk1 has quit IRC12:22
openstackgerritSergey Kolekonov proposed a change to openstack-infra/jenkins-job-builder: Description Setter Plugin support added  https://review.openstack.org/6067412:24
openstackgerritSergey Kolekonov proposed a change to openstack-infra/jenkins-job-builder: Description Setter Plugin support added  https://review.openstack.org/6067412:28
*** amotoki has joined #openstack-infra12:33
*** DennyZhang has joined #openstack-infra13:12
*** Loquacities has joined #openstack-infra13:14
*** avishayb has quit IRC13:15
*** Loquacity has quit IRC13:17
*** paul-- has joined #openstack-infra13:32
*** avishayb has joined #openstack-infra13:43
*** elasticio has joined #openstack-infra13:45
*** paul-- has quit IRC13:50
*** elasticio_ has joined #openstack-infra13:56
*** paul-- has joined #openstack-infra13:57
*** DennyZhang has quit IRC14:01
openstackgerritMax Rydahl Andersen proposed a change to openstack-infra/jenkins-job-builder: Add support for sidebar-links plugin via properties.  https://review.openstack.org/6071014:03
*** paul-- has quit IRC14:06
*** senk has joined #openstack-infra14:07
openstackgerritMax Rydahl Andersen proposed a change to openstack-infra/jenkins-job-builder: Add support for sidebar-links plugin via properties.  https://review.openstack.org/6071014:08
*** senk has quit IRC14:11
*** paul-- has joined #openstack-infra14:11
openstackgerritMax Rydahl Andersen proposed a change to openstack-infra/jenkins-job-builder: Add support for sidebar-links plugin via properties.  https://review.openstack.org/6071014:11
*** senk has joined #openstack-infra14:17
*** dkranz has quit IRC14:25
*** marun has quit IRC14:32
*** jcooley_ has joined #openstack-infra14:33
*** DennyZhang has joined #openstack-infra14:37
*** fbo_away is now known as fbo14:39
*** adalbas has joined #openstack-infra14:43
*** jcooley_ has quit IRC14:50
*** prad has joined #openstack-infra14:50
*** talluri has quit IRC14:51
*** DennyZhang has quit IRC14:54
*** prad has quit IRC14:54
*** paul-- has quit IRC14:57
*** fbo is now known as fbo_away15:09
*** dcramer_ has joined #openstack-infra15:10
*** senk has quit IRC15:10
*** senk has joined #openstack-infra15:12
*** fbo_away is now known as fbo15:13
*** SushilKM has joined #openstack-infra15:21
*** talluri has joined #openstack-infra15:22
*** tma996_ has quit IRC15:23
*** senk has quit IRC15:24
*** tma996 has joined #openstack-infra15:24
*** Ryan_Lane has joined #openstack-infra15:24
*** fbo is now known as fbo_away15:26
*** elasticio has quit IRC15:27
*** talluri has quit IRC15:27
*** elasticio_ has quit IRC15:27
*** tma996 has quit IRC15:39
*** SergeyLukjanov has joined #openstack-infra15:53
*** DennyZhang has joined #openstack-infra16:00
*** DennyZhang has quit IRC16:04
*** avishayb has quit IRC16:11
*** amotoki has quit IRC16:15
*** jcooley_ has joined #openstack-infra16:19
*** SushilKM has quit IRC16:20
*** leifmadsen has joined #openstack-infra16:23
*** talluri has joined #openstack-infra16:23
*** talluri has quit IRC16:28
*** SushilKM has joined #openstack-infra16:29
*** SushilKM has quit IRC16:46
*** nati_ueno has quit IRC16:55
*** yamahata_ has quit IRC17:04
*** adalbas has quit IRC17:05
*** lillie has joined #openstack-infra17:13
*** lillie is now known as Guest9883417:13
*** pvo has joined #openstack-infra17:14
*** dcramer_ has quit IRC17:15
*** yolanda has quit IRC17:23
*** AaronG_Zzz is now known as AaronGr17:25
*** dcramer_ has joined #openstack-infra17:29
*** talluri has joined #openstack-infra17:31
*** dkranz has joined #openstack-infra17:53
*** SergeyLukjanov is now known as _SergeyLukjanov17:55
anteayaAaronGr: so if I read the backscroll correctly, you are A-G, AaronG and AaronG_Zzz18:01
anteayakeeping track of your nick is going to be as entertaining as playing "where's monty"18:01
anteayaand welcome18:01
AaronGrThis should be the last set of name changes, was working out ZNC for playback, and didn't want to collide18:02
AaronGrand thanks :)18:02
pleia2welcome AaronGr :)18:02
AaronGr*waves*18:03
anteayaah18:04
anteayaAaronGrrrrr it is then18:05
anteayahey pleia218:05
anteayahow are your cats?18:05
*** _SergeyLukjanov is now known as SergeyLukjanov18:05
pleia2anteaya: poor critters have a cold, but mostly ok :)18:06
anteayaAaronGr: so you are in Victoria?18:06
anteayapleia2: how to you pick up they have a cold?18:06
anteayamine was sneezing last night but I did just turn the heat up 5 degrees C so I attributed it to that18:07
anteayashe isn't sneezing today18:07
AaronGranteaya:  Yes, just outside of it actually.18:07
anteayacool18:07
pleia2anteaya: boogers :)18:07
anteayaAaronGr: tell me sweet fairy tales of the weather18:07
anteayapleia2: ha ha ha18:07
anteayapleia2: lol18:07
anteayapleia2: well that is fairly definitive18:08
pleia2and watery eyes18:08
clarkbanteaya: it is really cold in this area too18:08
anteayaah okay18:08
AaronGr+1C and sunny. Not as nice as California at this time of the year, but I don't have to shovel.18:08
anteayaclarkb: hey there, how cold?18:08
pleia2AaronGr: I'm in san francisco, it's 3C right now D:18:08
anteayaclarkb: too early for football yet eh?18:08
pleia2everyone is freaking out18:08
clarkbanteaya: -7C was our low last night and ~-1C the high18:08
anteayaAaronGr: +1 is nice18:08
clarkbtoday is supposed to be warmer but I haven't checked18:08
AaronGrpleia2: wow, that does seem cold for there!18:09
anteayapleia2: freaking out about what? what is there to shovel at +3C?18:09
pleia2anteaya: too cold!18:09
anteayaclarkb: ah okay, so yeah cold18:09
pleia2but it is exciting to use my scarves and mittens :)18:09
*** jcooley_ has quit IRC18:09
anteayapleia2: scarves and mittens can be so pretty18:10
anteayaI got some new ones made from alpaca wool18:10
pleia2nice18:10
anteayaso light compared to sheep's wool18:10
anteayaI like to use my scarf on planes18:10
clarkbAaronGr: they are worried about freezing rain tuesday morning here. That could get fun with the hills18:10
anteayano idea why the airport is a sauna and the plane is a meat locker18:10
anteayaguess they like to keep travelers entertained18:11
AaronGrthat breeze around the ankles on some flights can certainly keep you awake.18:13
anteayaI actually looked at status.openstack.org/zuul about 20 minutes ago and not one thing was happening18:13
anteayaI couldn't believe my eyes18:13
anteayaI should have taken a screen shot18:13
anteayaAaronGr: never noticed the ankle breeze, I tend to keep my shoes on18:13
anteayabut I will look out for it, now that you mention it18:14
AaronGrit makes me regret my decision to wear short socks.18:14
anteayaah18:14
anteayaI'm not a short sock person18:14
anteayaif I wear socks they had better do a job, or I'm not wearing them18:14
anteayaI stayed at the hotel which is a front runner for the design summit in Paris18:15
anteayasteakhouse18:15
anteayajazz club18:15
anteayaon the #1 metro line18:15
anteayaI liked it18:15
*** dcramer_ has quit IRC18:16
anteayaobviously other than knowing the steak house is there, I have no further details on it18:16
anteayaclarkb: yeah, if you do have freezing rain, I suggest working from home18:16
anteayafreezing rain + hills + drivers who don't know freezing rain is a very bad combination18:17
*** bpokorny has joined #openstack-infra18:17
clarkbanteaya: re football, games started about 15 minutes ago. The one in philadelphia is being played in what looks like a blizzard18:22
anteayaclarkb: wow18:33
anteayayeah, snowed on the way home last night, just a few flurries18:33
anteayaclarkb: you went to a game a few weeks back, how was that?18:33
zaroAaronGr: hiho! i'm all about shorties.  and all the same so i don't need to sort them after laundry!18:38
anteayazaro: efficient laundry strategy ftw18:38
zaroclarkb: hey, can you check on something for me while your watching football18:38
zaroclarkb: i can't update submodule from our gerrit fork -> https://git.openstack.org/openstack-infra/gerrit18:39
zaroanteaya: yes, because my job is folding.18:40
clarkbzaro: we dont have submodules in our fork18:40
clarkbbecause we don't have the newer code. submodules may make this extra interesting. gerrit git prep and maybe zuul will need updating18:41
anteayazaro: ah18:42
zaroclarkb: well looks like more work to be able to build gerrit.18:42
* anteaya heads out to deliver chocolate to her sister and then to pile wood18:42
ekarlso2.8 live soon ? :D18:43
zaroekarlso: hopefully.18:44
zaroekarlso: but soon is very broad, i would say in a few months.18:45
*** ekarlso- has joined #openstack-infra18:49
*** jgriffit2 has joined #openstack-infra18:49
*** guitarza1 has joined #openstack-infra18:50
zaroclarkb: how about just replacing the submodules in our gerrit fork with git clones?18:51
*** mgagne has joined #openstack-infra18:52
*** mgagne has joined #openstack-infra18:52
*** elasticio has joined #openstack-infra18:52
*** bpokorny has quit IRC18:54
*** jgriffith has quit IRC18:54
*** dmakogon_ has quit IRC18:54
*** guitarzan has quit IRC18:54
*** hdd has quit IRC18:54
*** ekarlso has quit IRC18:54
*** bpokorny has joined #openstack-infra18:54
*** bpokorny has quit IRC18:54
clarkbthat doesn't help18:55
clarkbwe need to be able to deal with conflicts and should init submodules properly18:55
*** bpokorny has joined #openstack-infra18:55
*** bpokorny has quit IRC18:55
*** dmakogon_ has joined #openstack-infra18:57
zaroit seems like the same to me.  the submodules are only for plugins and those just point to the individual plugin projects.18:57
*** hdd has joined #openstack-infra18:57
zaroso during a build, we just replace the gerrit/plugins/* with git clones of the plugins.18:57
clarkbthat is basically what submodules do but the pointers are part of the repo so you can have conflicts18:59
clarkbthe amount of work is the same but on is correct18:59
*** jcooley_ has joined #openstack-infra19:00
zaroi know the right thing to do is update zuul and gerrit git prep but don't we want to do that after gerrit 2.8 upgrade?19:03
zarodoes the conflicts happen during an update or merge?  i'm talking about clone the plugins after everything updates and merges.19:04
clarkbit happens when you merge and doing the cloning like that would need to happen in ggp and zuul so may as well just deal with it correctly19:05
*** dkranz has quit IRC19:07
zarocan't it happen in the jenkins job?19:07
clarkbsure that is ggp19:08
*** fbo_away is now known as fbo19:12
zarono i mean in a shell after the ggp.19:13
zarojust a one off for gerrit only.19:13
clarkbbut why not in ggp? then all jobs do it correctly19:13
clarkbif you have to do it once do it right so that is works everywhere19:14
zarobecause at this point only gerrit uses submodules.19:14
clarkbthat doesnt matter19:14
*** AaronGr is now known as AaronG_Zzz19:14
clarkbsupporting submodules is the same for any project19:14
zarook.  in principle i agree we should make it work for all projects.  but need to leave it at that for now because kids are fast approaching..19:16
*** paul-- has joined #openstack-infra19:17
*** paul-- has quit IRC19:22
*** Daviey has quit IRC19:24
*** mriedem has quit IRC19:26
*** Daviey has joined #openstack-infra19:28
*** mgagne has quit IRC19:35
*** jcooley_ has quit IRC19:35
*** jcooley_ has joined #openstack-infra19:37
*** talluri has quit IRC19:44
*** mriedem has joined #openstack-infra19:47
*** davidhadas has joined #openstack-infra19:50
*** Daviey has quit IRC19:57
*** yolanda has joined #openstack-infra19:58
*** rfolco has joined #openstack-infra20:01
*** yolanda has quit IRC20:02
*** pcrews has joined #openstack-infra20:02
*** SergeyLukjanov has quit IRC20:14
*** fbo is now known as fbo_away20:16
*** thomasem has joined #openstack-infra20:21
*** cody-somerville has quit IRC20:23
*** Daviey has joined #openstack-infra20:29
*** dcramer_ has joined #openstack-infra20:29
fungiAaronGr: I hope you don't mind if in your head i pronounce your nick like "aaron burr with a g"20:30
fungier, in my head20:33
*** zul has quit IRC20:39
*** pcrews has quit IRC20:40
*** bknudson has quit IRC20:44
*** mgagne has joined #openstack-infra20:52
*** mgagne has joined #openstack-infra20:52
*** zul has joined #openstack-infra20:54
*** talluri has joined #openstack-infra20:56
*** rfolco has quit IRC20:57
*** bknudson has joined #openstack-infra20:59
*** thomasem has quit IRC20:59
*** talluri has quit IRC21:00
*** thomasem has joined #openstack-infra21:01
*** thomasem has quit IRC21:05
*** thomasem has joined #openstack-infra21:09
*** changbl has quit IRC21:12
clarkbmordred: you should reply to the keystoneclient test thread with your client testing plans21:14
*** jcooley_ has quit IRC21:22
lifelessclarkb: in https://review.openstack.org/#/c/53917/12/modules/openstack_project/files/jenkins_job_builder/config/devstack-gate.yaml21:25
lifelessclarkb: is it possible to do some metaprogramming to avoid what looks to me like copy-and-paste in all these job definitions ?21:26
clarkbnot with the JJB template system21:27
lifelessclarkb: also I have a q in https://review.openstack.org/#/c/53917/12/modules/openstack_project/files/zuul/layout.yaml21:27
clarkbif we went to jinja2 or similar I think we could21:28
*** ilyashakhat has quit IRC21:28
lifelessclarkb: what would jinja2 offer that's missing ?21:28
*** changbl has joined #openstack-infra21:28
Mithrandirturing completeness, I think?21:28
clarkbMithrandir: basically21:29
lifelessIt's not clear to me why thats needed.21:29
lifelessThe common code seems to be a strict prefix of all the jobs in question21:29
MithrandirI kinda want a better templating system for jjb, but jinja + yaml as done in salt stack just makes me sad.21:29
clarkblifeless: if something says postgres then set postgres arg, if something says neutron set neutron arg21:29
clarkband so on21:30
clarkbwe can only do string substitution in those templates21:30
lifelesssure21:30
lifeless%(POSTGRES_ARG)s with POSTGRES_ARG="" gets you a blank line21:30
clarkblifeless: responded to your comment in the layout file21:31
clarkblifeless: but then we copy pasta elsewhere. it is all ugly21:31
lifelessso, what I guess I'm missing is why we can't write a single shell body, with the union of all lines21:31
lifelessand a set of default values21:31
lifelessand pass in overrides21:32
lifelessat the moment we pass in overrides already21:32
clarkblifeless: we could, the branch-override is a step in that direciton21:32
lifelessbut write a customised shell body21:32
clarkbbut I don't want to do that in one go21:32
lifelessok21:32
clarkbtoo many moving parts and confusion (the existing override is already confusing people)21:32
lifelessyou know the guts better than  :)21:32
lifelessthan *I*21:33
lifelessMithrandir: I'm a huge skeptic of turing complete template languages; pystache/pybars are much much much easier to reason about IME21:33
lifelessnot to mention test.21:33
mgagneclarkb: can't you create a customer builder wrapping shell and pass argument to it from job-template?21:33
clarkblifeless: on that train of thought, imo jinja2 is better than crazy bash that we can't test at all without throwing it at jenkins21:33
Mithrandirjinja+yaml is unreadable.21:34
Mithrandircan't we fix it using includes?21:34
lifelessclarkb: and followup q in the layout21:34
clarkbincludes could cut back on some of the repition21:34
Mithrandiror add macros to the language?21:34
mgagneclarkb: just like coverage and docs macros are doing21:34
Mithrandir(which are basically the same thing)21:34
clarkbinclude d-g base ; include d-g publisher21:34
lifelessclarkb: the bash you refer to is devstack-gate?21:34
clarkbthen make the copy pasta as minimal as possible21:34
Mithrandirlifeless: I don't like them either.21:34
clarkblifeless: no the shell that wraps devstack-gate21:35
clarkblifeless: where you are indicating we can put overrides21:35
lifelessclarkb: oh right; so yeah rendering down to the right bash is a good idea; still doesn't need turing complete templates ;)21:35
lifelessMithrandir: when you say you don't like them, do you mean you don't like pystache/pybars ? [to what 'them' do you refer?]21:36
*** guitarza1 is now known as guitarzan21:36
clarkblifeless: and responded :)21:36
Mithrandirlifeless: I don't like turing complete templating languages either21:37
clarkbfwiw I am not a fan of it especially after seeing a lot of salt yaml. But I think you need something like that if you want to actually remove most of the repetition21:38
*** ianw has quit IRC21:38
clarkbyou could use includes but then you need a job template for each individual job still which is quite a bit of repetition21:38
clarkbyou can shove the logic into the jenins jobs themselves, but they really shouldn't care21:38
*** ianw has joined #openstack-infra21:38
*** CaptTofu has quit IRC21:41
*** CaptTofu has joined #openstack-infra21:41
*** jcooley_ has joined #openstack-infra21:41
lifelessclarkb: mmm I don't think you do21:43
lifelessclarkb: but we can noodle on that later21:43
lifelessclarkb: so is that check vs gate slave selection documented anywhere?21:43
*** jcooley_ has quit IRC21:43
mgagneclarkb: why not something like that? http://paste.openstack.org/show/54659/21:43
clarkbmgagne: because it shoves more of the logic into shell where it is already relogiced in devstack-gate21:44
*** jcooley_ has joined #openstack-infra21:44
clarkbmgagne: that would work but I am not a fan of it (did it with branch-override though)21:44
clarkblifeless: https://git.openstack.org/cgit/openstack-infra/config/tree/modules/openstack_project/files/jenkins_job_builder/config/projects.yaml#n104521:45
*** CaptTofu has quit IRC21:45
mgagneclarkb: http://www.technobunny.org/images/okay.png21:46
*** changbl has quit IRC21:47
lifelessclarkb: so only by internalising the ruleset?21:47
lifelessclarkb: I was meaning somewhere in the ci docs, for instance.21:47
clarkblifeless: no its not something in the ci docs, because the values change and we don't typically advertise the classes of slaves in that way21:47
clarkbit is configuration21:48
lifelessclarkb: which is code :)21:48
lifelessclarkb: which we document :>21:48
*** rongze has quit IRC21:50
*** changbl has joined #openstack-infra21:52
fungiit's a (hopefully temporary?) implementation detail stemming from one of our two current providers offering noticeably faster servers than the other22:07
fungii suspect we didn't think about documenting it because subconsciously we hope it will just go away22:08
fungias we'd rather not have special snowflakes there22:10
clarkbit is really interesting how the graph spikes at http://status.openstack.org/elastic-recheck/ on sunday. I am guessing only a handfully of jobs have run making the failures seem really bad22:10
fungithere was a fair uptick in new patchsets and approvals yesterday actually22:11
fungiand i was seeing a substantial percentage of those in the gate pipeline hitting random devstack/tempest/grenade issues22:11
*** Daviey has quit IRC22:19
*** rongze has joined #openstack-infra22:21
*** elasticio has quit IRC22:21
*** thomasem has quit IRC22:22
*** changbl has quit IRC22:22
*** changbl has joined #openstack-infra22:23
lifelessfungi: it's a barrier to learning about the system for folk that want to become contributors :)22:23
fungilifeless: i completely concur!22:23
fungiugh, the inode density on docs-draft  will probably require some fs tuning as it gets fuller. we're likely to run out of available inodes before available blocks there22:27
fungiaveraging 9kB/inode22:29
*** rongze has quit IRC22:29
fungiwell, not tuning per se (apparently can't increase the inode limit on a filesystem once it's created). i guess we can reduce the retention period for those and see about a cinder quota bump so we can recreate it on a fresh volume if/when it gets to be a problem22:39
fungianother reason i miss reiserfs22:39
clarkbmgagne: actually I wonder if we can use the env-inject stuff instead to reduce the env vars set by the shell things22:42
mgagneclarkb: stop reading my mind!!22:42
clarkbthen have a generic var value pair of things that individual jobs fill in22:42
mgagneclarkb: seriously, I just thought about it 2 minutes ago22:43
clarkbnice22:43
*** Daviey has joined #openstack-infra22:43
mgagneclarkb: please only read my openstack related thoughts, thanks22:43
mgagneclarkb: http://ci.openstack.org/jenkins-job-builder/properties.html#properties.inject and http://ci.openstack.org/jenkins-job-builder/wrappers.html#wrappers.inject exist22:46
lifelessfungi: yeah, but you need to hope reiser misses you :P <boom tish>22:46
mgagneclarkb: or http://ci.openstack.org/jenkins-job-builder/builders.html#builders.inject I don't know why you should one over the others.22:46
clarkbmgagne: yup there are subtle differences that zaro ran into22:46
clarkbmgagne: I don't remember what those differences are, but zaro should22:47
clarkbI thought we documented them22:47
* mgagne summons zaro22:47
fungiohhh, i totally set you up for that one22:47
*** jcooley_ has quit IRC22:48
mgagnecan't believe you can't increase inode limits on an ext filesystem...22:48
mgagneused to be able to increase it on netapp appliances ^^'22:49
lifelessmgagne: generally you just make it with a huge number of inodes22:49
lifelessthe overhead of more inodes is actually fairly low22:50
lifelessIIRC just the bitmap at the start22:50
fungiwe did make it with a huge number of inodes... just needed a huger (sic.) number of inodes22:50
mgagnelifeless: yes, but most of the time, you don't know you will run out of inodes someday :-/22:50
clarkbooh _david_ wrote a gerrit monitoring plugin \o/22:52
*** rongze has joined #openstack-infra22:55
*** rongze has quit IRC23:01
*** AaronG_Zzz is now known as AaronGr23:02
*** afazekas has quit IRC23:04
*** afazekas has joined #openstack-infra23:04
*** pcrews has joined #openstack-infra23:07
openstackgerritMax Rydahl Andersen proposed a change to openstack-infra/jenkins-job-builder: Add support for sidebar-links plugin via properties.  https://review.openstack.org/6071023:09
*** AaronGr is now known as AaronG_Zzz23:16
mordredclarkb: keystone thread responded to23:20
bknudsonmordred: thanks... this is something I've got some time to work on23:25
mordredbknudson: woot!23:25
mordredbknudson: honestly- if clarkb zaro and I can ever figure out what's up with that patch I posted23:26
mordredwe'll have a bunch more goodness that will make reasoning about keystone easier23:26
mordredbknudson: I've been ignoring your thing until I got that up and running23:26
clarkbbknudson: also, if mordred didn't stresss it enough, get the tests in tempest and 99% of everything else is take care of23:27
* mordred stresses the importance of beer23:27
bknudsonI'm trying to figure out if this is a matter of writing new tests or moving the existing tests from keystone23:28
bknudsonthe existing tests in keystone test the client functionality, but it's more of a side-effect... they're to test keystone.23:29
bknudsonbut I don't want to lose any test coverage, so that argues for starting with the keystone tests23:30
mordredI _believe_ that you probably can jut move them to tempest-ish23:31
*** yamahata_ has joined #openstack-infra23:43
*** rongze has joined #openstack-infra23:57

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