Saturday, 2014-01-25

sdagueso why did 62370,2 just restart?00:00
clarkbsdague: did it get reparented to avoid the git merge fail?00:00
*** ociuhandu_ has quit IRC00:00
clarkbjeblair: mordred: https://review.openstack.org/#/c/68782/ getting that approved would be excellent00:00
openstackgerritA change was merged to openstack-infra/elastic-recheck: Add fingerprint for bug 1271962  https://review.openstack.org/6903100:01
jeblairclarkb: i have like 5 million messages from jenkins to delete from the gerrit db.  i'm a bit busy.00:01
*** alexpilotti has joined #openstack-infra00:01
clarkbwoah00:02
*** pcrews has joined #openstack-infra00:02
openstackgerritPeter Portante proposed a change to openstack-infra/elastic-recheck: Add fingerprint for bug 1268631  https://review.openstack.org/6906800:02
portanteflashgordon: whack-a-mole00:02
flashgordonportante: don't remind me00:02
*** _ruhe is now known as ruhe00:04
jeblaironly 158700:05
clarkbsdague: also possible that the change ahead of it had a new patchset pushed to it00:05
jeblairto delete00:05
portanteflashgordon: what to do about double hits?00:05
clarkbcurrnetly digging through logs to see if I can give a real reason00:05
flashgordonportante: this is a little nitty but https://review.openstack.org/#/c/69067/00:05
*** praneshp has joined #openstack-infra00:05
flashgordonportante: not sure, maybe and another message:"" section00:06
flashgordonoh and great catc hon 6906700:06
*** flaper87 is now known as flaper87|afk00:07
*** sarob has joined #openstack-infra00:07
portanteflashgordon: thanks, luck really00:08
openstackgerritPeter Portante proposed a change to openstack-infra/elastic-recheck: Updated fingerprint for bug 1261182  https://review.openstack.org/6906700:09
clarkbsdague: http://paste.openstack.org/show/61870/ yup the merge fail00:11
*** thuc has quit IRC00:12
*** thuc has joined #openstack-infra00:12
clarkbjeblair: do we know why those comments happened?00:13
jeblairclarkb: yes00:14
*** ociuhandu_ has joined #openstack-infra00:14
jeblairclarkb: it's stopped.  i'll be more verbose in a bit.00:14
*** ociuhandu_ has quit IRC00:16
*** thuc has quit IRC00:17
*** boris-42 has quit IRC00:19
openstackgerritA change was merged to openstack-infra/elastic-recheck: Updated fingerprint for bug 1261182  https://review.openstack.org/6906700:22
vishydo you guys still use https://jenkins.openstack.org/ to backend the zuul jobs?00:22
vishyor has it been moved somewhere else?00:22
clarkbvishy: we do, but only a very small percentage of them00:23
clarkbvishy: we have 8 jenkins masters now, that is one of them00:23
openstackgerritJoshua Harlow proposed a change to openstack-dev/hacking: Add a tool that can fix the import order  https://review.openstack.org/6898800:25
*** ociuhandu_ has joined #openstack-infra00:27
jeblairclarkb: it seems that the check trigger was triggering itself and looping00:29
jeblairclarkb: i've cleaned up the duplicate messages in gerrit's db00:29
flashgordonportante: even with all your work the classifcation rate is dropping00:30
jeblairclarkb: and disabled puppet and reloaded zuul without that trigger in the config00:30
*** ociuhandu_ has quit IRC00:31
clarkbjeblair: ok00:31
portanteflashgordon: ugh00:31
jeblairclarkb: at least, for messages with > 5 copies00:31
portanteflashgordon: gotta head home, but will be available later to help, and hopefully on Monday00:32
flashgordonportante: thanks00:33
portanteflashgordon: though that honkin' list of neutron jobs went away00:33
flashgordonportante: :)00:33
portanteon the elastic recheck uncat page00:33
*** gyee has quit IRC00:33
*** UtahDave has quit IRC00:34
*** ruhe is now known as _ruhe00:37
*** sarob has quit IRC00:39
*** CaptTofu has joined #openstack-infra00:40
*** branen has joined #openstack-infra00:40
*** yamahata has joined #openstack-infra00:40
openstackgerritJames E. Blair proposed a change to openstack-infra/config: Revert "Ensure a recent (48h) check result"  https://review.openstack.org/6907100:40
jeblairclarkb: so it looks like if you leave the same vote value, the grantedon time doesn't get updated00:41
flashgordonwow 50 bugs on http://status.openstack.org/elastic-recheck/00:41
clarkbjeblair: interesting00:41
clarkbjeblair: that is gerrit not updating the value?00:41
jeblairclarkb: right00:41
jeblairclarkb: an easy fix would be to have the check queue also leave a 0 vote...00:42
jeblairclarkb: though that increases the chattiness... (2 msgs for every jenkins run)00:42
jeblairclarkb: or we could rework the feature so that it looks for the timstamp of a message....00:43
jeblairclarkb: ....though the messages aren't tied to the votes, so it'd be like "a verified vote from jenkins and also the newest message from jenkins is older than 48 hours)00:43
clarkbjeblair: I hate saying this. but what if we fix gerrit00:44
jeblairclarkb: is it a bug?00:44
clarkbjeblair: I would say so. If I explicitly leave a +1 vote the timestamp should update00:44
clarkbupstream will probably disagree with me though :)00:44
jeblairclarkb: want to make that change?  my employer has not signed the gerrit cla.00:46
jeblairclarkb: (or maybe ask em in irc first)00:46
*** melwitt has quit IRC00:46
clarkbjeblair: sure /me hops into #gerrit00:46
jeblairclarkb: it is true that in the command line, you can explicitly leave or not leave a vote when you leave a message.. so your suggestion makes sense there.00:46
jeblairclarkb: that is not the case for the webui, but that's not what we're dealing with00:47
clarkbjeblair: I am assuming I won't get much of a response at this end of the week00:50
jeblairclarkb: just double checked: same behavior in 2.800:50
fungimmm, good. i didn't miss much00:52
clarkbjeblair: you are in channel, let us see what happens00:52
* fungi popped back into that channel too, out of curiosity00:53
jeblairclarkb: i feel like the best short-med term solution is to make check chatty.  :/00:53
clarkbjeblair: wfm00:54
fungifor future reference, if anyone finds themselves stranded in the slc airport, the squatters brewery has opened a pub in concourse c00:54
clarkbjeblair: also the gerrit code baseis quite foreign to me now00:54
*** ok_delta__ has joined #openstack-infra00:55
jeblairclarkb: i'm going to think about the other option though, and do neither for now.  maybe it wouldn't be too bad to say "have heard from jenkins within this amount of time and also have a +1 vote from jenkins"00:55
*** dkehn has quit IRC00:56
clarkbjeblair: found the line of code causing the problem, I should be able to fix this easily if upstream will take the patch00:56
*** ativelkov has left #openstack-infra00:56
jeblairfungi: are your nodepool cleanup hacks still working after my oom adventure?00:56
*** ativelkov has joined #openstack-infra00:56
openstackgerritA change was merged to openstack-infra/devstack-gate: Adding the tripleo repositories to PROJECTS  https://review.openstack.org/6864500:57
*** ok_delta has quit IRC00:58
fungijeblair: i stopped doing oob deletes after we restarted nodepool yesterday, though i can start again (i did some spot deletes of stuff >5 hours old a couple times since)00:58
*** dkehn has joined #openstack-infra00:58
jeblairoh ok.  no worries00:59
openstackgerritA change was merged to openstack-infra/config: Revert "Ensure a recent (48h) check result"  https://review.openstack.org/6907100:59
fungiit was more important when we were flatlined at quota all the time, since deleting stuff which the periodic cleanup did't (or before it noticed) was buying us extra capacity00:59
fungisince clarkb's awesome throttle work, it's not really been as necessary01:00
*** thuc has joined #openstack-infra01:01
fungiand once we're cool with lifeless's delete queuing patches, it should hopefully be moot anyway01:01
*** thuc has quit IRC01:02
fungialso, looks like 68782 and 68990 are probably good to go, which ought to remove (or at least significantly reduce) one more cause of gate resets01:02
*** thuc has joined #openstack-infra01:02
fungii'm digging back into the mysql slow log patch now in earnest, which will also likely help with that01:03
openstackgerritJoshua Harlow proposed a change to openstack-dev/hacking: Add a tool that can fix the import order  https://review.openstack.org/6898801:07
openstackgerritJoshua Harlow proposed a change to openstack-dev/hacking: Add a tool that can fix the import order  https://review.openstack.org/6898801:08
*** SumitNaiksatam has quit IRC01:09
jeblairfungi: aprvd those 201:12
fungikeen01:12
clarkbwoot01:13
jeblairrestarted puppet on zuul01:14
*** CaptTofu has quit IRC01:15
openstackgerritJoshua Harlow proposed a change to openstack-dev/hacking: Add a tool that can fix the import order  https://review.openstack.org/6898801:15
jnollerif i offered candy, anyone able to +1 https://review.openstack.org/#/c/69008/601:15
mordredjnoller: what kind of candy?01:17
jnollermordred: any kind you want, or dogecoin stickers ;)01:17
*** CaptTofu has joined #openstack-infra01:18
clarkbre neutron kernel trouble. precise does have newer kernels available that we could try01:18
clarkbI wonder if that would possibly make cloud archive libvirt happy too01:19
clarkbsomething to ponder01:19
jnollerthanks jeblair01:19
clarkbmordred: for changes like 69008 should we go ahead and approve them and deal with fallout later?01:19
clarkbmordred: I am not sure what the current situation of manage projects is01:19
clarkbmordred: maybe we can have puppet set an at job to avoid puppet weirdness?01:20
clarkbmordred: that is a really crazy idea, but possibly it will work01:20
jnollerclarkb: are project jobs breaking things?01:21
jnollerif project jobs are hanging zuul / gate I'd rather wait on this and see if I can help on gate01:22
clarkbjnoller: no, manage-projects just fails hard to properly create projects01:22
clarkbjnoller: mordred cannot reproduce when run manually so current assumption is that hen puppet runs it does so in a weird way breaking it01:22
jnollerouch01:22
clarkbjnoller: so my silly idea is have puppet create an at job for now and not have puppet rtigger it directly :)01:23
jnollerclarkb: that's "eww" but effective01:23
jnollerI don't know enough puppet to dig in after the kids go to bed01:24
clarkbjeblair: 2.4 and 2.8 have completely different code paths for vote updates, but neither looks complicated. I will probably give fixing gerrit a shot01:24
openstackgerritJoshua Harlow proposed a change to openstack-dev/hacking: Add a tool that can fix the import order  https://review.openstack.org/6898801:24
mordredclarkb: it's turned off in puppet01:25
mordredclarkb: you can approve at will - we just need to manually run it01:25
clarkbmordred: awesome thanks01:25
mordredclarkb: what's the "vote update/fix gerrit" thing?01:26
clarkbmordred: when you vote +1 multiple times in jenkins without changing any values, the grantedon column in the DB doesn't update01:27
clarkbmordred: so jeblairs zuul stuff can't reliably check if jenkins voted more recently than yesterday01:27
clarkbmordred: I pinged in #gerrit and they seem to think that is a bug and sicne I am CLAd with gerrit I can give a fix for that a shot01:27
mordredclarkb: ah. excellent01:28
clarkbin fact it just does a rollback if the only thing that would change is the timestamp01:29
jnollerthanks you guys01:29
clarkbjnoller: note, once that merges and after puppet applies one of us will need to run the manage-projects script by hand01:30
jnolleryup01:30
jnollerjust saying thank you01:30
jnollerI'm opening the initial member bug01:30
*** yamahata has quit IRC01:31
clarkbI might also trick zaro into looking at it since he has well versed in gerrit testing atm01:31
*** markmcclain has joined #openstack-infra01:33
* clarkb walks home. AFK for a bit01:36
*** gokrokve has joined #openstack-infra01:37
fungidevstack patch for making query logging optional... https://review.openstack.org/6907501:38
*** dtroyer has quit IRC01:38
*** dtroyer has joined #openstack-infra01:39
openstackgerritA change was merged to openstack-infra/config: Add python-openstacksdk, see https://wiki.openstack.org/wiki/UnifiedSDK  https://review.openstack.org/6900801:39
*** thuc has quit IRC01:40
*** thuc has joined #openstack-infra01:40
*** wenlock has quit IRC01:42
*** jhesketh_ has joined #openstack-infra01:43
*** thuc has quit IRC01:45
fungiclarkb: if we're approving api/sdk projects into stackforge at the moment, https://review.openstack.org/61954 and https://review.openstack.org/62069 are probably ready to go too01:45
*** greghaynes has quit IRC01:46
*** SumitNaiksatam has joined #openstack-infra01:46
*** AlexF_ has joined #openstack-infra01:47
mordredfungi: I'm thinking once we land all of these I can log in and run by hand again and see if anything breaks for me01:49
clarkb++01:49
mordredjeblair: ^^ if you're around, could you un-2 that?01:50
mordredjeblair: https://review.openstack.org/#/c/62069/ that is01:50
fungii've still got a question out to mark c. about potential trademark issues with including the term "openstack" in the short names of unofficial projects, but i can't seem to get a timely answer so i'm in favor of just assuming it's fine until someone actually says it's not01:52
clarkbypu need to twitter at him01:54
fungii'm just not cool enough for teh tweeterz01:54
*** greghaynes has joined #openstack-infra01:55
*** CaptTofu has quit IRC01:56
flashgordonmordred: you had an idea about that one I think ^01:56
*** flashgordon is now known as jog001:57
funginova casual nick fridays only extend to the end of utc friday i guess01:57
mordredI think it's fine, personally01:57
fungii do too01:57
mordredthe concern is whether or not there is an implication that the thing is part of openstack, or is a thing intended to consume opensatck01:58
fungii would have liked to get a semi-official confirmation to that effect, but it doesn't strike me as something which risks brand confusion01:58
mordredit pretty clearly seems like the later to me, as it stands01:58
*** xchu has quit IRC01:58
jog0fungi: it extends to friday business hours and being 6pm here01:58
mordredright01:58
jog0OpenStack's X vs X for OpenStack01:59
jog0you see what DefCore wants to do?01:59
fungijog0: the issue being that in short project names there's limited real-estate to clarify the difference02:00
mordredjog0: no. what do they want to do?02:00
fungijog0: i recommend dialling into future defcore conference calls. after all, they reach out to the community by having an irc channel where they can post the join.me url for their meetings02:01
jog0mordred: see #5 http://robhirschfeld.com/2014/01/06/openstack-defcore/02:01
*** markmcclain has quit IRC02:02
jog0fungi: I defer to my boss for that kind of work usually. cough cough02:02
mordredugh02:02
fungialso, they seem to be conveniently scheduled at the same times as tc and project meetings, and don't appear on the official meeting schedule02:02
mordredyeah02:02
mordredthat's why I've stopped going to them02:02
jog0they want 'OpenStack Compute Project Nova ' not '“OpenStack Nova'02:02
mordredwhat I enjoy about programs vs. projects ...02:02
mordredis taht they're assuming programs are a concept that has anything to do with the board02:03
mordredwhich I believe would need to be asserted first02:03
jog0heh, or something a non dev needs to know about02:03
fungiit was also suggested on the last call that the board should scale back the official list of programs to just compute, storage and network02:03
mordredit's an interesting point they make though, tbh02:03
*** kraman has quit IRC02:04
mordredand one that we've also only danced aroudn a little bit02:04
mordredsuch as Deployment/TripleO being "official" but none of the projects in it being "integrated"02:04
jog0mordred: one massive issue with the official project thing: too long of a name02:04
mordredfungi: well - the board can't do that02:04
fungii know, that's why i found the suggestion amusing02:04
mordredfungi: so I'll be excited to see them attempt to try02:04
mordredwhat the board CAN do02:05
fungii think some board members (present company excluded) don't realize where their reach actually ends02:05
mordredis EXPAND the list of official things past nova and swift02:05
mordred:)02:05
jog0is write blog posts02:05
mordredjog0: yes. and write blog posts02:05
fungiblog posts fix everything02:05
jog0anyway mordred have fun with that one02:07
* jog0 wonders off while laughing at the board02:07
jog0wanders*02:08
clarkbjog0: wondering off seems like fun too02:10
clarkblike day dreaming02:10
fungimordred: there seems to be a lot of noise still about designing refstack... what if we just deployed havana branches of the integrated projects with mostly-default but minimally redundant configuration on top of the tripleo bm cloud and ran a subset of tempest tests against it?02:10
clarkbfungi: or just tripleo itself02:11
clarkbwhcih should be close to master02:11
fungithat seems to me to be at least 90% of the "challenge" met right there02:11
openstackgerritA change was merged to openstack-infra/config: Adding empty gce-api project to stackforge  https://review.openstack.org/6195402:13
clarkbfungi: so it looks like one of those changes you pointed out has been approved and the other has a jeblair -202:13
fungiand i say havana because the defcore committee still seems focused on coming up with havana validation requirements by the icehouse release (why they don't see benefit in developing in parallel with icehouse in time for the icehouse release is still a mystery to me)02:13
clarkbI figure we should try running by hand with the ones approved so far and deal with jeblair -2 in the future?02:13
fungisure02:13
clarkbfungi: hrm, ya seems like they should be looking forward02:13
clarkbsince we move and near c and havana is old02:13
*** changbl has joined #openstack-infra02:14
fungiespecially since havana will have about 9 months of security support left when icehouse releases02:14
*** oubiwann_ has joined #openstack-infra02:14
*** AlexF_ has quit IRC02:22
*** salv-orlando has quit IRC02:25
*** esker has joined #openstack-infra02:25
*** VijayT has quit IRC02:28
*** ArxCruz has quit IRC02:30
*** salv-orlando has joined #openstack-infra02:34
*** saper has quit IRC02:35
clarkbwe should install http://bropages.org/ on all of our servers02:36
clarkb>_>02:36
fungiwow that is so... just wow. next they'll make it capable of scraping stackexchange and ubuntu forums for questionable advice02:39
*** saper has joined #openstack-infra02:40
openstackgerritA change was merged to openstack-infra/config: Expands ldap to deal with nonstandard ldap configs.  https://review.openstack.org/6445602:44
jnollerdat moment when you keep refreshing stackforge02:44
mordredclarkb: um. wow02:44
mordredclarkb: think I should run the thing for jnoller ?02:44
jnollerlol02:45
jnollerI should be parenting02:45
jnollerbut they're cool right now02:45
mordredjnoller: bah02:45
*** emagana_ has quit IRC02:45
fungimordred: just say you did and then let him keep refreshing it for days02:45
mordredjnoller: stick em ina box with some cobras02:46
mordredjnoller: either they learn to deal with that situation or they don't02:46
clarkbmordred: I think you should go for it02:46
*** emagana has joined #openstack-infra02:46
mordredrunning02:47
mordredjnoller: /var/lib/git/stackforge/python-openstacksdk.git/02:50
mordredgah02:50
mordredjnoller: http://git.openstack.org/cgit/stackforge/python-openstacksdk02:50
mordredis what I meant02:50
*** emagana has quit IRC02:51
clarkbmordred: and the other project was created fine as well?02:51
clarkbmordred: I am seriously tempted to write a change that schedules an at job for manage-projects02:51
clarkbsuch a hack but if it works02:51
jnollermordred: i love you02:51
clarkbmordred: also, I noticed that our pull request closer bot doesn't seem to work anymore02:51
clarkbmordred: I think that may be related to the number of api requests manage-projects is making?02:52
*** VijayT has joined #openstack-infra02:52
mordredclarkb: haha02:52
mordredclarkb: well... funny story02:53
mordredI was thinking about making that be a zuul thing once I wrote the github zuul trigger thing02:53
mordredclarkb: because then instead of polling for things, we can just have zuul respond to github events and fire an action02:53
mordredwhich is hella more betterer02:53
clarkbmordred: neat02:53
*** jerryz has quit IRC02:53
fungiclarkb: the pull request closer is broken for a while. there are tracebacks in /var/spool/mail/gerrit2 for it, but i haven't had time to dig into the cause02:54
*** gokrokve has quit IRC02:54
*** gokrokve has joined #openstack-infra02:54
fungii've had a paste of it up in a browser tab at home for weeks, hoping i'd get far enough down the stack of other to-do tabs to exhume it02:55
mordredclarkb: yup. gce-api worked too02:56
*** gokrokve_ has joined #openstack-infra02:57
*** gokrokve has quit IRC02:59
*** AlexF_ has joined #openstack-infra02:59
*** wenlock has joined #openstack-infra03:01
*** gokrokve_ has quit IRC03:02
fungii will express again to no one in particular that it's really frustrating that doesn't "just work" when called in the scope of a puppet exec03:02
mordredyah03:03
*** rakhmerov has quit IRC03:05
fungiahh, time to board the next flight. bbiab03:07
*** pcrews has quit IRC03:07
clarkbfungi: have fun in pheonix03:07
clarkbI suppose not doing the rename works out because you will be in the wrong part of the coutrny03:07
*** oubiwann_ has quit IRC03:14
openstackgerritA change was merged to openstack/requirements: Allow use of oslo.messaging 1.3.0a4 from pypi  https://review.openstack.org/6804003:16
*** beekneemech has quit IRC03:21
*** gokrokve has joined #openstack-infra03:22
*** dcramer_ has joined #openstack-infra03:23
*** bnemec has joined #openstack-infra03:25
*** sarob has joined #openstack-infra03:26
*** sarob has quit IRC03:27
openstackgerritA change was merged to openstack/requirements: Add openstack-doc-tools  https://review.openstack.org/6536103:28
*** praneshp has quit IRC03:31
*** praneshp has joined #openstack-infra03:33
*** rakhmerov has joined #openstack-infra03:34
*** AlexF_ has quit IRC03:39
*** CaptTofu has joined #openstack-infra03:52
*** jasondotstar has joined #openstack-infra04:02
*** sarob has joined #openstack-infra04:05
*** coolsvap has quit IRC04:08
*** praneshp has quit IRC04:11
*** gokrokve has quit IRC04:21
*** gokrokve has joined #openstack-infra04:22
*** rwsu has quit IRC04:25
*** gokrokve has quit IRC04:26
*** esker has quit IRC04:26
*** senk has joined #openstack-infra04:26
*** esker has joined #openstack-infra04:27
*** mriedem has joined #openstack-infra04:27
*** sarob has quit IRC04:29
*** ok_delta__ has quit IRC04:30
*** senk has quit IRC04:31
*** mriedem has quit IRC04:36
*** mriedem has joined #openstack-infra04:41
*** SergeyLukjanov_ is now known as SergeyLukjanov04:42
*** esker has quit IRC04:49
*** mriedem has quit IRC04:51
*** VijayT has quit IRC04:55
*** rakhmerov has left #openstack-infra05:03
*** greghaynes has quit IRC05:05
*** greghaynes has joined #openstack-infra05:10
*** CaptTofu has quit IRC05:12
*** CaptTofu has joined #openstack-infra05:13
*** HenryG has quit IRC05:14
*** AaronGr is now known as AaronGr_Zzz05:16
*** CaptTofu has quit IRC05:18
clarkbfungi: http://www.ustream.tv/recorded/42950577 now we know what you do in utah05:19
*** dstanek has quit IRC05:22
fungithe wireless internet on my next flight had better not be broken like on the one i was just on. growl05:23
clarkbfungi: I'm sorry. also you should've helped with the drawing because lol05:23
fungiyeah, i didn't even realize they did that05:25
fungipretty hilarious05:25
fungii guess that's what claire was editing footage for yesterday when we were sitting quietly in the lounge getting work done05:26
*** pcrews has joined #openstack-infra05:26
*** pcrews has quit IRC05:36
*** SergeyLukjanov is now known as SergeyLukjanov_a05:36
*** SergeyLukjanov_a is now known as SergeyLukjanov_05:37
*** praneshp has joined #openstack-infra05:38
openstackgerritA change was merged to openstack/requirements: nova api validation fw requires jsonschema >= 2.0.0  https://review.openstack.org/6646405:40
openstackgerritA change was merged to openstack/requirements: Remove oauth2 requirement  https://review.openstack.org/6842205:40
openstackgerritA change was merged to openstack/requirements: Bump python-savannaclient version to >=0.4.1  https://review.openstack.org/6672305:55
*** SergeyLukjanov_ is now known as SergeyLukjanov06:08
*** SergeyLukjanov is now known as SergeyLukjanov_06:09
*** jasondotstar has quit IRC06:12
*** wenlock has quit IRC06:13
mordredAlex_Gaynor, jnoller: I just cookiecuttered your new repo. merry christmas06:30
*** coolsvap has joined #openstack-infra06:34
*** coolsvap has quit IRC06:34
*** harlowja is now known as harlowja_away06:52
*** CaptTofu has joined #openstack-infra07:14
*** CaptTofu has quit IRC07:18
*** morganfainberg is now known as morganfainberg|z07:29
*** markwash has quit IRC07:43
*** yamahata has joined #openstack-infra07:45
*** markwash has joined #openstack-infra08:02
*** denis_makogon_ has joined #openstack-infra08:03
openstackgerritA change was merged to openstack-infra/config: Add logging to #openstack-meeting-3  https://review.openstack.org/6715208:08
*** dkliban has quit IRC08:11
*** lttrl has joined #openstack-infra08:21
*** openstack has joined #openstack-infra08:22
*** yolanda_ has joined #openstack-infra08:33
*** lttrl has quit IRC08:35
*** afazekas has joined #openstack-infra09:04
*** CaptTofu has joined #openstack-infra09:15
*** luqas has joined #openstack-infra09:16
*** CaptTofu has quit IRC09:19
*** praneshp has quit IRC10:00
*** markwash has quit IRC10:01
*** max_lobur_afk is now known as max_lobur10:02
*** yamahata has quit IRC10:19
*** luqas has quit IRC10:31
*** boris-42 has joined #openstack-infra10:49
*** luqas has joined #openstack-infra11:01
*** CaptTofu has joined #openstack-infra11:15
fungiyay! my maildir for openstack mailing lists no longer claims >1500 unread messages11:17
*** CaptTofu has quit IRC11:20
*** wenlock has joined #openstack-infra11:26
*** wenlock has quit IRC11:30
*** coolsvap has joined #openstack-infra11:54
*** denis_makogon_ has quit IRC11:57
*** david-lyle has quit IRC12:04
sdaguefungi: craziness12:13
*** yolanda_ has quit IRC12:21
*** coolsvap has quit IRC12:30
*** yolanda_ has joined #openstack-infra12:32
*** coolsvap has joined #openstack-infra12:32
*** jasondotstar has joined #openstack-infra12:37
*** alexpilotti has quit IRC12:39
*** flaper87|afk is now known as flaper8712:44
*** yolanda_ has quit IRC12:50
*** yolanda_ has joined #openstack-infra12:52
*** CaptTofu has joined #openstack-infra12:53
*** yolanda_ has quit IRC12:57
*** CaptTofu has quit IRC12:57
*** CaptTofu has joined #openstack-infra13:04
*** coolsvap has quit IRC13:33
fungicraziness is that i still have another 4 hours in the phx airport. then i hop a plane which touches down in aus and bna on the way to rdu13:38
fungibut it was good for catching up on e-mail and bugs13:39
openstackgerritSean Dague proposed a change to openstack-infra/elastic-recheck: stop being rediculous with our time formats  https://review.openstack.org/6876513:40
*** nati_ueno has joined #openstack-infra13:40
fungithe gate also did a lot of catching up13:41
sdagueyep13:44
sdagueyeh, I noticed the fungi furious emailing13:44
sdaguethe sliding window definitely helped13:45
fungicertainly seems that way13:46
fungithat and fixing or mitigating lots of bugs13:46
openstackgerritA change was merged to openstack-infra/elastic-recheck: stop being rediculous with our time formats  https://review.openstack.org/6876513:47
sdagueyep13:47
anteayafungi: phx is not a bad airport13:48
sdaguefungi: also curious that the tripleo jobs are stacking up on waiting for something13:48
anteayajust sux that there are no beds in there13:48
fungianteaya: yeah, it's been pleasant. i found several places where they had televisions running cartoons13:48
*** dkliban has joined #openstack-infra13:48
sdaguenice13:48
fungialso, had the place mostly to myself. very peaceful except for the vacuum cleaners13:49
anteayafungi: yay for cartoons13:49
anteayaha ha ha13:49
fungisdague: gate-tripleo-deploy jobs use a node type specific to the tripleo bare metal cloud13:50
sdaguefungi: yep, that's what I expected13:50
sdagueso the tripleo cloud is not working to provide those resources?13:50
sdagueseems like waiting for 3+hrs probably means it's just down13:51
anteayajust looking at my inbox13:51
fungilooks like there are two tripleo-precise nodes building in the tripleo-test-cloud provider currently, according to nodepool list13:51
fungibut they may be failing to build13:51
anteayathe emails dealing with 3rd party test accounts has me scared13:52
anteayajust in terms of the number of them13:52
fungilifeless: http://paste.openstack.org/show/61879/13:53
fungianteaya: embrace the chaos13:53
* anteaya runs screaming into the abyss13:53
sdagueyeh, I think we need to realize that gerrit ui isn't really sufficient for this. And rethink how we are collecting results.13:54
fungiwhat frightened me was the thread on the -dev ml from some new driver author stating that they'd installed jenkins and wanted to know how they were supposed to configure it to contribute code to the project13:54
sdagueheh13:54
* fungi cries13:55
fungilifeless: and now with actual traceback http://paste.openstack.org/show/61880/13:57
fungithat's probabably more helpful13:58
fungiprobababababably13:58
anteayaso on that note, these new 3rd party testing accounts, do they have verification voting on or off by default?13:58
* anteaya tenses up13:58
fungiwe still haven't approved the documentation letting people know that all those instructions about how to vote won't actually work until some nebulous authority (not it!) confirms they're "stable"13:59
anteayaso that means they have voting turned on?13:59
anteaya:(13:59
fungiyep13:59
anteayaa sense another misarable week on the horizon14:00
* anteaya cries14:00
fungithe last teams whose service accounts i've disabled keep coming back to me trying to explain why it's okay to turn them back on again. i really don't want to be the decision maker on that since it doesn't affect me directly14:00
*** blitzrage is now known as leifmadsen14:01
fungimany of them seemed to be under the impression that if they weren't running voting systems by last week, their drivers were all going to be deleted14:02
anteayayeah14:03
anteayahence mikal's desire to get this issue in front of the tx14:03
anteayatc14:03
anteayawhich happened for the first time last week14:03
fungithus a scramble at the last moment to add numerous partially-working bots14:03
*** nati_ueno has quit IRC14:03
anteayabut needs more conversationi14:03
anteayayes14:03
*** nati_ueno has joined #openstack-infra14:04
anteayaand of the dozen or so just added, I have money on better than half of them not getting voting right14:04
anteayafor various and sundry reasons14:04
anteayawell when I get hit up, I will just come back here and request that voting get turned off on various accounts when it happens14:05
fungii've +2'd the documentation update for it, but don't want to go approving a proscriptive policy text as the only +214:06
anteayaright14:06
anteayaI hear that14:06
anteayayou are playing your role exactly as I rely on you to14:07
anteayathank you14:07
anteayamy frustration is with all these new account holders14:07
fungiso long as we have some consistent communication that we're retroactively revoking voting for all third-party service accounts, i'm fine doing it. but these people's configurations are going to break and fail to be able to comment on reviews from that point until they adjust their configuration to not try to vote any longer14:07
anteayait is like driving in snow in vancouver, I know what I am doing but I stay off the road anyway14:07
anteayaright14:08
anteayathey don't understand the difference between commenting and voting14:08
fungiand i've already tested out the "ptl approves voting" suggestion. getting ptls to respond to the driver authors e-mail pleas to reinstate voting rights seems to be a bit of a bottleneck14:09
anteayayes14:10
fungimaybe getting them approved as part of the weekly meeting for a particular program would be more constructive14:10
anteayaas I had figured it would14:10
anteayabut I had no data point to back up my thoughts14:10
anteayathat's an idea14:10
anteayawould force the 3rd party testing people to at least attend one meeting14:11
fungiat least it's a realtime conversation where there are enough people present to reach consensus, and meetbot can even be used to vote on them14:11
anteayaI agree14:11
anteayait is worth a try14:11
fungiand then we have a public record14:11
anteayaand takes the responsibility off of infra14:11
anteayayes14:11
fungiand then i'm perfectly happy to point the maintainers of those bots back to the deciding body and/or their decrees14:12
anteayamakes sense to me14:12
anteayathe thing is getting it in place14:12
anteayabecause here comes another wave14:13
fungiwell, the pain will spur people to action more quickly this time, perhaps14:13
*** CaptTofu has quit IRC14:13
*** CaptTofu has joined #openstack-infra14:14
fungicrikey! the gate's so shallow now that every change in it is currently being tested!14:14
fungid'oh, someone just added a new change to spite me14:14
anteayawoohoo14:14
anteayathey were listening14:15
fungiwe also have an inversion error or something in the enqueued time calculations... it very briefly showed as being in the gate for -1 hours and something14:15
anteayawe will see what action comes out of the pain this round14:15
anteayapersonally I am considering going back to bed and pulling the covers over my head14:15
anteayasome soft singing in the background couldn't hurt either14:15
anteayaand a cat14:16
anteayaha ah aha14:16
anteayathat would be our time travelling contributor14:16
*** CaptTofu has quit IRC14:18
*** dcramer_ has quit IRC14:20
sdaguefungi: yeh, that was me :)14:26
sdagueI'm going through some of my review queue14:26
fungii need to be doing that, but am a bit on the spacey side so i'm just going to stick to catching up on more of my e-mail backlog instead for now14:27
*** luqas has quit IRC14:38
*** mriedem has joined #openstack-infra14:41
*** mriedem is now known as mcrodermang14:44
*** nati_ueno has quit IRC14:44
*** nati_ueno has joined #openstack-infra14:44
*** DinaBelova_ is now known as DinaBelova14:45
anteayafungi: I'm going through my email and can't see the emails that are posted by people wanting the have voting turned on for their systems14:45
anteayado they use the [neutron] tag?14:46
pleia2anteaya: infra mailing list14:46
pleia2some folks also submit bugs against openstack-ci14:47
fungianteaya: the most recent one was e-mailing me privately, and then asked me to e-mail markmcclain (i told them to e-mail him instead and cc me if they liked, but agree cc'ing a ml would have been better)14:47
fungianteaya: but remember not all of these third-party systems are neutron-related. we've had misbehaving ones for nova drivers too14:48
anteayapleia2: yeah I'm reading the infra ml14:48
anteayayes14:48
pleia2ah, ok14:48
fungiand it sounds like we're likely to end up with similar requests from driver authors for cinder and ironic too14:49
pleia2I guess most on the ml have just been setting up accounts, not talking about voting14:49
anteayajust trying to track down the trail you mentioned of ptls being a bottle neck for not turning systems back on when they request it14:49
anteayapleia2: so far14:49
anteayaI have been making it a habit of looping in the infra ml when addressing issues with 3rd party systems14:49
fungianteaya: yeah, if the current one in question contacts me again asking for an update, i'll tell them to ask on the -dev ml and use a project tag for whatever their driver is in14:50
anteayafor lack of a better way to track conversation14:50
anteayathanks14:50
anteayaand if I miss it, do point me to it14:50
fungibecause not only do i not want to be the one making that decision, i also don't want to be a go-between14:50
anteayaI hear that14:51
anteayaI don't mind being the go between at this moment14:51
anteayasince most of teh pain comes from neutron14:51
*** CaptTofu has joined #openstack-infra14:51
anteayabut it isn't my decision either, unless some group decides they want me to exercise my judgement in that capacity14:51
anteayawhich noone has indicated they do14:52
fungihowever, many of these driver authors seem hesitant to speak in public irc channels or mailing lists and seem to want to use private messaging and personal e-mail messages. hopefully the project will break them of those tendencies14:52
anteayaagreed14:52
anteayathey fear of speaking in public leads to many of the problems they are experiencing and foisting on the rest of us14:52
anteayas/they fear/their fear14:53
*** oubiwann has quit IRC14:55
anteayaas a side note, on observation from code sprint is the folks that have the hardest time interacting remotely seem to have a higher tendencey of driving windows14:55
*** dcramer_ has joined #openstack-infra14:57
*** nati_ueno has quit IRC15:04
fungithat microsoft thing? ew15:06
anteayayeah15:07
anteayaso this neutron patch is in the gate on a reverify bug <bug number> https://review.openstack.org/#/c/66609/15:07
anteayado you want to see what it does or do you want it out?15:08
anteayaisolated tests are passing about 30% of the time still15:08
anteayanothing has been merged to neutron since last week: http://git.openstack.org/cgit/openstack/neutron/log/15:09
fungifun15:09
anteayayeah15:10
anteayaso snipe it out or leave it in?15:10
anteayait removes a dependency15:11
sdagueanteaya: if you don't think it's going to pass, I'd take it out15:11
anteayait has a 30% chance of passing isolated tests15:12
sdagueanteaya: collectively?15:12
sdagueor each one15:12
fungiit also might be worth pointing out to the people reverifying it that there's little hope of it merging until something is done about the failing jobs15:12
anteayaah, I don't know15:12
sdaguefungi: I've given up on educating people here15:12
anteayafungi: yeah, that was teh song I was singing all last week15:12
sdagueI honestly think reverify just needs to be removed15:12
anteayagot a bit of blow back from folks who thought I was overstepping by using gerrit comments to drive behaviour15:13
anteayamy response was what choice do I have, nothing else is working?15:13
sdaguefungi: yuo know what's about to happen?15:13
*** DinaBelova is now known as DinaBelova_15:14
fungi11-way merge?15:14
anteayahttps://review.openstack.org/#/c/67708/15:14
fungiahh, 6-way anyway. a couple of those 0-minute-remaining changes further down still have jobs running15:15
sdaguefungi: more importantly.... we're going to drop the gate down to changes that have been approved in the last 2 hrs15:15
dolphmi'm trying to test a log stash query and expecting to get the failed gate run i'm referencing... is there any reason why it wouldn't come back as a result? (delay in processing logs or something?)15:16
fungiyup15:16
fungidolphm: extremely recent result? if so, then yes there's a bit of a delay15:16
dolphmfungi: it's from a failed run 8 hours ago15:17
fungidolphm: oh, it should have been indexed long ago then. which log file? there are some logs which don't get loaded into logstash15:17
dolphmhttp://logs.openstack.org/46/63546/5/gate/gate-tempest-dsvm-postgres-full/37bd7e9/console.html15:18
fungiconsole.html should always be loaded15:18
dolphmand my query... http://logstash.openstack.org/#eyJmaWVsZHMiOltdLCJzZWFyY2giOiJtZXNzYWdlOlwiIFNTSEV4Y2VwdGlvbjogRXJyb3IgcmVhZGluZyBTU0ggcHJvdG9jb2wgYmFubmVyXCIgQU5EIG1lc3NhZ2U6XCIgRmFpbGVkIHRvIGVzdGFibGlzaCBhdXRoZW50aWNhdGVkIHNzaCBjb25uZWN0aW9uIHRvIGNpcnJvc1wiIEFORCBmaWxlbmFtZTpcImNvbnNvbGUuaHRtbFwiIiwidGltZWZyYW1lIjoiNjA0ODAwIiwiZ3JhcGhtb2RlIjoiY291bnQiLCJvZmZzZXQiOjAsInRpbWUiOnsidXNlcl9pbnRlcnZhbCI6MH0sInN0YW1wIjoxMzkwN15:18
dolphmjYzMDM3MDI5fQ==15:18
anteayaI've submitted a new patchset, so it should be out of the gate when the 25+ events are processed15:18
anteayasdague: have you seen 67708 yet?15:19
anteayait would be a fine time for fungi to approve it15:19
dolphm(short URL to the same query) http://bit.ly/1dXKTQQ15:20
fungianteaya: new patchset of what?15:20
sdaguedolphm: we were losing console logs for a while15:20
fungibut we shoudn't have lost any 8 hours ago, right?15:20
sdaguedolphm: so you can only match on single lines15:20
dolphmsdague: oooh...15:21
sdaguewhich is a limitation of logstash15:21
sdagueeach line is a document15:21
sdaguenot each log file15:21
anteayafungi: I sniped the neutron patch15:21
anteayaout of the gate15:21
dolphmharumph.15:21
fungianteaya: oh, that15:22
anteayayeah15:23
sdaguefungi: how do you feel about +A this - https://review.openstack.org/#/c/61892/15:23
sdagueas my d-g +2 still hasn't shown up ;)15:23
sdaguewe really do want the libvirt logs15:24
anteayahttps://review.openstack.org/#/c/67708/ <-- removes reverify entirely15:25
fungisdague: -1 (it will conflict, but it also needs to go into $BASE/logs now). i'll leave a comment15:26
fungisdague: i'm +2/a on the overall idea, fwiw15:29
sdagueok15:29
sdaguethat's been pretty long standing issue, so we need to get that collection in soon15:29
fungibut we changed that function up quite a bit yesterday chasing the enospc->jenkins offlines your slave during log collection bug which was causing some gate resets15:29
sdagueI need to catch up on the d-g changes... been ignoring everything the last week15:30
*** cody-somerville has quit IRC15:30
openstackgerritDolph Mathews proposed a change to openstack-infra/elastic-recheck: query for bug 1200731  https://review.openstack.org/6913215:33
*** jasondotstar has quit IRC15:35
*** SergeyLukjanov_ is now known as SergeyLukjanov15:35
sdaguetop of gate entered 1h 20min ago15:41
sdagueso nice when this stuff works15:41
dolphm(woo!)15:41
dolphmsdague: fungi: thanks for the tips regarding log stash15:41
sdaguedolphm: no problem15:41
fungidolphm: almost as good as stock tips15:41
*** pcrews has joined #openstack-infra15:43
*** DinaBelova_ is now known as DinaBelova15:43
sdagueyou don't want my stock tips15:44
anteayadon't invest in the Canadian dollar right now15:44
* fungi should have invested in icelandic currency when they crashed. would have made for a great vacation fund15:45
*** blamar has quit IRC15:45
*** SergeyLukjanov is now known as SergeyLukjanov_15:45
dolphmhow long does log stash retain records?15:46
*** cody-somerville has joined #openstack-infra15:46
fungidolphm: two weeks15:46
fungi(roughy)15:46
fungi(er, roughly)15:46
anteayafungi: yeah, no kidding15:47
fungicould have lived a month in one of those hot springs resorts for almost nothing15:48
anteayait would make a nice place to run away to15:51
anteayahard to keep the laptop dry in the hot springs15:52
dolphmso, when does /elastic-recheck/ earn a top level link on status.openstack.org?15:52
fungidolphm: i think they were working on the final feature-parity gap before replacing the current recheck page with it15:53
*** HenryG has joined #openstack-infra15:53
*** dstanek has joined #openstack-infra15:57
*** SergeyLukjanov_ is now known as SergeyLukjanov15:57
sdaguedolphm: yeh, honestly, https://review.openstack.org/#/q/status:open+project:openstack-infra/config+branch:master+topic:gatestatus,n,z needs to happen first as well15:59
*** SergeyLukjanov is now known as SergeyLukjanov_15:59
*** fallenpegasus has joined #openstack-infra16:02
*** thuc_ has joined #openstack-infra16:03
mordredsdague: I really want to break that stuff out into an app that's not just embedded in the config repo16:04
sdaguemordred: that's fine, submit the patches :)16:04
mordredsdague: I've been working on it :)16:05
anteayasdague: can you vote on this? https://review.openstack.org/#/c/68969/16:05
anteayaand then if sean likes it can you approve it please, mordred?16:06
sdagueanteaya: sure16:06
anteayathanks16:06
*** yamahata has joined #openstack-infra16:06
*** SergeyLukjanov_ is now known as SergeyLukjanov16:07
*** SergeyLukjanov is now known as SergeyLukjanov_a16:07
*** SergeyLukjanov_a is now known as SergeyLukjanov16:07
dolphmwhen you mark bugs as dupes in lp, is /rechecks/ smart enough to aggregate failures together?16:09
*** jasondotstar has joined #openstack-infra16:09
openstackgerritA change was merged to openstack-infra/config: Add logging to #openstack-gate  https://review.openstack.org/6896916:10
*** dstanek_afk has joined #openstack-infra16:11
anteayathanks mordred16:11
sdaguedolphm: I'm not sure16:13
sdague /rechecks/ is kind of crap16:13
*** dstanek has quit IRC16:13
sdaguemordred: are the new rax nodes all performance?16:14
mordredsdague: yeah16:15
sdaguebecause they kind of seem slow... though I haven't run numbers yet16:15
sdagueso I think one adjustment to the sliding window algorithm we need is to bump the window when anything in the window succeeds16:16
sdaguebecause otherwise we end up running very few d-g nodes, as we get chunks lined up ready to go, but we haven't bumped the window yet16:17
*** openstack has joined #openstack-infra16:19
*** alexpilotti has joined #openstack-infra16:22
*** lttrl has joined #openstack-infra16:23
jnollermordred: poke16:23
jnollersdague: all the rackspace nodes should be performance flavors; but what OS flavor did you guys pick?16:24
*** CaptTofu has quit IRC16:26
*** CaptTofu has joined #openstack-infra16:27
*** DinaBelova is now known as DinaBelova_16:27
*** dkliban has quit IRC16:28
jnollersdague mordred - you want the PV-HVM OS images16:28
*** lttrl has quit IRC16:29
jeblairclarkb: ack and cool on gerrit fix.16:30
jeblairmordred: -2 removed16:30
jeblairjnoller: looking16:30
sdaguejnoller: https://github.com/openstack-infra/config/blob/master/modules/openstack_project/templates/nodepool/nodepool.yaml.erb#L5216:31
*** CaptTofu has quit IRC16:31
*** SergeyLukjanov is now known as SergeyLukjanov_16:31
*** DinaBelova_ is now known as DinaBelova16:31
*** luqas has joined #openstack-infra16:32
jeblairjnoller: i only see 'Fedora 19 (Schrodinger's Cat) (PVHVM beta)' for that16:32
jeblairsdague: have you correlated slowness with zones?  istr that yesterday the slow nodes we were looking at were all from iad16:33
sdagueI haven't yet. I have a script, but I need to rework it to poll all the jenkins16:33
sdagueand ... it's saturday :)16:34
sdagueso I was just working through low complexity reviews that I felt like I got behind on16:34
jnollerjeblair: I can enable the pv-hvm images for the OS you need (see: http://developer.rackspace.com/blog/welcome-to-performance-cloud-servers-have-some-benchmarks.html)16:35
jnollerif you drop me an email (jesse.noller@rackspace.com) with the account # and which OS you all need I can flip the bit. They're still being rolled out16:36
jeblairjnoller: will it show up as an additional image?  that way we can spin up some test jobs with it to validate...16:36
jnollerYup!16:36
jnollerdifferent ID on the image16:37
jnollerit may take until monday unless the person with the accounts control is also working today :)16:37
*** max_lobur is now known as max_lobur_afk16:38
jnollermordred: just poking about https://review.openstack.org/#/c/69104/ vs https://review.openstack.org/#/c/69097/16:38
jeblairjnoller: cool, sounds great thanks; email sent16:39
jnollerjeblair: what OS16:41
sdaguejnoller: Ubuntu 12.0416:43
openstackgerritMonty Taylor proposed a change to openstack-infra/config: Unlaunchpadify projects.yaml  https://review.openstack.org/6218916:43
openstackgerritMonty Taylor proposed a change to openstack-infra/config: Track direct-release projects in projects.yaml  https://review.openstack.org/6219016:43
openstackgerritMonty Taylor proposed a change to openstack-infra/config: Split config from projects list  https://review.openstack.org/6218716:43
*** dcramer_ has quit IRC16:43
mordredclarkb: ok - I think I fixed that series up and should be good now - I also squashed two of them, because the jeepyb change, which is landed, handles both names, so there was no need for two steps16:44
mordredjnoller: oh! great - I didn't see that you'd done that - ignore mine16:45
mordredjnoller: I was merely trying to be helpful :)16:45
jnollermordred: Thank you, I really appreciate it16:45
jnollerI was just like huh what zuul git omg16:45
mordredjnoller: people are like that all the tie16:46
mordredtie16:46
*** DinaBelova is now known as DinaBelova_16:46
mordredtime16:46
mordreddammit16:47
mordredI need coffee16:47
anteayaI liked tie16:47
mordredjeblair: thanks on the -2 - btw - the current working theory is still sadly holding up16:47
jnollerjeblair: email sent - I'll let you know when they enable the image ID on the account16:47
mordredjeblair: I've now run a NUMBER of these by hand and all have worked - something about running via puppet is causing hives16:47
anteayayay 5 in post16:47
*** dstanek_afk has quit IRC16:50
mordredyay 502's from pypi!16:50
mordreddstufft: ^^ fwiw16:50
dstufftmordred: we're doing maintenance at the moment :]16:50
sdagueyet again, openstack-ci - the health checker of the internet16:52
openstackgerritA change was merged to openstack-infra/elastic-recheck: Clarify required parameters in query_builder  https://review.openstack.org/6775616:52
openstackgerritA change was merged to openstack-infra/elastic-recheck: Stop logging skipped bugs  https://review.openstack.org/6881016:53
jnolleroh god dstufft is here16:54
mordreddstufft: ok. as long as you know about it - not a huge deal here16:54
jnollero_O16:54
mordredjnoller: oh yeah. dstufft is our magical unicorn of pypi and pip16:55
jnollerhe's mine too16:55
*** dcramer_ has joined #openstack-infra16:55
mordredjnoller: I have literally no idea what we'd do without being able to poke him from time to time16:55
openstackgerritMonty Taylor proposed a change to openstack-infra/config: Account for both possible projects.yaml names  https://review.openstack.org/6913916:55
openstackgerritMonty Taylor proposed a change to openstack-infra/config: Unlaunchpadify projects.yaml  https://review.openstack.org/6218916:55
openstackgerritMonty Taylor proposed a change to openstack-infra/config: Track direct-release projects in projects.yaml  https://review.openstack.org/6219016:55
openstackgerritMonty Taylor proposed a change to openstack-infra/config: Split config from projects list  https://review.openstack.org/6218716:55
*** dstanek_afk has joined #openstack-infra17:00
mordredfungi: this one: https://review.openstack.org/#/c/69139/ need to land before the others can pass - because the test is in jenkins_slave_scripts and has to land before the repo can use it to test17:01
dstufftmordred: jnoller :D17:02
*** dstanek_afk has quit IRC17:02
dstufftmordred: we just moved PyPI off of a VM that radomly kernel paniced on OSUOSL hardware onto half a dozen racksapce cloud servers heh17:02
fungii think i've hit that zone where after you've been up for more than 24 hours suddenly you feel completely awake and clear-headed again17:04
fungibut now it's time to board another plane17:05
anteayaI hope you get some sleep17:07
anteayaor do you want to just stay awake?17:07
fungii just want to stay awake at this point17:10
anteayaokay17:11
*** elasticio has joined #openstack-infra17:11
anteayaI hope you stay awake then17:11
openstackgerritA change was merged to openstack-infra/config: Account for both possible projects.yaml names  https://review.openstack.org/6913917:14
clarkbfungi: I looked at grabbing configs eith logs in d-g and gave up because configs are too scattered. what I was hoping to do was tarball /etc/openstack/* and be done but config locations are static for many projrcts17:15
*** fallenpegasus has quit IRC17:19
*** fallenpegasus has joined #openstack-infra17:20
*** yamahata has quit IRC17:21
*** thuc_ has quit IRC17:21
*** markwash has joined #openstack-infra17:22
*** thuc has joined #openstack-infra17:22
*** DinaBelova_ is now known as DinaBelova17:22
*** kraman has joined #openstack-infra17:25
*** thuc has quit IRC17:26
*** fallenpegasus has quit IRC17:28
mordredclarkb: I'd REALLY like to poke people about that17:29
mordredyou know, next time I get half a second to spend on tilting at a new windmill17:29
openstackgerritMonty Taylor proposed a change to openstack-infra/config: Unlaunchpadify projects.yaml  https://review.openstack.org/6218917:31
openstackgerritMonty Taylor proposed a change to openstack-infra/config: Track direct-release projects in projects.yaml  https://review.openstack.org/6219017:31
openstackgerritMonty Taylor proposed a change to openstack-infra/config: Split config from projects list  https://review.openstack.org/6218717:31
mordredclarkb: also, those above shoudl have taken care of your and zaro's comments17:31
*** denis_makogon_ has joined #openstack-infra17:34
*** dcramer_ has quit IRC17:37
*** DinaBelova is now known as DinaBelova_17:38
anteayamordred: are you lacking for windmills?17:40
mordredanteaya: no. plenty of windmills17:40
anteayathat's what I had thought17:42
openstackgerritA change was merged to openstack-infra/elastic-recheck: Add --noirc to tox run job  https://review.openstack.org/6897217:48
openstackgerritA change was merged to openstack-infra/elastic-recheck: Add basic unit tests for the bot  https://review.openstack.org/6877817:48
*** jasondotstar has quit IRC17:52
sdaguejeblair / mordred: can we get a +A on this - https://review.openstack.org/#/c/63551/ - it would have helped on some debug this past week17:52
dtroyerclarkb, mordred:  re: moving configs - there is support in devstack to set XXX_CONF_DIR, you'd have to do each project though…I thought I has set it up to move tham all, wouldn't be hard to do though…17:56
*** kraman has quit IRC18:01
*** fbo_away is now known as fbo18:04
*** DinaBelova_ is now known as DinaBelova18:09
*** fallenpegasus has joined #openstack-infra18:11
*** SergeyLukjanov_ is now known as SergeyLukjanov18:12
*** luqas has quit IRC18:20
*** senk has joined #openstack-infra18:26
*** senk1 has joined #openstack-infra18:29
*** senk has quit IRC18:31
*** fallenpegasus has quit IRC18:32
openstackgerritMatt Riedemann proposed a change to openstack-infra/elastic-recheck: Add query for bug 1267326  https://review.openstack.org/6914418:34
*** coolsvap has joined #openstack-infra18:39
*** coolsvap has quit IRC18:42
lifelessfungi: is that still happening18:45
*** dkliban has joined #openstack-infra18:46
* fungi checks18:46
lifelessI'm going to guess so since nodepool just show two nodes in the head18:46
lifelessshot18:47
lifelessok, I see the failure18:47
lifelesshang a sec18:47
lifelessthe neutron dhcp agent has updated its list correctly18:51
lifelessfor both networks18:52
lifelessand it's answering on eth0 and eth1 correctly18:53
lifelessfungi: (nova console-log $uuid shows that cloud-init thinks networking isn't18:53
fungilifeless: still dying on "Exception: Timeout waiting for ssh access"18:53
fungisorry for the lag. i'm on airplane wireless (and will be for the next 6 hours or so)18:54
lifelessits ok18:54
lifelessso one possible thing would be it not signallings dnsmasq18:54
lifelessI'm going to restart the dhcp agent18:54
lifelesset voila18:55
lifelessso https://bugs.launchpad.net/neutron/+bug/127134418:55
fungii'll watch it for a sec to make sure they stick around18:56
*** oubiwann has joined #openstack-infra18:57
lifelessfungi:18:57
lifelessfungi: https://bugs.launchpad.net/neutron/+bug/127134418:58
* fungi nods18:59
*** cody-somerville has quit IRC18:59
fungii find it hard to believe openstack has bugs18:59
fungi</sarcasm>19:00
*** krtaylor has quit IRC19:00
*** coolsvap has joined #openstack-infra19:00
lifelessfungi: SHOCK AND HORROR19:00
clarkbfungi: both d-g changes failed in the gate early this morning19:04
clarkbfungi: I am going to be afk a bit today so probably can't shepherd them in19:04
*** emagana has joined #openstack-infra19:05
fungiclarkb: i'll be around--i'll reinject them19:05
*** praneshp has joined #openstack-infra19:05
fungilifeless: one of them (1208036) switched to used, so i suspect it's working19:06
fungithough it's cycling through its second slave quickly... i only ever see it in building state and then it's suddenly replaced by anoter19:09
jnollersdague jeblair I got the person on the horn, they should show up if you dump the image list from novaclient - but I'll confirm the image ID for you when I get home19:09
*** emagana has quit IRC19:10
*** ociuhandu has joined #openstack-infra19:10
fungiException: Timeout waiting for ssh access19:11
fungilifeless: is it possible something's going on with not being able to launch any subsequent servers after the first one?19:12
lifelessfungi: has it done more than one cycle?19:14
lifelessfungi: if so I'll punch the agent again19:14
*** miqui has quit IRC19:14
fungiit's done dozens19:16
lifelessme stabs19:17
*** praneshp has quit IRC19:17
*** thuc has joined #openstack-infra19:22
*** thuc has quit IRC19:27
*** coolsvap has quit IRC19:29
*** yolanda_ has joined #openstack-infra19:32
*** markwash has quit IRC19:35
lifelessfungi: both have ssh host keys now so I expect both to be ok19:39
fungilifeless: now there are two ready nodes which are 20 minutes old19:39
fungiheh19:39
lifelessgreat19:39
lifelessso19:39
lifeless*if* this happens again19:39
lifelessplease point any tripleo-cd-admin at the ci-overcloud control node and that bug19:39
lifelessor even just say 'there is a tripleo critical bug about this'19:39
lifelessand we'll fix19:40
lifelesss/if/when/19:40
fungilifeless: i'll ring the claxon in #tripleo next time, just wasn't sure whether you expected them to be working yet19:41
*** miqui has joined #openstack-infra19:42
lifelessfungi: I don't want to mess up your pretty logs19:42
lifeless(and yes, we expect everything to be working)19:43
lifelessfungi: hey, since you're stuck on a plane19:43
lifelessfungi: care to help me understand something?19:43
lifelessin http://logs.openstack.org/43/67343/7/check/gate-tripleo-deploy/5a7b282/console.html19:44
lifelessfungi: oh, *now* it shows me. NVM.19:44
lifelessfungi: have our changes to the ndoepool scripts not landed on the nodepool server?19:45
*** jcooley_ has joined #openstack-infra19:47
*** pcrews has quit IRC19:47
fungimmmpossibly?19:52
fungiwhich changs specifically?19:52
fungichanges19:52
fungilifeless: if you mean prep script changes, it may not have built new images since19:54
lifelessfungi: yeah we altered prepare_$stuff19:54
lifelessfungi: if puppet was disabled they wouldn't be on disk; otherwise could you delete the old image and make it build new?19:54
*** jcooley_ has quit IRC19:55
*** jcooley_ has joined #openstack-infra19:55
*** SergeyLukjanov is now known as SergeyLukjanov_a19:58
*** SergeyLukjanov_a is now known as SergeyLukjanov19:58
*** SergeyLukjanov is now known as SergeyLukjanov_a19:59
openstackgerritlifeless proposed a change to openstack-infra/config: Raise the tripleo-test-cloud max-servers.  https://review.openstack.org/6914919:59
*** jcooley_ has quit IRC20:00
*** jcooley_ has joined #openstack-infra20:00
fungilifeless: sure thing20:02
fungiprevious image was 17.14 hours old20:04
lifelessANCIENT20:04
*** jcooley_ has quit IRC20:04
lifelessfungi: 69149 should be innocuous too :)20:05
*** jcooley_ has joined #openstack-infra20:05
*** yolanda_ has quit IRC20:10
*** bknudson has quit IRC20:10
openstackgerritA change was merged to openstack-infra/config: Raise the tripleo-test-cloud max-servers.  https://review.openstack.org/6914920:12
openstackgerritA change was merged to openstack-infra/config: Build tripleo-fedora nodepool nodes  https://review.openstack.org/6880920:13
fungimight as well ^ while i keep tabs on the tripleo-precise image build20:15
lifelessthanks!20:15
lifelessnow I expect the fedora build to fail, pleia2 was looking at the script changes needed to support that properly20:17
*** praneshp has joined #openstack-infra20:19
*** jcooley_ has quit IRC20:21
fungiokay20:21
*** jcooley_ has joined #openstack-infra20:21
lifelesshmm20:21
lifelessI'd really love it if /config changes triggered a test node build20:21
lifelessI wonder how hard that would be20:22
lifelesswe'd need access to credentials on a cloud somewhere (I don't want to build in qemu :P)20:22
*** SergeyLukjanov_a is now known as SergeyLukjanov20:23
*** jcooley_ has quit IRC20:27
*** kraman has joined #openstack-infra20:28
*** krtaylor has joined #openstack-infra20:33
*** kraman has quit IRC20:35
*** praneshp_ has joined #openstack-infra20:41
fungitripleo-fedora image is now building too20:41
*** salv-orlando has quit IRC20:43
*** praneshp has quit IRC20:43
*** praneshp_ is now known as praneshp20:43
*** yolanda_ has joined #openstack-infra20:49
*** yolanda_ has quit IRC20:51
fungiyeah, it bailed out on the fedora image20:52
*** dkliban has quit IRC21:09
*** thuc has joined #openstack-infra21:23
*** DinaBelova is now known as DinaBelova_21:24
*** thuc has quit IRC21:37
*** fallenpegasus has joined #openstack-infra21:38
*** ok_delta has joined #openstack-infra21:39
*** alexpilotti has quit IRC21:40
*** persia has quit IRC21:45
*** persia has joined #openstack-infra21:46
*** persia is now known as Guest6592221:46
*** SergeyLukjanov is now known as SergeyLukjanov_21:48
*** fbo is now known as fbo_away21:48
*** Guest65922 has quit IRC21:50
*** persia__ has joined #openstack-infra21:51
*** denis_makogon_ has quit IRC22:06
*** boris-42 has quit IRC22:16
fungilifeless: how long would you expect a tripleo-precise image to take to build? it's approaching 2.5 hours in a building state22:27
*** fallenpegasus has quit IRC22:39
anteayafungi: what leg are you on now?22:49
anteayahave you left aus?22:49
fungianteaya: on the final bna->rdu stretch now22:50
anteayawoohoo22:52
anteayago you22:52
anteayahow much longer until you land?22:52
fungiabout an hour. then i need to go hunting for my luggage which probably arrived last night22:55
anteayaah22:56
anteayathe fun never stops22:56
*** flaper87 is now known as flaper87|afk22:58
openstackgerritRuslan Kamaldinov proposed a change to openstack-infra/reviewstats: Update openstack-infra/storyboard  https://review.openstack.org/6915023:00
openstackgerritRuslan Kamaldinov proposed a change to openstack-infra/reviewstats: Update Storyboard  https://review.openstack.org/6915023:01
*** dcramer_ has joined #openstack-infra23:06
*** chuck__ has joined #openstack-infra23:08
openstackgerritRuslan Kamaldinov proposed a change to openstack-infra/reviewstats: Update Storyboard  https://review.openstack.org/6915023:10
*** dcramer_ has quit IRC23:28
*** gokrokve has joined #openstack-infra23:31
*** thuc has joined #openstack-infra23:34
*** thuc has quit IRC23:39
dstufftmordred: Ok, almost 9 hours later, the migration is done now ;) If you see any problems now let us know!23:49
*** jnoller has quit IRC23:52
*** jnoller has joined #openstack-infra23:52
jnollersdague jeblair can you check out the OS images returned by nova list and see if you see ID 7818a834-d255-41cf-bd72-ac153b3d540623:53
jnollersdague jeblair - I'll send an email but essentially nova boot hello2 --flavor performance1-1 --image 7818a834-d255-41cf-bd72-ac153b3d5406 --key-name mykey --poll should work even if it's not show on image-list23:56
*** jnoller_ has joined #openstack-infra23:58
openstackgerritA change was merged to openstack-infra/elastic-recheck: Add query for bug 1267326  https://review.openstack.org/6914423:59

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