Thursday, 2013-12-05

*** rfolco_ has quit IRC00:00
openstackgerritA change was merged to openstack-infra/reviewstats: Fix CSV output  https://review.openstack.org/6010200:01
openstackgerritA change was merged to openstack-infra/reviewstats: update swift core  https://review.openstack.org/6005300:01
*** rongze has quit IRC00:02
openstackgerritClark Boylan proposed a change to openstack-infra/config: Remove tempest coverage job  https://review.openstack.org/5977500:03
clarkbjeblair: mordred: fungi: ^ I made a tiny edit to mtreinish's change, I plan to push through in a bit if no one else reviews00:03
*** rcleere has quit IRC00:03
*** rnirmal has joined #openstack-infra00:04
*** openstackgerrit has quit IRC00:06
*** rfolco_ has joined #openstack-infra00:06
*** openstackgerrit has joined #openstack-infra00:06
*** alexpilotti has joined #openstack-infra00:07
*** thomasem has joined #openstack-infra00:08
*** loq_mac has quit IRC00:08
*** sarob has quit IRC00:11
*** sarob has joined #openstack-infra00:11
*** rfolco_ has quit IRC00:12
jeblairclarkb: +4 https://review.openstack.org/#/c/59046/00:13
clarkbjeblair: ty, I will try to remember to approve this evening when no meetings are happening00:13
*** rfolco_ has joined #openstack-infra00:13
jeblairclarkb: aprvd coverage change00:14
openstackgerritJoshua Harlow proposed a change to openstack-infra/gerritlib: Add connection attempt limit  https://review.openstack.org/5298600:15
*** rnirmal has quit IRC00:15
clarkbjeblair: I have approved the companion change to d-g00:15
*** thomasem has quit IRC00:15
openstackgerritA change was merged to openstack-infra/config: Remove tempest coverage job  https://review.openstack.org/5977500:16
*** sarob has quit IRC00:16
openstackgerritA change was merged to openstack-infra/jeepyb: Ignore E125 and H flake8 violations.  https://review.openstack.org/5980200:18
*** pete5 has quit IRC00:18
*** mriedem has joined #openstack-infra00:18
*** dstufft has quit IRC00:18
*** rfolco_ has quit IRC00:20
*** CaptTofu has joined #openstack-infra00:20
*** dstufft has joined #openstack-infra00:21
*** matsuhashi has joined #openstack-infra00:21
openstackgerritJoshua Harlow proposed a change to openstack-infra/gerritlib: Add connection attempt limit  https://review.openstack.org/5298600:23
*** mfer has joined #openstack-infra00:24
*** esker has joined #openstack-infra00:27
*** alexpilotti has quit IRC00:28
*** andreaf has quit IRC00:30
openstackgerritA change was merged to openstack-infra/elastic-recheck: Remove queries for bugs 1225024 and 1225664  https://review.openstack.org/5981600:30
uvirtbotLaunchpad bug 1225024 in tempest "tempest.api.compute.admin.test_hosts.HostsAdminTestXML.test_list_hosts_with_zone[gate]  unexpected conductor service" [Undecided,Incomplete] https://launchpad.net/bugs/122502400:30
uvirtbotLaunchpad bug 1225664 in devstack "Swift: No space left on device" [Undecided,Fix committed] https://launchpad.net/bugs/122566400:30
*** andreaf has joined #openstack-infra00:31
*** rfolco_ has joined #openstack-infra00:31
*** esker has quit IRC00:31
*** dcramer_ has joined #openstack-infra00:33
*** mfer has quit IRC00:38
*** fifieldt_ has joined #openstack-infra00:38
*** mfer has joined #openstack-infra00:38
*** mfer has quit IRC00:39
dcramer_jeblair: So at long last, clouddocs-maven-plugin is a project on  stackforge: https://git.openstack.org/cgit/stackforge/clouddocs-maven-plugin/ Now I need permission to approve reviews.  I'll have a couple of others for this repo as well.00:41
*** reed has quit IRC00:43
*** dims has joined #openstack-infra00:44
*** dripton has quit IRC00:44
*** MarkAtwood has quit IRC00:44
*** dims has quit IRC00:45
*** reed has joined #openstack-infra00:45
*** reed has quit IRC00:45
*** reed has joined #openstack-infra00:45
*** dripton has joined #openstack-infra00:47
zaroclarkb: where is is the configuration for our on-demand jenkins slaves?00:49
clarkbzaro: in openstack-infra/config/modules/openstack_project/files/nodepool/scripts do it00:50
zaroclarkb: i want to add buck to it.00:50
clarkbzaro: we don't want buck on the on demand slaves00:50
fungidcramer_: i can add you as the initial core reviewer, presumably?00:50
clarkbzaro: adding buck should happen in openstack-infra/config/modules/jenkins/manifests/slave.pp00:50
fungidcramer_: you should now be able. also you can add/remove additional gerrit users to the core group at https://review.openstack.org/#/admin/groups/207,members00:51
zaroclarkb: no packages are available for buck :(00:52
zaroclarkb: can i do with exec command?00:52
zaroclarkb: the steps is to git clone, ant build, then create some soft links to the executable.00:53
clarkbugh00:54
clarkbit isn't maven isntallable or something like that?00:54
zarono, you need to build with ant.00:54
clarkbif not then yes, you need a vcsrepo to clone, then exec to ant build00:54
*** jgrimm has quit IRC00:55
zarodo you know if ant is available or does it need to be installed?00:56
fungii believe there are ubuntu packages for ant00:56
clarkbI don't know, you can check the list of packages in slave.pp00:56
* fungi confirms00:56
fungihttp://packages.ubuntu.com/precise/ant00:57
fungi1.8.200:57
clarkbbtw I am going to single +2 approve simple changes (like doc changes)00:58
zarofungi: huh?  does that mean that it's there by default or is it something that needs to be added to the slave.pp?00:58
zaroclarkb: is this a new policy?00:58
fungijust checked on one, and it's installed too http://paste.openstack.org/show/54465/00:58
fungizaro: it's an "emergency" policy because we're drowning in changes which need review/approval00:59
clarkbzaro: no, the queue is just so huge that I want to move things through it00:59
zarofungi: would the best practice be to add ant to slave.pp anyway? or just leave it out01:02
zaro?01:02
fungizaro: i suspect it's already being installed as a dependency of something else, if it's not explicitly listed01:03
*** markwash has quit IRC01:03
clarkbif we do need it for buck explicitly I would add it that way we can document what it is for01:03
clarkband no surprises when $otherthing changes01:03
zaroi would agree that it's be to explicitly add.01:04
fungithat's a worthwhile reason, agreed01:04
openstackgerritA change was merged to openstack-infra/zuul: Document zuul_url breaking change  https://review.openstack.org/6002801:04
zarocool. thanks gents. out for dinner now. see ya.01:05
*** Abhishe__ has quit IRC01:07
openstackgerritA change was merged to openstack-infra/config: create a selection of jobs to not use config drive  https://review.openstack.org/6006101:09
clarkbhttp://logs.openstack.org/00/60100/1/check/check-requirements-integration-dsvm/d4e873e/console.html d2to1 seems to still be a problem01:12
openstackgerritA change was merged to openstack-infra/config: Enable neutron on heat-slow experimental job  https://review.openstack.org/6008401:12
openstackgerritA change was merged to openstack-infra/config: Make havana devstack/tempest jobs non-voting  https://review.openstack.org/6009201:13
fungiclarkb: yeah, saw that01:13
fungiclarkb: and including d2to1 in the global requirements list did not cause it to be discovered by glance's preferred (oldish) version of oslo.messaging01:14
clarkbthis is very weird01:15
fungiand the revert of the change mordred thought might be causing that to emerge was punted from the gate by the error it was hoping to quell01:15
fungii'm sure fixing glance's requirements list to refer to an o.m which does not need d2to1 would "fix" it, but it wouldn't really *fix* it01:16
clarkbright01:17
fungiso where i think we're at is that pypi-mirror on the current state of openstack/requirements master branch does not add d2to1 to the tree, and even putting d2to1 in the global requirements seems not to change that01:19
fungii'll see if jenkins can tell me the last time that job succeeded01:20
*** krotscheck has quit IRC01:21
fungilatest successful run was https://jenkins01.openstack.org/job/check-requirements-integration-dsvm/4/console01:22
*** flaper87 is now known as flaper87|afk01:22
fungiand this seems to be the earliest failure unless jenkins purged earlier ones https://jenkins02.openstack.org/job/check-requirements-integration-dsvm/6/console01:23
fungiso sometime between 12:50 and 15:44 roughly01:24
fungii'm going to filter my irc log for gerritbot notifications and see what merged today before 15:4401:26
*** mfer has joined #openstack-infra01:26
fungi(well, yesterday now that utc midnight is behind us)01:26
*** ryanpetrello has joined #openstack-infra01:27
*** esker has joined #openstack-infra01:27
*** prad has quit IRC01:28
*** rongze has joined #openstack-infra01:28
*** rfolco_ has quit IRC01:31
clarkboh you know what01:31
*** matsuhashi has quit IRC01:31
* clarkb has a hunch that is being tracked down01:31
dcramer_fungi: Thanks!01:32
*** esker has quit IRC01:32
*** rongze has quit IRC01:33
*** matsuhas_ has joined #openstack-infra01:33
clarkbfungi: did we recently switch to pypi-mirror?01:33
clarkbor has it been in use for a while? (to build the mirror on the slave)01:34
fungiclarkb: that's what i was hoping to discover in my blank staring at datas here01:34
clarkbhmm november 16th so that can't be it01:34
mordredit's been in use for a while01:34
fungi06:03:06 openstack-dev/pbr: Use wheels for installation  https://review.openstack.org/4880301:35
mordredfungi: that's the last one before things broke?01:35
fungithat's maybe on the cusp of something which wouldn't have kicked in until a mirror rebuild, but i really think it would have impacted the 12:50 run if it had been the cause01:35
*** Mithrandir has quit IRC01:35
mordredwe also just reverted it01:35
fungiright, and it didn't help01:36
fungiafter that, skipping a couple of elastic-recheck query changes, we have01:36
fungi12:39:30 openstack/requirements: Allow Django 1.5 and 1.6  https://review.openstack.org/5992001:36
*** Mithrandir has joined #openstack-infra01:36
fungiwe know what that did, and i don't think it was responsible for this01:36
fungi12:39:36 openstack/requirements: Bump python-savannaclient version to >= 0.4.0  https://review.openstack.org/5965701:36
clarkbcould zaro's change have side effected something weirdly?01:36
*** nosnos has joined #openstack-infra01:36
fungiclarkb: maybe, but there were successful runs long after it went into affect01:37
fungier, effect01:37
fungianyway, after that savannaclient requirement change, there's nothing else reported of interest before we started getting the d2to1-induced braindeadedness in the requirements integration job01:38
*** prad_ has joined #openstack-infra01:38
fungicould it have changed the order in which things were processed to expose a latent problem there?01:39
*** sjing has joined #openstack-infra01:39
fungithat feels far-fetched to me, but...01:39
*** dims has joined #openstack-infra01:39
*** bingbu has joined #openstack-infra01:40
fungianyway, i think i've satisfied my curiosity that this bug didn't spontaneously spring from one of those changes merging01:40
*** herndon_ has joined #openstack-infra01:41
fungithough there were no mirror rebuild triggering changes before the django one, so the window of potential breakage (if we assume it was introduced by a mirror run) extends to the time when the periodic mirror jobs kicked off01:43
clarkbeven further back01:44
clarkbsince the periodic mirror jobs wouldn't be affected01:44
fungireally? why not?01:44
clarkb(this is glance trying to install against the built mirror)01:44
*** prad_ has quit IRC01:45
clarkbmirror job itself can talk to the world so d2to1 would be pulled in if necessary01:45
fungiright, but that job ran successfully as recently as 12:50 utc01:45
clarkbwhich job? the periodic-mirror-python2X or pbr integration?01:45
fungithe requirements integration job last successfully ran at 12:5001:46
clarkbgotcha01:46
fungiso something happened after that (or took effect after that) to cause this to begin cropping up01:46
fungii punted to maybe a mirror run dragging something in, but that job shouldn't use the mirror01:47
openstackgerritMathieu Gagné proposed a change to openstack-infra/config: gerritbot: Watch stable/havana on puppet-openstack projects  https://review.openstack.org/6012701:47
clarkbfungi: right, I think it is something self contained to how we build the mirror in the integration test01:48
*** dcramer_ has quit IRC01:49
*** prad_ has joined #openstack-infra01:52
fungiso, taking the behavior completely from that angle, i can see why glance's tarball dependency on oslo.messaging==1.2.0a11 while falsely claiming a desire for >=1.2.0a11 would cause the mirror script not to add d2to101:53
*** reed has quit IRC01:53
fungi*however* the question becomes, why now?01:53
fungiit's been that way, so...01:53
fungibest guess is something else was more directly depending on d2to1 up until today01:54
clarkbI think that is a good guess01:54
fungia second (or third or more) order transitive dependency potentially01:54
fungior, we were accidentally pointing the job at not-the-local-mirror so it was finding d2to1 when it wanted it01:55
fungibut i don't see anything which would have taken effect to bring about that change in behavior coincident to the time we witnessed01:56
*** fifieldt_ has quit IRC01:56
clarkbthat would seem more likely if the wheels stuff being reverted fixed our problems (because that change changed how we configure pip and pydistutils01:56
*** fifieldt_ has joined #openstack-infra01:56
fungiright01:57
fungii'm hoping something more concrete springs to mind, because i'm rapidly running out of brain cells for tonight01:58
clarkbI think transitive d2to1 somewhere is most likely01:58
*** guohliu has joined #openstack-infra02:00
*** ^d has quit IRC02:01
*** boris-42 has quit IRC02:02
*** prad_ has quit IRC02:05
*** melwitt has left #openstack-infra02:05
*** dcramer_ has joined #openstack-infra02:05
clarkbfungi: fwiw I am finding it incredibly difficult to be motivated to fix anything right now02:06
clarkbI think the dentist did that to me02:06
fungiclarkb: understandable02:06
*** mrodden has joined #openstack-infra02:06
fungithey take out small chunks of your soul with their demonic power tools02:06
openstackgerritClark Boylan proposed a change to openstack-infra/config: Rename #openstack-metering to #openstack-ceilometer  https://review.openstack.org/5904602:07
clarkbI am reapproving that ^ fixed a merge conflict02:07
fungianyway, i'm going to suggest that our previous mirror-building assumptions were based on an additive principle, whereby projects depending on older stuff was fine because we didn't remove it, just added newer things02:07
fungithat coupled with the idea that we could ignore tarball links in requirements lists when building "the one mirror"02:09
fungibecause, at least at one point, whatever they needed had been listed and added02:10
*** markwash has joined #openstack-infra02:10
fungithe new "buildamirror from-scratch![tm]" makes the assumption that things really use the versions they're claiming, and tarball links break that assumption02:10
openstackgerritA change was merged to openstack-infra/config: Rename #openstack-metering to #openstack-ceilometer  https://review.openstack.org/5904602:10
fungii can see where wheels of prereleases in pypi, so we can throw away all the tarball links, will help tremendously here02:11
fungibut we're obviously not there yet02:12
harlowjaqq for u guys, when enabling coverage job in jenkins, are the outputs of coverage at a known browsable location?02:13
lifelessis https://review.openstack.org/#/c/60063/ urgent?02:13
lifeless(assuming it probably is as i's a revert)02:13
harlowja*not in jenkins, i mean in zuul (or review.openstack.org) - my fault02:14
fungiharlowja: they're... findable at least, but you need to know the sha of the commit which triggered them02:14
harlowjakk02:14
harlowjanot posted on the review?02:14
fungilifeless: it *was* urgent, except it failed to merge due to the bug it was hoped to fix, so now less urgent02:15
fungilifeless: it's basically the bug we were discussing just above02:15
clarkbharlowja: no because they run after merge02:15
lifelessso if it doesn't solve the bug, we shouldn't land it, right?02:15
clarkbyou can run them pre merge if you like as well but we don't do that much02:15
clarkb(we need to look at that again imo)02:15
harlowjaah, k, clarkb logs.openstack.org/$hash_here (?)02:16
fungilifeless: we may still want to revert it, but not just now02:16
clarkbharlowja: logs.openstack.org/FIRSTTWOCHARSOFHASH/ENTIREHASH02:16
harlowjakk02:16
fungiharlowja: there's an additional layer, because there are too many hashes, so directory browsing gets very impossible02:16
harlowjasure, makes sense02:16
lifelessfungi: want to remove your +2 ?02:16
fungilifeless: i'm pondering02:16
*** openstack has joined #openstack-infra02:17
fungilifeless: yeah, i -1'd it just now, but see you did as well. cool02:18
*** herndon_ has quit IRC02:21
lifelessclarkb: fungi: do you need more eyes on the pbr thing?02:23
openstackgerritPranesh Pandurangan proposed a change to openstack-infra/config: Added new project, entropy Entropy is a framework for creating audit and repair scripts for openstack  https://review.openstack.org/6013202:23
clarkblifeless: I am about ready to sleep on it02:23
clarkbI have pizza for dinner that is going cold02:23
lifelesshow wide spread is the impact?02:23
fungilifeless: fresh eyes in a hemisphere where it is not sleepytime would be very, very welcome02:23
clarkblifeless: right now only openstack/requirements and pbr are affected iirc02:24
lifelessok02:24
clarkbthough everything will be hurting if we need another requirements change in real quick like we had with django02:24
fungilifeless: for now, it blocks only new patches to openstack/requirements, openstack-dev/pbr and openstack-infra/{jeepyb,pypi-mirror} i believe02:24
*** openstackgerrit has quit IRC02:24
*** openstackgerrit has joined #openstack-infra02:24
fungilifeless: the only failing job from this is  check-requirements-integration-dsvm02:25
fungiso projects which depend on that for gating are impacted02:25
fungiwe don't know how far beyond that job the potential impact spreads, but there is a good chance it only affects the integration test script02:26
*** esker has joined #openstack-infra02:27
fungiin particular, we're attempting to suss out why it seems that a fresh pypi-mirror run no longer adds d2to1 (which glance indirectly still needs because of a tarball link dependency on an older oslo.messaging which declared a setup_requires on it), but was not suffering from this prior to ~noon utc yesterday02:27
fungiour best guess, as described a few minutes ago in scrollback, is that up until that time some second-order transitive dependency declared a requirement on d2to but a newer version stopped doing so and now the mirror script no longer thinks it needs to be pulled in02:29
fungier, on d2to102:29
*** oubiwann_ has joined #openstack-infra02:30
lifelessso the options are02:31
lifelessa) transitive dep fail, there is a bug02:31
fungiat least following the code, there's nothing in pypi-mirror which would ignore glance's declared requirement of oslo.messaging>=1.2.0a11 (current versions of which no longer want d2to1) and instead process the tarball link it carries to 1.2.0a11 itself (which wants d2to1)02:31
lifelessb) known limitation rears it's head and we should add d2to1 explicitly somewhere02:31
*** esker has quit IRC02:32
fungii think it's likely to be a combination of both...02:32
fungiwe could consider the behavior "a" something undesirable, and just go fix glance (we expect this to become moot when we get wheels uploading to pypi)02:32
fungii tried "b" adding d2to1 explicitly to global-requirements.txt in openstack/requirements though and that change failed on the same error, so i'm not *positive* i've followed the code as accurately as i'd hoped02:33
*** esker has joined #openstack-infra02:34
fungibut, running out of steam02:34
*** bingbu has quit IRC02:34
fungiand not so urgent i'm going to pull an all-nighter and then be useless tomorrow02:34
*** jcooley_ has joined #openstack-infra02:35
*** ryanpetrello has quit IRC02:35
*** matsuhas_ has quit IRC02:35
*** esker has quit IRC02:35
*** sjing has quit IRC02:35
*** ryanpetrello has joined #openstack-infra02:36
*** esker has joined #openstack-infra02:36
*** dims has quit IRC02:37
* fungi still has advisories which need writing up and security fixes to test. don't want those to miss the stable release boat next week02:37
*** sjing has joined #openstack-infra02:38
*** sandy__ has quit IRC02:43
openstackgerritA change was merged to openstack-infra/config: Enable requirement update proposal jobs  https://review.openstack.org/5985502:43
*** Ryan_Lane has joined #openstack-infra02:44
*** DennyZhang has joined #openstack-infra02:46
*** Ryan_Lane has quit IRC02:49
*** mfer has quit IRC02:50
*** ryanpetrello has quit IRC02:51
*** rongze has joined #openstack-infra02:53
*** vkozhukalov has joined #openstack-infra02:55
*** mfer has joined #openstack-infra02:55
*** rongze_ has joined #openstack-infra02:55
*** jamesmcarthur has joined #openstack-infra02:56
*** mrodden has quit IRC02:57
*** rongze has quit IRC02:58
*** sandy__ has joined #openstack-infra02:59
openstackgerritMathieu Gagné proposed a change to openstack-infra/config: Mark puppet-unit-3.2/3 jobs as voting  https://review.openstack.org/6014002:59
openstackgerritlifeless proposed a change to openstack-infra/reviewstats: Use six to handle cpickle.  https://review.openstack.org/6011403:09
*** HenryG has quit IRC03:11
openstackgerritA change was merged to openstack-infra/reviewstats: Remove text blobs from CSV output  https://review.openstack.org/6010303:14
*** mfer has quit IRC03:15
*** chandankumar has joined #openstack-infra03:17
*** jcooley_ has quit IRC03:17
*** markwash has quit IRC03:18
*** mfer has joined #openstack-infra03:20
*** sdake_ has quit IRC03:21
*** dripton has quit IRC03:21
*** CaptTofu has quit IRC03:24
*** resker has joined #openstack-infra03:33
*** esker has quit IRC03:35
*** CaptTofu has joined #openstack-infra03:36
*** mfer has quit IRC03:38
*** mfer has joined #openstack-infra03:39
*** mfer has quit IRC03:41
*** DennyZhang has quit IRC03:42
*** DennyZha` has joined #openstack-infra03:42
*** sarob has joined #openstack-infra03:45
*** Ryan_Lane has joined #openstack-infra03:45
*** mriedem has quit IRC03:47
*** Ryan_Lane has quit IRC03:50
*** dcramer_ has quit IRC03:53
*** matsuhashi has joined #openstack-infra04:14
*** michchap_ has joined #openstack-infra04:24
*** michchap has quit IRC04:26
*** sdake_ has joined #openstack-infra04:29
*** MarkAtwood has joined #openstack-infra04:30
fungibug https://bugs.launchpad.net/bugs/125793104:32
uvirtbotLaunchpad bug 1257931 in openstack-ci "recent import of stevedore to stackforge failed to import the git repo" [Undecided,New]04:32
*** fifieldt has quit IRC04:33
fungimakes me think mordred's manage-projects refactor didn't completely eradicate that issue04:34
openstackgerritA change was merged to openstack-infra/devstack-gate: Drop TEMPEST_LARGE_OPS_NUMBER by 33% to 100  https://review.openstack.org/6000004:34
clarkbdoesn't look like it04:35
clarkbI wonder if we are also hitting a puppet exec timeout04:35
*** rongze_ has quit IRC04:35
*** michchap has joined #openstack-infra04:35
*** CaptTofu has quit IRC04:36
*** CaptTofu has joined #openstack-infra04:36
*** michchap_ has quit IRC04:38
*** oubiwann_ has quit IRC04:40
*** mfer has joined #openstack-infra04:44
*** markwash has joined #openstack-infra04:45
*** Ryan_Lane has joined #openstack-infra04:46
*** mfer has quit IRC04:48
*** pcrews has quit IRC04:51
*** Ryan_Lane has quit IRC04:51
*** amotoki has joined #openstack-infra04:53
*** sarob has quit IRC04:59
*** marun has joined #openstack-infra05:03
*** aardvark has joined #openstack-infra05:15
*** mgagne1 has joined #openstack-infra05:15
*** mgagne1 has quit IRC05:15
*** mgagne1 has joined #openstack-infra05:15
*** mgagne has quit IRC05:17
*** wusui has quit IRC05:18
*** fifieldt_ has quit IRC05:21
*** DennyZha` has quit IRC05:24
*** changbl has joined #openstack-infra05:25
*** sarob has joined #openstack-infra05:27
*** sarob has quit IRC05:28
*** sarob has joined #openstack-infra05:28
*** jamesmcarthur has quit IRC05:29
*** sarob has quit IRC05:30
*** sarob has joined #openstack-infra05:32
*** sarob has quit IRC05:33
*** sarob has joined #openstack-infra05:34
*** dkliban_ has joined #openstack-infra05:35
*** sarob has quit IRC05:39
*** talluri has joined #openstack-infra05:43
*** nicedice has quit IRC05:46
*** Ryan_Lane has joined #openstack-infra05:47
*** sdake_ has quit IRC05:49
*** Ryan_Lane has quit IRC05:51
*** mrodden has joined #openstack-infra05:52
*** yamahata_ has quit IRC05:55
*** DennyZha` has joined #openstack-infra05:57
*** DennyZha` has quit IRC06:01
openstackgerritA change was merged to openstack-infra/devstack-gate: Rip out the tempest coverage runs from devstack-gate  https://review.openstack.org/5977706:03
*** paul-- has quit IRC06:06
*** matsuhashi has quit IRC06:14
*** matsuhashi has joined #openstack-infra06:14
*** rpodolyaka1 has joined #openstack-infra06:17
*** rcleere has joined #openstack-infra06:19
*** guohliu has quit IRC06:29
*** guohliu has joined #openstack-infra06:30
*** Abhishek has joined #openstack-infra06:34
*** guohliu has quit IRC06:35
*** slong_ has quit IRC06:35
*** slong_ has joined #openstack-infra06:36
*** guohliu has joined #openstack-infra06:36
*** jkyle has quit IRC06:37
*** rcleere has quit IRC06:41
*** Ryan_Lane has joined #openstack-infra06:44
*** jkyle has joined #openstack-infra06:44
openstackgerritStuart Warren proposed a change to openstack-infra/jenkins-job-builder: Add Mercurial plugin feature to jenkins-job-builder  https://review.openstack.org/6006606:44
*** vkozhukalov has quit IRC06:48
*** Ryan_Lane has quit IRC06:50
*** pblaho has joined #openstack-infra06:50
*** rongze has joined #openstack-infra06:54
*** Abhishek has quit IRC06:56
*** jpich has joined #openstack-infra07:00
*** paul-- has joined #openstack-infra07:01
*** gyee has quit IRC07:06
*** Ryan_Lane has joined #openstack-infra07:09
*** rpodolyaka1 has quit IRC07:11
*** SergeyLukjanov has joined #openstack-infra07:12
*** denis_makogon has joined #openstack-infra07:15
openstackgerritStuart Warren proposed a change to openstack-infra/jenkins-job-builder: Add Mercurial plugin feature to jenkins-job-builder  https://review.openstack.org/6006607:15
*** matsuhashi has quit IRC07:17
*** Ryan_Lane has quit IRC07:19
openstackgerritMax Rydahl Andersen proposed a change to openstack-infra/jenkins-job-builder: add private-repository to maven-target  https://review.openstack.org/6016107:22
openstackgerritQianLin proposed a change to openstack-infra/nodepool: Fix the issue which raise FloatingIpPoolNotFound  https://review.openstack.org/5433107:23
*** nati_ueno has quit IRC07:24
*** jcoufal has joined #openstack-infra07:30
*** boris-42 has joined #openstack-infra07:35
*** SergeyLukjanov has quit IRC07:41
openstackgerritMax Rydahl Andersen proposed a change to openstack-infra/jenkins-job-builder: add private-repository to maven-target  https://review.openstack.org/6016107:43
*** SergeyLukjanov has joined #openstack-infra07:48
*** jpich has quit IRC07:56
*** dkliban_ has quit IRC08:04
*** k4n0 has joined #openstack-infra08:07
*** harlowja has quit IRC08:08
*** paul-- has quit IRC08:08
k4n0Hey guys, i have odd situation here, for my review https://review.openstack.org/#/c/59860/, all python 2.6 unit tests are failing, but all python 2.7 pass. Seems to be some bug on the CI side08:08
*** flaper87|afk is now known as flaper8708:09
*** DinaBelova has joined #openstack-infra08:10
*** che-arne has quit IRC08:12
openstackgerritAndreas Jaeger proposed a change to openstack-infra/config: Add workInProgress Gerrit ACL for Documentation  https://review.openstack.org/6017108:12
*** osanchez has joined #openstack-infra08:15
*** nati_ueno has joined #openstack-infra08:15
openstackgerritMax Rydahl Andersen proposed a change to openstack-infra/jenkins-job-builder: add private-repository to maven-target  https://review.openstack.org/6016108:22
*** vipul is now known as vipul-away08:27
*** DinaBelova has quit IRC08:27
*** Ryan_Lane has joined #openstack-infra08:29
*** vkozhukalov has joined #openstack-infra08:32
*** hashar has joined #openstack-infra08:34
*** salv-orlando has joined #openstack-infra08:34
*** SergeyLukjanov has quit IRC08:36
*** dizquierdo has joined #openstack-infra08:37
*** Ryan_Lane has quit IRC08:37
*** vipul-away is now known as vipul08:37
*** nprivalova has joined #openstack-infra08:38
*** SlickNik has quit IRC08:39
*** SlickNik has joined #openstack-infra08:40
*** Ryan_Lane has joined #openstack-infra08:42
*** denis_makogon has quit IRC08:47
*** mancdaz_away is now known as mancdaz08:47
*** sjing has quit IRC08:49
*** MarkAtwood has quit IRC08:50
*** MarkAtwood has joined #openstack-infra08:52
*** MarkAtwood has quit IRC08:52
*** boris-42 has quit IRC08:57
*** nati_ueno has quit IRC08:57
*** boris-42 has joined #openstack-infra09:16
*** yamahata_ has joined #openstack-infra09:16
*** nati_ueno has joined #openstack-infra09:22
*** jpich has joined #openstack-infra09:24
*** inkerra has quit IRC09:29
*** jcoufal has quit IRC09:30
*** k4n0 has left #openstack-infra09:32
*** Ryan_Lane has quit IRC09:32
*** fbo_away is now known as fbo09:33
*** avishayb has joined #openstack-infra09:35
*** rossella_s has joined #openstack-infra09:40
*** Ryan_Lane has joined #openstack-infra09:53
*** markmcclain has joined #openstack-infra09:54
*** johnthetubaguy has joined #openstack-infra09:56
*** jcoufal has joined #openstack-infra09:58
*** yamahata_ has quit IRC09:58
*** nprivalova has quit IRC09:58
*** markmcclain has quit IRC09:59
*** Ryan_Lane has quit IRC09:59
*** johnthetubaguy1 has joined #openstack-infra09:59
*** johnthetubaguy has quit IRC10:00
*** Ryan_Lane has joined #openstack-infra10:00
*** Ryan_Lane has quit IRC10:01
*** Ryan_Lane has joined #openstack-infra10:03
*** Ryan_Lane has quit IRC10:03
*** nati_ueno has quit IRC10:06
*** mancdaz is now known as mancdaz_away10:07
*** osanchez has quit IRC10:09
*** nprivalova has joined #openstack-infra10:09
*** mancdaz_away is now known as mancdaz10:12
*** derekh has joined #openstack-infra10:13
*** che-arne has joined #openstack-infra10:25
*** michchap has quit IRC10:30
*** michchap has joined #openstack-infra10:31
*** che-arne has quit IRC10:31
*** nprivalova has quit IRC10:32
*** rfolco has joined #openstack-infra10:45
*** Ryan_Lane has joined #openstack-infra10:45
openstackgerritMax Rydahl Andersen proposed a change to openstack-infra/jenkins-job-builder: add private-repository to maven-target  https://review.openstack.org/6016110:47
*** boris-42 has quit IRC10:47
*** hashar has quit IRC10:51
*** dizquierdo has quit IRC10:53
*** markmc has joined #openstack-infra10:55
*** nprivalova has joined #openstack-infra10:58
*** yamahata_ has joined #openstack-infra11:01
*** boris-42 has joined #openstack-infra11:02
*** Ryan_Lane has quit IRC11:09
*** rongze has quit IRC11:12
openstackgerritJaroslav Henner proposed a change to openstack-infra/jenkins-job-builder: Add seealso to batch_tasks from promoted_build.  https://review.openstack.org/5747311:13
*** hashar has joined #openstack-infra11:14
*** mdenny has quit IRC11:15
*** yamahata_ has quit IRC11:15
*** senk has joined #openstack-infra11:21
*** guohliu has quit IRC11:22
*** senk has quit IRC11:25
*** senk has joined #openstack-infra11:25
*** senk has quit IRC11:26
*** hashar has quit IRC11:31
*** Ryan_Lane has joined #openstack-infra11:31
*** HenryG has joined #openstack-infra11:32
*** jcoufal has quit IRC11:36
*** wenlock has joined #openstack-infra11:36
*** senk has joined #openstack-infra11:44
*** osanchez has joined #openstack-infra11:46
*** ArxCruz has joined #openstack-infra11:48
*** senk has quit IRC11:49
*** senk has joined #openstack-infra11:49
*** jamesmcarthur has joined #openstack-infra11:53
*** fbo is now known as fbo_away11:54
*** rongze has joined #openstack-infra11:57
*** weshay has joined #openstack-infra11:58
*** weshay has quit IRC12:02
*** weshay has joined #openstack-infra12:02
*** Ryan_Lane has quit IRC12:04
*** weshay has quit IRC12:11
*** weshay has joined #openstack-infra12:11
*** rongze has quit IRC12:17
*** senk has quit IRC12:20
*** rongze has joined #openstack-infra12:35
*** hashar has joined #openstack-infra12:36
*** thomasem has joined #openstack-infra12:37
*** jamesmcarthur has quit IRC12:41
*** rongze_ has joined #openstack-infra12:41
*** rongze has quit IRC12:42
*** nosnos has quit IRC12:45
*** talluri has quit IRC12:45
*** talluri has joined #openstack-infra12:46
*** dizquierdo has joined #openstack-infra12:46
*** talluri has quit IRC12:50
*** dims has joined #openstack-infra12:51
*** CaptTofu has quit IRC12:51
*** CaptTofu has joined #openstack-infra12:51
*** xchu has joined #openstack-infra12:59
*** senk has joined #openstack-infra12:59
*** senk has quit IRC13:03
*** senk has joined #openstack-infra13:03
*** salv-orlando_ has joined #openstack-infra13:06
*** avishayb has quit IRC13:07
*** dkranz has quit IRC13:08
*** avishayb has joined #openstack-infra13:08
*** dolphm has joined #openstack-infra13:08
*** salv-orlando has quit IRC13:09
*** salv-orlando_ is now known as salv-orlando13:09
*** resker has quit IRC13:10
*** esker has joined #openstack-infra13:10
*** prad has joined #openstack-infra13:13
*** Ryan_Lane has joined #openstack-infra13:14
*** loq_mac has joined #openstack-infra13:14
*** esker has quit IRC13:15
*** DinaBelova has joined #openstack-infra13:15
*** paul-- has joined #openstack-infra13:16
openstackgerritNikita Konovalov proposed a change to openstack-infra/storyboard: Added basic popup messages  https://review.openstack.org/5970613:16
*** prad has quit IRC13:18
*** tma996 has joined #openstack-infra13:18
*** paul-- has quit IRC13:21
*** yamahata_ has joined #openstack-infra13:25
*** CaptTofu has quit IRC13:25
*** CaptTofu has joined #openstack-infra13:25
*** yamahata_ has quit IRC13:26
*** yamahata_ has joined #openstack-infra13:26
*** loq_mac has quit IRC13:26
*** SergeyLukjanov has joined #openstack-infra13:26
*** avishayb has quit IRC13:28
*** fbo_away is now known as fbo13:29
*** mattymo has joined #openstack-infra13:32
sdaguestoryboard question (for whenever people are up): do we have a CD version of it up anywhere? (like sbbeta.openstack.org?)13:34
*** CaptTofu has quit IRC13:35
SergeyLukjanovsdague, hey13:35
SergeyLukjanovsdague, there are plans for doing it13:35
*** CaptTofu has joined #openstack-infra13:35
*** changbl has quit IRC13:35
sdagueSergeyLukjanov: cool, let me know when it happens. I'd love to play with it13:35
SergeyLukjanovsdague, sure13:36
sdaguewhile I don't think I'll end up a significant contributor, I'd like to start being able to write tooling against the rest api early to make sure that it could do the things I need.13:37
sdagueso it should provide some good feedback13:37
openstackgerritNikita Konovalov proposed a change to openstack-infra/storyboard: Added basic popup messages  https://review.openstack.org/5970613:38
*** avishayb has joined #openstack-infra13:39
* jd__ read "CD version" as "CD-ROM version", you should have seen his face13:43
SergeyLukjanovjd__ :)13:43
SergeyLukjanovsdague, plan for nearest future is to have CD for current storyboard PoC, but it'll require rewrite, so, feedback will be very useful13:44
*** dkranz has joined #openstack-infra13:44
*** fbo is now known as fbo_away13:45
*** dkranz has quit IRC13:50
sdaguejd__: yes, I think we should make slackstack13:52
sdagueand send it out on 53 floppies13:52
jd__haha13:52
SergeyLukjanovis it still possible to buy usb floppy drive?13:52
*** fbo_away is now known as fbo13:54
sdaguehttp://www.amazon.com/Sabrent-External-Floppy-Drive-FL-UDRV/dp/B00E9MD700/ref=sr_1_1?ie=UTF8&qid=1386251654&sr=8-1&keywords=usb+floppy+drive13:54
sdagueyes13:54
SergeyLukjanovheh13:55
*** smarcet has joined #openstack-infra13:55
mordredzomg. that would be an awesome thing for a summit - a floppy-based installation system13:55
*** NikitaKonovalov has joined #openstack-infra13:56
SergeyLukjanovand vintage series on 5.25'' floppies13:56
mordred++13:56
openstackgerritMartin Mágr proposed a change to openstack-infra/config: Add packstack to gerritbot  https://review.openstack.org/6024013:58
*** Ryan_Lane has quit IRC13:58
sdagueor 8.5"13:58
sdagueI had one of those for a while, though no reader13:59
sdaguefound it in a supply closet in the physics dept in college13:59
*** dprince has joined #openstack-infra13:59
*** jcoufal has joined #openstack-infra14:03
*** senk has quit IRC14:04
*** senk has joined #openstack-infra14:12
*** yamahata_ has quit IRC14:12
sdaguemordred: well, I figured out why I couldn't update my profile. In some version of the rewrite of that page, they started using urls that load javascript and have affiliation in the url14:12
sdaguewhich adblock doesn't like14:12
*** yamahata_ has joined #openstack-infra14:12
*** julim has joined #openstack-infra14:12
mordrednice14:12
*** dims has quit IRC14:12
sdaguesadly - http://www.openstack.org/foundation/technical-committee/ isn't updated from that data14:12
*** fbo is now known as fbo_away14:12
*** Ryan_Lane has joined #openstack-infra14:12
*** dims has joined #openstack-infra14:12
*** mriedem has joined #openstack-infra14:12
*** heyongli has joined #openstack-infra14:12
*** ruhe has joined #openstack-infra14:12
mordredsdague, SergeyLukjanov: I know that neither of you have the power over this - but:14:12
mordredhttp://www.mirantis.com/blog/over-1500-donated-bare-metal-servers-mirantis-softlayer-and-ibm-new-benefits-for-the-openstack-community/14:12
mordredit would be really neat if softlayer would donate 1500 servers to openstack ci - we could really use it14:12
SergeyLukjanovsdague, storyboard discussions are living in #storyboard channel14:12
SergeyLukjanovmordred, yep, it looks pretty easy to utilize all of them :)14:12
mordredSergeyLukjanov: we're very good at using all of the servers :)14:13
sdaguemordred: yeh, honestly, before I left the softlayer folks said they were going to provide that. But they were really slow in getting back about anything.14:13
mordredmaybe I should follow up - would it be worth pinging todd?14:14
*** fbo_away is now known as fbo14:15
sdaguehe's in a different enough area, I don't think so. mriedem do you know who's community active that moved over into one of the softlayer teams?14:16
*** senk has quit IRC14:16
sdagueat least mriedem would have access to org charts to figure it out14:16
*** dolphm has quit IRC14:16
mriedemsdague: talking about mjfork?14:16
mriedemsdague: mjfork moved to softlayer14:16
sdagueyeh, he's in one of those teams14:16
sdaguemriedem: I thought someone that actually was doing more code did to14:17
mriedemsdague: i just read about the mirantis + softlayer infra thing yesterday. maybe brian cline?14:17
mriedembriancline: ping!14:17
mriedemworlds are colliding!14:18
sdaguebriancline is new to me14:18
mriedemme too as of yesterday when he was asking me questions in the internal forums14:18
mriedembut he comes from softlayer and is apparently in this channel14:19
*** afazekas has quit IRC14:20
sdaguemriedem: well, if you could run down a contact it would be awesome. mjfork told me they'd be happy to add resources to the pool, but actually carving off the api creds seems to have gotten lost14:21
*** loq_mac has joined #openstack-infra14:21
openstackgerritA change was merged to openstack-infra/reviewstats: Use six to handle cpickle.  https://review.openstack.org/6011414:22
*** prad has joined #openstack-infra14:23
mriedemsdague: yeah, i'll run down brian cline and michael fork14:23
*** dolphm has joined #openstack-infra14:25
mriedemwell you're probably not going to like what i'm hearing14:29
mriedemsounds like the mirantis article is a bit messed up14:29
*** oubiwann_ has joined #openstack-infra14:29
mriedemsdague: mordred: i guess in "SoftLayer will supply the infrastructure to conduct OpenStack benchmark testing using over 1,500 bare metal servers in order to improve scalability and perform other previously impossible OpenStack testing."14:30
mriedem"will supply" is not correct o-)14:30
sdagueheh14:30
sdaguewhat... news on the internet not correct :)14:31
mriedemhttp://paste.openstack.org/show/54513/14:31
*** markmcclain has joined #openstack-infra14:32
mordrednice14:32
sdagueah... sigh14:32
*** pabelanger has quit IRC14:32
mriedemyeah, ibm f*cks up again14:32
mriedemyay!14:32
sdaguepr machines and their lack of caring about reality14:33
*** loquacities has joined #openstack-infra14:33
sdagueso, that being said, mjfork did say they'd be interested in contributing something previously (maybe not 1500 servers)14:33
sdagueso it would be good to ask him if that's still on the table14:33
sdagueand if so, how to move forward, and if not, we can drop it14:34
*** loquacities has quit IRC14:34
mordredhell, just getting a rack added to the lifeless pool would be awesome- 1500 added to the lifeless pool would, well, be really really awesome14:34
*** loquacities has joined #openstack-infra14:34
mriedemsdague: mjfork said he's working on something but can't commit to anything14:35
*** loq_mac has quit IRC14:35
*** mancdaz is now known as mancdaz_away14:35
*** johnthetubaguy1 is now known as johnthetubaguy14:35
*** dkranz has joined #openstack-infra14:36
sdaguemriedem: cool14:37
sdaguemriedem: would be nice to know when he'd figure it out by, just so we know when to annoy him again14:37
mriedemhe said he wished you wouldn't have brought it up :)14:37
fungisdague--for a while i've wanted a side project named openslack. close enough14:38
sdagueheh14:40
sdaguewell he's not on IRC, so I can't hit him directly14:40
*** jamesmcarthur has joined #openstack-infra14:40
mattymofungi, will you have time in the next hour or so to discuss moving the rest of Fuel to Stackforge? We're split across fuel-web, fuel-main, fuel-ostf, but our main puppet section is still in github14:40
mriedemsdague: having other people from ibm in irc would make my life a lot easier14:41
fungimattymo: i can discuss a little. i've overcommitted myself though so i need to pretend to be scarce and get some actual work done too ;)14:41
sdaguefwiw, I did legitimately look for his nick in #openstack before poking yuo14:41
zulmordred:  so what do we need to do go to get mox3 in gerrit?14:41
mriedemsdague: i think once he moved to softlayer he's been off openstack a bit14:42
mattymofungi, so you're going to pretend to listen to my request? :)14:42
mordredzul: just a patch to put it in. we've decided that we are going to keep usin git?14:42
mriedems/a bit//14:42
zulmordred:  yeah14:43
*** amotoki_ has joined #openstack-infra14:44
*** dolphm has quit IRC14:48
*** dolphm has joined #openstack-infra14:48
*** yaguang has joined #openstack-infra14:48
openstackgerritRoman Prykhodchenko proposed a change to openstack-infra/config: Adds devstack-gate tests for Ironic  https://review.openstack.org/5391714:48
fungimattymo: nah, it won't take much time to talk through--what specifically do you need to get clarification on?14:49
*** loquacities has quit IRC14:49
*** markmcclain has quit IRC14:49
*** loq_mac has joined #openstack-infra14:50
*** sdague has left #openstack-infra14:53
*** dolphm has quit IRC14:53
*** sdague has joined #openstack-infra14:53
*** mfer has joined #openstack-infra14:53
*** osanchez has quit IRC14:53
*** osanchez has joined #openstack-infra14:53
HenryGDoes gerrit have a way of ordering the review queue by priority of the bug(s) they reference?14:54
fungiHenryG: gerrit knows nothing about bug priorities14:54
*** dcramer_ has joined #openstack-infra14:54
fungiHenryG: the closest we have is http://status.openstack.org/reviews/14:55
*** loq_mac has quit IRC14:56
*** fbo is now known as fbo_away14:56
*** jcoufal has quit IRC14:56
*** katyafervent has quit IRC14:57
mattymofungi, we had some concerns about moving fuel library to stackforge previously14:57
mattymobut now we've really accelerated our dev cycle with gerrit, gate, formal review process, notifications, irc bots, etc... it's awesome14:58
fungimattymo: great!14:58
fungii'm glad it's working outy14:58
mattymoyeah I thought I hated gerrit but you learn to love it14:58
fungier, out14:58
mattymothe original concern about moving Fuel to Stackforge was it directly conflicts with puppet-openstack14:58
sdaguejamesmcarthur: so I sorted out the affiliation page14:58
fungii hate gerrit with every fiber of my being, but it's still a great tool14:58
mattymowe're slowly refactoring Fuel to use upstream puppet modules, but this process is drawn out (and I'm doing it alone) and we won't get there till probably May14:59
sdagueturns out, some combination of updates to that page + updates to adblock, means adblock blocks the js load. It might be good to try to detect that to warn folks14:59
sdaguejamesmcarthur: I'd still like to get http://www.openstack.org/foundation/technical-committee/ updated with the new profile14:59
*** loq_mac has joined #openstack-infra14:59
fungimattymo: sounds like a good synergy though. once you get there, you're not the only one taking care of all that puppet14:59
mattymofungi, I was wondering if we could simply move https://github.com/Mirantis/fuel to stackforge with the pledge that we'll move to a modular format as our dev cycle permits and play nicer with the other puppet groups14:59
*** dripton has joined #openstack-infra15:01
mattymoI believe a good date for that change is early next week, but we want to get some sort of infra sign-off first so we can find a "stopping point" before the move15:01
fungimattymo: i wouldn't anticipate any objections to that plan. the only people who are likely to care at all will be infra (because we have to review the patches to import it) and maybe the openstack puppet community, but you make a good argument for working with them to dissipate that over time15:01
mattymofungi, who all would be gatekeepers besides yourself?15:02
mattymofor the initial repo creation15:02
*** loq_mac has quit IRC15:02
*** sdake has quit IRC15:02
*** loq_mac has joined #openstack-infra15:03
fungime, clarkb, jeblair and mordred are the core reviewers for infra15:03
*** sdake has joined #openstack-infra15:03
mattymofungi, should I start this process with a proposed date and an email to openstack-infra?15:03
*** ryanpetrello has joined #openstack-infra15:04
fungimattymo: if you're hoping for rapid turn-around on the review, we can't really promise anything as far as prioritizing stackforge reviews (we're a bit behind still). i don't think it would take too terribly long though. e-mail can't hurt (the openstack-infra@lists.openstack.org ml is probably your best bet)15:04
mattymook great15:04
mattymothat's enough info15:05
mattymothanks15:05
fungimy pleasure15:05
*** loq_mac has quit IRC15:07
*** heyongli has quit IRC15:07
*** pblaho has quit IRC15:08
*** Ryan_Lane has quit IRC15:09
*** senk has joined #openstack-infra15:12
*** afazekas has joined #openstack-infra15:13
*** Ryan_Lane has joined #openstack-infra15:14
*** Ryan_Lane has quit IRC15:15
*** NikitaKonovalov has quit IRC15:15
*** yamahata_ has quit IRC15:17
*** senk has quit IRC15:17
*** yamahata_ has joined #openstack-infra15:17
*** senk has joined #openstack-infra15:17
*** xchu has quit IRC15:17
*** yolanda has quit IRC15:19
*** senk has quit IRC15:20
*** avishayb has quit IRC15:20
*** Ryan_Lane has joined #openstack-infra15:21
*** fbo_away is now known as fbo15:22
*** NikitaKonovalov has joined #openstack-infra15:23
*** sandy__ has quit IRC15:23
*** marun has quit IRC15:23
*** alcabrera has joined #openstack-infra15:23
*** datsun180b has joined #openstack-infra15:23
*** jergerber has joined #openstack-infra15:27
*** avishayb has joined #openstack-infra15:27
*** Ryan_Lane has quit IRC15:29
*** jgrimm has joined #openstack-infra15:31
*** sandy__ has joined #openstack-infra15:35
*** senk has joined #openstack-infra15:40
*** sarob has joined #openstack-infra15:43
*** senk has quit IRC15:45
mikalSo... My team has an internal debate at the moment. Should turbo-hipster only report on changes if tests fail? Or should it always report?15:45
fungimikal: smokestack always reports, so if that's the pattern you want to follow...15:46
jeblairmikal: if it's infrastructure is reliable, it should always report -- even if it fails on a nondeterministic bug, that's helpful because it helps expose them.15:46
jeblairmikal: (and always reporting will help people who want to know whether it has run; otherwise they might wait a long time)15:47
anteayamikal: I do agree for report the findings of the tests15:47
anteayas/do/too15:47
mikalYeah, the concern is that people will get annoyed with lots of bots say "me too!"15:47
mikalI worry about reviewers blocking reviews waiting for comments that never come15:47
mikalBut I am trying to represent the other side here fairly too15:48
jeblairmikal: (my opinion; we don't have a policy or anything)15:48
jeblairmikal: "too many things have tested this patch!" will be an interesting thing to hear from a developer!15:48
*** mancdaz_away is now known as mancdaz15:50
*** NikitaKonovalov has quit IRC15:51
mikaljeblair: I do think we're running into that scenario -- there are lots of CI systems being written at the moment15:51
mikalWe can deal with that when we get there though15:51
jeblairmikal: yes, i do wish we could better represent that in gerrit.15:53
*** SergeyLukjanov has quit IRC15:54
*** DinaBelova has quit IRC15:54
openstackgerritA change was merged to openstack-infra/config: Zuul: set gear log level to INFO  https://review.openstack.org/6009515:56
jeblairlifeless: i'm not sure if you read post-merge gerrit comments, but i left you some more info there ^15:56
*** yaguang has quit IRC15:57
*** rcleere has joined #openstack-infra15:57
*** jcooley_ has joined #openstack-infra15:58
fungimordred: i've got some additional analysis in bug 1257931 now, so we can see where things might still be racy15:58
uvirtbotLaunchpad bug 1257931 in openstack-ci "recent import of stevedore to stackforge failed to import the git repo" [High,In progress] https://launchpad.net/bugs/125793115:58
openstackgerritA change was merged to openstack-infra/config: Nodepool: cache all gerrit repos  https://review.openstack.org/6004015:59
openstackgerritA change was merged to openstack-infra/config: Add zuul pack-refs cron  https://review.openstack.org/5954315:59
*** jcooley_ has quit IRC15:59
jeblairfungi: do we have logs from those manage-projects runs?  istr its logs may be getting useful15:59
*** vkozhukalov has quit IRC16:00
fungijeblair: yeah, i'm starting to dig into them now and find if we got real detail for those16:00
mordredfungi: spectacular16:00
fungii'm mainly just surprised that those are the only handful of affected projects so far, given all the ones we've added in the past week or so16:01
*** dkliban_ has joined #openstack-infra16:03
*** pabelanger has joined #openstack-infra16:03
*** wenlock_ has joined #openstack-infra16:04
jeblairfungi: also, keep in mind clark's sleep after group creation patch merged on 2013-12-04 19:07:13 UTC16:04
anteayaI'm trying to get gerrit query to return results based on topic, if I want all the topics beginning with bp/ so far topic:bp, topic:bp/ and topic:bp/* all fail to return the results I am looking to see16:05
jeblairfungi: which is likely after these creations16:05
fungioh, yep16:05
anteayathoughts on what I should be using for syntax?16:05
*** mrodden has quit IRC16:05
*** jcooley_ has joined #openstack-infra16:06
anteayaI'm working off this page: https://gerrit.googlecode.com/svn/documentation/2.1.6/cmd-query.html16:06
*** hashar has quit IRC16:06
*** yamahata_ has quit IRC16:06
fungianteaya: well, for starters, that's not the version of gerrit we're running16:06
*** rnirmal has joined #openstack-infra16:06
jeblairanteaya: topic:^bp/.*16:07
anteayajeblair: thanks16:07
jeblairanteaya: https://review.openstack.org/Documentation/user-search.html16:07
anteayafungi: ah, good to know16:07
anteayajeblair: thanks, will click16:07
fungijeblair beat me to the url16:07
anteayahe is fast that way16:07
*** wenlock has quit IRC16:07
*** wenlock_ is now known as wenlock16:08
fungiis there any default behavior for logging.getLogger("foo") without passing any configuration?16:09
fungilike, does it just dump to syslog by default?16:09
fungibecause if so, i'm not finding any log for manage-projects16:10
*** boris-42 has quit IRC16:11
*** miqui has joined #openstack-infra16:11
*** herndon has joined #openstack-infra16:12
*** jcooley_ has quit IRC16:12
*** jcooley_ has joined #openstack-infra16:13
*** rossella_s has quit IRC16:15
*** pabelanger has quit IRC16:15
*** changbl has joined #openstack-infra16:17
sdaguefungi: it's definitely not syslog by default16:18
*** jcooley_ has quit IRC16:18
*** mgagne1 is now known as mgagne16:18
jeblairfungi: i believe the app needs to do some kind of configuration.  calling logging.basicConfig is a good start (that will dump it to stderr)16:18
jeblairfungi: it looks like manage-projects does basicConfig at level ERROR by default (unless you pass in -v)16:19
fungiokay, that's what i was vague on. so basicConfig will spew on stderr at least16:20
*** mrodden has joined #openstack-infra16:20
fungii suspect puppet agent does not actually regurgitate stderr from children however16:20
jeblairit might be stdout.  i'm not certain about that.16:21
*** nprivalova has quit IRC16:22
jeblairfungi: it's stderr.16:22
*** ndipanov has quit IRC16:24
*** pabelanger has joined #openstack-infra16:28
*** ^d has joined #openstack-infra16:29
*** ^d has joined #openstack-infra16:29
*** che-arne has joined #openstack-infra16:30
*** SergeyLukjanov has joined #openstack-infra16:30
SpamapShmm.. shouldn't "Partial-Bug: #xxxxx" result in a note on the bug about the review? https://review.openstack.org/#/c/60079 did not.16:30
*** SergeyLukjanov is now known as _SergeyLukjanov16:31
*** _SergeyLukjanov is now known as SergeyLukjanov16:31
jeblairSpamapS: tripleo mappings in jeepyb need updating16:32
maurosrhey someone know if I can get all the env vars from the tests log?16:32
*** SergeyLukjanov is now known as _SergeyLukjanov16:32
SpamapSjeblair: ah ty16:33
*** _SergeyLukjanov is now known as SergeyLukjanov16:33
jeblairSpamapS: in progress16:33
openstackgerritJames E. Blair proposed a change to openstack-infra/jeepyb: Update tripleo project name mappings  https://review.openstack.org/6027816:33
*** dolphm has joined #openstack-infra16:34
*** tma996 has quit IRC16:35
*** pcrews has joined #openstack-infra16:37
*** rongze_ has quit IRC16:37
fungiwell, either the puppet agent is swallowing the stderr from manage-projects or there wasn't any error output when those projects were added. also, we're running manage-projects *a lot*16:39
fungisince we have a recursively managed directly we're ensuring absent, puppet thinks it should refresh manage-projects on every agent run16:40
fungiplus the cron job which runs it every half hour16:41
fungiwe're basically running manage-projects 6 times an hour, give or take16:41
jeblairfungi: wow, you'd think that would mean run it if it _changes_, not run it if things are still absent.16:41
fungivery much seems to mean the latter16:42
fungii'm going to try and clean this up real quick with a change. hopefully this will help16:42
fungiand also i'll repair the three affected stackforge projects16:42
*** pete5 has joined #openstack-infra16:43
SpamapSjeblair: oh even more ty!16:44
*** syerrapragada1 has joined #openstack-infra16:46
*** senk has joined #openstack-infra16:49
*** syerrapragada has quit IRC16:49
*** syerrapragada has joined #openstack-infra16:50
ruhecan someone point me to a script/config which mirrors all the gerrit commits (and change requests) to github?16:50
fungiruhe: it's a built-in feature of gerrit16:50
*** senk1 has joined #openstack-infra16:51
fungiruhe: gerrit has a replication config file which can contain patterns of various git remotes where you want things replicated16:51
*** jcooley_ has joined #openstack-infra16:51
ruhefungi: thank you!16:52
*** rossella_s has joined #openstack-infra16:52
sdaguejeblair: have things calmed down enough that you'd be up for landing the er web console flip over?16:52
*** syerrapragada1 has quit IRC16:52
fungiruhe: https://review.openstack.org/Documentation/config-replication.html16:52
jeblairsdague: yes!16:52
*** senk has quit IRC16:53
jeblairsdague: oh, we need to land this first, yeah? https://review.openstack.org/#/c/59289/16:53
*** reaperhulk has joined #openstack-infra16:54
sdaguejeblair: yes16:54
sdaguethat should be non destructive to the environment16:54
*** jcooley_ has quit IRC16:54
sdagueso if you just push go on that, let that land, then the config change should bring everything up on it16:54
*** jcooley_ has joined #openstack-infra16:55
*** dolphm has quit IRC16:55
fungizaro: interestingly, i'm finding no references to gerrit's replication config in the 2.7 docs... did that change significantly?16:56
*** jcooley_ has quit IRC16:57
openstackgerritA change was merged to openstack-infra/elastic-recheck: add support for installing the web dashboard  https://review.openstack.org/5928916:57
*** jbryce has joined #openstack-infra16:57
openstackgerritA change was merged to openstack-infra/elastic-recheck: Cleanup graphite graph  https://review.openstack.org/5982916:57
openstackgerritA change was merged to openstack-infra/elastic-recheck: Remove check jobs from graphite graph  https://review.openstack.org/5983016:57
openstackgerritA change was merged to openstack-infra/elastic-recheck: Add more gate jobs to graphite graph  https://review.openstack.org/5983116:57
fungizaro: nevermind. seems it became a plugin... http://gerrit-documentation.googlecode.com/svn/ReleaseNotes/ReleaseNotes-2.5.html#replication16:57
mordredfungi: we're going to have to learn more about plugins aren't we16:57
fungizaro: obviously we'll want to add that, i suspect16:57
*** jcooley_ has joined #openstack-infra16:57
fungiyup16:58
*** avishayb has quit IRC16:58
*** andreaf has quit IRC17:00
*** mdenny has joined #openstack-infra17:01
*** andreaf has joined #openstack-infra17:01
openstackgerritA change was merged to openstack-infra/elastic-recheck: make ResultSet actually inherit from list  https://review.openstack.org/5980517:01
*** fbo is now known as fbo_away17:01
*** jcooley_ has quit IRC17:02
jeblairi'm updating the devstack-precise nodepool images in az1 and the bare-precise in rax-ord to get the new cached git repos17:03
*** gyee has joined #openstack-infra17:03
*** rnirmal has quit IRC17:05
*** rnirmal has joined #openstack-infra17:05
*** jcooley_ has joined #openstack-infra17:06
*** markmc has quit IRC17:06
fungiexcellent17:07
pleia2good morning17:07
zaromorning pleia217:07
*** syerrapragada1 has joined #openstack-infra17:07
zarogood morning as weel17:08
*** dolphm has joined #openstack-infra17:08
*** krtaylor has quit IRC17:09
*** CaptTofu has quit IRC17:10
*** senk1 has quit IRC17:10
*** CaptTofu has joined #openstack-infra17:10
*** jswarren has joined #openstack-infra17:10
*** syerrapragada has quit IRC17:10
*** jswarren has quit IRC17:11
*** dolphm has quit IRC17:11
*** CaptTofu has quit IRC17:12
*** CaptTofu has joined #openstack-infra17:12
*** jswarren has joined #openstack-infra17:12
*** jamesmcarthur has quit IRC17:12
*** julim has quit IRC17:16
*** julim has joined #openstack-infra17:17
*** dolphm has joined #openstack-infra17:19
*** jamesmcarthur has joined #openstack-infra17:21
jeblairsdague: https://review.openstack.org/#/c/59294/ aprvd17:23
*** reaperhulk has left #openstack-infra17:24
openstackgerritA change was merged to openstack-infra/config: move to elastic-recheck web console in share  https://review.openstack.org/5929417:25
*** johnthetubaguy has quit IRC17:26
*** pcrews_ has joined #openstack-infra17:26
fungiokay, so to reduce the manage-projects churn, i'm first going to need to add a --sync-upstream option which does nothing but call sync_upstream() for projects, since that's why we have a cronjob firing manage-projects twice an hour right now17:26
*** chandankumar has quit IRC17:26
jeblairfungi: is 'fix up something that failed during the last run' something that we want the cron to do too?17:27
fungithat's possible17:27
mordredthat's what the recent patches were aimed at allowing to happen17:27
jeblairfungi: i thought that may have been some of the reasoning for the cron, but i'm not certain.17:27
fungiif "eventually consistent" is something we're shooting for17:27
mordredit's for both17:27
mordredbecause github17:27
fungiokay, so maybe just fix the puppet stupidity where it thinks recursively absent content is something to refresh the manage-projects exec for on every run17:28
fungimordred: do you expect manage-projects to operate sanely if two copies of it run simultaneously?17:28
mordredfungi: hrm. probably not, no17:29
fungiotherwise we may want a locking mechanism so that the cron job and puppet-triggerred runs don't step on one another17:29
mordredit does git branch resetting17:29
*** rcleere has quit IRC17:32
*** rcleere has joined #openstack-infra17:33
*** dolphm has quit IRC17:34
*** mancdaz is now known as mancdaz_away17:35
openstackgerritSergey Lukjanov proposed a change to openstack-infra/config: Fix Savanna d-g jobs  https://review.openstack.org/6029617:36
*** markwash has quit IRC17:38
*** sarob has quit IRC17:39
*** sarob has joined #openstack-infra17:40
mgagnefungi: there should already be a lock mechanism for puppet17:40
mgagnefungi: unless you are not referring to puppet itself but something triggered by puppet and a cron17:40
mgagnefungi: nvm, I should read more than 2 lines of backlog17:41
openstackgerritSergey Lukjanov proposed a change to openstack-infra/config: Add gate-climate-devstack-dsvm  https://review.openstack.org/5870217:41
mordredmgagne: :)17:42
zaroclarkb: yah! i can finally get into my owa email.17:43
*** sarob has quit IRC17:44
openstackgerritSergey Lukjanov proposed a change to openstack-infra/config: Fix Savanna d-g jobs  https://review.openstack.org/6029617:46
jeblairfungi: does launchpad parameter exist in projects.yaml?  i have forgotten the status of that move.17:47
*** jcooley_ has quit IRC17:47
*** osanchez has quit IRC17:47
*** jcooley_ has joined #openstack-infra17:48
openstackgerritJames E. Blair proposed a change to openstack-infra/jeepyb: Update tripleo project name mappings  https://review.openstack.org/6027817:48
openstackgerritRuslan Kamaldinov proposed a change to openstack-infra/config: Add #heat to eavesdrop  https://review.openstack.org/6029817:48
*** sarob has joined #openstack-infra17:49
*** krtaylor has joined #openstack-infra17:50
fungijeblair: good question. doesn't look like any are in there yet, though i believe it's supposed to work17:51
*** NikitaKonovalov has joined #openstack-infra17:52
fungispeaking of projects.yaml and alphabetization, i just noticed we merged at least one more out-of-order entry to it this week17:52
*** jcooley_ has quit IRC17:52
fungii keep forgetting to check for that17:53
jeblairfungi: i am bad at remembering to check that. :(17:53
clarkbmorning17:55
pleia2humans are bad at that17:55
clarkbzaro: good to know, I will try it on my end too17:55
pleia2morning clarkb17:55
*** jcooley_ has joined #openstack-infra17:56
*** senk has joined #openstack-infra17:56
*** sarob has quit IRC17:57
openstackgerritJames E. Blair proposed a change to openstack-infra/jeepyb: Update tripleo project name mappings  https://review.openstack.org/6027817:58
*** sarob has joined #openstack-infra17:58
jeblairmost difficult change i've written in quite some time!17:58
jeblairif anyone is looking for somewhat low-hanging-fruit, finishing the move of this info to projects.yaml would be well appreciated!17:58
SergeyLukjanovjeblair, it'll require some freeze for projects.yaml17:59
zaroclarkb: bad part is that we now have to read all those17:59
jeblairSergeyLukjanov: we can do that17:59
*** ArxCruz has quit IRC17:59
SergeyLukjanovwhat do you think about moving launchpad mappings out of the jeepyb too?17:59
SergeyLukjanovit should be moved to another file18:00
jeblairSergeyLukjanov: yes, i think it should all move to projects.yaml18:00
fungii believe the way you coded it, we can incrementally add launchpad and direct release mappings for individual projects into projects.yaml if desired, and those will override the ones listed in jeepyb's code18:00
SergeyLukjanovmy patch that *should* read settings from projects.yaml was landed, but it should be tested more before using18:00
fungiand then just remove those data structures from the code at the end18:01
SergeyLukjanovfungi, yup18:01
*** senk has quit IRC18:01
SergeyLukjanovfungi, I'm thinking about how to test it to be sure that it'll work ok18:01
jeblairSergeyLukjanov: how about moving -infra proects to projects.yaml ?18:01
fungiwe do like to dogfood our own changes before inflicting them on the rest of the project sometimes. sounds like a good idea to me18:02
*** nati_ueno has joined #openstack-infra18:02
SergeyLukjanovgood idea, I'll do it tonight18:02
*** sarob has quit IRC18:02
sdaguejeblair: so apparently something was not right with that change - http://status.openstack.org/elastic-recheck/18:03
*** rossella_s has quit IRC18:03
*** reed has joined #openstack-infra18:03
sdagueserver 500 error18:03
jeblairsdague: yes, i think there's some confusion in the puppet config18:03
jeblairsdague: the elastic_recheck module is not used on static.o.o, because we don't run the irc bot there.18:04
*** senk has joined #openstack-infra18:04
openstackgerritA change was merged to openstack-infra/jenkins-job-builder: Add support for batch_tasks.  https://review.openstack.org/5746918:04
sdaguejeblair: ah... oops18:04
sdaguewait, how did it make the graphs?18:04
sdagueit has to be over there18:04
jeblairsdague: we may want to extend the elastic_recheck module to have elastic_recheck::bot and elastic_recheck::graphs or something18:04
jeblairsdague: we just install the elastic_recheck package from the repo, without the other supporting stuff.18:05
jeblairsdague: (not using the e_r puppet module)18:05
sdaguejeblair: right, so that's ok18:05
jeblairsdague: yeah, but it doesn't get /var/lib/elastic_recheck created because that's in the module18:06
sdagueah... oops18:06
sdagueok, I'm going to eat some lunch, then try to fix it18:07
jeblairsdague: ok.  i think splitting the module into the bot, dashboard, and common components is the way to go, that way we can have it on different servers as now, or colocate as needed.18:07
sdagueyep18:07
*** vkozhukalov has joined #openstack-infra18:08
*** jbryce has quit IRC18:09
jeblaira job is running on an az1 node built from the new image, and it seems to be functioning18:10
*** derekh has quit IRC18:10
*** ArxCruz has joined #openstack-infra18:11
*** jerryz has joined #openstack-infra18:12
*** markwash has joined #openstack-infra18:12
fungimgagne: weird puppet question... if i have a directory with recurse/replace/purge true and then subscribe an exec to that directory, is it expected that the exec will get a triggered refresh on every puppet agent run even when there are no files actually added/changed/removed in that dir?18:12
clarkbsdague: jeblair: I like that plan18:12
*** harlowja has joined #openstack-infra18:14
mgagnefungi: damn, that's one good question... I'll have to dig for that one.18:15
fungimgagne: the docs are a little vague on that point, but it seems to be the behavior we're getting18:15
mgagnefungi: is it what you are expecting?18:15
fungimgagne: we want to run a command when contents of a puppet-managed directory change18:16
fungimgagne: and not run it when the contents don't change18:16
fungimgagne: but with recurse/replace/purge true puppet seems to remove all contents, add them back again, and then counts that as a change to trigger refresh on teh exec we subscribed18:17
mgagnefungi: The doc I usually use doesn't convert this use case: http://ai.stanford.edu/~llao/puppet/puppet-trainning.html18:17
*** ryanpetrello has quit IRC18:17
openstackgerritJames E. Blair proposed a change to openstack-infra/config: Have gerrit-git-prep clone from /opt/git if it exists  https://review.openstack.org/6030418:17
*** nicedice has joined #openstack-infra18:18
openstackgerritJames E. Blair proposed a change to openstack-infra/config: Have gerrit-git-prep clone from /opt/git if it exists  https://review.openstack.org/6030418:18
clarkbfungi: that is really interesting behavior from purge18:18
mgagnefungi: I have a sandwich and a mouse in the other hand and I'm in a meeting. I'll check asap18:19
*** shardy is now known as shardy_afk18:19
fungimgagne: there's no rush. i'll keep poking at it, just didn't know whether that was known behavior18:19
jeblairmgagne: i read that as a sandwich and a mousse.18:19
jeblairwhich sounds like an interesting meeting.18:20
mgagnejeblair: a moose sandwich =)18:20
*** yolanda has joined #openstack-infra18:22
*** dizquierdo has quit IRC18:23
fungimgagne: nevermind--i missed a critical bit of detail18:24
*** Hefeweizen has quit IRC18:24
fungiseems that i was looking at the one run where it started managing this directory and removed all (well, most) of the unneeded contents, but then it kept re-triggering because without force it will refuse to remove empty directories18:25
fungiso apparently when it tries and fails to remove something, that triggers a refresh on subscribers18:25
clarkbfungi: did I miss an option when I made that change?18:25
mgagnefungi: oh, true https://github.com/puppetlabs/puppet/blob/master/lib/puppet/type/file.rb#L18418:26
fungiclarkb: i think you have to use force if you want directories removed18:26
lifelessjeblair: I do when they stand out, but I get 80 + gerrit mails a day - I don't read the majority of tem18:26
lifelessjeblair: so thanks for pointing that out18:26
clarkbfungi: which dir is this? the gerrit acl dir?18:27
fungiclarkb: yeah18:27
clarkbI think you are correct, sorry about missing that18:27
fungiclarkb: on every agent run it's trying to remove the heat-api and openstack-ci directories18:27
fungiclarkb: no worries. i think i've done the exact same thing elsewhere18:27
fungior at least i know i already had to look that behavior up once18:27
clarkbfungi: we should probably add force to the other purges to avoid this problem in the future (though they don't contain dirs iirc)18:28
fungii'd just never seen its effects where subscribers to the directory were concerned18:28
*** sarob has joined #openstack-infra18:28
fungiclarkb: can do. cahnge inbound in moments18:29
fungiclarkb: seems we were already doing it in modules/asterisk/manifests/function/customdir.pp but nowhere else18:32
*** sarob has quit IRC18:33
clarkbpabelanger and russellb and jeblair know how to puppet18:34
*** melwitt has joined #openstack-infra18:34
*** dkranz has quit IRC18:35
russellbi do?18:35
clarkbrussellb: you wrote the asterisk stuff I thought :)18:35
russellbi did18:35
russellbi stumbled my way through it :)18:36
*** dkranz has joined #openstack-infra18:36
pabelangerclarkb: fungi: Yup, that is how I do it today with asterisk, if the content of a folder changes, I notify a server to reload a specific module18:36
lifelessjeblair: I need to start organising (or delegating organisation of) the tripleo midcycle meetup soonish; would like to know what you thought of my question in the infra mid cycle meetup thread18:37
fungipabelanger: yeah, and if force isn't set and there are cruft subdirectories then that will reload every time the agent run and fails to remove them18:37
pabelangerfungi: where is the code?18:38
*** krotscheck has joined #openstack-infra18:39
pabelangerfungi: so, if I understand you have files in the directory that is not manage by puppet?18:39
fungipabelanger: yes, and want puppet to remove them18:40
jeblairlifeless: colocating may be a good idea.  what i'm struggling with now is the possibility of infra, tripleo, and storyboard meetups all drawing for our attention (along with other similarly timed events)18:40
pabelangerfungi: purge => true, recurse => true18:40
fungipabelanger: in cases where there may have previously been subdirectories under it managed by puppet which were subsequently deleted from the repository some tim elater18:41
pabelangerAnd force of course18:41
lifelessjeblair: yeah, thats precisely it18:41
openstackgerritJeremy Stanley proposed a change to openstack-infra/config: Use force when purging  https://review.openstack.org/6031318:42
jeblairlifeless: at this point, i need to sift through the responses and see what kind of meetup would help achieve our goals.  not having one is an option too (there are a few events where some of us will be together, so we can try to put more emphasis on those)18:42
fungipabelanger: yep, we arrived at that as the solution ^18:42
*** che-arne has quit IRC18:42
sdagueoh, man, we dup the er stuff all over again in static.pp18:43
pabelangerfungi: There is a performance hit from puppets POV with that, just a heads up18:44
pabelangerso, expect puppet to run longer on nodes18:44
fungipabelanger: i'll keep an eye out for it. thanks!18:44
*** sdake has quit IRC18:44
*** sdake has joined #openstack-infra18:45
jeblairlifeless: i'll try to be more concrete next week.18:45
fungipabelanger: if it becomes a problem, i guess we can fall back on manually removing empty directories18:45
pabelangerfungi: it's more with recursive directories that puppet knows about. For example, that's why I didn't example it under the asterisk logic18:46
pabelangerI was having puppet issues with load18:46
lifelessjeblair: thanks!18:46
pabelangerbut that might have changed with the version we are using now, ever tested it again18:46
fungipabelanger: force, specifically, was causing higher load?18:47
pabelangerright18:47
pabelangerbecause it would have to search and compare to know puppet files / directories before deleting18:47
pabelangerIIRC18:47
fungipabelanger: i would expect purge to be the culprit for that18:48
pabelangermaybe the puppetlabs guys can comment, like I said, it's been a while since I tested that18:48
*** dkranz has quit IRC18:48
pabelangerwell, I think if purge is on, and force is off, it won't actually delete a folder18:48
pabelangerand go past the first layer18:48
pabelangerfungi: it was the checksum process18:49
pabelangerthat is what it was18:49
fungiit seems to purge file contents successfully within those subdirectories, just leaves the empty directories behind and throws an error about them18:49
*** Abhishe__ has joined #openstack-infra18:51
clarkbis pbr integration testing still a problem?18:52
lifelessI did not get time to look into it18:52
lifelessso I expect so18:53
openstackgerritSean Dague proposed a change to openstack-infra/config: factor out bot code to elastic_recheck::bot  https://review.openstack.org/6031818:54
sdaguejeblair: so I only split out ::bot18:54
openstackgerritJames E. Blair proposed a change to openstack-infra/config: Switch rax nodes to performance flavors  https://review.openstack.org/6031918:54
sdaguebecause the er config is a bit too muddled for a cleaner separation. I'll clean some on the er side for that at some point.18:55
jeblairfungi, clarkb: would you please review https://review.openstack.org/#/c/60304/ and https://review.openstack.org/#/c/60319/ ?  watching those roll out should keep me occupied for a while18:55
jeblairfungi, clarkb: and i want the g-g-p change to be included in the image rebuilds that will be triggered.18:56
fungiclarkb: yeah, most recent run still fails on glance->oslo.messaging->d2to1 https://jenkins01.openstack.org/job/check-requirements-integration-dsvm/6/consoleText18:56
jeblairhrm, actually, there may not be image rebuilds triggered; i may have to do that manually.18:56
jeblairsdague: ack18:56
clarkbjeblair: sdague: is there any concern putting ssh keys on static.o.o since jenkins can get ot static.o.o?18:57
clarkboh nevermind releasestatus already does that18:58
clarkbignore me18:58
sdagueclarkb: good question, but I saw other keys, so I didn't even think twice about it18:58
jeblairclarkb, sdague: i was just looking at that, and i don't really like that we're adding keys there that aren't used18:58
clarkbjeblair: I agree18:58
*** sarob has joined #openstack-infra18:58
*** yolanda has quit IRC18:58
sdagueyeh, so I think we need to do some er cleanup before that's going to be easy to separate18:59
*** Loquacity has quit IRC18:59
sdaguebut, my brain is a little foggy atm. Going to go for a walk to reenergize18:59
clarkbjeblair: I just approved the g-g-p change, should I wait on the nodepool config change or is it ok to approve that now too?19:00
jeblairsdague: i don't think it's as bad as you think.  i think you can already remove a lot of stuff from the init.pp file that you just moved into bot.pp, including the keys.19:00
*** Loquacity has joined #openstack-infra19:00
jeblairsdague: will leave a comment for when you are reenergized :)19:00
clarkbnodepool config change ahs my +219:01
fungiclarkb: i'm taking longer than i should to figure out where the name-filter parameter was introduced into nodepool's parser, but feel free to skip me19:01
openstackgerritA change was merged to openstack-infra/config: Have gerrit-git-prep clone from /opt/git if it exists  https://review.openstack.org/6030419:01
clarkbfungi: no no, you should do that :)19:01
clarkbmordred wrote the change, recent git log should show it19:01
*** dkranz has joined #openstack-infra19:02
jeblairfungi: and i restarted nodepool yesterday, so running nodepool should support it19:02
clarkb28a1c41400d40cf9eecdfd102b98b8ffa43052a4 is the commit sha19:02
fungiyup, lgtm19:03
*** sarob has quit IRC19:03
fungii just didn't remember it because i hadn't reviewed that change19:03
fungithat's what i get for falling behind19:03
*** sarob has joined #openstack-infra19:03
jeblairfungi, clarkb: | performance2-120 | 120 GB Performance      | 122880    | 40   | 1200      |      | 32    | 10000.0     | N/A       |19:04
fungiright on19:05
openstackgerritA change was merged to openstack-infra/config: Switch rax nodes to performance flavors  https://review.openstack.org/6031919:05
clarkbjeblair: I should move all of the elasticsearch stuff to those flavors :)19:05
clarkb0.90.3 is holding up well so far, don't think we quite need to update at least not yet19:06
jeblairhttp://paste.openstack.org/show/54542/  full list for easy reference19:06
clarkbhmm the disk is different, would need to move the data to cinder block volumes19:07
clarkbanyways good to know thanks for pasting that19:07
*** sdake_ has joined #openstack-infra19:07
jeblairclarkb: ssd cinder volumes? :)19:07
*** derekh has joined #openstack-infra19:08
*** sarob has quit IRC19:08
*** ^d has quit IRC19:08
*** ^d has joined #openstack-infra19:08
*** NikitaKonovalov has quit IRC19:10
openstackgerritlifeless proposed a change to openstack-infra/reviewstats: Add a bug statistics tool.  https://review.openstack.org/5890319:13
*** thomasm_ has joined #openstack-infra19:13
*** thomasem has quit IRC19:14
jeblairBadRequest: Instance type's disk is too small for requested image. (HTTP 400) (Request-ID: req-64298f0b-ccdc-47da-ae83-fae3e6fcb926)19:15
jeblairokay, so, that makes sense.  i'll kick off new image builds19:15
fungiright, probably need a separate base image?19:15
jog0I assume someone is aware http://status.openstack.org/elastic-recheck/ is 500 erroring19:15
*** yolanda has joined #openstack-infra19:16
jeblairjog0: yeah, sdague was working on fixing it, but he stepped out.  if you like we can revert the change that broke it19:16
lifelessI wish one could -1 a review comment19:16
mgagnelifeless: is there a reddit plugin for gerrit adding upvote/downvote for comemnts? =)19:17
fungilifeless: i'm willing to bet you want to -1 one of my comments ;)19:17
lifelessfungi: no19:18
lifelessfungi: thing in solum19:18
fungiahh19:18
lifelessthere is a suggestion that there be data in the DB that isn't exposed in the API19:18
lifelesswith the comment19:18
lifeless'if a platform operator needs this they should read the DB directly.19:18
lifelessWhich sends a huge raft of warning signals down my spine19:19
lifelessnext we'll be expecting operators to log into machines to manually analyse problems19:19
*** jbryce has joined #openstack-infra19:19
clarkblifeless: that is what gerrit does :)19:19
lifeless*thats not how scale works*19:19
clarkband it is painful19:20
openstackgerritDarragh O'Reilly proposed a change to openstack-infra/os-loganalyze: Provide a script that starts a simple WSGI server  https://review.openstack.org/6032519:20
lifelessyup19:20
fungiwe hateses it19:20
clarkbfungi: I am staring at pbr integration test test logs and ugh19:20
clarkbpython setup.py sdist works (which makes sense) then we take that sdist and try to install it in a different virtualenv and that fails19:20
fungiclarkb: right. so returning to last night's question, do we fix glance and just bank on wheel uploads to pypi soon? or do we make pypi-mirror deal with tarball urls, or option three none of the above?19:21
*** ^d has quit IRC19:21
jog0jeblair: yeah that would be helpful. unless sdague finishes lunch soon19:21
clarkbfungi: I am leaning towards deal with tarball urls19:22
clarkb(we should also fix glance, but that doesn't fix the larger problem)19:22
fungiclarkb: an optional --dont-ignore-tarball-urls flag to pypi mirror which we can run with in the integration test and then kiwf asap when wheel uploads arrive?19:23
jog0jeblair: now that that page is the e-r repo can we in theory make a test to make sure that page at least renders and doesn't return a 500?19:23
openstackgerritJames E. Blair proposed a change to openstack-infra/config: Revert "move to elastic-recheck web console in share"  https://review.openstack.org/6032619:23
jog0jeblair: oh never mind19:23
jeblairjog0: ?19:23
jeblairjog0: oh, i see.19:23
jog0jeblair: I thought it was a bad e-r patch not a config .. ignore me19:23
jeblairjog0: there's your revert ^19:23
clarkbfungi: I am not sure why they would be exclusive19:23
mtreinishfungi: hey a quick question, how do I update the meeting ical feed?19:24
mtreinishwe're changing the qa meeting time19:24
jeblairmtreinish: ask ttx19:24
*** pabelanger has quit IRC19:25
clarkbfungi: actually wait. This is a symptom of broken dependencies right? do we expect anyone to use tarball links at all anymore?19:25
*** derekh has quit IRC19:25
clarkbI guess we do for oslo19:25
mtreinishjeblair: ok thanks19:25
fungiclarkb: we expect them to until we upload wheels for prereleases to pypi, then we want to remove them with extreme prejudice19:25
mtreinishI guess it's a bit late for ttx right now19:25
clarkbmordred: are you around? want to talk pbr integration testing and wheels and tarballs19:25
fungimtreinish: yeah, i have no details on that ical feed. i think he may be the only one managing it19:26
*** pabelanger has joined #openstack-infra19:26
fungiclarkb: which is why if we do add tarball url support to pypi-mirror, i'd like to see it removed again (either via revert or by hiding the code path behind a switch)19:27
mikalOne last question before I turn on turbo hipster commenting on reviews...19:27
clarkbfungi: gotcha (I remember now that wheels require new pip which won't install alpha releases by default which means we can avoid tarball releases)19:27
mikalWe can't expose the logs of runs yet because of security concerns, so instead we're going to zero vote on everything with a comment. Is that ok with people?19:27
fungiclarkb: because i expect it to add complexity (though maybe not--maybe we just turn off the filter which ignores them)19:27
*** jerryz has quit IRC19:27
openstackgerritSergey Lukjanov proposed a change to openstack-infra/config: Use launchpad prop to store link to the lp project  https://review.openstack.org/6032919:28
fungimikal: only if the comment includes instructions for what form they need to mail you for a paper copy of the logs? ;)19:28
clarkbmikal: I would be ok with that if the comment included info on why it isn't a +/-119:28
*** yolanda has quit IRC19:28
jeblairmikal: i think publicly accessible logs are a pretty strong requirement; it's basically not actionable information for a developer19:28
mikalYeah, it will say "test failed" with a log URL. The log URL (for now) redirects to a page explaining our concerns.19:28
mikalAnd who to contact to get help.19:29
mikalWe're working on fixing this, but I don't want to block on it19:29
*** jerryz has joined #openstack-infra19:29
clarkbjeblair: I definitely agree especially with -1s19:29
mikalYeah, which is why we're going to zero vote for now19:29
*** jerryz has quit IRC19:29
mikalWe feel we can't -1 without public logs19:29
clarkbbut a 0 is something anyone can do to provide whatever info19:29
mikalOur problem is this is real user data, so we're worried about leaking things in the logs19:29
jeblairmikal: i won't throw a fit if you start doing this experimentally, but i don't think it's tenable in the long run.  but it sounds like you know that, so.  :)19:30
mikaljeblair: yes, we agree here19:30
krtaylorspeaking of public logs, is there any space to copy results up to a log server?19:30
clarkbno19:30
mikalIts about timing... Do we block while we fix the log thing or not?19:30
clarkb:( unless fungi found the fountain of disk the other day19:30
*** jerryz has joined #openstack-infra19:31
krtaylorhehheh19:31
fungiclarkb: no, but i've gotten very, very good with pvmove19:31
clarkbkrtaylor: we are limited to 1TB block devices and can only have a total of 10TB19:31
*** rpodolyaka1 has joined #openstack-infra19:31
SergeyLukjanovwhere is the elastic-recheck bug tracker?19:31
SergeyLukjanovI mean which LP project used for it?19:31
mikalclarkb: until you get to swift... Which I will hassle Josh about.19:31
clarkbSergeyLukjanov: openstack-ci19:31
SergeyLukjanovok, thx19:31
krtaylorok, I am also concerned with making our results available vs seeing any results at all, timely that mikal asked19:32
jeblairmikal: yeah, if/when he's ready to work on that, poke him to hop on irc, and we can make a blueprint or etherpad or whatever to make sure he has what he needs19:32
fungiclarkb: as soon as i'm done dissecting this manage-projects traceback/exception, i've got 1tb more to shuffle onto new cinder volumes and then the last of the 0.5tb volumes can be removed (so the server will be using only 1tb volumes at that point)19:32
mikaljeblair: yeah... I was mean and force him to work on TH first though19:33
mikaljeblair: we're in a pretty target rich environment and its hard to get everything done19:33
dprincesdague: any ideas on the grenade failure here: https://review.openstack.org/#/c/53874/19:34
clarkbfungi: woot19:34
openstackgerritSergey Lukjanov proposed a change to openstack-infra/config: Use launchpad prop to store link to the lp project  https://review.openstack.org/6032919:34
clarkbfungi: also http://git.openstack.org/cgit/openstack/requirements/tree/dev-requirements.txt says we can use older oslo.messaging19:34
clarkbfungi: so if we were to try and fix that in glance we would have to fix it in requiremetns first which currently is broken19:34
fungiclarkb: well, that's annoying19:35
fungii mean, annoying i didn't spot it before, anyway19:35
openstackgerritSergey Lukjanov proposed a change to openstack-infra/jeepyb: Remove openstack-infra projects mappings to LP  https://review.openstack.org/6033219:35
fungiclarkb: so if we do have pypi-mirror deal with tarball urls, then that would be the place to have it look for them at any rate19:36
clarkbfungi: right, I think we will have to make pypi-mirror smarter first if we want to do this in a gat efriendly way19:36
fungiso that it would be consistent with what we're enforcing19:36
brianclinemriedem: sdague: pong! (by pigeon carrier, sorry)19:37
clarkbI will look at adding that flag to pypi-mirror19:37
fungibriancline: they were just trying to convince you to join the tripleo bare-metal cloud with lots of shiny-fast hardware i think19:38
*** yolanda has joined #openstack-infra19:39
*** jpich has quit IRC19:40
*** yolanda has quit IRC19:41
SergeyLukjanovjeblair, fungi, I've moved mappings to lp of openstack-infra projects to the project.yaml (https://review.openstack.org/#/c/60329/ and https://review.openstack.org/#/c/60332/)19:42
brianclinefungi: for openstack CI?19:43
fungibriancline: yep19:44
fungiSergeyLukjanov: awesome! taking a look19:44
SergeyLukjanovI've moved only mappings, the next step is move release options19:44
*** jcooley_ has quit IRC19:44
SergeyLukjanovand other projects :)19:44
fungiSergeyLukjanov: it's a great first step so we can make sure it's working like we expect19:45
SergeyLukjanovfungi, I hope it'll work ;)19:45
brianclinefungi: I'll ask about it -- it was something I personally really wanted to do as far back as 14 months ago, but as is usually the case, other stuff came up and it fell off everyone's radar19:46
*** jcoufal has joined #openstack-infra19:46
fungibriancline: i know how it is. if you manage to get it to happen, it'll be quite exciting though!19:46
brianclinewe did contribute some resources to Swift to do their global clustering testing, so it certainly seems reasonable to me, I just have to push for approval. I'll move it back up the list for sure19:47
*** hashar has joined #openstack-infra19:48
fungimordred: so i think i found why manage-projects breakage is happening. steps seem to be "first init a repo, then if there's an upstream defined clone it there" (except git will patently refuse to do the latter--directory already exists)19:48
fungiat least that's what the debug output says is happening. i'm starting to follow the source now to confirm19:49
brianclinemordred: fungi: sdague: just to get an idea, what quantity and caliber of CI resources are we talking? (can take it offline if needed)19:50
fungibriancline: probably a larger discussion, and possibly better in the context of what the tripleo team is already managing for the first zone/region of the bare-metal cloud19:51
fungii believe sdague stepped out for a bit, so i don't want to speak for whatever he wanted to discuss though19:52
openstackgerritA change was merged to openstack-infra/config: Revert "move to elastic-recheck web console in share"  https://review.openstack.org/6032619:54
clarkbfungi: I am starting to grok this problem much better now. Biggest issue is that we are not installing each project to get the actual dependencies. This coupled with not installing any tarball deps to get transitive dependencies means we end up with an incomplete mirror (I think all of this has been said, but it is really clicking in my head now)19:55
clarkbfungi: though, if that is the problem, why did explicitly stating d2to1 is a dependency not help?19:55
brianclinefungi: no worries, I'll shoot out an email19:55
*** ^demon has joined #openstack-infra19:55
*** ^demon has joined #openstack-infra19:55
*** ^demon is now known as ^d19:56
fungiclarkb: yep, that's what i was trying (but probably failing) to point out19:56
fungiclarkb: and agreed, not sure why my attempt to add d2to1 to the global reqs didn't "fix" it19:56
fungiwhich means maybe we're wrong19:56
clarkbunless maybe the mirror is being built with the wrong version of global reqs19:57
ttxmtreinish: if you update the meetings wiki page, I'll get a notification and update the feed19:57
fungimordred: red herring on the manage-projects thing. if i'm clearing git repos to re-run it, i have to remember to delete the /var/lib/jeepyb repo copies now too19:57
*** elasticio has joined #openstack-infra19:57
*** jcooley_ has joined #openstack-infra20:01
clarkbit uses the requirements repo put in place by devstack-gate so I don't think that hunch is the problem20:01
mriedembriancline: hey, i was also talking with michael fork this morning, so you might want to circle back with him20:04
*** jerryz_ has joined #openstack-infra20:04
mtreinishttx: ok thanks, I've already updated the wiki20:04
*** jerryz has quit IRC20:07
*** melwitt has quit IRC20:08
*** melwitt1 has joined #openstack-infra20:08
clarkbfungi: I think I see the problem20:09
*** adalbas has quit IRC20:09
*** rfolco has quit IRC20:09
fungiclarkb: i'm on the edge of my seat20:10
*** atiwari has joined #openstack-infra20:10
*** dolphm has joined #openstack-infra20:10
*** dstanek_afk has joined #openstack-infra20:10
*** dstanek_afk has quit IRC20:10
clarkbfungi: getting a link together20:11
clarkbfungi: http://git.openstack.org/cgit/openstack-infra/pypi-mirror/tree/pypi_mirror/cmd/run_mirror.py#n214 branch processing there coupled with http://git.openstack.org/cgit/openstack-dev/pbr/tree/tools/integration.sh#n116 means we are not testing using the version of the repo put in place by devstack-gate20:13
sdaguebriancline: back now.20:13
clarkbfungi: which is fail sauce20:13
clarkbfungi: and I think is going to take some unravelling to fix. I am starving though, going to grab some food then tackle that20:13
fungioh, that's... okay20:13
clarkbfungi: ya20:13
*** jerryz has joined #openstack-infra20:13
fungiright20:13
sdagueso, honestly, a starting point would probably be an account with some small quota on it for us to play around. Because current CI is all openstack clouds, so we'd have to write some automation against softlayer to make it fit our model20:14
*** sarob has joined #openstack-infra20:14
*** ^d is now known as ^demon|lunch20:14
clarkbcurrent idea in my head is maybe add and argument to run-mirror that just runs in the cwd rather than doing branch manipulation20:14
sdaguedprince: I was talking that through with maurosr this morning on -qa20:14
clarkbfungi: but I am open to other ideas20:14
sdagueI think there is a complicated lock step we have to do, he's off working on the order.20:15
fungiclarkb: i'll need to ponder that20:15
sdaguethere was some confusion about how zuul did refs and why we have to do things a certain way20:15
fungiclarkb: but sounds like an acceptable alternative20:15
*** denis_makogon has joined #openstack-infra20:16
*** julim has quit IRC20:16
dprincesdague: okay. well hopefully that is progress. I looked over the logs myself and wasn't quite sure why it wasn't working20:16
*** jerryz_ has quit IRC20:17
jeblairclarkb, fungi  on line 225 of run-mirror, there is 'if not self.args.no_update:'20:19
jeblairclarkb, fungi: once i unravel the double negative, i wonder if that would help us...20:19
*** sarob has quit IRC20:19
fungiha20:19
*** gyee has quit IRC20:19
jeblairfungi: so maybe add '--no-update' to http://git.openstack.org/cgit/openstack-dev/pbr/tree/tools/integration.sh#n116 ?20:20
jeblairperformance run of postgres-full in progress at: https://jenkins02.openstack.org/job/check-tempest-dsvm-postgres-full/799/20:22
*** adalbas has joined #openstack-infra20:23
fungiokay, finally all cleaned up from the manage-projects freakout. repositories on review, zuul and all the git servers appear operable20:24
*** senk has quit IRC20:25
*** senk has joined #openstack-infra20:25
*** Abhishe__ has quit IRC20:25
sdaguedprince: yeh, I haven't had time to dive into it. Honestly we need to get a few more folks that grok the upgrade pieces. It's a pretty skeleton crew of us on it right now20:27
brianclinemriedem: awesome, will do20:33
*** jcooley_ has quit IRC20:35
fungiclarkb: right, so that script was originally intended to get a requirements list from a branch of a project (or iterating through all branches) and update a mirror... trying to run it to gate a change to the list of requirements means that we need to stop retrieving that separately from git and use the one which gets set up for the test instead20:36
ArxCruzALL: When I ran tox -efull I'm getting this error on tempest: http://paste.openstack.org/show/54546/ does anyone know what's wrong? there's a lot of \xfd\x12\xc4\xe0\xb3)\x01@i@^t code there20:36
fungiArxCruz: it looks like discover might be breaking20:38
*** jcooley_ has joined #openstack-infra20:41
fungiclarkb: and looking closer, i think i agree with jeblair that --no-update seems to skip resetting to the branch20:41
*** dprince has quit IRC20:41
clarkbwill look after food20:41
openstackgerritJames E. Blair proposed a change to openstack-infra/config: Switch some infra jobs to use bare precise nodes  https://review.openstack.org/6034420:44
ArxCruzfungi: just pip install discover ? I already did it and the error remains :(20:44
SpamapSclarkb: https://review.openstack.org/#/c/60100/ .. I don't really know what to do to move this forward. recheck no bug? Investigate d2to1?20:46
openstackgerritA change was merged to openstack-infra/reviewstats: Add a bug statistics tool.  https://review.openstack.org/5890320:48
*** adalbas has quit IRC20:49
*** hashar has quit IRC20:49
*** senk has quit IRC20:50
*** ywu has joined #openstack-infra20:50
jeblairfungi, clarkb, mordred, sdague: https://jenkins02.openstack.org/job/check-tempest-dsvm-postgres-full/799/ took 35 minutes on a rackspace performance node (quick look at history suggest that job takes ~40mins on hpcloud and ~50 on rax standard +/- 5 mins)20:54
*** ^demon|lunch is now known as ^d20:55
fungiArxCruz: i'm thinking testr.run discover might be throwing an exception or something. just a guess but seems like that's exhibited similar behavior in the past20:56
openstackgerritJames E. Blair proposed a change to openstack-infra/config: Switch some infra jobs to use bare precise nodes  https://review.openstack.org/6034420:56
*** julim has joined #openstack-infra20:56
openstackgerritJames E. Blair proposed a change to openstack-infra/devstack-gate: Use the ephemeral disk on rackspace performance nodes  https://review.openstack.org/6034720:57
*** Abhishek_ has joined #openstack-infra20:57
*** rpodolyaka1 has left #openstack-infra20:58
openstackgerritJames E. Blair proposed a change to openstack-infra/devstack-gate: Use the ephemeral disk on rackspace performance nodes  https://review.openstack.org/6034720:59
*** jcooley_ has quit IRC20:59
*** datsun180b has quit IRC21:01
openstackgerritKhai Do proposed a change to openstack-infra/config: Setup builds for openstack Gerrit ver 2.8  https://review.openstack.org/6034821:03
jeblairclarkb, fungi: also, bare-precise nodes seem to be able to run nova-pep8: https://jenkins01.openstack.org/job/dev-nova-pep8/4/console21:03
*** melwitt1 is now known as melwitt21:03
clarkbjeblair: woot21:04
clarkbSpamapS: I am working on it now. Found at least one problem with the test that may have allowed the breakage through21:04
clarkbSpamapS: going to correct that then force d2to1 in I think21:05
SpamapSclarkb: oh cool thanks. Let me know if there is anything I can do.21:05
* SpamapS breaking stuff as usual21:05
*** lcestari has quit IRC21:07
*** dkranz has quit IRC21:08
*** senk has joined #openstack-infra21:08
*** krtaylor has quit IRC21:11
*** alcabrera has quit IRC21:11
openstackgerritClark Boylan proposed a change to openstack-dev/pbr: Build mirror using proper requirements.  https://review.openstack.org/6034921:12
openstackgerritClark Boylan proposed a change to openstack-dev/pbr: Force d2to1 into the integration test mirror.  https://review.openstack.org/6035021:12
clarkbI think the order on that is wrong, but I wanted to see what happens when d2to1 isn't forced into the mirror. I will rebase and flip the order if the test for 60349 fails21:12
*** senk has quit IRC21:13
*** senk has joined #openstack-infra21:13
jeblairclarkb: one of those is running on a rax performance node.  considering the run time is _so long_ i'm interested in the results there..  :)21:14
*** hennystupbebe has joined #openstack-infra21:14
jeblairclarkb: oh, which suggests to me:  i think you should write two more copies of those changes (with new change ids) in the other order and go ahead and push them up21:14
clarkbjeblair: will do21:15
jeblairclarkb: that way, both alternatives will be tested in approx 1 hour (instead of 2) and you can abandon whichever you don't like.21:15
openstackgerritKhai Do proposed a change to openstack-infra/config: Setup builds for openstack Gerrit ver 2.8  https://review.openstack.org/6034821:15
openstackgerritClark Boylan proposed a change to openstack-dev/pbr: Force d2to1 into the integration test mirror.  https://review.openstack.org/6035121:15
openstackgerritClark Boylan proposed a change to openstack-dev/pbr: Build mirror using proper requirements.  https://review.openstack.org/6035221:15
clarkbjeblair: ^ there we go21:16
*** hennystupbebe has quit IRC21:16
*** FallenPegasus has joined #openstack-infra21:16
clarkbSpamapS: ^ should get d2to1 sorted21:17
*** smarcet has left #openstack-infra21:17
jeblairclarkb: want to do one with both changes squashed?21:17
clarkbjeblair: sure21:18
*** FallenPegasus has quit IRC21:18
*** FallenPegasus has joined #openstack-infra21:19
*** dkranz has joined #openstack-infra21:19
openstackgerritClark Boylan proposed a change to openstack-dev/pbr: Build mirror using proper requirements.  https://review.openstack.org/6035421:20
clarkband squashed ^21:20
*** dolphm has quit IRC21:22
openstackgerritJames E. Blair proposed a change to openstack-infra/zuul: Remove unecessary debug message  https://review.openstack.org/6035521:22
*** gyee has joined #openstack-infra21:23
notmynamejeblair: clarkb: I need to tag swift-bench so it goes through the -infra process to get pushed to pypi. This is the first time this repo has been "released". Anything to watch out for?21:28
jeblairnotmyname: let's double check that openstackci has perms on pypi and the jobs are setup in zuul21:29
clarkbnotmyname: make sure you have tagged the correct commit before pushing it. And make sure that commit is upstream (so no accidental local commits getting tagged_21:29
notmynameclarkb: of course :-) that's no different than normal releases21:29
jeblairnotmyname: swift-bench lacks the necessary jenkins jobs for this to happen21:30
notmynamejeblair: ok, what needs to be done from my side?21:30
jeblairnotmyname: pypi looks good though21:30
jeblairnotmyname: i'll hack out a change real quick21:30
*** fifieldt has joined #openstack-infra21:32
jeblairnotmyname: does swift-bench have sphinx docs?21:32
notmynamejeblair: no it does not, right now21:33
*** FallenPegasus has quit IRC21:33
openstackgerritJames E. Blair proposed a change to openstack-infra/config: Run swift-bench build/release jobs  https://review.openstack.org/6035621:34
*** pabelanger has quit IRC21:34
jeblairnotmyname: the acls in gerrit also look good, so the only thing lacking should be that ^21:34
*** dkranz has quit IRC21:35
* clarkb reviews21:35
notmynamethanks21:35
jeblairack; i see an error21:35
clarkbhmm did I miss it?21:35
clarkbshame on me21:35
jeblair(which i think the test will catch)21:35
sdaguejeblair: awesome21:35
clarkbdo you need to include another set of templates?21:36
openstackgerritJames E. Blair proposed a change to openstack-infra/config: Run swift-bench build/release jobs  https://review.openstack.org/6035621:36
clarkbnevermind I see it now21:36
sdaguejeblair: so can we flip our rax quota to performance nodes?21:36
clarkbpython-swift-bench21:36
clarkbsdague: I think we have21:36
lifelessquestion: I have two new projects to make; is it better to wait for a git tree to be ready, or can we push one up once it's there?21:36
clarkbsdague: what we need to do is let rax run gate jobs21:36
jeblairsdague, clarkb: indeed we have; i expect most of the standard nodes will have aged out now21:36
clarkblifeless: if you can push one up via review then now is fine. If you don't want to bootstrap in gerrit I would wait21:37
jeblairsdague, clarkb: i think we want to get the use-ephemeral-disk change merged first; we're running with no swap right now.21:37
*** yamahata_ has joined #openstack-infra21:37
clarkbjeblair: that should be fine21:38
*** jswarren has quit IRC21:38
clarkbhpcloud doesn't give us swap but d-g sorts it out right?21:38
jeblairclarkb: yes, but only on hpcloud -- the change is to d-g to also do that on rax21:38
jeblairclarkb: cause they have different disk names :(21:38
* clarkb looks21:38
jeblairclarkb: https://review.openstack.org/#/c/60347/  (it's testing)21:39
clarkbjeblair: I think there is a tyop21:39
lifelessclarkb: it's the split out from nova of scheduler21:40
*** sdake has quit IRC21:40
lifelessclarkb: so we're looking at filter-tree stuff21:40
clarkblifeless: I would wait then21:40
openstackgerritA change was merged to openstack-infra/config: Run swift-bench build/release jobs  https://review.openstack.org/6035621:40
lifelessclarkb: for the api; probably a new tree for the client, as there is just marshalling code for that21:40
clarkbotherwise one of us needs to force push the contents into the project21:40
*** sdake has joined #openstack-infra21:40
*** miqui has quit IRC21:40
*** sdake has quit IRC21:40
*** sdake has joined #openstack-infra21:40
clarkband really there is little value if having an empty project then doing a force push21:40
notmynamejeblair: is the post-mirror-python33 right? we don't support py33 at all21:42
jeblairnotmyname: since other things may be updated i think it's good to update all the mirrors21:43
funginotmyname: doesn't matter. just kicks off another run of the python33 mirror job in case you did, and in case you pulled in different dependencies on py33 than another python rev21:43
jeblairlifeless: let me know if you want me to do the filter-branch21:44
lifelessjeblair: there is a nova volunteer to do it, may I point them at you for advice?21:44
lifelessjeblair: your advice could be 'let me do it' :)21:44
jeblairlifeless: sure :)21:44
lifelessthanks!21:44
*** dkranz has joined #openstack-infra21:47
sdaguejeblair: https://review.openstack.org/#/c/60318/ can yuo look back in there21:50
sdaguewhile the vars aren't used directly, they are passed to the template21:50
clarkbjeblair: did you see my comment on https://review.openstack.org/#/c/60347/21:51
clarkbsdague: we can check in the template if vars are defined before using them21:51
notmynamejeblair: thanks. does that mean it's ready to tag now?21:51
clarkband only dump data in there if the data is available21:51
jeblairsdague: i reverted the config change, you may want to rebase that on an un-revert21:52
clarkbnotmyname: I would wait until top of the hour (to make sure puppet applies that on zuul first)21:52
sdaguejeblair: will do21:52
*** gyee has quit IRC21:52
notmynameclarkb: ok. I wanted to push one more commit anyway. thanks21:52
jeblairnotmyname: looks like the config is in place, so you should be good to go, but let's use that extra commit you want to push to make sure it's building tarballs correctly21:53
notmynamejeblair: thanks21:53
jeblairnotmyname: after it merges, you should see a swift-bench dir at http://tarballs.openstack.org/21:54
jeblairnotmyname: if not: problem.21:54
notmynamejeblair: thanks. I'll check when it lands21:54
*** gyee has joined #openstack-infra21:55
*** pabelanger has joined #openstack-infra21:56
openstackgerritPranesh Pandurangan proposed a change to openstack-infra/config: Fixed typo in zuul layout file  https://review.openstack.org/6036321:56
*** atiwari has quit IRC21:56
*** mfer has quit IRC21:56
openstackgerritPranesh Pandurangan proposed a change to openstack-infra/config: Fix more typos in zuul layout file  https://review.openstack.org/6036521:57
*** SergeyLukjanov has quit IRC21:58
*** SergeyLukjanov has joined #openstack-infra21:59
fungii'm starting to think that experience contributing to existing openstack or stackforge projects should be a prerequisite for adding a project to stackforge22:01
fungi"adding your project to stackforge" should pretty much never be your first experience pushing a change to gerrit22:01
clarkb++22:01
sdaguejeblair: did that revert go through?22:01
jeblairsdague: yes22:01
clarkbbut being inclusive is a good thing too22:01
sdagueok, git rebase is getting all weird on me22:02
openstackgerritJames E. Blair proposed a change to openstack-infra/devstack-gate: Use the ephemeral disk on rackspace performance nodes  https://review.openstack.org/6034722:02
openstackgerritPranesh Pandurangan proposed a change to openstack-infra/config: Added new project, entropy  https://review.openstack.org/6013222:03
*** melwitt has quit IRC22:04
*** melwitt1 has joined #openstack-infra22:04
*** melwitt has joined #openstack-infra22:05
*** melwitt1 has quit IRC22:05
*** jamesmcarthur has quit IRC22:05
fungisdague: in what ways? is it finding repeated patterns and merging in the wrong places?22:06
fungifwiw i've had gnu patch do the same to me in the past22:07
openstackgerritJames E. Blair proposed a change to openstack-infra/zuul: Document zuul client  https://review.openstack.org/6036622:07
openstackgerritSean Dague proposed a change to openstack-infra/config: move to elastic-recheck web console in share  https://review.openstack.org/6031822:07
sdaguefungi: I think I got it22:07
sdagueI had to trigger another revert22:07
*** DennyZhang has joined #openstack-infra22:10
jeblairsdague: i'm still not excited by all the extra stuff ending up on static.o.o.. i'm confused about why the base class needs to write the config file?22:10
*** denis_makogon has quit IRC22:10
jeblairsdague: why not have the bot class write that?22:10
jeblairsdague: it doesn't look like we have a config file on static now22:10
*** denis_makogon has joined #openstack-infra22:10
clarkblooks like forcing d2to1 in is going to be required22:10
clarkbsince that is the case, I almost prefer the squashed commit22:11
*** jcooley_ has joined #openstack-infra22:12
jeblairclarkb: (looks like the two changes in that pair took ~same amount of time on a rax performance and hpcloud node)22:13
notmynamejeblair: how long should it take for swift-bench to show up in tarballs?22:13
openstackgerritA change was merged to openstack-infra/jeepyb: Update tripleo project name mappings  https://review.openstack.org/6027822:13
sdaguejeblair: ah, true, I'll flip that22:14
jeblairnotmyname: like a minute.  it's there now: http://tarballs.openstack.org/swift-bench/22:14
*** jbryce has quit IRC22:14
notmynamejeblair: great!22:14
notmynamejeblair: so that means' everything is good, right?22:14
openstackgerritPranesh Pandurangan proposed a change to openstack-infra/config: Added new project, entropy  https://review.openstack.org/6013222:15
jeblairnotmyname: i think so.  i'd say go for it!22:15
notmynamejeblair: great. thanks for your help (and clarkb) :-)22:15
jeblairnotmyname: np22:16
*** ruhe has quit IRC22:17
*** jcoufal has quit IRC22:17
openstackgerritSean Dague proposed a change to openstack-infra/config: move to elastic-recheck web console in share  https://review.openstack.org/6031822:18
*** jgrimm has quit IRC22:18
sdagueok, the only question I have in there is the include in ::bot22:18
sdagueso see if I got that right22:18
openstackgerritA change was merged to openstack-infra/zuul: Document zuul client  https://review.openstack.org/6036622:18
clarkbfungi: jeblair: btw after reading a bunch of run_mirror.py I think we should try limping along until wheels happen. Doing special fancy stuff for tarballs will be a pain since ideally we don't put the tarballs on the mirror instead we just have their dependencies in the mirror22:20
sdagueclarkb: when you get a chance, can you swing back around and see if I answered you here correctly - https://review.openstack.org/#/c/59822/ (it didn't trigger a reset on your -1 so you might not see it)22:21
clarkbsdague: looking22:22
fungiclarkb: right, one of the basic assumptions in that integration job is that our projects don't depend explicitly on more than one version of our other projects at a time, and declaring one thing with a requirements spec while actually using a different version via a tarball link breaks that very badly22:22
notmynamejeblair: ok, I've tagged and pushed the tag (signed of course) to gerrit22:23
*** elasticio has quit IRC22:23
jeblairnotmyname: you can see progress here http://status.openstack.org/zuul/22:23
notmynamethanks22:23
notmynameit takes an hour to push to a mirror?22:23
clarkbsdague: for __getattr__ isn't hit an instanec of Hit? if so you can just add __getitem__ to the Hit class22:24
jeblairnotmyname: https://pypi.python.org/pypi/swift-bench/1.0  looks like a thing22:24
fungiclarkb: i'd be in favor of force-merging an openstack/requirements change for dev-requirements.txt to up the oslo.messaging tarball rev, then follow with updates to the requirements files of affected projects22:24
sdagueclarkb: I suppose I could, I just wasn't treating hit like a dictionary :)22:24
clarkbsdague: for _hits, _hits is literally used nowhere else and is private. I don't see what it gives us22:24
sdagueclarkb: it's used in the face_discover22:25
funginotmyname: it doesn't take an hour to push it to the mirror, it takes an hour to regenerate, satisfy and download all dependencies throughout our project requirements and then update the mirror with the results22:25
notmynameok :-)22:25
jeblairnotmyname: that ^, but why 27 is 1 hour and 26 and 33 are 15 minutes is a good question...22:25
clarkbsdague: can you link to that? I don't see it22:25
sdagueoh actually, sorry you are right22:26
jeblairor at least 2.6 vs 2.7... seems strange to me.22:26
sdagueI went around on this a few iterations22:26
sdagueand confused myself22:26
fungijeblair: my guess is that 2.7 is working and 2.6 and 3.3 are spontaneously breaking but maybe not returning a nonzero result22:26
fungi(and so bailing early)22:26
clarkbsdague: both changes would be cleanup and the __getattr__ vs __getiem__ doesn't really bother me. Was just a thought22:26
jeblairfungi: of course! :)22:26
clarkbsdague: _hits concerned me more because it appeared to be dead code22:26
sdagueyep, you are right22:27
*** DennyZhang has quit IRC22:27
fungii'm looking to see if my theory on mirror26 failures holds any merit22:27
fungithe recent job history at https://review.openstack.org/#/c/60132/2 suggests it does actually return nonzero (most of the time)22:28
fungier, wrong paste buffer22:28
fungihttps://jenkins.openstack.org/job/post-mirror-python26/22:28
fungiactually, 2.6 seems to be consistently failing, 2.7 and 3.3 seem to be consistently passing22:29
*** thomasm_ has quit IRC22:29
funginope, the consistent failures for 2.6 appear to be 50-minute job timeouts22:30
fungier, 60-minute22:30
clarkbjeblair: fungi: I have abandoned the two pbr changes that are in the wrong order, that leaves us with the two changes in the proper order or the squashed change22:30
fungicool22:30
*** sarob has joined #openstack-infra22:31
*** changbl has quit IRC22:32
klrmnlifeless / clarkb: is there a way to get testr / subunit to collect and report the stdout and/or do something with —verbose?22:34
clarkbklrmn: I think lifeless left acomment showing you how to do it22:35
jeblairsdague: i'm not positive about that include.  i would think it should work, but clarkb did something else in the etherpad-lite module (expecting you to include etherpad_lite and include etherpad_line::foo22:35
openstackgerritA change was merged to openstack-infra/config: gerritbot: Watch stable/havana on puppet-openstack projects  https://review.openstack.org/6012722:35
clarkbklrmn: testr run --subunit | $SUBUNITPARSER one parser that comes with subunit is subunit2pyunit so you can do testr run --subunit | subunit2pyunit22:35
jeblairsdague: if no one has a definite answer on that, i'm okay trying it and fixing it if needed.22:36
openstackgerritSean Dague proposed a change to openstack-infra/elastic-recheck: implementation of FacetSet for client side nested facets  https://review.openstack.org/5982222:36
openstackgerritSean Dague proposed a change to openstack-infra/elastic-recheck: port check_success to facets  https://review.openstack.org/5983222:36
clarkbklrmn: is the desire just to have every test listed as it runs?22:36
klrmnclarkb: i have my resources telling me how long they take to set things up outputting to stdout22:36
sdagueclarkb: ok, I fixed both of your comments.22:37
klrmnclarkb: but it would be really nice to know that something is actually happening22:37
lifelessklrmn: hi22:37
*** SergeyLukjanov has quit IRC22:37
clarkbklrmn: do you want to see that data realtime or just look at it post run? (I really don't know why so many people want to watch data spew all over their terminals :P)22:37
lifelessklrmn: testr only processes things that are encapsulated in subunit; other content is passed through22:38
*** mriedem has quit IRC22:38
*** SergeyLukjanov has joined #openstack-infra22:38
*** rnirmal has quit IRC22:38
lifelessklrmn: it's up to the runner to do the encapsulation (thats what subunit.run does, for python unittests, for instance)22:38
openstackgerritSean Dague proposed a change to openstack-infra/config: move to elastic-recheck web console in share  https://review.openstack.org/6031822:38
*** adalbas has joined #openstack-infra22:39
sdaguejeblair: ok, purged that last left over bit in static.pp22:39
klrmnclarkb: when you have a test suite that takes hours to run, it's better to know that it got hung up an hour in than it is to wake up in the morning and discover it's still spinnging22:39
klrmnspinning even22:39
clarkbklrmn: test timeouts combat that problem22:40
jeblairclarkb: i wish timeouts worked for zuul22:41
clarkbjeblair: I think we can make them work by making threads daemon threads or by using the forceful timeout22:42
clarkbiirc zuul's test suite has non daemon threads so one of them dying doesn't kill everything else22:42
klrmnlifeless: i seem to be kinda slow at figuring out how to swap out the various parts of things22:42
*** SergeyLukjanov has quit IRC22:42
openstackgerritA change was merged to openstack-infra/config: Fix Savanna d-g jobs  https://review.openstack.org/6029622:43
lifelessklrmn: No, I suspect you're working in a corner that isn't super well serviced today22:44
lifelessklrmn: what do you want to see, ideally?22:44
*** adalbas has quit IRC22:44
klrmnlifeless: it should be noted that most of my tests use selenium and a UI, which introduces a *lot* of variable latency22:45
lifelessklrmn: are you collaborating with tim ansell's subunit runner for selenium ?22:45
*** yamahata_ has quit IRC22:45
klrmnlifeless: i'm looking at needing to profile my tests to figure out what the hell is taking so long22:45
klrmnlifeless: i had no idea that even existed, and also have no idea how difficult it would be to retrofit it into my project22:46
klrmnlifeless: google didn't bring me anything in searching for 'python selenium subunit22:47
lifelesshttps://github.com/mithro/runner-testharness.js22:48
*** openstackgerrit has quit IRC22:48
*** openstackgerrit has joined #openstack-infra22:48
openstackgerritSean Dague proposed a change to openstack-infra/elastic-recheck: implementation of FacetSet for client side nested facets  https://review.openstack.org/5982222:49
openstackgerritSean Dague proposed a change to openstack-infra/elastic-recheck: port check_success to facets  https://review.openstack.org/5983222:49
klrmnlifeless: i think i have sub-classed enough to use the TestLoader from testresources with SubUnitTestProgram (but if there's an example out there i should look at or use instead, this'd be a great time)22:49
clarkbjeblair: fungi: the pbr tools/integration.sh test fixes have all reported their check results back. The squash commit failed in the postgres test due to known bugs. The other two passed. I think we should pick one set and approve them22:49
fungiclarkb: do you have a preference for one?22:50
klrmnlifeless: it is unclear if i should be using SubunitTestRunner or some other library's TestRunner22:50
notmynameI like the graph at the top of http://status.openstack.org/elastic-recheck/22:50
lifelessklrmn: are your tests python code or js code?22:50
klrmnlifeless: python22:50
jeblairnotmyname: jog0 did that ^22:50
lifelessklrmn: then subunit.run should be fine22:51
clarkbfungi: slight preference for the squash as chances it hits more non deterministic failures are lower, but the two commits are more easiliy git bisectable22:51
jeblairclarkb: i also lean toward playing the odds and going with squash22:51
clarkbjeblair: wfm22:51
fungii'm not too concerned about being able to bisect between those two adjacent changes, really22:51
notmynamejog0: nice work. what about summing them together (in some fancy baysean statistics way) to give an overall patch failure rate?22:52
clarkbnotmyname: note for the summing that half of them are check and the other half gate. Don't want to add those two sets together22:52
klrmnlifeless: is there a way to use subunit.run with testresources.OptimisingTestSuite22:53
klrmnwithout subclassing SubunitTestProgram to replace the Loader sent to ._do_discovery?22:53
lifelessklrmn: yes, the load_tests hook22:54
lifelessbut http://bugs.python.org/issue16662 will interact with the obvious approach for that22:55
*** senk has quit IRC22:55
klrmnlifeless: there are hooks? in which project? where is the documentation?22:56
lifelessklrmn: upstream python22:56
fungiclarkb: reviewed and approved 60354 (hopefully that was the right one)22:56
*** adalbas has joined #openstack-infra22:56
fungiclarkb: seemed to be the combination of the other two at any rate22:56
openstackgerritSean Dague proposed a change to openstack-infra/config: move to elastic-recheck web console in share  https://review.openstack.org/6031822:57
clarkbfungi: that was the one, thanks22:57
*** sarob has quit IRC22:57
*** sarob has joined #openstack-infra22:58
*** sandy__ has quit IRC22:59
*** sarob_ has joined #openstack-infra23:01
*** sarob has quit IRC23:01
*** nati_uen_ has joined #openstack-infra23:04
clarkbsdague: https://review.openstack.org/#/c/60149/ didn't that break everything right around havana feature freeze?23:06
clarkbanteaya: ^23:06
*** nati_ueno has quit IRC23:07
sdagueclarkb: yes... but just because it added more load to the neutron q / db system23:07
*** rcleere has quit IRC23:07
sdagueso it all deadlocked23:07
*** oubiwann_ has quit IRC23:07
openstackgerritElizabeth Krumbach Joseph proposed a change to openstack-infra/config: Configure pretty formatting for README files  https://review.openstack.org/6037523:08
clarkbsdague: should we avoid enabling it on stable/havana then?23:08
sdagueclarkb: oh, yes, for sure23:08
clarkbI am semi ok with dealing with pain on master23:08
sdaguebut that's a master change23:08
* clarkb leaves a comment23:08
clarkbsdague: it is both23:09
sdagueoh23:09
sdagueyeh, definitely not havana23:09
*** SergeyLukjanov has joined #openstack-infra23:11
*** sandy__ has joined #openstack-infra23:12
*** mriedem has joined #openstack-infra23:13
*** jcooley_ has quit IRC23:14
*** jcoufal-mob has joined #openstack-infra23:16
clarkbsdague: source the .tox venv then do `python setup.py egg_info` and see if you get a better stacktrace23:16
clarkber23:16
clarkbwow23:16
clarkbsdague: ignroe me23:16
openstackgerritSean Dague proposed a change to openstack-infra/config: move to elastic-recheck web console in share  https://review.openstack.org/6031823:17
sdagueheh23:17
sdagueok23:17
*** SergeyLukjanov has quit IRC23:18
openstackgerritA change was merged to openstack-infra/config: Use launchpad prop to store link to the lp project  https://review.openstack.org/6032923:19
*** mattymo has quit IRC23:19
openstackgerritJames E. Blair proposed a change to openstack-infra/devstack-gate: Use the ephemeral disk on rackspace performance nodes  https://review.openstack.org/6034723:19
*** dolphm has joined #openstack-infra23:20
*** mattymo has joined #openstack-infra23:20
fungiheh... -n vs -z is definitely a world of difference23:22
jeblairfungi: yeah, but it didn't break hpcloud, which i don't understand.  thus the -x and WIP.23:23
openstackgerritlifeless proposed a change to openstack-infra/config: Add Gantt repositories.  https://review.openstack.org/6038023:23
*** melwitt has quit IRC23:23
* fungi nods. i'm curious to see what it spews now23:23
*** jcooley_ has joined #openstack-infra23:23
*** melwitt has joined #openstack-infra23:24
*** jcoufal-mob has quit IRC23:24
*** prad has quit IRC23:25
*** dolphm has quit IRC23:25
fungiprobably on everyone's mind, but i find it amusing that rackspace performance nodes are running as fast as hpcloud's legacy region, and hpcloud's new region runs as slow as non-performance rackspace23:25
*** dkranz has quit IRC23:25
fungiwondering whether those numbers will hold steady23:25
jeblairfungi: it had occurred to me.  that may in fact be the thing that prompted me to pick this up.  :)23:26
*** vkozhukalov has quit IRC23:26
fungii wondered23:26
*** sarob_ has quit IRC23:31
*** sarob has joined #openstack-infra23:31
*** krtaylor has joined #openstack-infra23:31
*** nati_uen_ has quit IRC23:34
fungimordred: we haven't reverted anything that https://review.openstack.org/58153 needs, right?23:34
*** nati_ueno has joined #openstack-infra23:34
*** sarob_ has joined #openstack-infra23:34
*** sarob_ has joined #openstack-infra23:35
jeblairI have squatted gantt on launchpad, pypi and #openstack-gantt on freenode23:35
jeblair(and python-ganttclient)23:35
fungiclearly i'm behind on my ml reading23:35
*** sarob has quit IRC23:35
clarkbgantt?23:36
jeblairfungi, clarkb: 23:26 < openstackgerrit> lifeless proposed a change to openstack-infra/config: Add Gantt repositories.  https://review.openstack.org/6038023:36
fungioho23:36
fungii should look up from my keyboard a little more often apparently23:36
lifelessjeblair: awesome, thank you!23:36
jeblairlifeless: np, much easier to do this earlier rather than later.  :)23:37
*** denis_makogon has quit IRC23:38
jeblairlifeless: has anyone given lauren or jonathan a heads up on this yet?23:40
openstackgerritA change was merged to openstack-infra/config: Add gate-manila-devstack job  https://review.openstack.org/5798723:40
lifelessjeblair: nup23:41
lifelessit's not well thought out, just a holding name for code23:41
lifelessif it is feasible long term, cool23:41
*** asalkeld has joined #openstack-infra23:41
notmynamethis is just getting silly https://review.openstack.org/#/c/60343/23:41
openstackgerritA change was merged to openstack-infra/config: Initial configuration for python-barbicanclient  https://review.openstack.org/6010123:42
*** jcooley_ has quit IRC23:42
jeblairlifeless: i'll send an email and cc you, russelb; any other nova folks?  we can let them know it's early days, but they still might be interested.23:42
lifelessjeblair: thats very kind; thank you. I was just about to offer to do it myself.23:43
*** jcooley_ has joined #openstack-infra23:43
lifelessjeblair: the name was put forward in the nova IRC meeting my morning, so I think nova are sufficiently notified23:44
asalkeldHi, how do I track down  a readthedocs.org upload? I got this in: https://github.com/openstack-infra/config/commit/55eac64f376b2f5dd73ac4e169f05617197d7e24 but still no docs23:46
asalkeldthe docs are getting built: http://docs-draft.openstack.org/24/58424/5/gate/gate-solum-docs/686f2c4/doc/build/html/23:46
openstackgerritKhai Do proposed a change to openstack-infra/config: Setup builds for openstack Gerrit ver 2.8  https://review.openstack.org/6034823:46
asalkeldbut nothing here: https://readthedocs.org/projects/solum/23:47
*** openstackgerrit has quit IRC23:47
*** openstackgerrit has joined #openstack-infra23:47
jeblairasalkeld: logs.openstack.org/SHORT_SHA where SHORT_SHA is the last 7 chars of the git commit you're interested in (note, it's the merge commit if that's applicable)23:47
asalkeldok23:47
clarkbI think it might be firsttwocharsofsha/entiresha now23:48
jeblairasalkeld: whoops ^ sorry23:48
jeblairthere's a git subcommand to do that automatically somewhere...23:49
asalkeldso: https://github.com/stackforge/solum/commit/ca1f3ccdbea8744a089197af6d802ac28b25d9e2 -> http://logs.openstack.org/ca1f3ccdbea8744a089197af6d802ac28b25d9e223:51
asalkeld40423:51
jeblairasalkeld: http://logs.openstack.org/ca/ca1f3ccdbea8744a089197af6d802ac28b25d9e2/23:51
asalkeldaaa23:52
jeblairasalkeld: sorry, it's the first 2 chars of the sha, then the sha23:52
jeblairasalkeld: and it looks like there was no run of the rtfd hook23:52
asalkeldyeah, only coverage23:52
lifelessjeblair: a while ago I naively pushed up a commit adding me to having root access on -infra machines.23:52
*** herndon has quit IRC23:52
lifelessjeblair: and you said ' hey we'd love to have you eventually, but this probably isn't the time for it yet'23:53
lifelessjeblair: which was entirely appropriate as I didn't want that :)23:53
jeblairlifeless: excellent, we almost never give root access to people who don't want it.  :)23:53
lifelessjeblair: however, I would like that at some point, to get wider coverage for ops things; could you, when you have a minute, let me know if there is anything I should be doing that I'm not, to head down that path23:54
*** geekinutah has left #openstack-infra23:54
jeblairlifeless: root is (theoretically) a subset of -core (though currently a complete subset in practice)23:55
*** jcooley_ has quit IRC23:55
sdaguejeblair: so, with any luck, https://review.openstack.org/#/c/60318/ is right now. Once that lands I'll build up a patch to reorder the graphs on the page by hits23:55
jeblairlifeless: i think the main thing to do would be to become familiar/comfortable with the infra/config repo and puppet changes there...23:55
lifelessjeblair: righto; so I have a parallel copy of that infra deployed, and review in that tree daily23:56
*** senk has joined #openstack-infra23:56
fungithat sounds fairly familiar23:56
lifelessI need to do a bunch of updates to my copy though :(23:56
fungilifeless: by the way, those were excellent docs you put together for deploying all the bits23:57
lifelessfungi: thank you23:57
fungii have steered several people to them already23:57

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