Tuesday, 2015-06-23

*** markvoelker has joined #openstack-infra00:00
*** r-daneel has quit IRC00:01
*** laughterwym has joined #openstack-infra00:02
*** mtanino has joined #openstack-infra00:03
tchaypois the data in the gitdm repo still used by anyone?00:03
openstackgerritMerged openstack-infra/project-config: Add lbaasv1 api job to the check queue, non-voting  https://review.openstack.org/18779100:05
*** xyang1 has quit IRC00:06
openstackgerritArkadiy Kraminsky proposed openstack-infra/puppet-zuul: add notification for existing pidfile  https://review.openstack.org/19444900:07
*** SumitNaiksatam has quit IRC00:09
*** yamamoto has joined #openstack-infra00:10
openstackgerritMerged openstack-infra/zuul: Don't send reports without any job  https://review.openstack.org/18838300:11
*** ashleighfarnham has quit IRC00:12
*** Longgeek has joined #openstack-infra00:13
*** fallenpe_ has quit IRC00:13
*** esmute__ has quit IRC00:13
*** asettle has joined #openstack-infra00:15
*** puranamr has joined #openstack-infra00:15
*** fallenpegasus has joined #openstack-infra00:17
openstackgerritKurt Taylor proposed openstack-infra/infra-specs: Host a CI monitoring dashboard service  https://review.openstack.org/19443700:19
*** claudiub has joined #openstack-infra00:19
ianwdstufft: if you have any thought on https://bugs.launchpad.net/devstack/+bug/1467569 i would be glad to hear it.  python 2.7.9 on fedora includes the rewheel stuff, which makes python pkg dependent on python-pip00:20
openstackLaunchpad bug 1467569 in devstack "f22: python-pip removal pull python too" [Undecided,New]00:20
ianwour old method of purging system python-pip packages therefore attempts to remove the world00:21
dstufftprobably can't do that anymore00:21
dstufftsince python-pip and python have circular dependencies00:22
*** abitha has quit IRC00:24
ianwdstufft: yeah, i got that far :)  is "python -m ensurepip" in some ways a replacement for get-pip.py?  or do we just have to use the system pip and that's it?00:25
ianwistr we got to this point as system packaged pip had issues, so we started grabbing the latest00:25
*** Swami has quit IRC00:25
dstufftensurepip and get-pip.py are different in that ensurepip uses the bundled pip while get-pip.py pulls the latest from PyPI00:26
ianwdstufft: so you'd think that leaving python-pip package alone, and just calling get-pip.py is probably the right thing?00:27
*** doude_ has joined #openstack-infra00:27
*** doude has quit IRC00:28
dstufftI forgot to update it with the latest release (oops :|) but there is a PyPA COPR for the latest pip00:28
dstufftI need to automate that still00:28
ianwhttps://git.openstack.org/cgit/openstack-dev/devstack/tree/tools/install_pip.sh <- the right thing for this to do00:28
*** jasondotstar has quit IRC00:29
*** jamielennox|away is now known as jamielennox00:29
dstufftianw: if I can figure out how to get some automation around it, the "Right" thing to do is probably: dnf copr enable pypa/pypa && yum update00:29
dstufftI guess I should probably figure out how to do a jenkins or get right with the idea of travis pushing releases00:31
openstackgerritKurt Taylor proposed openstack-infra/infra-specs: Host a CI monitoring dashboard service  https://review.openstack.org/19443700:32
*** esker has quit IRC00:32
*** resker has joined #openstack-infra00:36
*** ociuhandu has quit IRC00:38
*** puranamr has quit IRC00:38
*** zz_dimtruck is now known as dimtruck00:39
ianwdstufft: ok, is this a long way down your todo list?  i think we just need to skip this on f22 at this point.00:40
dstufftianw: I can manually release it now00:40
*** mriedem has quit IRC00:43
*** nelsnelson has quit IRC00:44
*** Longgeek has quit IRC00:46
*** fallenpegasus has quit IRC00:47
openstackgerritMatt Riedemann proposed openstack-infra/project-config: Use DEVSTACK_GATE_SETTINGS in tempest-dsvm-cells job definition  https://review.openstack.org/19441000:47
*** fallenpegasus has joined #openstack-infra00:49
ianwdstufft: i'm not sure it matters right now, it's just those periods where we hit a bug and quickly need to get the latest version to get ci working again00:50
dstufftianw: well I intended to make sure it was up to date00:50
dstufftI just forgot :)00:50
*** nihilifer has quit IRC00:51
*** rlandy has quit IRC00:51
*** yamamoto has quit IRC00:52
lifelessok https://review.openstack.org/#/c/189250/ is green again00:52
lifelessbut it needs https://review.openstack.org/#/c/194401/ reviewed00:53
*** DericHorn-HP has joined #openstack-infra00:53
clarkbthe eplite table encoding conversion is now documebted updtream and we should expect a longish downtime to make that change00:53
clarkbunless mysqlers know how to make it faster00:53
*** bradjones has quit IRC00:54
*** dims has quit IRC00:55
openstackgerritMerged openstack-infra/zuul: Check that empty reports aren't sent  https://review.openstack.org/19158800:56
*** dims has joined #openstack-infra00:57
*** bradjones has joined #openstack-infra00:59
*** bradjones has quit IRC00:59
*** bradjones has joined #openstack-infra00:59
*** puranamr has joined #openstack-infra00:59
*** resker has quit IRC01:00
*** yamamoto has joined #openstack-infra01:02
*** dboik has joined #openstack-infra01:02
*** puranamr has quit IRC01:04
dstufftianw: ok, they are building and will be live in a bit01:06
dstufftianw: https://copr.fedoraproject.org/coprs/pypa/pypa/builds/01:06
*** bradjones has quit IRC01:07
*** dboik has quit IRC01:07
*** esmute__ has joined #openstack-infra01:07
*** bradjones has joined #openstack-infra01:09
*** bradjones has quit IRC01:09
*** bradjones has joined #openstack-infra01:09
*** esker has joined #openstack-infra01:10
*** yamamoto has quit IRC01:10
*** esmute__ has quit IRC01:11
clarkbhttps://github.com/ether/etherpad-lite/wiki/How-to-use-Etherpad-Lite-with-MySQL documents the table conversion and seems to indicate the etherpad lite isn't going to do db migrations properly for us (at least not yet)01:14
*** esmute__ has joined #openstack-infra01:14
*** Longgeek has joined #openstack-infra01:15
*** jasondotstar has joined #openstack-infra01:15
*** Kennan has quit IRC01:16
*** Kennan has joined #openstack-infra01:16
*** esker has quit IRC01:17
*** stevemar has joined #openstack-infra01:19
*** cing has joined #openstack-infra01:21
*** patrickeast has quit IRC01:23
*** esmute__ has quit IRC01:24
*** sdake has joined #openstack-infra01:24
*** asselin_ has joined #openstack-infra01:26
*** esker has joined #openstack-infra01:27
*** sdake_ has joined #openstack-infra01:27
*** shardy has quit IRC01:27
*** shardy has joined #openstack-infra01:28
*** sdake has quit IRC01:31
*** yamahata has quit IRC01:33
*** larainema has joined #openstack-infra01:34
*** asselin__ has joined #openstack-infra01:36
*** esmute__ has joined #openstack-infra01:36
*** esker has quit IRC01:37
*** asselin_ has quit IRC01:39
*** alexpilotti has quit IRC01:40
*** esmute__ has quit IRC01:40
lifelessfungi: https://review.openstack.org/#/c/194401/  is what needs approved next01:42
*** SumitNaiksatam has joined #openstack-infra01:42
*** esmute__ has joined #openstack-infra01:44
*** otter768 has joined #openstack-infra01:44
*** asettle has quit IRC01:44
*** yamamoto has joined #openstack-infra01:46
*** esmute__ has quit IRC01:47
*** otter768 has quit IRC01:49
*** esmute__ has joined #openstack-infra01:49
*** julim has quit IRC01:53
*** sdake_ has quit IRC01:58
*** sqchen has joined #openstack-infra01:58
*** hdd has joined #openstack-infra01:58
*** sdake has joined #openstack-infra01:58
*** esmute__ has quit IRC01:58
*** esmute__ has joined #openstack-infra01:59
sqchenanybody here can help with configuing gate job for horizon?02:00
clarkbsqchen: usually best to just ask then we can chime in as able02:01
clarkblifeless: that particular change did fail a test and got a -1, not sure if related to change though02:02
*** unicell1 has quit IRC02:02
lifelessclarkb: its not related02:03
lifelessclarkb: flaky test in kilo02:03
lifelessclarkb: elastic-recheck spotted it and I've rechecked it02:03
clarkblifeless: as far as the change goes it looks like its just code move and we still install pbr to that venv02:03
clarkbso I don't grok the diff + commit message02:04
lifelessclarkb: see, thats why bash is a bad language02:04
lifelessclarkb: the variable scope is moved out of the function02:04
*** claudiub has quit IRC02:04
clarkboh is PIP_VIRTUAL_ENV a thing pip uses directly?02:04
clarkbaffecting the later pip runs02:05
lifelessclarkb: yup02:05
lifelessstate is nasty02:05
clarkbI have no +2 but gave you a +102:07
*** tlian has joined #openstack-infra02:07
*** sdake has quit IRC02:07
lifelessthank you!02:10
*** tjones1 has quit IRC02:10
*** patrickeast has joined #openstack-infra02:14
*** ddieterly has joined #openstack-infra02:16
*** jerryz has quit IRC02:20
openstackgerritKhai Do proposed stackforge/python-jenkins: WIP: support scriptText api to execute groovy scripts on the server  https://review.openstack.org/19446802:24
openstackgerritKhai Do proposed stackforge/python-jenkins: Support scriptText api to execute groovy scripts on the server  https://review.openstack.org/19446802:26
*** DericHorn-HP has quit IRC02:30
*** sqchen has quit IRC02:31
*** sqchen has joined #openstack-infra02:33
*** dims has quit IRC02:36
*** DericHorn-HP has joined #openstack-infra02:39
*** fallenpegasus has quit IRC02:42
*** fallenpegasus has joined #openstack-infra02:44
*** AzherKhan has joined #openstack-infra02:47
*** jasondotstar has quit IRC02:47
*** fallenpegasus has quit IRC02:47
openstackgerritliusheng proposed openstack/requirements: Bump oslo.messaging to greater than or equal to 1.11.0  https://review.openstack.org/19447502:48
*** sqchen has quit IRC02:51
*** asettle has joined #openstack-infra02:51
*** sqchen has joined #openstack-infra02:51
*** Longgeek has quit IRC02:55
*** juggler has quit IRC02:56
*** juggler has joined #openstack-infra02:57
*** DericHorn-HP has quit IRC02:57
*** hdd has quit IRC02:57
*** bdemers has joined #openstack-infra02:57
*** puranamr has joined #openstack-infra02:58
openstackgerritIan Wienand proposed openstack-infra/infra-specs: Spec for diskimage-builder image caching  https://review.openstack.org/19447702:59
ianwgreghaynes: ^ not sure about where to store the images, nodepool host seems fine but others probably disagree03:00
*** arxcruz has quit IRC03:02
*** puranamr has quit IRC03:05
*** puranamr has joined #openstack-infra03:05
openstackgerritIan Wienand proposed openstack-infra/infra-specs: Spec for diskimage-builder image caching  https://review.openstack.org/19447703:06
*** tlian has quit IRC03:16
*** tlian has joined #openstack-infra03:17
*** ivar-laz_ has joined #openstack-infra03:19
*** erlon has quit IRC03:20
*** jogo has joined #openstack-infra03:22
*** Longgeek has joined #openstack-infra03:23
*** ivar-lazzaro has quit IRC03:23
*** ctlaugh has quit IRC03:23
*** ivar-laz_ has quit IRC03:24
*** achanda has joined #openstack-infra03:25
*** sdake has joined #openstack-infra03:25
*** sdake_ has joined #openstack-infra03:27
*** prometheanfire has joined #openstack-infra03:27
prometheanfireso, about getting something into disk-image-builder...03:28
prometheanfireI'm generating stage4 images for making my openstack qcow2 files03:28
greghaynesianw: hrm, I wonder if we could use dib itself to cache the images...d03:29
prometheanfireI could publish them officially for gentoo probably03:29
* prometheanfire is infra there03:29
greghaynesprometheanfire: fun03:29
prometheanfireall that would be needed is to run the last half of this script I think, which might make my 'disk image builder thing' the easiest distro.  https://github.com/prometheanfire/gentoo-cloud-prep/blob/master/03-prep-that-image.sh03:30
prometheanfiregonna look into it by the 7th, hopefully have a review in by then03:30
greghaynesprometheanfire: yep, so that sould be super easy - check out root.d on http://docs.openstack.org/developer/diskimage-builder/developer/developing_elements.html#phase-subdirectories03:31
greghaynesprometheanfire: basically you just have to generate your base rootfs in a dir passed in by an environment var03:31
prometheanfirethen I can tell people to test openstack (non-voting) on gentoo, that's a stretch :P03:31
*** sdake has quit IRC03:31
*** sdake__ has joined #openstack-infra03:31
*** fedexo has joined #openstack-infra03:32
prometheanfireya, all my stuff would be outside chroot too, since that's done by catalyst on gentoo's infra03:32
greghaynesprometheanfire: heh, well I would be happy to accept a DIB test to gate on gentoo image builds passing03:32
greghaynesso, thats a start ;)03:32
*** bdemers has quit IRC03:32
*** bdemers has joined #openstack-infra03:33
*** bdemers has quit IRC03:34
prometheanfireoddly, I think that getting an 'offical' openstack gentoo image thing is going to be easier then getting an on gentoo's side03:34
prometheanfirethe stage4's are all in catalyst, so that meets with the reqs of gentoo-infra03:34
*** bdemers has joined #openstack-infra03:34
prometheanfirethe qcow2 stuff catalyst can't do that I know of03:34
prometheanfireso I can't make those official yet03:34
jeblairoregon/washington/new york peeps: there's a strong geomagnetic storm; you might be able to see the aurora.  http://www.swpc.noaa.gov/03:35
prometheanfirewish I was that far north :(03:35
davidlenwellyeah .. nothing from the bay area.. boo03:35
*** sdake_ has quit IRC03:35
greghayneswhen it eventally gets dark here, that is03:36
prometheanfirenothing in satx, but that's a given03:37
*** DericHorn-HP has joined #openstack-infra03:38
*** DericHorn-HP has quit IRC03:38
*** achanda has quit IRC03:39
tonybjhesketh: ping?03:39
greghaynesianw: so crazy idea maybe - What if we make some tooling that does preemptive caching for a set of elements and we run that for the set of DIB elements we are building for, we then rsync the cache dir up to a mirror location and also have a tool that will export the proper source-repository overrides to direct downloads at the mirror03:40
*** larainema has quit IRC03:40
greghaynesianw: should be doable with the information in elements already, and then we dont have to write out any new info on what files we want to mirror03:40
*** laughterwym has quit IRC03:40
prometheanfiregreghaynes: signed? don't like plaintext :P03:41
greghaynesprometheanfire: well, this isnt for public consumption03:41
jrolljeblair: too far north for the bay? :(03:42
prometheanfireas long as it doesn't traverse the public internet then fine, if it traverses the net...03:42
*** laughterwym has joined #openstack-infra03:43
*** asselin__ has quit IRC03:43
*** harlowja has quit IRC03:43
*** tlian has quit IRC03:44
*** otter768 has joined #openstack-infra03:45
openstackgerritEmilien Macchi proposed openstack-infra/project-config: puppet: experiment upgrade job with beaker  https://review.openstack.org/19429303:45
jheskethtonyb: pong03:46
*** coolsvap|away is now known as coolsvap03:47
tonybDo you have time to help me work out why files that should be published to eavesdrop.o.o aren't?03:47
tonybjhesketh: my current theories are a) they are but there is a ServerAlias hiding them ; or b) they're not being uploaded beacuse the scp publisher needs a '-r' flag03:48
tonybjhesketh: of course I could be on crack and I really doubt that option 2 is it ....03:48
*** spzala has quit IRC03:48
jheskethtonyb: do you have a link to the publish job(s)?03:50
greghaynesianw: replied on spec, sorry if its a bit hard to parse03:50
tonybjhesketh: I can. (assuming I'm looking at the right thing03:50
*** otter768 has quit IRC03:50
tonybjhesketh: http://git.openstack.org/cgit/openstack-infra/project-config/tree/jenkins/jobs/infra.yaml#n11903:51
*** ddieterly has quit IRC03:51
tonybjhesketh: in review: https://review.openstack.org/#/c/193123/ we added a 'calendars' directory that we need along with *.ics and *.html03:52
jheskethtonyb: err sorry, I meant in jenkins03:52
tonybjhesketh: Umm I don't know how to get that.03:52
tonybjhesketh: I can link to the output from the Jenkins gate job.03:52
tonybjhesketh: http://logs.openstack.org/23/193123/1/gate/gate-irc-meetings-tox-ical/b4dad66/03:53
tonybbut I'm sure that's not what you mean03:53
jheskethah right, it's in the post queue03:53
tonybjhesketh: feel free to tell me how to find the thing you *actually* want03:53
* tonyb likes to learn03:54
*** abitha has joined #openstack-infra03:54
jheskethtonyb: so I'm not sure if there's a better way, but because the output isn't published anywhere (well other than the final scp location) you have to visit each jenkins0[1-7].o.o to find recent irc-meetings-publish runs03:54
jheskethwhich I'm doing currently03:55
tonybjhesketh: okay your start at j1, I'll start a j7 and meet you in the middle03:55
jheskeththis looks to be most recent: https://jenkins03.openstack.org/job/irc-meetings-publish/11/console03:56
jheskethit appears to have worked03:57
jheskethmaybe I'll use my root powers for the first time to take a look!03:57
tonybjhesketh: it uploaded the files but not the dir03:57
tonybif I read the last 5 lines correctly03:57
jheskethtonyb: right, but the index.html also still links to meetings, not calendars03:57
*** larainema has joined #openstack-infra03:58
tonybjhesketh: No.03:58
tonybjhesketh: and from the URL you gave me: 2015-06-22 04:08:32.477 |   /home/jenkins/workspace/irc-meetings-publish$ /home/jenkins/workspace/irc-meetings-publish/.tox/ical/bin/yaml2ical -y meetings -i output/calendars/ -f"03:59
*** Longgeek has quit IRC04:00
*** Longgeek has joined #openstack-infra04:00
jheskethtonyb: that's correct isn't it?04:02
*** harlowja_at_home has joined #openstack-infra04:02
tonybYeah they build job seems okay and the generated index.html points to the correct calendars/*.ics but that dir isn't accessable from here04:03
*** david-lyle_ has quit IRC04:03
tonybjhesketh: I may be talking in circles04:04
*** jamespage has quit IRC04:05
*** gnuoy has quit IRC04:05
jheskethtonyb: the index.html on eavesdrop has a mod time the same as the upload04:05
jheskethbut  contains 'meetings' in the url04:05
jheskethtonyb: actually so does http://logs.openstack.org/23/193123/1/gate/gate-irc-meetings-tox-ical/b4dad66/site-index.html04:05
*** gnuoy has joined #openstack-infra04:06
*** jamespage has joined #openstack-infra04:06
jheskethtonyb: actually, does the job use a different template to that in the irc-meetings folder04:06
* jhesketh thinks it does04:06
tonybjhesketh: hang on.  Where are you seeing meetings in the URL?04:06
tonybjhesketh: I don't see that here at all04:06
jheskethtonyb: nevermind, I'm looking in the wrong spot04:07
jheskethtonyb: on the bright side, it just looks like the calendars aren't being uploaded?04:08
jheskethie the folder04:08
tonybjhesketh: okay.04:08
tonybwow so my option 2 was correct04:08
tonybjhesketh: can I add 'recurse: yes' or somethign similar to the yaml I posed befoer?04:08
tonybjhesketh: where can I see the options that the scp publisher takes?04:09
jheskethtonyb: likely, yes04:09
jheskethtonyb: look for other jobs or at the jenkins plugin04:09
* jhesketh looks too04:09
*** unicell has joined #openstack-infra04:09
tonybjhesketh: keep-hierarchy: true ?04:11
*** dimtruck is now known as zz_dimtruck04:11
jheskethtonyb: I /think/ that'll make it upload 'output' into a folder called 'output'04:13
tonybjhesketh: okay04:13
jheskethtonyb: it's odd though, because I assume the job was working before and uploading the icals?04:14
*** mtanino has quit IRC04:15
tonybjhesketh: no it was uploading *the* ical (that has all the meetings in it) but not the individual icals04:15
tonybjhesketh: that's a 'recent' change04:15
*** jogo has quit IRC04:16
tonybjhesketh: we /could/ just put the icals in the / with the index.html but that's gross04:16
jheskethtonyb: so I think we need to do something like http://git.openstack.org/cgit/openstack-infra/project-config/tree/jenkins/jobs/macros.yaml#n66204:18
jheskethwhere we need the ** to make it recursive04:18
*** abitha has quit IRC04:18
jheskethtonyb: and I think we do need keep-hierarchy as it'll apply to the ** part (looking at the output of a dsvm job)04:19
*** mestery has joined #openstack-infra04:19
tonybjhesketh: I was just going to ask you that after relising that the scp plugin wan't something "we" write04:19
*** kaisers has quit IRC04:19
tonybjhesketh: So how can I test this?04:19
*** jogo has joined #openstack-infra04:20
jheskethtonyb: with difficulty... it'd probably involve setting up a jenkins to test on04:21
tonybjhesketh: Umm yeah, Well I suppose I'll just write up the problem and suggest a patch and see where the review leads me.04:21
jheskethtonyb: we could make the gate job scp its results to the static log server like it used to before we started uploading the swift04:21
*** kaisers has joined #openstack-infra04:22
jheskethtonyb: fungi setup  the scp dest so he probably knows more about it :-)04:22
*** sdake__ has quit IRC04:24
*** ildikov has quit IRC04:24
*** ParsectiX has joined #openstack-infra04:25
tonybjhesketh: okay.  We'll see what he says overnight.04:25
*** Longgeek has quit IRC04:29
*** sqchen has left #openstack-infra04:32
*** harlowja_at_home has quit IRC04:32
*** Longgeek has joined #openstack-infra04:32
openstackgerritTony Breeds proposed openstack-infra/project-config: Have irc-meetings-publish also publish directories  https://review.openstack.org/19448804:35
tonybjhesketh: ^^04:35
*** puranamr has quit IRC04:38
ianwgreghaynes: i was thinking to minimize the downloading every time04:39
*** husanu has quit IRC04:39
*** husanu has joined #openstack-infra04:41
greghaynesianw: yes, we should be able to do the same amount of downloads either way - im thinking just do the source-repositories parts, not a full dib run04:41
*** hdd has joined #openstack-infra04:42
*** bdemers has quit IRC04:44
*** bdemers has joined #openstack-infra04:45
ianwgreghaynes: ok, source-repositories is a new one for me ... do you mean we port builds to use that element to get their qcows?04:46
*** yfried__ has quit IRC04:48
*** linkedinyou has quit IRC04:49
*** mestery has quit IRC04:49
*** husanu has quit IRC04:50
*** shardy has quit IRC04:51
*** shardy has joined #openstack-infra04:51
*** husanu has joined #openstack-infra04:52
*** ddieterly has joined #openstack-infra04:52
greghaynesianw: yes. right now theres not a consistent way to override images that are used for the distro elements, so im thinking we make that consistent, then just use the same tooling dib uses to download images to mirror and override them04:56
greghaynesand in theory source-repositories should be able to do that for us04:56
*** ddieterly has quit IRC04:56
ianwok, yeah at the moment things use "cache-url" to grab the url they construct in some way04:57
ianwi guess also having the images mirrored in one place only partially solves the issue -- it stops us hitting upstream mirrors, but if the mirror is on rax we still have the connectivity issue to hp04:58
ianwso like pypi we probably need per region mirrors04:58
*** puranamr has joined #openstack-infra04:58
greghaynesyea, and then well want some smart logic for how were overriding url's rather than a static list of resource -> env var we need to set04:59
greghaynesIll play with making the distro elements use source-repositories so I can try and remember what the issues are around that05:00
*** asettle has quit IRC05:00
*** bhunter71 has quit IRC05:00
*** husanu has quit IRC05:01
ianwgreghaynes: so do you think we should repurpose the pypi mirror into a more generic "devel" or "ci" mirror05:01
*** unicell1 has joined #openstack-infra05:02
*** husanu has joined #openstack-infra05:03
ianwmaybe we want some other type of mirror too, and it seems 95% of the infrastructure in terms of puppet setup, etc, are the same05:03
greghayneshrm, it does seem like all of that should be in the same place, it is going to scale out in the same way05:03
*** puranamr has quit IRC05:03
*** sabeen2 has joined #openstack-infra05:03
*** unicell has quit IRC05:04
*** sabeen has quit IRC05:05
*** krtaylor has quit IRC05:05
*** flwang has quit IRC05:05
*** armax has joined #openstack-infra05:08
*** zz_dimtruck is now known as dimtruck05:11
*** boris-42 has quit IRC05:12
*** ParsectiX has quit IRC05:18
*** woodster_ has quit IRC05:21
*** deepakcs has joined #openstack-infra05:22
*** dimtruck is now known as zz_dimtruck05:22
*** alexsyip has joined #openstack-infra05:23
*** patrickeast has quit IRC05:23
*** patrickeast has joined #openstack-infra05:24
*** hdd has quit IRC05:24
*** krtaylor has joined #openstack-infra05:24
*** sabeen2 has quit IRC05:27
*** alexsyip has quit IRC05:28
*** nikil89_ has joined #openstack-infra05:31
*** fawadkhaliq has joined #openstack-infra05:38
*** enikanorov2 has joined #openstack-infra05:39
*** abregman has joined #openstack-infra05:39
*** ildikov has joined #openstack-infra05:41
*** larainema has quit IRC05:41
*** larainema has joined #openstack-infra05:41
* larainema 05:41
jheskethtonyb: thanks for that patch, small nit that I'd like fixed if possible though please05:45
*** otter768 has joined #openstack-infra05:46
openstackgerritWayne Warren proposed openstack-infra/infra-specs: Add JJB DSL Roadmap spec.  https://review.openstack.org/19449705:46
openstackgerritTony Breeds proposed openstack-infra/project-config: Have irc-meetings-publish also publish directories  https://review.openstack.org/19448805:47
*** fedexo has quit IRC05:47
openstackgerritDavid Lenwell proposed openstack-infra/infra-specs: Spec to host refstack.org in infra  https://review.openstack.org/18820705:48
*** otter768 has quit IRC05:51
*** yfried__ has joined #openstack-infra05:52
*** larainema has quit IRC05:53
*** ddieterly has joined #openstack-infra05:53
*** salv-orlando has joined #openstack-infra05:54
*** mrmartin has joined #openstack-infra05:54
*** stevemar has quit IRC05:54
*** stevemar has joined #openstack-infra05:54
*** larainema_ has joined #openstack-infra05:55
*** larainema has joined #openstack-infra05:55
*** patrickeast has quit IRC05:56
*** husanu has quit IRC05:56
*** salv-orlando has quit IRC05:56
*** husanu has joined #openstack-infra05:58
*** ddieterly has quit IRC05:58
*** salv-orlando has joined #openstack-infra05:58
tonybjhesketh: fixed05:59
tonybjhesketh: Well changed I wasn't game to remove it altogether.  Only because you seemed unsure.05:59
jheskethtonyb: heh, I was a little, but lgtm06:00
openstackgerritDavid Lenwell proposed openstack-infra/infra-specs: Spec to host refstack.org in infra  https://review.openstack.org/18820706:00
tonybjhesketh: \o/06:00
tonybif this lands I will have earned by "infra badge"06:00
*** laughterwym_ has joined #openstack-infra06:03
*** laughterwym has quit IRC06:05
openstackgerritJoshua Hesketh proposed openstack-infra/system-config: Symlink yaml2ical calendar folder to apache dir  https://review.openstack.org/19450406:08
jheskethtonyb: I noticed ^ would also be needed06:08
tonybjhesketh: Ahh okay that makes sense.06:10
*** stevemar2 has joined #openstack-infra06:10
tonybjhesketh: Thanks for digging more into it.06:11
*** Kennan has quit IRC06:11
*** Kennan2 has joined #openstack-infra06:11
jheskethno trouble :-)06:12
*** stevemar has quit IRC06:12
*** yamada-h has joined #openstack-infra06:12
*** _nadya_ has joined #openstack-infra06:14
*** mohankumar has joined #openstack-infra06:15
*** hichihara has quit IRC06:15
*** stevemar2 is now known as stevemar06:17
openstackgerritIan Wienand proposed openstack-infra/infra-specs: Spec for diskimage-builder image caching  https://review.openstack.org/19447706:19
*** yamada-h has quit IRC06:20
*** yamada-h has joined #openstack-infra06:20
*** yamada-h has quit IRC06:22
*** nikil89_ is now known as naggappan06:23
*** mrmartin has quit IRC06:24
*** mrmartin has joined #openstack-infra06:26
*** mrunge has joined #openstack-infra06:31
*** ParsectiX has joined #openstack-infra06:34
*** dtantsur|afk is now known as dtantsur06:36
*** k4n0 has joined #openstack-infra06:37
*** ihrachyshka has joined #openstack-infra06:38
*** mrmartin has quit IRC06:39
*** markvoelker has quit IRC06:40
tchaypolooking at the graphite graphs for the stable jobs, some of them only ever seem to have had success06:41
*** kushal has joined #openstack-infra06:46
tchaypofor instance, stats.zuul.pipeline.periodic-stable.job.periodic-ceilometer-python27-havana06:47
*** armax has quit IRC06:49
*** jtomasek has joined #openstack-infra06:49
*** fallenpegasus has joined #openstack-infra06:51
*** mmedvede has quit IRC06:53
Kennan2hi tchaypo06:53
*** Kennan2 is now known as Kennan06:53
*** kushal has quit IRC06:54
*** ddieterly has joined #openstack-infra06:54
*** fawadkhaliq has quit IRC06:55
*** mpaolino has joined #openstack-infra06:55
Kennanlifeless there?06:55
*** fawadkhaliq has joined #openstack-infra06:55
*** Longgeek_ has joined #openstack-infra06:56
*** kushal has joined #openstack-infra06:56
*** ddieterly has quit IRC06:58
*** Longgeek has quit IRC06:59
*** yfried__ has quit IRC07:01
*** Ala has joined #openstack-infra07:01
lifelessKennan: hi07:02
*** macjack has quit IRC07:03
Kennanhi lifeless do you know anything about build atomic image in infra?07:03
Kennanthis https://fedorapeople.org/groups/ have some images, Our atomic image is old, I am trying to find ways to how to build that07:04
lifelessKennan: I don't, sorry.07:05
lifelessKennan: I know they exist is about all07:05
KennanOK lifeless anyone here know about fedora atomic image?07:05
*** yfried__ has joined #openstack-infra07:06
openstackgerritSpencer Krum proposed openstack-infra/puppet-iptables: Support side-loading iptables rules  https://review.openstack.org/19384707:07
lifelessKennan: some of the fedora folk might07:08
*** laughterwym has joined #openstack-infra07:09
*** jlanoux has joined #openstack-infra07:10
*** shardy has quit IRC07:11
*** shardy has joined #openstack-infra07:11
*** laughterwym_ has quit IRC07:12
Kennanyes, I am not sure where can find them. lifeless :)07:12
*** sshnaidm has joined #openstack-infra07:13
openstackgerritSpencer Krum proposed openstack-infra/project-config: Create a puppet module for os_client_config  https://review.openstack.org/18889507:13
*** sergmelikyan has joined #openstack-infra07:13
*** HeOS has quit IRC07:13
lifelessKennan: perhaps in a fedora channel ?07:18
*** samuelBartel has joined #openstack-infra07:19
openstackgerritSpencer Krum proposed openstack-infra/system-config: Alpha-sort modules.env  https://review.openstack.org/19124507:20
openstackgerritSpencer Krum proposed openstack-infra/project-config: Add beaker-localhost and beaker-nodepool hypervisors  https://review.openstack.org/18575407:21
openstackgerritFatih Degirmenci proposed openstack-infra/jenkins-job-builder: Add Google Cloud Storage Plugin support  https://review.openstack.org/19239707:22
openstackgerritSpencer Krum proposed openstack-infra/system-config: Bring in module to manage update-alternatives  https://review.openstack.org/17155607:22
openstackgerritSpencer Krum proposed openstack-infra/system-config: Add puppet-community/archive module  https://review.openstack.org/16456907:22
openstackgerritSpencer Krum proposed openstack-infra/system-config: Load in nanliu/git module  https://review.openstack.org/16435507:22
*** bexelbie has joined #openstack-infra07:24
openstackgerritSteve Kowalik proposed openstack-infra/project-config: Sprinke in Horizon translations for Zanata  https://review.openstack.org/19452507:25
openstackgerritSteve Kowalik proposed openstack-infra/project-config: Sprinkle in Horizon translations for Zanata  https://review.openstack.org/19452507:26
*** boris-42 has joined #openstack-infra07:32
openstackgerritSpencer Krum proposed openstack-infra/system-config: Set the openstack_project::users_install class to run in 'setup' stage  https://review.openstack.org/16104807:32
*** cpallares has quit IRC07:33
openstackgerritSpencer Krum proposed openstack-infra/system-config: Use Puppet Stages to setup users first  https://review.openstack.org/16104807:33
StevenKAJaeger: I absolutely love your work WRT horizon and .pot files, thanks!07:34
nibalizerStevenK: what is your middle name07:34
nibalizeri just realized we have the same 2 letter intials07:34
*** arxcruz has joined #openstack-infra07:35
StevenKnibalizer: "Adam"07:36
*** ihrachyshka has quit IRC07:37
*** cpallares has joined #openstack-infra07:37
*** salv-orl_ has joined #openstack-infra07:37
*** shardy has quit IRC07:38
*** shardy has joined #openstack-infra07:38
*** fawadkhaliq has quit IRC07:39
*** salv-orlando has quit IRC07:40
mrungeKennan, please ask at #fedora-devel07:40
*** bexelbie_ has joined #openstack-infra07:41
*** dizquierdo has joined #openstack-infra07:41
*** markvoelker has joined #openstack-infra07:41
Kennanyes mrunge Thanks asked not got response07:41
mrungeKennan, it's a bit early for most folks there, please be patient, stay in the channel07:42
mrungeyou also might ask in #fedora-cloud07:42
mrungeKennan, if you didn't got an answer within the next 2-3 hours, please ping me07:42
openstackgerritSpencer Krum proposed openstack-infra/project-config: Create a puppet module for os_client_config  https://review.openstack.org/18889507:42
*** stevemar has quit IRC07:43
*** mpavone has joined #openstack-infra07:43
KennanOK mrunge Thanks07:43
*** fawadkhaliq has joined #openstack-infra07:44
*** Hal has joined #openstack-infra07:44
*** Hal is now known as Guest6026907:44
*** devvesa has joined #openstack-infra07:45
*** markvoelker has quit IRC07:46
*** linkedinyou has joined #openstack-infra07:46
*** otter768 has joined #openstack-infra07:47
*** markus_z has joined #openstack-infra07:47
*** chlong has quit IRC07:48
*** claudiub has joined #openstack-infra07:51
*** otter768 has quit IRC07:51
*** masayukig has quit IRC07:52
*** linkedinyou has quit IRC07:53
*** persia has quit IRC07:53
*** persia has joined #openstack-infra07:53
*** persia has quit IRC07:53
*** persia has joined #openstack-infra07:53
*** flepied1 has quit IRC07:54
*** masayukig has joined #openstack-infra07:54
*** cjwei has quit IRC07:55
*** ddieterly has joined #openstack-infra07:55
ttxhmm, request timeouts on review.openstack.org ?07:58
*** salv-orl_ has quit IRC07:58
ttxhmm, it's back, ignore me07:59
*** ddieterly has quit IRC07:59
*** salv-orlando has joined #openstack-infra08:00
*** claudiub has quit IRC08:00
mrungettx, it's painful slow right now.08:00
*** dtantsur is now known as dtantsur|brb08:01
dtantsur|brbyeah, the same for me08:01
*** mpaolino has quit IRC08:01
*** nfedotov has joined #openstack-infra08:01
*** mpaolino has joined #openstack-infra08:03
*** rossella_s has joined #openstack-infra08:04
*** u_glide has joined #openstack-infra08:05
*** claudiub has joined #openstack-infra08:05
u_glideIs review.openstack.org alive? :)08:06
AJaegerStevenK: glad to hear!08:06
*** claudiub|2 has joined #openstack-infra08:07
*** sshnaidm has quit IRC08:07
nfedotovI can not open https://review.openstack.org/  either08:08
nfedotovOh. it works now08:09
*** hashar has joined #openstack-infra08:09
AJaegerStevenK: I was really happy to get all these patches merged yesterday.08:10
*** eezhova has quit IRC08:10
*** claudiub has quit IRC08:10
StevenKAJaeger: Yeah, quite! \o/08:10
*** claudiub has joined #openstack-infra08:14
*** claudiub|2 has quit IRC08:16
*** fhubik has joined #openstack-infra08:16
lifelesswe've had multiple reports now of folk missing the 12.0.0a0 tag in their nova repos, accessed over https://git.openstack.org/08:18
*** sergmelikyan has quit IRC08:19
*** sshnaidm has joined #openstack-infra08:19
lifelessI'm wondering if someone could check that its present in all the mirror hosts08:19
lifelesssince a bad copy there would propogate the missingness08:20
clarkbanyone can check, http://git0X.openstack.org:8080/cgit/openstack/nova looks like 01 is missing it08:21
lifelessclarkb: thanks!08:21
clarkba force replication may resolve it but tags are likely special so that may not be the case08:22
lifelessclarkb: fetch --tags will fix it08:22
*** eezhova has joined #openstack-infra08:22
clarkbbut I need to pass out now because twins are sleeping but ya 01 is the only off node08:23
lifelessjhesketh: ready to exercise some root ?08:23
*** sshnaidm has quit IRC08:24
*** fhubik is now known as fhubik_afk08:26
*** hashar has quit IRC08:28
*** hashar has joined #openstack-infra08:29
anteayaclarkb: night clark, happy sleep08:29
*** ihrachyshka has joined #openstack-infra08:30
*** hashar has quit IRC08:30
*** rossella_s has quit IRC08:30
*** rossella_s has joined #openstack-infra08:30
*** e0ne has joined #openstack-infra08:32
*** sergmelikyan has joined #openstack-infra08:32
*** larainema has quit IRC08:33
*** larainema_ is now known as larainema08:33
*** madong_ has joined #openstack-infra08:33
*** flepied has joined #openstack-infra08:34
*** flepied1 has joined #openstack-infra08:34
*** fhubik_afk is now known as fhubik08:35
*** linkedinyou has joined #openstack-infra08:35
*** hashar has joined #openstack-infra08:36
*** linkedinyou has quit IRC08:36
*** HeOS has joined #openstack-infra08:37
*** e0ne is now known as e0ne_08:38
*** flepied has quit IRC08:39
*** yfried__ is now known as yfried|afk08:43
*** e0ne_ has quit IRC08:44
*** HeOS has quit IRC08:44
*** HeOS has joined #openstack-infra08:45
*** e0ne has joined #openstack-infra08:48
*** dguerri` is now known as dguerri08:48
openstackgerritJames Page proposed openstack/os-testr: Misc Python 3 compatibility fixes  https://review.openstack.org/19455108:49
*** yfried|afk is now known as yfried__08:49
*** fallenpegasus has quit IRC08:50
*** pelix has joined #openstack-infra08:51
openstackgerritSteve Kowalik proposed openstack-infra/project-config: Sprinkle in Horizon translations for Zanata  https://review.openstack.org/19452508:51
*** dtantsur|brb is now known as dtantsur08:51
*** hashar has quit IRC08:53
*** laughterwym has quit IRC08:54
*** shardy_ has joined #openstack-infra08:54
*** shardy has quit IRC08:55
*** fabo has quit IRC08:55
*** ddieterly has joined #openstack-infra08:55
*** hashar has joined #openstack-infra08:55
*** fabo has joined #openstack-infra08:57
*** cdent has joined #openstack-infra08:58
*** shardy_ has quit IRC09:00
*** ddieterly has quit IRC09:00
*** HeOS has quit IRC09:01
*** shardy has joined #openstack-infra09:01
jheskethlifeless: sure, what's up09:01
*** HeOS has joined #openstack-infra09:02
lifelessjhesketh: we have some tags that weren't mirrored properly09:02
lifelesson git0109:02
*** chen12 has joined #openstack-infra09:03
*** mpavone has quit IRC09:03
*** park__heijlong has joined #openstack-infra09:04
lifelessjhesketh: so needs a log into the box and mirror the tags across; separately may want investigation into how etc etc09:04
*** e0ne is now known as e0ne_09:06
*** dtantsur is now known as dtantsur|brb09:07
ttxseems to be git01-specific09:08
lifelessalready been analysed :)09:08
ttxis git01 actually even part of the rotation ?09:08
lifelesswould need to check haproxy's live config to know09:09
openstackgerritSpencer Krum proposed openstack-infra/system-config: Alpha-sort modules.env  https://review.openstack.org/19124509:09
lifelessbut I would assume so since we got onto this from missing tag reports09:09
*** juggler has quit IRC09:10
*** juggler_ has joined #openstack-infra09:10
jheskethshould be part of the haxproxy09:10
ttxweird that it still gets regular commits :) Oh well09:11
*** yfried__ is now known as yfried|afk09:11
*** e0ne_ has quit IRC09:11
*** yfried|afk is now known as yfried__09:11
openstackgerritSpencer Krum proposed openstack-infra/system-config: Use Puppet Stages to setup users first  https://review.openstack.org/16104809:11
openstackgerritJens Rosenboom proposed openstack-infra/system-config: Change doc link to the parsed version for gerrit  https://review.openstack.org/19172409:13
*** sergmelikyan has quit IRC09:14
*** yfried has joined #openstack-infra09:15
*** e0ne has joined #openstack-infra09:15
openstackgerrityolanda.robla proposed openstack-infra/nodepool: Update nodepool to use secure.conf file  https://review.openstack.org/18976209:16
*** fhubik is now known as fhubik_afk09:17
*** markvoelker has joined #openstack-infra09:17
jheskethlifeless: err, so those tags all appear to be there to me09:18
*** yamahata has joined #openstack-infra09:18
jheskethis it possible they have gotten in sync since you filed the bug?09:18
*** u_glide has quit IRC09:19
*** yfried__ has quit IRC09:19
openstackgerrityolanda.robla proposed openstack-infra/puppet-nodepool: Add creation of secure.conf file  https://review.openstack.org/18832509:19
lifelessjhesketh: unless clarkb decided he had time... or our friendly russian sysadmin09:20
*** fhubik_afk is now known as fhubik09:20
jheskethwould have thought they'd weigh in...09:21
jheskethlifeless: can you check if the tag(s) are there for you please09:21
*** markvoelker has quit IRC09:22
AJaegerShit, just sent internal email to openstack-infra ;(09:22
*** afazekas has joined #openstack-infra09:22
*** jcoufal has joined #openstack-infra09:23
*** afazekas has quit IRC09:23
*** afazekas has joined #openstack-infra09:24
openstackgerrityolanda.robla proposed openstack-infra/puppet-jeepyb: Separate the fetch remotes log dir and file  https://review.openstack.org/19218709:24
lifelessAJaeger: whups09:24
openstackgerritThomas Bechtold proposed openstack-dev/pbr: Use string for 'OS_TEST_TIMEOUT' default  https://review.openstack.org/19456409:24
*** mancdaz has quit IRC09:25
*** mancdaz has joined #openstack-infra09:25
*** ghostpl has joined #openstack-infra09:26
lifelessjhesketh: so I see it in a fresh tag09:27
lifelessof horizon from 0109:27
lifelesschen12: how did you test for thepresence/absence of the tags?09:27
openstackgerritFabien Boucher proposed openstack-infra/puppet-openstackci: Add generic zuul manifests  https://review.openstack.org/17597009:28
*** fhubik is now known as fhubik_afk09:30
*** alexpilotti has joined #openstack-infra09:32
*** Alexandra_ has joined #openstack-infra09:34
*** rguillebert has joined #openstack-infra09:35
*** Alexandra_ has joined #openstack-infra09:35
*** jasondotstar has joined #openstack-infra09:37
openstackgerritRuslan Kamaldinov proposed openstack/requirements: Add yaql  https://review.openstack.org/19318009:38
Kennanmrunge: there?09:40
*** nihilifer_ has joined #openstack-infra09:40
mrungeKennan, yupp09:41
openstackgerritFabio Porcedda proposed openstack-infra/git-review: Add "-p"/"--pull" option to merge a change on the local branch  https://review.openstack.org/19409409:41
KennanI got one echo, he gave me an example https://github.com/cgwalters/fedora-atomic-work/blob/master/fedora-atomic-base.json09:41
Kennanbut I still not know what can follow for build new image for atomic09:41
*** fhubik_afk is now known as fhubik09:42
Kennanmrunge the http://www.projectatomic.io/blog/2014/08/build-your-own-atomic-centos-or-fedora/ have old information09:42
Kennanwhich is diffcult to follow09:42
Kennando you know any better guide or link for ? mrunge:09:42
*** laughterwym has joined #openstack-infra09:43
mrungeKennan, actually, I thought, you'd be asking in #fedora-cloud09:45
KiallJust me or is gerrt down again?09:45
mrungeas that's the working group producing the images09:45
Kennanok let me try if possible to get right echo there. thanks mrunge09:46
mrungeKennan, have your seen this? http://community.redhat.com/blog/2014/12/fedora-21-fedora-goes-atomic/09:46
openstackgerrityogananth subramanian proposed openstack/diskimage-builder: Allow using local image for ubuntu  https://review.openstack.org/19457409:46
*** nihilifer_ has quit IRC09:46
KiallYep - Gerrit UI is down again - http://paste.openstack.org/show/316268/ - pings, but UI won't load09:46
*** nihilifer has joined #openstack-infra09:47
*** tosky has joined #openstack-infra09:47
mrungeKennan, which basically means, you could get an image from here: https://getfedora.org/cloud/download/09:47
Kennanhi mrunge:09:47
mrungeKennan, sorry: this one: https://getfedora.org/cloud/download/atomic.html09:47
*** otter768 has joined #openstack-infra09:48
*** marzif has joined #openstack-infra09:48
ekarlso+1 Kiall gerrit is dead :p09:48
Kennanright now, we built our image on atomic(as fedora 21 atomci image not updated Since Dec 2014 for long time)09:48
*** k4n0 has quit IRC09:48
KennanI found our new image (intergrated docker verison 1.5) is old09:48
KennanI want to intergated with 1.6 version09:48
Kennanso I need to build a new image, that;s the reason I asked about how to build atomic image09:49
Kennanmrunge: I think we can not directly use fedora cloud image(21) which is old09:49
HeOSHello, folks. review.openstack.org is unavailable now. :(09:49
Kiallyea, seems none of the infra folks are online / reading at the moment09:50
*** nihilifer_ has joined #openstack-infra09:50
Kiall"Proof" - since sdague didn't believe there was an issue yesterday :P http://paste.openstack.org/show/316271/09:50
*** sergmelikyan has joined #openstack-infra09:50
*** oomichi has joined #openstack-infra09:51
*** kashyap has joined #openstack-infra09:51
mrungeKennan, http://www.projectatomic.io/community/ mentions an irc channel: #atomic09:51
johnthetubaguy+1 for gerrit being a bit dead right now09:51
*** shardy has quit IRC09:52
johnthetubaguyah, good point, its the slight gap and aus has dinner, and the US are still sleeping09:52
Kennanok mrunge: thanks09:52
*** shardy has joined #openstack-infra09:52
*** otter768 has quit IRC09:53
kashyapThanks folks, Gerrit seems to be back up!09:53
*** nihilifer has quit IRC09:54
*** oomichi has quit IRC09:54
*** vlaza has joined #openstack-infra09:55
*** ddieterly has joined #openstack-infra09:56
*** oomichi has joined #openstack-infra09:57
StevenKAJaeger: Can you peer at https://review.openstack.org/#/c/194525/ ? I'm concerned that gate-infra-puppet-apply-centos6 passed in patchset 2, and failed with patchset 309:58
StevenKAJaeger: (And the difference between the two patchesets is two characters of whitespace)09:58
*** chen12 has quit IRC10:00
*** nihilifer_ has quit IRC10:00
*** chen12 has joined #openstack-infra10:00
*** ddieterly has quit IRC10:01
*** Guest60269 has quit IRC10:01
*** ghostpl has quit IRC10:03
wznoinskfungi: when you're here, please let me know if I can pass your details to my infra team in regards the email delays we've spoke about shortly a couple of days ago (Too many recipients received this hour)10:05
*** dims has joined #openstack-infra10:06
*** ghostpl has joined #openstack-infra10:07
*** e0ne is now known as e0ne_10:10
*** yamamoto has quit IRC10:11
openstackgerritSerg Melikyan proposed openstack/requirements: Add python-muranoclient to requirements  https://review.openstack.org/17720410:12
*** fawadkhaliq has quit IRC10:15
openstackgerritThomas Bechtold proposed openstack-dev/pbr: Use string for 'OS_TEST_TIMEOUT' default  https://review.openstack.org/19456410:17
*** cing has quit IRC10:20
*** e0ne_ has quit IRC10:20
sdagueKiall: I was one of the people that reported the outage yesterday :)10:21
Kiallsdague: Oh, confused so :)10:23
AJaegerStevenK: WE have on the centos6 image a known git timeout issue. If there's a failure with 30mins, it is the timeout. Run "recheck"10:24
*** yfried is now known as yfried|afk10:24
*** serg_melikyan has joined #openstack-infra10:26
*** Mmike has quit IRC10:28
*** Mmike has joined #openstack-infra10:28
StevenKAJaeger: Ah ha! 10:28
*** dtantsur|brb is now known as dtantsur10:29
*** salv-orlando has quit IRC10:29
*** sergmelikyan has quit IRC10:30
AJaegerStevenK: But better update as suggested ;)10:30
*** oomichi has quit IRC10:31
openstackgerritSteve Kowalik proposed openstack-infra/project-config: Sprinkle in Horizon translations for Zanata  https://review.openstack.org/19452510:31
*** yfried|afk is now known as yfried10:34
*** hdd has joined #openstack-infra10:38
*** mrmartin has joined #openstack-infra10:39
*** ociuhandu has joined #openstack-infra10:40
openstackgerritWitold Bedyk proposed openstack-infra/project-config: added project stackforge/monasca-log-api  https://review.openstack.org/19459510:42
*** kashyap has left #openstack-infra10:45
*** serg_melikyan has quit IRC10:48
wznoinskinfra - is there a gerrit group you could add my CI accounts so they don't get emails about comments on each change the comment on? I can't find such option in gerrit settings... correct me if I'm wrong10:49
*** yamamoto has joined #openstack-infra10:50
*** mrunge has quit IRC10:53
*** kashyap has joined #openstack-infra10:53
*** AzherKhan has quit IRC10:53
*** mrunge has joined #openstack-infra10:53
*** yfried is now known as yfried|afk10:54
*** yamamoto has quit IRC10:55
*** coolsvap is now known as coolsvap|away10:55
*** yamahata has quit IRC10:55
*** yfried|afk is now known as yfried10:56
*** ddieterly has joined #openstack-infra10:57
kashyapCan anyone provide a pointer to a git repo of all the Jenkins' jobs?11:00
*** jamielennox is now known as jamielennox|away11:00
*** mohankumar has quit IRC11:01
*** sergmelikyan has joined #openstack-infra11:01
* kashyap scours in git.openstack.org11:01
*** ddieterly has quit IRC11:01
AJaegerkashyap: project-config repo, directory jenkins/jobs11:01
*** radez is now known as radez_g0n311:01
AJaegerkashyap: see also http://docs.openstack.org/infra/11:02
kashyapAJaeger: Yep, thank you.11:02
*** fhubik is now known as fhubik_afk11:02
*** yamamoto has joined #openstack-infra11:03
AJaegerwznoinsk: yes, see http://docs.openstack.org/infra/system-config/third_party.html#creating-a-service-account11:04
*** markvoelker has joined #openstack-infra11:06
*** yfried is now known as yfried|afk11:09
*** e0ne has joined #openstack-infra11:09
*** markvoelker has quit IRC11:11
*** Alexandra_ has quit IRC11:11
*** mpavone has joined #openstack-infra11:11
*** ZZelle has quit IRC11:14
*** ZZelle has joined #openstack-infra11:14
* kashyap wonders if folks here use 'lp-tools' to manage Launchpad bugs from CLI. 11:14
*** yfried|afk is now known as yfried11:15
samueldmqI've a quick question regarding shade11:18
samueldmqsince I found this https://github.com/openstack-infra/shade/blob/18cf7c2ccbddfdf296b93f0f36a7355a0d8c1e36/shade/tests/functional/hooks/post_test_hook.sh#L1811:18
samueldmqI suppose the funtional tests are supposed to be run against a devstack11:19
*** radez_g0n3 is now known as radez11:19
*** baoli has joined #openstack-infra11:23
openstackgerritMerged openstack-infra/project-config: Allow Merge Commits to fuel-docs repo  https://review.openstack.org/19433811:23
*** yamamoto has quit IRC11:24
*** abregman has quit IRC11:25
*** abregman has joined #openstack-infra11:25
*** shardy has quit IRC11:25
*** shardy has joined #openstack-infra11:26
wznoinskAJaeger: thanks, as far as I understand this group's settings tho is that 'my CI account's comments will not be emailed to others' and what I am looking for to not to get emails when someone else is adding a comment, btw. my Intel Networking CI is already is in that group11:26
*** hdd has quit IRC11:26
*** yamamoto has joined #openstack-infra11:27
*** yamamoto_ has joined #openstack-infra11:28
*** HeOS has quit IRC11:28
*** HeOS has joined #openstack-infra11:28
*** jcoufal has quit IRC11:29
*** HeOS_ has joined #openstack-infra11:29
*** salv-orlando has joined #openstack-infra11:30
*** hdd has joined #openstack-infra11:30
*** radez is now known as radez_g0n311:31
*** yamamoto has quit IRC11:32
*** salv-orlando has quit IRC11:36
*** hdd has quit IRC11:36
*** yamamoto_ has quit IRC11:36
*** e0ne is now known as e0ne_11:36
*** ociuhandu has quit IRC11:39
openstackgerritMerged openstack-infra/release-tools: Update milestone.sh to support new versioning  https://review.openstack.org/19406811:41
AJaegerwznoinsk: In that case I can't help. Others might be - or join the third-party CI meeting and ask others11:41
*** marcusvrn has joined #openstack-infra11:41
mordredsamueldmq: yes - they do run against a devstack11:41
mordredsamueldmq: however, they should also be runnable against a live cloud account too (if you don't mind the tests creating and deleting things on that account)11:42
wznoinskAJaeger: sure, thanks for the try anyways, I'll speak to them yes11:42
fungiwznoinsk: sure, they can reach me at fungi@yuggoth.org if they have any questions about our mailserver logs11:42
*** cburroughs has joined #openstack-infra11:43
samueldmqmordred: nice thanks, I'll create a VM in our internal cloud and try it out11:43
*** yamamoto has joined #openstack-infra11:43
samueldmqmordred: just against a VM for now, need to ask people here to run against the live cloud11:44
samueldmqmordred: thanks11:44
*** mrmartin has quit IRC11:44
*** hashar has quit IRC11:45
openstackgerritFabio Porcedda proposed openstack-infra/git-review: Add "-p"/"--pull" option to merge a change on the local branch  https://review.openstack.org/19409411:46
*** Hal has joined #openstack-infra11:46
*** sdake has joined #openstack-infra11:46
*** e0ne_ has quit IRC11:46
*** Hal is now known as Guest3991211:46
fungiwznoinsk: as for getting gerrit to not e-mail you updated for changes on which you've commented, there's a long-standing bug report/feature request asking for that capability https://code.google.com/p/gerrit/issues/detail?id=109711:47
*** sdake_ has joined #openstack-infra11:47
*** shardy has quit IRC11:47
*** otter768 has joined #openstack-infra11:49
*** dprince has joined #openstack-infra11:49
*** otter768 has quit IRC11:50
*** otter768 has joined #openstack-infra11:50
openstackgerritMerged openstack-infra/project-config: Run sudo /w -H in akanda post_test_hook  https://review.openstack.org/19428211:50
*** sdake has quit IRC11:50
wznoinskfungi: thanks, I was trying to get the number of emails to our gateway down as we're hitting a /h limit but it looks like the only option is to increase the limit on our side11:52
wznoinskmy infra would like a contact person details to contact about that, could I give your email address?11:52
*** salv-orlando has joined #openstack-infra11:54
*** shardy has joined #openstack-infra11:54
*** sdake_ has quit IRC11:54
*** husanu has quit IRC11:56
*** fawadkhaliq has joined #openstack-infra11:57
*** husanu has joined #openstack-infra11:58
*** ddieterly has joined #openstack-infra11:58
*** ParsectiX has quit IRC11:58
*** bexelbie has quit IRC11:59
*** bexelbie_ has quit IRC11:59
*** ociuhandu has joined #openstack-infra11:59
wznoinskfungi: see your reply above, ignore pls11:59
*** mpavone has quit IRC12:00
*** otter768 has quit IRC12:00
*** hashar has joined #openstack-infra12:01
*** zz_dimtruck is now known as dimtruck12:02
ZZellefungi, hi12:02
*** ddieterly has quit IRC12:02
*** weshay has joined #openstack-infra12:03
*** bdemers has quit IRC12:04
*** markvoelker has joined #openstack-infra12:05
*** mtanino has joined #openstack-infra12:05
*** jasondotstar has quit IRC12:06
*** fhubik_afk is now known as fhubik12:06
fungiZZelle: hey there, what do you need?12:06
*** ParsectiX has joined #openstack-infra12:07
*** bexelbie_ has joined #openstack-infra12:07
*** bexelbie has joined #openstack-infra12:07
ZZellefungi, in neutron tox.ini you define the env variable VIRTUAL_ENV (https://github.com/openstack/neutron/blob/master/tox.ini#L7)12:07
*** Ala has quit IRC12:08
*** kushal has quit IRC12:08
fungii do?12:08
ZZellefungi, but lots of neutron jobs overload setenv without defining VIRTUAL_ENV (for example: https://github.com/openstack/neutron/blob/master/tox.ini#L36-L41)12:08
ZZellefungi, oups, i was mordred :s12:09
*** kushal has joined #openstack-infra12:09
mordredZZelle: wow, you were me? that must be scary12:10
Clinti'm seeing double12:10
*** dimtruck is now known as zz_dimtruck12:11
*** gordc has joined #openstack-infra12:12
*** pc_m has joined #openstack-infra12:12
*** shardy has quit IRC12:12
ZZellemordred, i am wondering if we need to ensure VIRTUAL_ENV is set for every job12:13
*** sergmelikyan has quit IRC12:13
ZZelle(tox job)12:13
fungiZZelle: anyway, if you're referring to implications of that situation on https://review.openstack.org/193262 then yes, that probably does cause some of those tox envs to miss the pymysql dburi override. if those envs don't run things which make use of oslo.db opportunistic backend detection then it's probably not important anyway12:13
*** Ala has joined #openstack-infra12:13
fungiespecially if oslo.db tags a new release very soon12:14
*** kushal has quit IRC12:14
ttxfungi: you probably already noticed that, but Gerrit seems to have... variable performance this morning.12:15
*** claudiub has quit IRC12:15
mordredfungi: yup. I was just about to say what ttx said12:15
*** dguerri is now known as dguerri`12:15
*** dguerri` is now known as dguerri12:15
fungittx: yes, the webui stopped responding a few times yesterday too12:15
ttxsometimes it's fast, sometimes it hangs.12:16
fungiwe've been so far unable to find any indication in the logs of the cause. i'm starting to wonder if there's a qos issue in rackspace12:16
ttxcurrent status: hanging.12:16
fungiyep, when this happens, apache is logging no new requests in its access log and no errors in its error log12:17
fungiso the https connections aren't making it to the server, i think12:17
ttxcan't tie that to any CPU or network usage peak either12:17
*** shardy has joined #openstack-infra12:18
StevenKHuh, mtr hits the machine, but nc -vz is hanging12:18
fungii'll get a ticket opened with rackspace shortly12:18
fungiStevenK: yeah, weird thing is the ssh api is still responsive too12:19
fungibut connections aren't reaching apache at all12:19
StevenKsteven@undermined:~% nc -vz review.openstack.org https12:19
StevenKConnection to review.openstack.org 443 port [tcp/https] succeeded!12:19
StevenKSo it did work, but nc only does SYN / SYNACK / ACK, but it took a while12:19
*** dizquierdo is now known as dizquierdo_afk12:19
fungiand now it's logging connections again12:20
*** bauzas has joined #openstack-infra12:20
fungiso you may have simply caught it going back to normal there12:20
StevenKRight at the end12:21
*** sergmelikyan has joined #openstack-infra12:21
StevenKIt could well be rate limiting for 443 traffic, like you say.12:21
StevenKI don't like it when machines act differently for ICMP versus TCP traffic. :-/12:21
fungiin this case it's acting differently for ssh vs https traffic even12:22
StevenKYeah, which is even stranger.12:23
*** claudiub has joined #openstack-infra12:24
*** woodster_ has joined #openstack-infra12:25
* AJaeger managed to briefly comment on a patch using the webui. Seems I have to try gerrty soon...12:25
*** mpavone has joined #openstack-infra12:26
*** aysyanne has joined #openstack-infra12:26
*** agentle has joined #openstack-infra12:26
ttxcould be some weird filtering12:27
*** kgiusti has joined #openstack-infra12:27
ttxor ids12:27
fungiis it hung for everyone else again12:27
*** sergmelikyan has quit IRC12:27
ttxit is hung for me12:28
ttxgot briefly on hit12:28
ttxone hit 2 min ago12:28
fungii've got tcpdump running and my https requests aren't reaching the interface at all12:28
*** sergmelikyan has joined #openstack-infra12:28
*** bswartz has quit IRC12:28
ttxyeah, likely not Gerrit fault... this time.12:28
fungiokay, now it just got through12:28
*** chlong has joined #openstack-infra12:31
*** kushal has joined #openstack-infra12:31
*** fawadkhaliq has quit IRC12:32
*** sergmelikyan has quit IRC12:32
*** yfried is now known as yfried|afk12:33
*** yfried|afk is now known as yfried12:34
*** sergmelikyan has joined #openstack-infra12:35
*** agentle has quit IRC12:35
*** bknudson has quit IRC12:35
*** agentle has joined #openstack-infra12:36
*** mrmartin has joined #openstack-infra12:36
*** fhubik is now known as fhubik_afk12:37
*** fhubik_afk is now known as fhubik12:37
*** dizquierdo_afk has quit IRC12:37
*** e0ne has joined #openstack-infra12:37
openstackgerritFabien Boucher proposed openstack-infra/puppet-openstackci: Add generic zuul manifests  https://review.openstack.org/17597012:37
*** ldnunes has joined #openstack-infra12:39
fungii've opened rackspace ticket 150623-ord-0000826 about this12:39
*** dizquierdo_afk has joined #openstack-infra12:39
*** radez_g0n3 is now known as radez12:39
*** boris-42 has quit IRC12:42
*** mwagner_afk has joined #openstack-infra12:43
*** yamamoto has quit IRC12:45
wznoinskfungi: is there a set of IP's that send emails for review.openstack.org?12:46
*** rfolco has joined #openstack-infra12:47
*** sshnaidm has joined #openstack-infra12:49
*** amotoki has joined #openstack-infra12:51
*** coolsvap|away is now known as coolsvap12:51
*** salv-orlando has quit IRC12:52
*** linkedinyou has joined #openstack-infra12:56
clarkbwznoinsk just the A and AAAA recorded IPs in dns12:57
* clarkb notes new traffic on gerrit list about the packfile error issue thibg12:57
*** ddieterly has joined #openstack-infra12:58
*** rlandy has joined #openstack-infra13:00
*** pc_m has quit IRC13:01
*** coolsvap is now known as coolsvap|away13:01
*** xyang1 has joined #openstack-infra13:02
*** mpavone has quit IRC13:02
*** u_glide has joined #openstack-infra13:02
*** ddieterly has quit IRC13:03
*** dkranz has joined #openstack-infra13:04
*** tlian has joined #openstack-infra13:06
*** shardy has quit IRC13:07
*** shardy has joined #openstack-infra13:07
*** shardy has quit IRC13:07
*** sergmelikyan has quit IRC13:07
ZZellemordred, could you provide some references why we have VIRTUAL_ENV in tox.ini (https://github.com/openstack/neutron/blob/master/tox.ini#L7)13:08
*** coolsvap|away is now known as coolsvap13:08
*** peristeri has joined #openstack-infra13:08
*** sshnaidm has quit IRC13:08
*** sshnaidm has joined #openstack-infra13:09
mordredZZelle: I do not remember why we set that. if clarkb is still around, he might13:17
*** changbl_ has quit IRC13:17
ZZellemordred, clarkb, it seems related to https://github.com/openstack/oslo-incubator/blob/2ca2b4b49367d6033bdd2c5e47d17392e7aba318/openstack/common/setup.py#L80-L8713:17
ZZellemordred, clarkb, which has been removed but perhaps there are other usages13:17
*** jasondotstar has joined #openstack-infra13:17
*** kushal has quit IRC13:17
*** bswartz has joined #openstack-infra13:17
mordredSO - I think it may have been working aroudn something .. normally virtualenv itself sets that env var13:17
clarkbI would check git log -p tox.ini13:17
*** sergmelikyan has joined #openstack-infra13:17
*** deepakcs has quit IRC13:17
*** openstack has quit IRC13:17
*** openstack has joined #openstack-infra13:19
dstufftmore zeros == better than13:20
mordredttx: I believe this might be by design and I think I remember talking about it months ago13:20
mordredbut I do not remember the details ... do you fungi ?13:20
dstufftI think it is by design, I recall it being talked about too13:20
*** shardy has quit IRC13:20
ttxwell, if that's by design I would have welcomed a heads-up13:20
fungittx: pbr considers to be the normalized form of that tag13:20
mordredttx: I thought you were in the conversation13:20
ttxespecially since we've been tweaking versions lately13:20
dstufftI think it was so the beta marker was in it's own "part"13:20
*** esker has joined #openstack-infra13:20
*** shardy has joined #openstack-infra13:20
dstufftto make it easier to parse13:20
dstufftor something13:20
ttxI don't remember pushing 2015.1.0.0rc1 things though13:21
clarkbits based on an interpretation of pep440 iirc13:21
clarkbso the normal form is more correct because 440 says so13:21
fungittx: it would only be since the pbr 0.11 release13:21
dstufftpep 440 is more than happy to have 3.0.0b113:21
dstufftthe extra 0 is a pbr thing13:21
clarkbsee my comment13:21
ttxI just don't think generating versions that are different from the one in the tag to be sane13:21
clarkbon the linked change it is a 440 thing13:21
*** dizquierdo_afk is now known as dizquierdo13:22
ttxwhat am I supposed to push as a tag there ?13:22
mordredttx: I agree with you - I do not believe I fully grokked that we would _generate_ versions on tagged commits13:22
clarkbttx is the correct tag13:22
ttxbecause that extra 0 won't sort correctly with the version in the tag, it's not harmless13:22
clarkbaccording to pbr's interpretation of pep44013:22
ttxso I should have pushed ?13:23
*** scheuran has joined #openstack-infra13:23
ttxand then I would have gotten what I pushed in ?13:23
ttxhrm. So what now ? Should we clean up the 3.0.0b1 tag ?13:24
ttx(from sahara repo ?)13:24
ttxand repush ?13:24
fungittx: yeah, the debate right now is whether we configure pbr-specific pre-release pipelines to only fire jobs on pbr-normal-form pre-release tags to avoid further confusion13:24
ttxfungi: ack -- because that is confusing alright13:24
dhellmannfungi: ++ to reducing confusion13:25
ttx(in addition to being dead ugly)13:25
openstackgerritAndrey Pavlov proposed openstack/requirements: Add botocore to requirements  https://review.openstack.org/17233513:25
fungihowever if we do that, we probably want the pbr-specific pre-release pipeline separate from the normal pre-release pipeline so that we aren't dictating pre-release tag semantics to non-pbr projects13:25
ttxshould we clean up the accidental tag before it spreads out ?13:25
fungiwe can't, really13:26
wznoinskclarkb: thanks13:26
*** sergmelikyan has quit IRC13:26
ttxif I retag, will the tarball in tarballs.o.o be overwritten ?13:26
fungiand configuring filters in the zuul layout to be more strict isn't going to prevent people from pushing any old signed tag into gerrit anyway13:27
fungittx: yes13:27
*** eharney has joined #openstack-infra13:27
ttxI guess I should retag then13:27
*** mriedem has joined #openstack-infra13:27
ttxdhellmann: thoughts ?13:27
sdaguedstufft: question about the built in wheel cache in pip... does pip prefer content there over upstream on an install -U ?13:27
clarkbright so 440 says normal form is somever.XaY and pbr has taken a generic approach to normalizing that13:27
dhellmannttx: yes, let's retag at least so it's consistent13:27
ttxI'll modify milestone.sh so it generates extra 0s13:27
clarkbso even if you start with that it separates the prerelease specifier onto its own dotted thing13:28
dstufftsdague: No, if a wheel is uploaded to PyPI it will prefer that instead of the cached wheel, and if a new version is uploaded it will prefer that13:28
sdaguedstufft: hmmm....13:28
dstufftor at least, that's how it _should_ work and is intended to work13:28
dstufftif it doesn't it's a bug13:28
sdaguedstufft: even if the wheel cache would satisfy it?13:28
dhellmanndstufft, sdague : I think we've seen a couple of folks in the last week or so have cache-related upgrade issues (at least deleting their cache allowed them to upgrade)13:28
dhellmannI haven't been able to reproduce the problems locally to debug better :-/13:29
sdagueso, the situation I seemed to get into was that I had python-openstackclient 1.0.4 previously installed. I uninstalled it.13:29
fungiit's possible that when they run a job which tests, say, oslo.messaging from master branch tip, pip is caching a wheel of a dev version that's newer than what's on pypi, then later pypi reuses that because it satisfies the version constraints13:30
sdaguebut pip install -U python-openstackclient >=1.0.3 gave me 1.0.413:30
sdaguenot 1.613:30
sdaguewhich causes issues13:30
fungibut yeah, the cache symptoms you're seeing can't be what i'm describing13:30
dhellmannyou know, the problem we had reported late last week was the same package13:30
*** dustins has joined #openstack-infra13:31
dstufftsdague: yes, the wheel cache happens after we select a potentional file, essentially we resolve a file to download from all of the available versions (ignoring the fact that there is a wheel cache), and once we've selected a link to download and install we check if that link is a wheel, if it is already a wheel we use that, if it's a sdist we check in the cache (keyed by the full url) to see if we've already built a wheel13:31
zarofungi: hi, i've been vacationing since last week so won't be in meeting today.13:31
dhellmannfungi, weren't you involved in helping to debug that in here last week?13:31
clarkbthe pep pages are almost ubreadable using mobile chrome...13:31
dstufftsdague: hrm13:31
*** svasheka has joined #openstack-infra13:31
sdaguedstufft: what happens if a permissions issue happens somewhere in that process?13:31
clarkbthey auto scroll to the top13:31
sdaguelike ~/.pip became owned by root because sudo13:31
fungizaro: okay, thanks for the heads up. also clarkb says there's been some further discussion on the jgit corrupt packfile issue13:32
*** bdemers has joined #openstack-infra13:32
*** sdake_ has joined #openstack-infra13:32
clarkbfungi zaro just more people affected13:32
*** fawadkhaliq has joined #openstack-infra13:33
fungidhellmann: yeah, the person reporting the issue kept getting a version much older than what was on pypi, until they deleted their pip cache13:33
clarkbwe should maybe write the list with a big warning to not upgrade gerrit...13:33
*** ayoung has joined #openstack-infra13:33
dstufftsdague: if it couldn't _write_ a wheel, it should fall back to installing from sdist I think (at least, if any error occurs during setup.py bdist_wheel, not sure if the actual writing to the final location is inside that try: except offhand, but if it's not then it should flat out bomb out), if it can't _read_a wheel, it should bomb out I think13:33
dstufftsdague: are yoju able to reproduce this right now?13:33
*** sergmelikyan has joined #openstack-infra13:34
sdaguedstufft: no, I hacked past it because this was preventing me from getting any work done13:34
zarogreat news.  i was going peer into the code when i return but wasn't sure i would get far.13:34
sdagueI can probably build a reproduce env13:34
dstufftmy guess is something caused /simple/python-openstackclient/ to get cached for much longer than it should13:34
wznoinsksdague: is that the cause .pip dir didn't exist before and it was created by root (hence owned)? would that happen if you already had .pip directory?13:34
openstackgerritThierry Carrez proposed openstack-infra/release-tools: Pre-versions are now $VERSION.0b1  https://review.openstack.org/19465913:34
zarofungi, clarkb : should be back end of next week.13:35
sdaguewznoinsk: I'm pretty sure .pip did not exist in the stack user before I got started with these devstack vagrants13:35
sdaguethey were pretty blank boxes13:35
dstufftsdague: if you can reproduce that would be great13:35
mrmartinfungi: hi13:36
wznoinsksdague: then that would be expected .pip was owned by root after sudo that created it...13:36
*** amitgandhinz has joined #openstack-infra13:36
dstufftif the cache dir is owned by someone other than the current user we disable the cache completely IIRC13:36
mordredyes. that is the behavior13:37
sdaguewell, I'm getting a behavior that's as if building the cache is disabled, but reading from it is not13:37
dstufftthough we might not disable it for the wheel cache13:38
* dstufft looks13:38
*** fawadkhaliq has quit IRC13:39
dstufftyes the wheel cache in particular isn't being disabled if we don't own the dir13:40
dstufftgot missed13:40
dstufftthough that still shouldn't have the effect you're seeing13:40
dstufftsince the wheel cache happens _after_ pip has resolved things to a link13:41
*** tvanderwiel has joined #openstack-infra13:41
openstackgerritDoug Hellmann proposed openstack-infra/release-tools: Only use tput when we are writing to a terminal  https://review.openstack.org/19466113:41
openstackgerritMerged openstack-infra/release-tools: Pre-versions are now $VERSION.0b1  https://review.openstack.org/19465913:41
mrmartinfungi: this patch: https://review.openstack.org/#/c/186217/ changed the groups-core to groups-release pushSignedTag, and I'm not a member of groups-release, just groups-core so I lost the ability to deliver new releases to groups portal13:41
sdaguedstufft: well, do you think you have a handle on the bug, or do you still need a reproduce? Just so I know if I should spend the time trying to build this env13:41
*** ddieterly has joined #openstack-infra13:41
dstufftsdague: if you can reproduce that would be useful. I'm not sure what would be causing the issue you're seeing13:42
mrmartinfungi: can you add me to the proper group? I like to roll-out a security fix13:42
*** tonytan4ever has joined #openstack-infra13:42
fungimrmartin: yep, i saw you asking about that in scrollback. we should double-check with jeblair so he can officially delegate it now that we have the groups redone for our governance reorganization. i don't think it will be a problem, but i'd rather leave that as a ptl task so he has proper oversight13:42
*** fhubik_afk is now known as fhubik13:43
wznoinski just got 'server unavailable' from gerrit http13:43
mrmartingreat, thnx13:43
wznoinskcan anyone check? not sure whether it's just me13:43
marcusvrnwznoinsk: it's not working here too13:45
mordredwznoinsk: we are having intermittend as-of-yet unknown issues with http to gerrit13:45
vlazasame to me, gerrit unavailable13:45
wznoinskhttp://www.isitdownrightnow.com/review.openstack.org.html, phew13:45
openstackgerritThierry Carrez proposed openstack-infra/release-tools: upload_release.py should use $VERSION.0b1 too  https://review.openstack.org/19466213:46
*** yamamoto has joined #openstack-infra13:46
*** coolsvap is now known as coolsvap|away13:46
sdaguedstufft: ok, I'll build a second vagrant and see if I can guess how I got into this state13:46
*** bknudson has joined #openstack-infra13:48
*** fallenpegasus has joined #openstack-infra13:48
*** yfried is now known as yfried|afk13:48
fungiwznoinsk: vlaza: marcusvrn: yes, i have a ticket open with rackspace support asking them to look into it. we're having intermittent periods of no https connections reaching the server13:50
*** tosky has quit IRC13:51
*** yamamoto has quit IRC13:51
*** dkranz has quit IRC13:52
*** salv-orlando has joined #openstack-infra13:52
*** agentle has joined #openstack-infra13:53
*** yfried|afk is now known as yfried13:53
*** sergmelikyan has quit IRC13:53
*** bhunter71 has joined #openstack-infra13:54
*** bknudson has quit IRC13:54
*** mpavone has joined #openstack-infra13:54
*** sergmelikyan has joined #openstack-infra13:55
*** xyang1 has quit IRC13:56
jeblairjhesketh, tonyb: all our software is documented :)  http://docs.openstack.org/infra/jenkins-job-builder/publishers.html#publishers.scp13:56
*** bexelbie has quit IRC13:56
jheskethjeblair: heh, jjb's documentation is better than the plugins themselves13:57
*** bexelbie_ has quit IRC13:57
*** salv-orlando has quit IRC13:57
*** dkranz has joined #openstack-infra13:57
*** cdent has quit IRC13:58
*** cdent has joined #openstack-infra13:58
*** fallenpegasus has quit IRC13:58
*** bexelbie has joined #openstack-infra13:59
*** baoli has quit IRC13:59
marcusvrnfungi: ok. thanks for the feedback14:00
*** rwsu has joined #openstack-infra14:01
fungihrm, rackspace says they have no port-specific filtering or proxying14:01
*** otter768 has joined #openstack-infra14:01
fungii'll see if i can collect some additional details the next time this happens14:01
*** agentle has quit IRC14:01
*** yamamoto has joined #openstack-infra14:02
*** baoli has joined #openstack-infra14:02
openstackgerritFeodor Tersin proposed openstack-infra/project-config: Add Neutron vpn service for ec2api Neutron gating job  https://review.openstack.org/19467014:03
jeblairjhesketh: yeah, it's almost a required side effect of the way we build jjb :)14:03
*** naggappan has quit IRC14:04
*** ParsectiX has quit IRC14:04
*** otter768 has quit IRC14:05
jeblairfungi, ttx, clarkb, mordred: i feel that changing the zuul config to restrict to pbr-semver is unecessary because it will not prevent people from pushing noncompliant tags in the first place; it will only cause a tarball not to be built.  we can, if we choose, delete the tarball which will have about the same effect as not having run the job.14:05
jeblairfungi, ttx, clarkb, mordred: meanwhile, having that restriction will confuse other people who don't use pbr-semver versions and wonder why their jobs didn't run (or don't know which pipeline to use)14:06
openstackgerritMatthew Treinish proposed openstack-infra/devstack-gate: Fix the setup_workspace to handle missing ZUUL_BRANCH  https://review.openstack.org/19430214:07
jeblairi will put that in the review; but i wanted to mention it here because the events of the last 2 days have brought clarity to me on the issue :)14:07
*** sigmavirus24_awa is now known as sigmavirus2414:07
mtreinishfungi, jeblair: ^^^ I think that should work now14:07
fungijeblair: well, it causes the tarball/wheel to be built with a different name from the tag, and then not uploaded to pypi because of the mismatch14:07
fungibut it does end up on tarballs.o.o14:07
*** ddieterly has quit IRC14:08
*** bknudson has joined #openstack-infra14:08
fungii'm generally not in favor of that patch either, but proposed it as a straw man to see if anyone else has better ideas about how we can solve this14:08
jeblairfungi: right, but not running the job would merely cause a tarball not to be created at all.  we can get to the same end-point by deleting it from tarballs.o.o14:08
jeblairfungi: tag review :(14:08
jeblairi heard.... someone... was working on that.... who was it....14:09
*** zz_jgrimm is now known as jgrimm14:09
dhellmannjeblair, fungi, clarkb : is paste.openstack.org down? I'm seeing timeouts14:09
fungidhellmann: hem... mysqld is chewing up an entire cpu on the paste server14:10
*** bexelbie has quit IRC14:11
*** bexelbie_ has joined #openstack-infra14:11
fungijeblair: what was the symptom you saw yesterday which caused you to restart something on the paste server?14:11
*** Hal has joined #openstack-infra14:11
*** Hal is now known as Guest3615014:12
jeblairfungi, ttx: do we want to (or have we already) deleted the tag?  it is a thing we can do, and the consequences are not quite as dire if we have no plans to re-issue the tag.  otoh, probably won't hurt to leave it.14:12
jeblairfungi: similar to this.14:12
*** sdake_ has quit IRC14:12
fungiwhat did you restart? mysql?14:12
jeblairfungi: lodgeit14:12
*** jrniemijr has joined #openstack-infra14:12
*** zz_dimtruck is now known as dimtruck14:13
*** Guest39912 has quit IRC14:13
*** fhubik is now known as fhubik_afk14:16
*** marun has joined #openstack-infra14:16
fungiaha, upstart job named openstack-paste14:16
jeblairAJaeger: that's pretty much the most benign internal email i've ever seen accidentally leaked to a public list14:16
fungiokay, that seems to have terminated whatever was beating up mysqld14:16
fungidhellmann: ^14:16
funginevermind--now it's gone right back to eating an entire processor again14:17
Clintmaybe it's a 4-octet utf-8 character14:18
*** doug-fish has joined #openstack-infra14:18
fungiwe're getting a lot of http queries from a browser reporting itself as "Java/1.8.0_45"14:20
jeblairfungi: does it seem like the quantity is the problem, or is there perhaps a single bad paste that's being hit?14:21
openstackgerritThierry Carrez proposed openstack-infra/release-tools: Fix version determination in corner cases  https://review.openstack.org/19467514:22
*** marun has quit IRC14:22
fungijeblair: it's about one request a second for random pastes. like it's a crawler maybe14:22
dhellmannfungi: thanks!14:22
* dhellmann keeps reading14:22
dhellmannfungi: :-/14:22
*** spzala has joined #openstack-infra14:23
*** hichihara has joined #openstack-infra14:23
fungithe ip address is a customer in "eunetworks" (seems to be a germany-based isp)14:23
*** bexelbie has joined #openstack-infra14:24
openstackgerritMerged openstack-infra/release-tools: upload_release.py should use $VERSION.0b1 too  https://review.openstack.org/19466214:25
*** dboik has joined #openstack-infra14:25
jeblairfungi: what's with the base64 looking urls?14:26
fungijeblair: no clue14:27
jeblairthey're all 50214:27
openstackgerritMerged openstack-infra/release-tools: Fix version determination in corner cases  https://review.openstack.org/19467514:27
jeblairor is everything 502 now14:27
jeblairfungi: oh, i think those are all timing out, i suspect this client is behaving very poorly and is flooding us with requests14:28
jeblairfungi: so maybe let's block the ip?14:29
sdaguedstufft: would the fact that we still specify --download-cache impact things?14:29
fungiyeah, i think it's a crawler. i'm working on an iptables rule to block it14:29
dstufftsdague: it's completely ignored since 614:29
sdagueok, it generates a deprecation warning that doesn't make it clear that it's ignored, but that's fine14:29
*** lennyb_ has joined #openstack-infra14:29
mrmartinjeblair: re, please add me to the groups-release group on gerrit acl. this change: https://review.openstack.org/#/c/186217/ upgraded the ACL's, and I lost my ability to do a groups portal release.14:30
*** jasondotstar has quit IRC14:31
*** ghostpl has quit IRC14:31
*** nihilifer has joined #openstack-infra14:31
*** dustins_ has joined #openstack-infra14:32
*** fhubik_afk is now known as fhubik14:33
*** sergmelikyan has quit IRC14:33
jeblairmrmartin: what's your gerrit username?14:33
*** soren has joined #openstack-infra14:33
mrmartinjeblair: marton-kiss14:33
*** soren has quit IRC14:33
*** soren has joined #openstack-infra14:33
fungiokay, paste.o.o seems to be behaving better now that the offending client has been blocked and the service restarted again14:33
*** Guest36150 has quit IRC14:34
*** dustins has quit IRC14:35
jeblairmrmartin: done!14:35
jeblairmrmartin: sorry about that14:35
mrmartinjeblair: thnx!14:35
*** afazekas has quit IRC14:36
fungiwe've got other spiders in the access logs on paste.o.o too, and it looks like we're not serving a /robots.txt14:36
mordredoh. we should serve one of those14:37
*** signed8bit has joined #openstack-infra14:38
jeblairwhat should it say?14:38
*** annegentle has joined #openstack-infra14:38
Clint"go home, Java/1.8.0_45, you're drunk"14:38
*** annegentle is now known as Guest2057414:38
fungiyeah, i'm not convinced that we should disallow spidering of the paste site. it doesn't have a content index, so in theory only pastes referenced from somewhere else would get spidered14:40
fungiperhaps that same spider was indexing our newly-htmlified irc logs14:40
Clintyou could put Crawl-delay: 5 or something14:40
*** vlaza has quit IRC14:41
*** bdemers has quit IRC14:41
fungiyeah, 290 hits from that same client address in the eavesdrop apache logs for today14:42
fungithe bot wasn't providing referrer headers, so not certain that's where it was getting them, but it's a good bet14:42
fungialso, if it's not a well-behaved bot (for example someone crawling looking for security vulnerabilities to exploit) then a robots.txt is unlikely to help at all14:43
fungiand given that this client lacked even a well-formed reverse dns entry, that seems possible14:44
*** e0ne is now known as e0ne_14:44
*** xyang1 has joined #openstack-infra14:44
*** ghostpl has joined #openstack-infra14:46
*** jasondotstar has joined #openstack-infra14:46
*** stevemar has joined #openstack-infra14:46
*** bdemers has joined #openstack-infra14:47
*** thedodd has joined #openstack-infra14:47
*** flepied1 has quit IRC14:47
*** Guest20574 is now known as agentle14:47
*** sergmelikyan has joined #openstack-infra14:47
*** lennyb_ has quit IRC14:47
*** shardy has quit IRC14:47
*** shardy has joined #openstack-infra14:48
*** e0ne_ is now known as e0ne14:49
*** heyongli has quit IRC14:49
*** tosky has joined #openstack-infra14:50
jeblairfungi: they may have just been using openstack.14:50
*** heyongli has joined #openstack-infra14:50
*** sdake has joined #openstack-infra14:51
openstackgerritDoug Hellmann proposed openstack-infra/release-tools: Add script for targeting bugs from the command line  https://review.openstack.org/19468814:51
*** _nadya_ has quit IRC14:52
*** Ala has quit IRC14:52
*** erlon has joined #openstack-infra14:52
*** _nadya_ has joined #openstack-infra14:53
openstackgerritDoug Hellmann proposed openstack-infra/release-tools: Add --no-clear flag to adjust_blueprints.py  https://review.openstack.org/19468914:55
*** Hal has joined #openstack-infra14:55
*** Hal is now known as Guest1532614:56
*** yamamoto has quit IRC14:56
*** _nadya_ has quit IRC14:58
openstackgerritSean Dague proposed openstack/requirements: bump python-openstack client to 1.6.0  https://review.openstack.org/19469214:58
*** yamahata has joined #openstack-infra14:59
*** sergmelikyan has quit IRC15:00
*** annegentle_ has joined #openstack-infra15:00
*** sergmelikyan has joined #openstack-infra15:00
*** agentle has quit IRC15:00
*** unicell has joined #openstack-infra15:01
*** shardy has quit IRC15:01
*** yamamoto has joined #openstack-infra15:01
*** ddieterly has joined #openstack-infra15:02
*** unicell1 has quit IRC15:02
*** shardy has joined #openstack-infra15:02
*** sabeen1 has joined #openstack-infra15:03
*** sarob has joined #openstack-infra15:03
*** samuelBartel has quit IRC15:04
*** hdd has joined #openstack-infra15:04
*** bauzas is now known as JustinBieber15:04
*** JustinBieber is now known as NotJustinBieber15:05
*** Ala has joined #openstack-infra15:05
*** NotJustinBieber is now known as bauzas15:06
*** ghostpl has quit IRC15:06
*** claudiub has quit IRC15:07
*** fhubik is now known as fhubik_afk15:07
*** r-daneel has joined #openstack-infra15:07
*** mrunge has quit IRC15:08
*** fhubik_afk is now known as fhubik15:08
*** arxcruz has quit IRC15:10
*** claudiub has joined #openstack-infra15:10
*** sergmelikyan has quit IRC15:10
*** sarob has quit IRC15:10
*** isviridov_away is now known as isviridov15:10
*** ghostpl has joined #openstack-infra15:11
*** sshnaidm has quit IRC15:13
openstackgerritMatt Riedemann proposed openstack-infra/elastic-recheck: Add query for grenade + neutron + glance bug 1467972  https://review.openstack.org/19470415:15
openstackbug 1467972 in grenade "new (liberty) side of check-grenade-dsvm-neutron fails with auth error in g-api" [Undecided,New] https://launchpad.net/bugs/146797215:15
*** sabeen2 has joined #openstack-infra15:16
*** samuelBartel has joined #openstack-infra15:16
*** sshnaidm has joined #openstack-infra15:16
*** dannywilson has joined #openstack-infra15:16
*** esker has quit IRC15:16
*** Ala has quit IRC15:16
*** shardy has quit IRC15:18
*** shardy has joined #openstack-infra15:18
*** sabeen1 has quit IRC15:19
*** dannywil_ has joined #openstack-infra15:19
*** dannywilson has quit IRC15:20
openstackgerritNeil Jerram proposed openstack-infra/project-config: Add stackforge/networking-calico project  https://review.openstack.org/19470915:21
*** erikmwilson_ has joined #openstack-infra15:22
*** fhubik has quit IRC15:23
*** sshnaidm has quit IRC15:25
openstackgerritWayne Warren proposed openstack-infra/infra-specs: Add JJB DSL Roadmap spec.  https://review.openstack.org/19449715:25
*** erikmwilson is now known as Guest4084015:25
*** erikmwilson_ is now known as erikmwilson15:25
*** erikmwilson_ has joined #openstack-infra15:25
sdaguedstufft: well, I'm 2 hours into a reproduce, but it looks like this will be at least an all day afair. So I'm going to bail on that. If I see it again I'll try to freeze the env to look at it15:25
*** mpaolino has quit IRC15:25
dstufftsdague: thanks, and no problem15:25
*** amotoki has quit IRC15:26
*** sshnaidm has joined #openstack-infra15:26
*** pblaho has quit IRC15:26
*** changbl_ has joined #openstack-infra15:28
*** yuikotakada has joined #openstack-infra15:28
*** sarob has joined #openstack-infra15:29
*** flepied has joined #openstack-infra15:29
*** yuikotakada has left #openstack-infra15:31
*** annegentle_ is now known as agentle15:31
*** yfried has quit IRC15:32
*** armax has joined #openstack-infra15:33
*** mestery has joined #openstack-infra15:34
*** amotoki has joined #openstack-infra15:35
*** mestery_ has joined #openstack-infra15:35
*** dimtruck is now known as zz_dimtruck15:36
*** shardy has quit IRC15:36
*** shardy has joined #openstack-infra15:37
*** hashar has quit IRC15:37
*** HeOS has quit IRC15:37
openstackgerritAleksandra Fedorova proposed openstack-infra/project-config: Fix fuel-docs acls to allow merge commit  https://review.openstack.org/19471715:38
*** kushal has joined #openstack-infra15:38
*** mestery has quit IRC15:39
*** mestery_ is now known as mestery15:40
*** ihrachyshka has quit IRC15:42
*** sdake has quit IRC15:42
*** shardy has quit IRC15:42
*** zz_dimtruck is now known as dimtruck15:42
greghaynesmordred: For putting in a req to replace NIC's in hp1, is there info on where / how to do that somewhere?15:43
*** marun has joined #openstack-infra15:43
*** shardy has joined #openstack-infra15:43
jeblairgreghaynes: (when you find out the answer, we should add it to the system-config documentation for infra-cloud)15:43
scheuranHi, I submitted a request for a new stackforge project a week ago. Is it possible to get that approved soon?15:44
greghaynesjeblair: definitely15:44
sdaguedstufft: so, another interesting thing. pip install python-openstackclient>=1.0.3 when 1.0.4 is installed is not a no op15:45
sdagueit reinstalls all the dependencies to versions15:45
sdaguespecified in python-openstackclient package15:45
dstufftsdague: without a -U?15:45
*** e0ne is now known as e0ne_15:46
*** e0ne_ is now known as e0ne15:47
sdaguedstufft: without -U15:47
*** ajmiller has joined #openstack-infra15:48
*** cdent has quit IRC15:49
dstufftsdague: oh right, it's not supposed to be a no-op, it re-verifies that all of the dependencies are installed (and if they don't match the version specifier, will do something to make them match)15:49
openstackgerritPaul Belanger proposed openstack-infra/grafyaml: Add datasource functionality  https://review.openstack.org/19426515:49
*** cdent has joined #openstack-infra15:49
dstufftsdague: is that the command that was failing with seeing an old python-openstackclient?15:49
sdaguedstufft: this is actually a different issue15:49
dstufftbecause that has a manual wheel cache too15:49
sdaguethough, that's how we call things in devstack, so we probably want to address that as well15:50
sdaguethe manual wheel cache bit15:50
sdaguebut, this non "no-op" behavior is actually currently masking a requirements issue15:50
*** bdemers has quit IRC15:50
*** armax has quit IRC15:50
sdagueor more importantly, let one slip in15:50
*** hichihara has quit IRC15:51
*** marun has quit IRC15:51
thingeemordred: http://lists.openstack.org/pipermail/openstack-dev/2015-June/067795.html15:52
*** puranamr has joined #openstack-infra15:53
dtantsurhey folks, if you have some time, have a look at https://review.openstack.org/#/c/193941/ please, we're really looking forward to our first devstack check job :)15:53
*** bdemers has joined #openstack-infra15:53
*** fawadkhaliq has joined #openstack-infra15:53
greghaynesfungi: any idea if the rax dib images worked last night?15:54
*** abregman has quit IRC15:54
greghaynesfungi: hrm, actually, we might not have updated nodepool15:54
fungithingee: i'll reply on the ml, but having cinderclient stop working with existing cinder deployments (and expecting deployers to adjust their configs to accommodate it) seems problematic15:54
*** marun has joined #openstack-infra15:55
fungithingee: even worse, expecting deployers to apply a patch to their systems to accommodate new client behavior15:55
*** erikmwil_ has joined #openstack-infra15:56
*** laughterwym has quit IRC15:56
*** devvesa has quit IRC15:56
fungisort of the opposite of having backward-compatible clients15:56
thingeefungi: I agree. I was talking to ttx this morning about it and the other concern was removing version discovery after it has been released. I'm not sure how many people started adjusting their service catalogs with no versions set for cinder.15:56
*** erikmwil_ has quit IRC15:56
*** erikmwil_ has joined #openstack-infra15:57
fungiyeah, quite the pickle15:57
clarkbcan turn it off by default?15:57
jeblairgreghaynes: logs here http://nodepool.openstack.org/15:57
*** ghostpl has quit IRC15:57
*** erikmwilson has quit IRC15:57
*** erikmwil_ is now known as erikmwilson15:57
jeblairgreghaynes: and yes, i don't think anyone restarted nodepool15:57
*** rwsu has quit IRC15:58
jeblairgreghaynes: running since Jun19.  want me to restart now?15:58
fungiyeah, last started june 1915:58
greghaynesactually, we might not have to - it was a glean release which is pulled in by DIB so it should have just pulled in the new version on image build15:58
openstackgerritMarton Kiss proposed openstack-infra/groups: Stop importing data from groups.json  https://review.openstack.org/19408215:58
fungiimage updates may be in process right now15:58
*** ZZelle is now known as ZZelle-away15:58
funginot sure if nodepoold recovers sanely from a restart in the middle of image updates15:58
jeblairanyone want to implement the remote image builder for nodepool? :)15:59
*** unicell has quit IRC15:59
*** dboik_ has joined #openstack-infra15:59
*** dboik_ has quit IRC15:59
jeblairgreghaynes: aren't you infra-clouding? :)15:59
greghaynesjust give me 10mins15:59
*** dboik_ has joined #openstack-infra15:59
jeblairoh, okay, i guess we can spare you for 10 mins :)15:59
*** gordc_ has joined #openstack-infra15:59
*** gordc has quit IRC16:00
fungiwe did restart it after the gear release at least, yeah?16:00
*** gordc_ is now known as gordc16:00
*** chen12 has quit IRC16:00
fungitiming is close, so i'm assuming so16:00
jeblairfungi: not sure, but zuul is the important part of that one anyway16:00
jeblairbug was geard-side16:00
*** chen12 has joined #openstack-infra16:01
*** dtantsur is now known as dtantsur|afk16:01
*** yamamoto has quit IRC16:01
fungifor some reason i thought we were causing socket buffering issues in nodepoold while image uploads to glance were happening16:01
*** otter768 has joined #openstack-infra16:02
*** juggler_ is now known as juggler16:02
fungiand that was causing the gear client to not successfully submit admin requests16:02
greghayneshrm, that does bring up a good point... the upload was happening from nodepool16:02
* fungi completely misunderstood the problem there16:02
jeblairfungi: first half is correct, but the full tcp windows on nodepool.o.o were causing geard to stop sending the response16:02
greghaynesoh, windows were effecting the other side16:02
*** bdemers has quit IRC16:03
jeblairright.  but then the client timed out because it never got a response16:03
fungithe tcp version of spooky action at a distance16:03
*** dboik has quit IRC16:03
* jeblair thinks all networking is spooky action at a distance16:03
*** bexelbie has quit IRC16:03
*** jgrimm is now known as zz_jgrimm16:03
*** ghostpl has joined #openstack-infra16:04
*** bdemers has joined #openstack-infra16:04
*** armax has joined #openstack-infra16:05
*** yfried has joined #openstack-infra16:05
*** sdake has joined #openstack-infra16:05
*** samuelBartel has quit IRC16:06
*** ayoung is now known as ayoung-lunch16:06
openstackgerritSean Dague proposed openstack/requirements: bump python-openstack client to 1.5.0  https://review.openstack.org/19469216:06
jeblairis there a reason that we aren't using dib for all image types on hpcloud?16:06
*** otter768 has quit IRC16:07
jeblair(or are we?)16:07
greghaynesI thought we were for most, not 100% sure its all16:07
jeblairit just looks like we have some -dib suffixes still16:07
*** SumitNaiksatam has quit IRC16:08
jeblairit looks like we have both devstack-trusty and devstack-trusty-dib in hpcloud16:08
*** armax has quit IRC16:08
*** yamamoto has joined #openstack-infra16:08
*** SumitNaiksatam has joined #openstack-infra16:08
jeblairer wait that's wrong16:08
jeblairthe only dib images we have configured are  devstack-centos7-dib and devstack-fedora21-dib16:09
jeblairand we have a non-dib version of centos7, but only dib of f2116:10
greghaynesdevstack-trusty looks like its dib16:10
greghaynesor you mean only *-dib images?16:11
jeblairgreghaynes: oh you are right, we've moved that over completely on hp16:11
fungithe devstack variants are dib'able, the bare variants are not yet. back to working now on that bit16:11
greghaynesfungi: oh? Why arent the bare?16:11
openstackgerritSpencer Krum proposed openstack-infra/system-config: Bring in module to manage update-alternatives  https://review.openstack.org/17155616:12
jeblairfungi: ah, that's a piece i was missing, thanks.16:12
fungigreghaynes: database setup16:12
openstackgerritSpencer Krum proposed openstack-infra/system-config: Add puppet-community/archive module  https://review.openstack.org/16456916:12
greghaynesah, I didnt realize that was required for -bare16:12
fungii'm trying to follow the twisty path of dib elements to figure out when pip is getting reinstalled in cache-devstack16:12
*** achanda has joined #openstack-infra16:12
jeblairfungi: or we could punt on that and move it to job setup.16:12
fungihow does http://git.openstack.org/cgit/openstack-infra/project-config/tree/nodepool/elements/cache-devstack/source-repository-pip get invoked?16:12
jeblairwhich i think we wanted to do eventually anyway?16:12
greghaynesfungi: check out the source-repositories element README16:13
*** nihilifer has quit IRC16:13
mordredjeblair: ++16:13
greghaynesfungi: all that does is download that url to that location, and allow overriding16:13
*** RichardRaseley has joined #openstack-infra16:13
fungijeblair: that's the plan, but we need bindep installing all the distro reqs for databases et al16:13
nibalizerpabelanger: so I think this is actually going to fix the users/packages/uids problem https://review.openstack.org/#/c/161048/16:13
nibalizerinteresed in your thoughts16:14
jeblairfungi: oh, okay, gotcha.16:14
fungiso trying to make sure i create the bindep env after pip is installed, but still early enough to cache distro packages16:14
fungii may need to reorder some of the extsting element install.d components16:14
greghaynesfungi: ah, so youll want to find the thing that exec's get-pip.py, that just does the download16:14
openstackgerritMerged openstack-infra/groups: Stop importing data from groups.json  https://review.openstack.org/19408216:14
mordredjeblair, fungi: as a stepping stone, couldln't we just make a jjb macros that does "apt-get install mysql-server" ? - like, we don't have to have the final solution for general package install to get mysql installed16:14
*** agentle has quit IRC16:15
*** bexelbie_ has quit IRC16:15
*** nfedotov has quit IRC16:16
fungimordred: sure, if we want to just go ahead and create bare-.* images via dib rather than moving the jobs to devstack-.* workers instead16:16
fungii've basically already got the plumbing for that working as an experimental job example16:16
*** tonytan4ever has quit IRC16:17
greghaynesfungi: looks like that sticks get-pip in a place where its pre-cached for devstack/tools/install_pip to run it16:17
*** ddieterly has quit IRC16:17
*** scheuran has quit IRC16:17
fungigreghaynes: oh, so it's not actually used to update pip on the image16:18
pabelangernibalizer, Ya, I think stages are the best.  In fact, I was going to work on something related to apt-get update in a new stage16:18
jeblairianw: what's the difference between devstack-f21 and devstack-fedora21-dib ?16:18
greghaynesfungi: I dont think so, no16:18
*** esker has joined #openstack-infra16:18
pabelangerbut I can give a quick run and see how it works16:18
greghaynesfungi: meaning I couldnt find anything running get-pip using grep16:18
*** maxskew has joined #openstack-infra16:18
fungigreghaynes: thanks, that makes things simpler i think16:18
openstackgerritPaul Belanger proposed openstack-infra/system-config: Create grafana.o.o under -infra  https://review.openstack.org/17920816:18
jeblairpabelanger: !16:19
pabelangeryolanda, ^ I think that should get you want you need16:19
clarkbI thought stages were an anti pattern in puppet16:19
nibalizerpabelanger: are you seeing apt-get update problems?16:19
nibalizerlike around ppas or whatever?16:20
clarkbgreghaynes fungi correct thats just a cache step for devstack it does not install pip, the puppet install should install pip16:20
*** mpavone has quit IRC16:20
clarkbusing install_puppet.sh16:20
greghaynesoh fun16:20
pabelangernibalizer, in the past, I did. but nothing recently. I did notice something today about apt-get update during mid run16:20
*** ashleighfarnham has joined #openstack-infra16:21
*** esker has quit IRC16:21
pabelangernibalizer, what I would do in the past was create a bootstrap stage in puppet: https://github.com/kickstandproject/kickstandproject-ci-puppet/blob/master/modules/kickstandproject/manifests/node/base/bootstrap.pp16:21
pabelangerwhich ran before main16:21
mordredgreghaynes, fungi: the dib elements are in need of a re-org - but I've been waiting until we stopped having non-dib images16:21
pabelangerseemed to fix all my apt issues16:21
*** samueldmq has quit IRC16:21
*** yamamoto has quit IRC16:21
jeblairtimrc, yolanda, nibalizer, or anyone else: interested in reviewing grafyaml patches?16:21
mordredbecause there are places where we're split how we are to make maintaining two parallel implemntations of the images easier16:22
greghaynesyea, makes sense16:22
greghaynesget all moved over then start making dib-specific fixes16:22
mordredand once we only have _one_ area and it's the dib elements, refactoring them into nicely reusable chunks should be easier16:22
nibalizerjeblair: sure, i've got some plane time today so ill try to look at them then16:22
*** marun has quit IRC16:23
nibalizerpabelanger: that doesn't help when one of your puppet tasks is to add an apt repo16:23
*** salv-orlando has joined #openstack-infra16:23
nibalizeranother cute trick is to add a fact that runs apt-get update16:23
nibalizerbut the modern apt module is supposed to DTRT16:24
*** madong_ has quit IRC16:24
pabelangernibalizer, right, for me I already moved my apt tasks into bootstrap, and didn't have the puppet-module manage it. Not the best, but worked for me16:24
*** samueldmq has joined #openstack-infra16:24
*** yfried has quit IRC16:24
clarkbya could the provider trap errors, update then retry?16:24
*** madong_ has joined #openstack-infra16:24
*** sshnaidm has quit IRC16:24
pabelangerHowever, I'm not seeing apt errors right now.  More about getting a new stage going16:24
*** yamahata has quit IRC16:24
pabelangereither way, just brain dumping16:25
*** yamamoto has joined #openstack-infra16:26
*** yfried has joined #openstack-infra16:28
mordredhave i mentioned that a provider that does dpkg --set-selections foo and then apt-get dselect-upgrade would be neat16:28
*** esmute__ has quit IRC16:28
*** mestery has quit IRC16:28
mordredbecause then there would be one and only one run of "please change the set of installed packages16:28
*** e0ne has quit IRC16:29
mordredand other tasks could depend on "packages up to date"16:29
*** Swami has joined #openstack-infra16:29
*** krtaylor has quit IRC16:29
*** markus_z has quit IRC16:29
* mordred waits for nibalizer to go rage code that ...16:29
*** puranamr has quit IRC16:30
*** rossella_s has quit IRC16:30
*** dims has quit IRC16:30
nibalizerhaha one does not simply make puppet more efficient16:30
*** rossella_s has joined #openstack-infra16:31
nibalizerthere have been millions of suggestions to make packages batch and they all eventually get shot down, it sucks16:31
*** puranamr has joined #openstack-infra16:31
nibalizermordred: would you like to reconsider your -1 on https://review.openstack.org/#/c/177320/ ? I'm pretty sure you and clark are just reading it wrong, it already does what we want16:32
*** dims has joined #openstack-infra16:33
mordrednibalizer: done16:34
greghaynesmordred: hey, did you see my Q about hp server NIC's - any idea where I should look for info on where we file tickets / what info they need?16:34
mordredah! let me send you an email intro16:34
greghaynesmordred: ok, ty16:35
greghaynesmordred: SpamapS ^ you should probably be on there too16:35
jeblairnibalizer: oh rock on with progress on that :)16:36
*** yamamoto has quit IRC16:38
*** jlanoux has quit IRC16:38
*** chlong has quit IRC16:39
openstackgerritGeoff Arnold proposed openstack-infra/irc-meetings: Add Mercador project meeting to IRC.  https://review.openstack.org/19473816:40
*** jerryz has joined #openstack-infra16:40
mordredgreghaynes: oh - oops, I just emailed you16:40
fungigreghaynes: i hate to keep asking dense dib questions, but can you enlighten me as to when the source repositories are cloned? is it before or after install.d stages run?16:40
greghaynesmordred: ill just forward replies to SpamapS / add him if I reply16:41
mordredgreghaynes: k. I just replied all and added him16:41
greghaynesfungi: psh, ask away on DIB16:41
*** mrmartin has quit IRC16:41
*** geoffarnold has joined #openstack-infra16:41
greghaynesfungi: it happens in extra-data.d phase16:41
fungigreghaynes: does that run before or after install.d?16:41
*** hdd has quit IRC16:41
*** fawadk has joined #openstack-infra16:42
fungior concurrent with, interleaved by priority number?16:42
greghaynesfungi: before, the idea is that is a phase right after you get rootfs installed where elements can copy in extra stuff to the chroot16:42
fungigreghaynes: okay, perfect--thanks again!16:42
*** fawadkhaliq has quit IRC16:42
greghaynesfungi: http://git.openstack.org/cgit/openstack/diskimage-builder/tree/elements/source-repositories/extra-data.d/98-source-repositories is the codes, FWIW16:43
*** davideagnello has quit IRC16:43
*** jasondotstar has quit IRC16:43
mordredgreghaynes: http://git.openstack.org/cgit/openstack/diskimage-builder/tree/README.rst#n28416:43
mordredfungi: http://git.openstack.org/cgit/openstack/diskimage-builder/tree/README.rst#n28416:43
mordredfungi: you may find that a useful thing to keep bookmarked16:43
fungigreghaynes: ahh, perfect16:44
fungimordred: yep, will rtfm for a bit16:44
greghayneswe even have pretty docs for that part now http://docs.openstack.org/developer/diskimage-builder/16:44
*** davideagnello has joined #openstack-infra16:44
mordredhttp://docs.openstack.org/developer/diskimage-builder/developer/developing_elements.html#phase-subdirectories then16:45
pleia2so, I've got this zanata review in the queue https://review.openstack.org/#/c/193677/ and when it lands I'll need to set up the infra account, do we have an infra openstackid account? or have any thoughts there?16:45
fungipleia2: interesting. no afaik we don't have any service accounts for our infra yet (unless groups.o.o uses one maybe)16:46
fungimrmartin isn't around at the moment16:46
funginor smarcet16:46
*** baoli has quit IRC16:46
*** SumitNaiksatam has quit IRC16:49
*** unicell has joined #openstack-infra16:49
*** hdd has joined #openstack-infra16:50
pleia2fungi: hm, maybe a meeting topic?16:50
*** marzif_ has joined #openstack-infra16:50
*** ayoung-lunch is now known as ayoung-burp16:50
fungipleia2: yep, good idea16:50
pleia2k, adding16:50
fungimaybe one of them will be present for the meeting16:51
*** baoli has joined #openstack-infra16:51
*** krtaylor has joined #openstack-infra16:51
*** AzherKhan has joined #openstack-infra16:51
*** ayoung-burp is now known as ayoung16:51
*** ildikov has quit IRC16:51
*** esmute__ has joined #openstack-infra16:52
*** shardy has quit IRC16:52
*** jasondotstar has joined #openstack-infra16:52
*** shardy has joined #openstack-infra16:53
*** yfried has quit IRC16:53
jeblairpleia2, fungi: i guess we could make one with one of our email addresses? (is maybe one option to consider)16:53
*** kdas_ has joined #openstack-infra16:54
pleia2could do16:55
*** yamamoto has joined #openstack-infra16:55
*** marzif has quit IRC16:56
*** sergmelikyan has joined #openstack-infra16:56
SpamapSTheJulia: finally getting around to trying your gPXE patch.16:56
*** cdent has left #openstack-infra16:56
*** kushal has quit IRC16:58
fungijeblair: seems like a reasonable option if all we need is a normal user account16:59
fungiwe can presumably change it later16:59
*** BobBall is now known as BobBall_AWOL16:59
*** shardy has quit IRC17:00
*** shardy has joined #openstack-infra17:00
*** patrickeast has joined #openstack-infra17:00
*** esker has joined #openstack-infra17:00
*** zz_jgrimm is now known as jgrimm17:01
*** dontalton has joined #openstack-infra17:01
openstackgerritSpencer Krum proposed openstack-infra/system-config: Load in nanliu/git module  https://review.openstack.org/16435517:01
openstackgerritPaul Belanger proposed openstack-infra/system-config: WIP: Create dotfiles type for testing environments  https://review.openstack.org/19474417:01
rbradforcould I get some help please re: jenkins gates defined in zuul/layout.yaml17:03
*** baoli has quit IRC17:03
*** husanu has quit IRC17:03
pabelangerjeblair, So, just added a simple patch to puppet to allow additional dotfiles to be installed when the environment is flagged as something other then production.17:04
pabelangerwould be interested in peoples thoughts17:04
*** esker has quit IRC17:04
pabelangereg: puppet hackers :)17:05
*** husanu has joined #openstack-infra17:05
*** shardy has quit IRC17:05
samueldmq /query sdague17:05
samueldmqoh sorry :-)17:05
*** shardy has joined #openstack-infra17:06
*** fawadk has quit IRC17:06
mordred    fail('Only RedHat family supported, and require redhat-lsb package or lsb module.')17:07
*** rkukura has quit IRC17:07
jeblairpabelanger, mordred: whenever someone says "dotfile" i think graphviz.17:08
*** DericHorn-HP has joined #openstack-infra17:08
*** rkukura has joined #openstack-infra17:08
*** rguillebert has quit IRC17:08
mordredjeblair: ++17:09
jeblairpabelanger: that feels like a bit of special privilege since you happen to have root on a machine; this doesn't look like it would scale out to others...17:09
*** dizquierdo has quit IRC17:10
jeblairpabelanger: whenever i'm testing with puppet, i spin up nodes via a mechanism that already sets up my user account before puppet is run, so it installs anything my account needs prior to that17:11
*** bdemers has quit IRC17:12
pabelangerjeblair, Right, this is unique to people that are already setup in users.pp. Which, doesn't scale like you said.17:12
*** bdemers has joined #openstack-infra17:13
*** fitoduarte has joined #openstack-infra17:13
pabelangerjeblair, Right, actually, I'm hoping the playbook to spin up nodes could use this code17:13
pabelangersome ability to extend the playbook to include external things for testing mode17:13
openstackgerritArkadiy Kraminsky proposed openstack-infra/puppet-zuul: add notification for existing pidfile  https://review.openstack.org/19444917:14
pabelangerso, the location of it is likely wrong, but if people are open to the idea, then we can figure out how to do it17:14
pabelangerfor me, manually copying that over every time is tiresome :)17:14
jeblairpabelanger: that sounds like a good use case (then i can retire my script too)17:14
*** pelix has quit IRC17:15
*** esmute__ has quit IRC17:15
jeblairpabelanger: but assuming you have a local.pp version of site.pp, you could add a stanza like what you propose to that, right?17:15
pabelangerjeblair, right. For the most part, I don't have a local.pp right now.17:16
*** AzherKhan has quit IRC17:16
jeblairpabelanger: oh, do you have a puppetmaster?17:16
pabelangerno, I just do 'puppet apply' and things get installed17:17
nibalizeri sshfs from laptop to testnode so I have my vim and my git configuration17:17
nibalizerbut can quickly iterate17:17
pabelangerright now, for grafana.o.o, I realize a localuser with my sshkey17:17
pabelangerlike you said, I have that info in system-config17:17
nibalizermordred: yes thats a fail message?17:17
*** mrmartin has joined #openstack-infra17:17
jeblairpabelanger: ah, the defaults in site.pp won't actually work for a lot of servers, i think, so generally we expect people to copy them and replace with something17:18
pabelangerjeblair, for the most part, its been okay for me17:18
jeblairyeah, i could see that working for some hosts17:18
TheJuliaSpamapS: yay17:18
pabelangerjeblair, however.... I am thinking we need a testing hiera.yaml included to we can mock up environments17:19
pabelangerneed to spend some time on building that out17:19
pabelangerright now, I don't have hiera.yaml17:19
jeblairpabelanger: check with nibalizer who's working on moving more data into a public hiera file17:19
jeblairpabelanger: http://specs.openstack.org/openstack-infra/infra-specs/specs/public_hiera.html17:20
*** dboik_ has quit IRC17:20
openstackgerritMonty Taylor proposed openstack-infra/system-config: Trim user specific package a bit  https://review.openstack.org/19475117:20
mordrednibalizer: that's in the gitmodule you just proposed - it only supports redhat right now17:20
mordrednibalizer: otherwise, it looks great!17:20
*** AzherKhan has joined #openstack-infra17:20
*** dguerri is now known as dguerri`17:21
nibalizerpabelanger: https://review.openstack.org/#/c/177320/  would have hiera source a directory of data17:21
nibalizermordred: thats only the git class, becuase redhat is bad at installing git17:22
nibalizerdebian doesn't suffer this ailment17:22
nibalizerthe key is the git { 'somerepo': } resource17:22
mordrednibalizer: yah - but that resource autorequires the git class17:22
mordrednibalizer: so the git class needs to not throw an error when it's instantiated17:22
nibalizeri mean I use this thing on debian a lot17:23
*** bdemers has quit IRC17:23
nibalizeri think autorequires just adds a require if the thing is already in the graph|catalog thing17:23
*** esker has joined #openstack-infra17:23
*** esker has quit IRC17:23
mordrednibalizer: AH. gotcha17:23
mordredthat makes sense then17:23
nibalizermordred++ for reivewing libraries before they go in though17:23
*** esmute__ has joined #openstack-infra17:24
*** esker has joined #openstack-infra17:24
mordrednibalizer: fwiw, it seems like it would be dead-simple to add support for debians though, just saying17:24
*** dboik has joined #openstack-infra17:24
*** yamamoto has quit IRC17:24
AJaegerjeblair: it should have gone to SUSE's infra team but somehow thunderbird autocompletion tricked me. Seems I use openstack-infra more often ;)17:25
pabelangernibalizer, will check it out17:25
mordrednibalizer: btw - since public hiera is on the brain today - have you looked at https://review.openstack.org/#/c/179653/3/tools/remap_manifest.py ?17:25
*** mestery has joined #openstack-infra17:25
pabelangernibalizer, do you have public yaml files now?17:25
nibalizeryea, i decided since I can't see the common.yaml file I have no context for what that code does17:25
mordrednibalizer: it's the script I wrote to split an existing hiera file into our new dir structure17:25
mordrednibalizer: well, you can see the gozer one17:25
mordrednibalizer: and the script _Should_ work there too17:26
geoffarnoldQuick process question:  I just submitted https://review.openstack.org/#/c/194738/ to schedule an IRC meeting, and it's passed the tox-ical gate. Is the eavesdrop setup automated, or does it require further action on my part?17:26
*** esmute__ has quit IRC17:26
* nibalizer ponders17:26
mordrednibalizer: concept being that it should read manifests/site.pp, figure out which nodes want which keys, and split the file accordingly17:26
*** yfried has joined #openstack-infra17:26
*** harlowja has joined #openstack-infra17:26
*** patrickeast_ has joined #openstack-infra17:27
*** ildikov has joined #openstack-infra17:27
mordredgeoffarnold: you will still have to start the meeting at meeting time17:27
*** esker has quit IRC17:27
mordredgeoffarnold: btw - I am in favor of enabling the federation of services ...17:27
geoffarnoldThat I understand. Just looking at the infra setup - eavesdrop, wiki, etc.17:28
jeblairwow, we do have meetbot documented here: http://docs.openstack.org/infra/system-config/irc.html#id3 but we did not document the #startmeeting command17:28
jeblairmeetbot is documented here https://wiki.debian.org/MeetBot17:28
jeblairbut we added the meeting name argument, so we should mention that17:29
mordredjeblair: maybe we should document that :)17:29
*** bswartz has quit IRC17:29
*** whitenoise has left #openstack-infra17:29
* jeblair documents17:29
mrmartinfungi, I've fixed the solr mirror pkg deprecated issue in ask-staging.o.o patch, may I ask you to do a review here: https://review.openstack.org/#/c/172880/ ?17:29
*** baoli has joined #openstack-infra17:29
* mordred considers jeblair a fountain of documentation llamas17:29
fungimrmartin: thanks! i'll give it another try shortly17:30
mrmartinfungi: just the solr package we used was removed from the mirrors, so it must work now17:30
fungimrmartin: also while you were off-channel pleia2 had a question you might be able to answer about having a service account for zanata to authenticate users against openstackid17:31
*** rwsu has joined #openstack-infra17:31
mrmartinoh great17:31
openstackgerritMerged openstack-infra/irc-meetings: Add meeting for Magnum Networking Subgroup  https://review.openstack.org/19369417:31
fungiyou got that working with groups.o.o so presumably have a better idea of what's involved than i do17:31
mrmartinyeap, openid works there well17:31
* mrmartin happy to help to finalize zanata integration17:32
pleia2mrmartin: I added it to the meeting, pretty much just looking for how we should handle service accounts WRT openstackid17:32
*** gyee_ has quit IRC17:33
mrmartinwhat you mean on service account?17:33
*** rwsu has quit IRC17:33
SpamapSgreghaynes: looks like TheJulia found a way to make bifrost work for us. :)17:33
openstackgerritJames E. Blair proposed openstack-infra/system-config: Document startmeeting in our meetbot fork  https://review.openstack.org/19475617:33
pleia2mrmartin: an openstackid user for zanata that only runs infra-controlled scripts17:33
jeblairgeoffarnold, mordred: https://review.openstack.org/19475617:33
mrmartinpleia2: I guess we need to create a fake profile entry for that in openstack.org profile17:34
pleia2mrmartin: ok17:34
mrmartinpleia2: but it is the easy part, I guess only web authentication was implemented for actual openstackid version17:34
pleia2zanata supports openstackid already17:35
*** flepied has quit IRC17:35
pleia2so we're fine with auth, this is all about the user we create17:35
AJaegerpleia2: currently I create new projects in transifex. How do you want to do this with Zanata? I'm happy to pass the job on...17:35
*** esmute__ has joined #openstack-infra17:35
pleia2AJaeger: for new ones it may still be manual, StevenK's scripts will automatically pull over the current ones17:35
*** amitgandhinz has quit IRC17:35
mrmartinok, than someone can create the fake account in openstack.org profile, and we have a special role field, that you can assign for the account to mark somehow it is a script, not a real human17:36
*** gyee has joined #openstack-infra17:36
jeblair(eventually we want to automate creating new projects)17:36
geoffarnoldjeblair: This name has to match the "project:" in the yaml file?17:36
pleia2mrmartin: ok, great17:36
jeblairgeoffarnold: no, but it can, and it should be consistent from week to week17:36
*** amitgandhinz has joined #openstack-infra17:36
jeblairgeoffarnold: it will be the name of the directory here: http://eavesdrop.openstack.org/meetings/17:36
*** ghostpl has quit IRC17:36
mrmartinpleia2: an infra-core with an access to openstack.org database can do it quickly, it is just two sql insert17:37
AJaegerpleia2: great to hear for the migration case!17:37
pleia2mrmartin: great, we've started by aiming this at openstackid-dev17:37
mrmartinyeah, it is the same there17:38
pleia2mrmartin: are the user databases synced between -dev and prod?17:38
AJaegerpleia2: for new ones we need to eventually find a work flow and responsible persons17:38
geoffarnoldso "#startmeeting networking_advanced_services" with the underscores, case sensitive17:38
pleia2AJaeger: yeah, and we have an outstanding bug in zanata because currently any user can create projects (aaahhh)17:38
*** amitgandhinz has quit IRC17:38
mrmartinpleia2: I guess it is a manual process, and two db was not synced for a while17:38
*** esmute__ has quit IRC17:38
*** amitgandhinz has joined #openstack-infra17:39
*** flepied has joined #openstack-infra17:39
AJaegerpleia2: oops ;(17:39
pleia2mrmartin: ok, good to know17:39
geoffarnoldjeblair: I'd like to see that requirement actually in the doc update17:39
*** rwsu has joined #openstack-infra17:39
*** ZZelle has joined #openstack-infra17:39
pleia2mrmartin: after I have the account set up I'll give you a ping so I can make the database change, thanks for your help17:40
pleia2AJaeger: yeah, I'm worried it may actually be a show stopper for us with this version of Zanata, they're working on the bug with different roles/admin levels but it'll be in a future version17:41
pleia2but I'm sure we'll bump into more issues when we go into testing mode with the translators in a couple weeks, so I'm plowing on ahead17:41
*** rwsu has quit IRC17:42
*** rwsu has joined #openstack-infra17:43
*** tonytan4ever has joined #openstack-infra17:43
jeblairgeoffarnold: it is lowercased and unusual characters translated to underscores automatically17:44
*** ghostpl has joined #openstack-infra17:44
*** marun has joined #openstack-infra17:44
pleia2doesn't production nodepool need pretty much full database access? re: https://review.openstack.org/#/c/170699/3/doc/source/installation.rst17:44
*** RichardRaseley has quit IRC17:44
geoffarnoldThanks. Just trying to minimize the amount of folklore   :-)17:44
jeblairgeoffarnold: so "#startmeeting Networking Advanced Services" is fine.  or "#startmeeting networking advanced services" will both log to "networking_advanced_services"17:45
jeblairgeoffarnold: i'll update the docs17:45
*** linkedinyou has quit IRC17:47
openstackgerritJames E. Blair proposed openstack-infra/system-config: Document startmeeting in our meetbot fork  https://review.openstack.org/19475617:47
jeblairgeoffarnold: ^17:47
*** patrickeast_ has quit IRC17:47
*** sergmelikyan has quit IRC17:49
AJaegerpleia2: yeah, let's move forward and see what else we need to get fixed..17:49
*** ghostpl has quit IRC17:49
*** bdemers has joined #openstack-infra17:49
*** armax has joined #openstack-infra17:50
mrmartinjeblair: if I like to create a specs repo for the groups portal, am I need to get an approval from TC? or just simply create the patch that infra can approve?17:52
*** bswartz has joined #openstack-infra17:52
jeblairmrmartin: since groups is part of the infra project, you can use the infra-specs repo and run it through the infrastructure council process: http://docs.openstack.org/infra/system-config/project.html#teams17:54
jeblairmrmartin: and also here: http://specs.openstack.org/openstack-infra/infra-specs/readme.html17:54
*** fawadkhaliq has joined #openstack-infra17:54
mrmartingreat, thnx17:55
*** liam__ has joined #openstack-infra17:55
*** amitgandhinz has quit IRC17:56
jeblairmordred: have you been doing ipv6 related things in nodepool?17:56
jeblairmordred: should you review this?  https://review.openstack.org/15617817:57
openstackgerritMerged openstack-infra/nodepool: Convert to use latest statsd version  https://review.openstack.org/17189017:57
*** juggler has quit IRC17:57
jeblairmordred: do you want to clean up https://review.openstack.org/136589 or should we find a new owner?17:58
*** amitgandhinz has joined #openstack-infra17:58
*** esmute__ has joined #openstack-infra17:58
jeblairgreghaynes: what do you think the future holds for https://review.openstack.org/177043 ?17:59
*** mpaolino has joined #openstack-infra17:59
mordredjeblair: looking17:59
*** mtanino has quit IRC18:00
*** mestery has quit IRC18:00
mordredjeblair: I think the approach in that patch is a fine one - I'm going to similarly make a parameter in shade that you can specify if you want ipv6 instead of ipv4 for some things18:00
*** mestery has joined #openstack-infra18:01
*** bdemers has quit IRC18:01
mordredjeblair: although I also want to just look in /proc/net/if_inet6 to see if there are any routable ipv6 interfaces, and if so, use ipv6 by default18:01
mordredbut that might be too aggressive18:01
openstackgerritMerged openstack-infra/shade: Convert ironicclient node.get() call to Task  https://review.openstack.org/19353518:02
openstackgerritSergey Skripnick proposed openstack-infra/project-config: Fix experimental rally-cue job  https://review.openstack.org/19476818:02
*** mwagner_afk has quit IRC18:02
*** bdemers has joined #openstack-infra18:02
asselinperisteri, hi...can you elaborate more on your suggestion?18:02
*** otter768 has joined #openstack-infra18:02
peristeriasselin, A document "How to get openstack CI working for you". In other words, what's in it for third parties.18:02
openstackgerritMerged openstack-infra/nodepool: Remove unused 'reset' setting from images  https://review.openstack.org/16513618:03
*** ociuhandu has quit IRC18:03
*** kdas_ is now known as kushal18:03
*** kushal has quit IRC18:04
*** kushal has joined #openstack-infra18:04
*** ericksonfgds is now known as ericksonsantos18:04
*** ivar-lazzaro has joined #openstack-infra18:04
*** enikanorov2 has quit IRC18:05
*** armax has quit IRC18:05
*** achanda_ has joined #openstack-infra18:06
asselinperisteri, ok yeah..there's a lot of information documented but not that easy to find.18:06
asselinperisteri, what do you mean by " what's in it for third parties."?18:06
SpamapSgreghaynes: so, good news, I can get bifrost to boot _one machine_. Bad news: our servers ignore ipmi commands to change boot order, so we'll have to be careful about that as we boot all the rest. ;)18:07
peristeriasselin, The document can be an overview of the current architecture and maybe point out areas of interest where interested people. Like a informal introduction documentation18:07
*** jasondotstar has quit IRC18:07
*** otter768 has quit IRC18:07
*** esmute__ has quit IRC18:07
peristeriei: a walk-through18:07
mordredjeblair: I +2d the nodepool change, but did not +A in case you wanted to look18:08
*** e0ne has joined #openstack-infra18:08
asselinperisteri, this is a really good blog and what I started with:http://www.joinfu.com/2014/01/understanding-the-openstack-ci-system/18:08
jeblairmordred: yeah, i have not reviewed, just skimming changes18:08
*** achanda has quit IRC18:08
asselinperisteri, some details are out of date, but the basics are still relevant18:09
*** mestery has quit IRC18:10
*** tosky has quit IRC18:11
harlowjacould i request a infra folk to look over https://review.openstack.org/#/c/194367/ so that it can go in and merge and unblock some of those libraries gates :)18:11
harlowja*or even 2 infra folk(s)18:11
harlowjaor 3!18:11
harlowja*but no more, lol18:11
*** armax has joined #openstack-infra18:13
jeblairharlowja: there are 2 failing non-voting jobs; do you know if they are benign?18:13
peristeriasselin, the post looks like a great reference document.18:14
*** jasondotstar has joined #openstack-infra18:14
*** mpaolino has quit IRC18:14
*** esmute__ has joined #openstack-infra18:15
harlowjajeblair i would assume so, since those libraries aren't used in juno18:15
krtaylorperisteri, have you seen http://docs.openstack.org/infra/system-config/third_party.html18:15
mgagneIn nodepool, this part assumes the public IP address can be found in the network named public (or private): https://github.com/openstack-infra/nodepool/blob/master/nodepool/provider_manager.py#L4018:15
mgagnewould it be reasonable to make it configurable?18:15
krtaylorperisteri, and http://docs.openstack.org/infra/system-config/running-your-own.html18:15
greghaynesSpamapS: \O/18:15
mgagneespecially since you can override the network on which the nodes are spawn18:16
jeblairmgagne: chat with mordred or SpamapS about that; i believe that logic is moving to shade18:16
greghaynesSpamapS: well, you only need to boot one for now to get a clean new bm host up, right?18:16
harlowjajeblair and or they seem to be 'timeout: timed out' so timeout stuff talking to cirros18:16
*** bexelbie_ has joined #openstack-infra18:16
mgagneI don't know shade, I only trying to make it work18:16
*** fawadkhaliq has quit IRC18:16
*** ghostpl has joined #openstack-infra18:17
*** achanda_ has quit IRC18:17
SpamapSmgagne: yes, shade is growing smarter methods for that.18:17
*** amotoki has quit IRC18:17
*** achanda has joined #openstack-infra18:17
peristerikrtaylor, yes. But there's a lot of implicit knowledge required to understand. The post asselin can complement this document.18:18
*** armax has quit IRC18:18
jeblairharlowja: i don't know openstack; if you're sure they are benign, leave a comment indicating that so i can point to that if things go wrong and i'll +2 it.  :)18:18
mordredmgagne: yah - so - the logic to make it work is VERY long once you get in to it - we're just now landing it in the shade library18:19
SpamapSmgagne: nodepool will consume shade rather than implement this privately.18:19
mordredmgagne: it took us at least a good month of wrangling to finally figure out all of the ways in which clouds make finding teh ip of a server difficult18:19
mordredmgagne: but yes - I agree with you - the current code in nodepool is deficient18:19
mgagnealright, that's good stuff I guess. I'm trying to make it work today with my cloud. =)18:19
mgagneI guess I'll just fork and wait18:20
mordredmgagne: does your cloud fill in the AccessIPv4 value in nova?18:20
*** enikanorov2 has joined #openstack-infra18:20
jeblairmgagne: what's the name of your cloud's network?18:21
mgagneno. we manage the cloud and the network name is non-deterministic18:21
krtaylorperisteri, ok, also see: https://www.ibm.com/developerworks/community/blogs/fe313521-2e95-46f2-817d-44a4f27eba32/entry/building_your_openstack_3rd_party_ci_system1?lang=en18:21
mgagneit depends on the tenant and for various reason, it cannot be renamed18:21
mordredmgagne: are you using neutron? and does the network in question support the router:external attribute?18:21
peristerikrtaylor, I'm looking at it right now.18:21
mgagnewe don't use router18:22
mgagneit's pure provider networks with hardware routers18:22
jeblairmgagne: that makes sense for my question; but is that also an answer to mordred's question about AccessIPv4?18:22
mordredbut you should have a neutron network object, yes?18:22
mgagnethere is no AccessIPv4 populated18:22
openstackgerritMerged openstack-infra/project-config: Create puppet-openstack-integration repo  https://review.openstack.org/19428718:22
harlowjajeblair ok, i signed the release form (aka the review) for u18:22
mgagnebecause AFAIK Nova uses a fixed network name to populate it18:22
mordredmgagne: that is correct -so if you have one per-tenant, it's not going to work with that nova code18:23
jeblairharlowja: stamped and approved!  :)18:23
*** esmute__ has quit IRC18:23
mgagnethat's why I was wondering if net-label found in nodepool.yaml could be used to fetch the IP address18:24
mordredmgagne: no reason it couldn't, since it's already there - that's a find idea18:24
mtreinishjeblair: any thoughts on https://review.openstack.org/#/c/194302/ I think I finally got the logic right18:24
mordredmgagne: if you neutron net-show your network, do you get something liek this:18:24
mordredmgagne: http://paste.openstack.org/show/31731918:24
*** yamamoto has joined #openstack-infra18:25
openstackgerritRyan Bourgeois proposed openstack-dev/pbr: Support git://, git+ssh://, git+https:// without -e flag. Closes-Bug: 1467683 Change-Id: I09163c8072ecbee5c1a24951511c75009511b6f8  https://review.openstack.org/19440518:25
openstackbug 1467683 in PBR "PBR Ignores git://, git+ssh://, git+https:// Requirement Lines" [Wishlist,Confirmed] https://launchpad.net/bugs/1467683 - Assigned to Ryan Bourgeois (bluedragonx)18:25
*** dangers_away is now known as dangers18:25
mordredhrm. ok18:27
mordredmgagne: well, ok then - we may have to get fancier to support you - you have successfully added a new use case18:27
*** e0ne is now known as e0ne_18:27
mordreddguerri`: ^^18:27
mgagnemordred: my use case looks to be supported when creating the node but not after when the validation is done :-/18:28
openstackgerritMerged openstack-infra/shade: Convert ironicclient node.update() call to Task  https://review.openstack.org/19355318:28
*** dguerri` is now known as dguerri18:28
mordredmgagne: I think we can't only assume net-id is the network to use for the public ip, because on the tripleo cloud and on HP, we use netid to tell it what network to boot on - but the VM still needs a floating ip18:28
mordredso the ip from the specified network in both of those cases would be wrong18:28
mordredI think we'd need another option, as you originally suggested - that is maybe "interface-network" or something - that can be configured to tell it what network to use to get the IP it should use to talk to the VM18:29
*** dguerri is now known as dguerri`18:29
mordredmgagne: I'd like to ponder that for a just a few to make sure we're not just adding cruft - but I think it's likely the only way18:29
*** yamamoto has quit IRC18:30
mordredmgagne: this may actualy want to be an os-client-config option as well18:30
*** angela-s has joined #openstack-infra18:30
mgagneit also goes to show that the accessIPv4 extension in Nova doesn't support all use cases =)18:30
*** juggler_ has joined #openstack-infra18:31
*** e0ne_ has quit IRC18:32
*** isviridov is now known as isviridov_away18:32
mordredoh golly18:33
mordredmgagne: http://git.openstack.org/cgit/openstack-infra/shade/tree/shade/meta.py#n73 this is what we have so far18:33
mordredmgagne: it does not support your use case yet18:34
*** bexelbie_ has quit IRC18:34
mordredso - I think the logic change would be "if network_name: search_networks(network_name): else: search_networks(filters={'router:external': True})18:34
mgagneI think I would be happy with a config for L10318:34
mordredor yeah - we could just make key_name a param that defaults to public18:35
mordredand not have to do more new logic18:35
mordredI like that18:35
*** kushal has quit IRC18:35
greghaynesmtreinish: dropped some unimportant bash-fu on that patch ;)18:35
*** reed has quit IRC18:36
mgagnemordred: get_server_external_ipv4 vs get_server_public_ip  ?18:36
*** markmcclain has quit IRC18:36
mtreinishgreghaynes: heh, yeah my bash-fu is weak18:36
mtreinishI just iterate until I find something that works18:36
mtreinishthen I let other people tell me how I really should do it18:37
mordredmgagne: yah - some people have ipv6 even :)18:37
*** puranamr has quit IRC18:37
*** ryanpetrello has quit IRC18:37
mgagnemordred: my network is private so fallback to is_globally_routable_ipv4 would fail too18:37
*** puranamr has joined #openstack-infra18:37
mordredthis makes sense18:38
mordredmgagne: we talked some about your use case before, but we had no concrete examples, so we did not handle it (didn't want to get extra complex if we didn't need to)18:38
mordredbut- I think it's very easy to support you18:38
greghaynesmtreinish: hah, well then success!18:38
mgagnesure, I understand18:38
mordredin fact...18:38
mordredwe could start using get_server_external_ipv4 in nodepool pretty soon18:39
mgagneIMO, it's not an unusual case if people are running private cloud to run their CI infra18:39
mordredmgagne: yah - I think we were mostly thinkig that they'd still want the IP that gets them _off- of their cloud - but we were wrong18:39
mordredShrews: ^^ if you missed it18:40
openstackgerritMatthew Treinish proposed openstack-infra/devstack-gate: Fix the setup_workspace to handle missing ZUUL_BRANCH  https://review.openstack.org/19430218:40
mtreinishgreghaynes: ^^^ I had to recheck anyway18:41
lifelessdims: dhellmann: mordred: reviews please https://review.openstack.org/#/c/189250/18:41
*** e0ne has joined #openstack-infra18:41
*** puranamr has quit IRC18:41
lifelessthats an empirical review18:41
lifelesse.g. I found out via trial and error18:41
mgagnemordred: It however looks like you *could* have multiple networks in the provider config which could make the logic a bit more difficult.18:42
*** marzif_ has quit IRC18:43
openstackgerritMonty Taylor proposed openstack-infra/shade: Allow specifying the name of the network for IPs  https://review.openstack.org/19477918:44
mordredmgagne, Shrews, SpamapS: ^^18:44
mgagnemordred: unless we add an attribute to the network definition in nodepool.yaml to tell that this particular network will be used to access the instance via SSH18:44
*** puranamr has joined #openstack-infra18:44
mordredmgagne: well, I'm thinking actually adding a setting to clouds.yaml - it's an attribute of the cloud account, really. "what network should be used for ips"18:45
mgagnemordred: I'm just not sure how we can retrieve that info later, maybe storing it in the instance metadata18:45
Shrewsmordred: this IP logic is getting quite complex18:45
mgagneit is madness18:46
mordredmgagne: and yes - we can totally set it in the instance metadata if need be - although all of our interactions with the server will be via things that have called get_server_public_ipv4 - so it'll always be in teh server dict we have18:46
mordredShrews: yes. clouds are hard18:46
*** puranamr has quit IRC18:46
mgagnemordred: make_server_dict(server) is server a Nova "object" ?18:47
*** puranamr has joined #openstack-infra18:48
mordredmgagne: yes18:48
*** esmute__ has joined #openstack-infra18:48
mordredmgagne: well, it's one stripped down to be a dict, but yes - all the fields should be the same18:48
mordredmgagne: except for ones we add to the dict in shade18:48
mgagnethanks, I'll see what I can do18:49
* mgagne pulls out duct tape18:49
*** puranamr has quit IRC18:50
*** rwsu has quit IRC18:51
*** nihilifer has joined #openstack-infra18:52
*** dguerri` is now known as dguerri18:53
jogowould it be possible to get gdb attached to nova-compute to help debug https://bugs.launchpad.net/nova/+bug/146230518:53
openstackLaunchpad bug 1462305 in OpenStack Compute (nova) "multi-node test causes nova-compute to lockup" [High,In progress] - Assigned to Joe Gordon (jogo)18:53
*** rwsu has joined #openstack-infra18:53
mtreinishjogo: just do it with devstack18:53
jogowill first have to find a case where that bug is happening for a job that hasn't finished yet18:53
mtreinishthat's what launches nova18:53
dguerrimordred mgagne: interesting use case :)18:53
jogomtreinish: well this only happens in multinode and happens infrequently18:54
mgagnetranslation: you guys are weird :D18:54
jogomtreinish: I could run gdb every time I  guess18:54
mordredmgagne: :)18:54
jogojust like the GMR stuff18:54
mtreinishjogo: just a wip, we have the multinode on experimental on devstack right?18:54
*** dguerri is now known as dguerri`18:54
mtreinishjust recheck until it hits18:54
*** baoli has quit IRC18:54
jogomtreinish: yes, but this doesn't happen that often18:54
jogowould have to run 10/20 jobs18:54
mtreinishjogo: self replicating patch?18:55
jogomtreinish: hehe that is an option18:55
mtreinishyou totally should do that for your last day18:55
*** baoli has joined #openstack-infra18:56
*** puranamr has joined #openstack-infra18:56
*** bdemers has quit IRC18:56
*** bdemers has joined #openstack-infra18:57
jogoI would rather get multinode working18:58
*** shardy has quit IRC19:00
*** DericHorn-HP has quit IRC19:00
*** shardy has joined #openstack-infra19:00
pleia2meeting time19:01
*** DericHorn-HP has joined #openstack-infra19:02
*** ramashri has joined #openstack-infra19:05
*** rwsu has quit IRC19:05
*** maxskew has quit IRC19:07
*** DericHorn-HP_ has joined #openstack-infra19:07
*** Longgeek_ has quit IRC19:07
*** DericHorn-HP has quit IRC19:07
*** DericHorn-HP_ is now known as DericHorn-HP19:07
*** esmute__ has quit IRC19:08
*** esmute__ has joined #openstack-infra19:10
*** markmcclain has joined #openstack-infra19:11
*** shardy has quit IRC19:11
*** shardy has joined #openstack-infra19:12
*** markmcclain has quit IRC19:12
*** markmcclain has joined #openstack-infra19:13
*** reed has joined #openstack-infra19:13
*** rwsu has joined #openstack-infra19:13
SpamapSgreghaynes: success! I have a box up now, and I can document how to repeat.19:16
SpamapSmordred: ^19:16
greghaynes\O/ we can boot computers19:16
mordredSpamapS: woot!19:16
*** tim_o has joined #openstack-infra19:17
greghaynesIts sad how much effort it takes to do that19:17
greghaynesbut yay19:17
timrcjeblair, I'm having mega connection issues, but yeah, I'm interested in reviewing those patches.19:18
SpamapSgreghaynes: especially since we spent 2 years doing that.. on these specific computers. ;)19:18
timrcyolanda, nibalizer Congrats on infra-puppet core!19:19
*** openstackgerrit has quit IRC19:21
*** dizquierdo has joined #openstack-infra19:22
*** openstackgerrit has joined #openstack-infra19:22
*** esmute__ has quit IRC19:23
*** esmute__ has joined #openstack-infra19:24
*** ryanpetrello has joined #openstack-infra19:24
*** yamahata has joined #openstack-infra19:24
*** achanda has quit IRC19:26
*** jtomasek has quit IRC19:26
davideagnellohello, we have a change that needs to get merged for our rally job to work:  https://review.openstack.org/#/c/194768/  Thank you19:26
*** husanu has quit IRC19:26
*** husanu has joined #openstack-infra19:28
timrcpabelanger, I'll start looking at grafyaml patches today.19:28
pabelangertimrc, cool. I think it is in good shape19:28
*** sabeen2 has quit IRC19:29
pabelangerwell, the puppet-grafana patchset is too19:29
*** sabeen has joined #openstack-infra19:29
*** EricGonczer_ has joined #openstack-infra19:30
*** ramashri has quit IRC19:30
timrcpabelanger, I fell off the face of the earth for a minute working on internal stuff, but I'm hoping the internal stuff that was done will make doing the external stuff easier now :)19:30
*** fifieldt_ has joined #openstack-infra19:31
pabelangertimrc, Ya, I hear that19:32
mordredtimrc: \o/19:32
*** lhcheng has joined #openstack-infra19:33
lhchenghi folks19:33
*** fifieldt has quit IRC19:34
lhchengit has been reported than one of the horizon pypi packages contains .git files. Is this something infra can help with?19:34
openstackLaunchpad bug 1451567 in OpenStack Dashboard (Horizon) "XStatic-Magic-Search pypi package has errors" [High,Confirmed]19:34
*** SumitNaiksatam has joined #openstack-infra19:34
lhchengthe author of the package is on vacation19:34
david-lylelhcheng: I think we can just tag a new version of that and let the publish job post the correct parts19:35
*** esmute__ has quit IRC19:35
clarkbwell bug is likely in setup.py so that needs fixing too19:35
fungilhcheng: david-lyle: yeah, just make sure that you check the tarball created by running `tox -e venv python setup.py sdist`19:36
fungiand fix whatever needs fixing in that repo until it gives you a tarball whose contents you're satisfied with19:36
lhchengfungi: thanks, let me try that out first.19:36
fungilhcheng: david-lyle: since that's what our tarball job will run, and what it will ultimately upload when you push a new tag19:37
*** esmute__ has joined #openstack-infra19:37
*** Longgeek has joined #openstack-infra19:38
*** dprince has quit IRC19:38
*** puranamr has quit IRC19:38
*** puranamr has joined #openstack-infra19:39
mgagnemordred: my slaves are now added to Jenkins so I'm happy: https://github.com/mgagne/nodepool/tree/access_network19:39
pabelangertimrc, just created https://github.com/grafana/grafana/issues/2218 to help with api_key issue19:40
mordredmgagne: yay!19:40
mgagnepabelanger: "this is convenient" :D19:40
*** tonytan4ever has quit IRC19:40
mgagneoh misread :P19:41
*** rwsu has quit IRC19:41
*** tonytan4ever has joined #openstack-infra19:41
lhchengfungi david-lyle: running `tox -e venv python setup.py sdist` returned an error: "ERROR: toxini file 'tox.ini' not found"19:42
*** esmute__ has quit IRC19:42
lhchengdavid-lyle: maybe we don't have a publish job yet ?19:42
lhchengdavid-lyle: and the author just manually uploaded the package to pypi19:42
*** puranamr has quit IRC19:43
david-lylemaybe that's the reason microversion aren't getting published either19:43
*** achanda has joined #openstack-infra19:43
david-lylethose horizon guys are nuts19:44
*** jtomasek has joined #openstack-infra19:44
fungidavid-lyle: i've heard that too. if you know any of them, we can probably set them straight ;)19:45
lhchengclarkb: setup.py seems consistent with other xstatic packages, either it is correct or all setup.py are wrong :P19:45
*** rwsu has joined #openstack-infra19:45
*** tonytan4ever has quit IRC19:45
lhchengfungi: yeah, those horizon guys are trouble makers19:46
*** tonytan4ever has joined #openstack-infra19:46
*** hashar has joined #openstack-infra19:46
fungilhcheng: well, the setup.py may be fine. with no tox.ini though our tarball jobs won't run. they call tox19:47
asselinperisteri, fyi, use my fork of the repo since it works/actively maintained: https://github.com/rasselin/os-ext-testing. I've been slowly migrating it to use the new puppet-openstackci modules19:47
peristeriasselin, thanks.19:48
lhchengfungi: check couple of  xstatic-* packages, and all doesn't have a tox.ini file.  hmm..  so we didn't have any publish jobs at all :(19:50
*** tonytan4ever has quit IRC19:51
*** rwsu has quit IRC19:51
*** Rockyg has joined #openstack-infra19:51
david-lylethat explains the my xstatic issue from yesterday19:51
david-lylelooks like we have some tox-ing to do19:51
*** tonytan4ever has joined #openstack-infra19:52
*** sdake has quit IRC19:54
*** esmute__ has joined #openstack-infra19:56
*** _nadya_ has joined #openstack-infra19:57
dhellmannhave there been any issues with gerrit being very very slow today? I'm trying to figure out if the timeouts I'm seeing are related to something on my end of the connection.19:59
openstackgerritSylvain Bauza proposed openstack-infra/elastic-recheck: add query for py27 bug 1468087  https://review.openstack.org/19479719:59
openstackbug 1468087 in OpenStack Compute (nova) "py27 job failing on test_save_objectfield_missing_instance_row" [High,Confirmed] https://launchpad.net/bugs/146808719:59
bknudsongerrit is a slowpoke20:00
dansmithdhellmann: it's not you20:00
dansmithwell, not just you.. you might be the overall cause I guess, I dunno :P20:00
dhellmanndansmith: ok, good, thanks20:00
dhellmanndansmith: sssh, don't tell anyone20:01
bknudson"who smelt it dealt it"20:01
*** mrmartin has quit IRC20:01
fungii have a sinking suspicion something is going on with the compute node where out gerrit vm is running20:01
fungithough rackspace didn't spot anything wrong20:01
mtreinishfungi: when you get a sec could you take a look at: https://review.openstack.org/194302 I'd really like to get it in before the next round of periodic jobs kick off20:03
*** otter768 has joined #openstack-infra20:03
*** weshay has quit IRC20:03
*** annegentle has joined #openstack-infra20:04
*** weshay has joined #openstack-infra20:04
*** cburroughs has quit IRC20:05
*** e0ne is now known as e0ne_20:05
*** e0ne_ is now known as e0ne20:05
*** achanda has quit IRC20:06
openstackgerritMatthew Treinish proposed openstack-infra/project-config: Add python3 jobs to os-testr  https://review.openstack.org/19479820:06
*** reed has quit IRC20:07
*** jtomasek has quit IRC20:07
fungimtreinish: lgtm but should also get jeblair to weigh in since he untangled this the last time we made okay-looking changes to the branch fallbacks20:08
*** reed has joined #openstack-infra20:08
*** greghaynes has quit IRC20:08
*** kgiusti has quit IRC20:08
*** otter768 has quit IRC20:09
*** _nadya_ has quit IRC20:09
*** _nadya_ has joined #openstack-infra20:09
mtreinishfungi: ok, thanks. Don't worry, I'll bug jeblair about it again :)20:10
*** ramashri has joined #openstack-infra20:12
*** puranamr has joined #openstack-infra20:12
bknudsonare there docs for how to merge a feature branch?20:14
*** _nadya_ has quit IRC20:14
*** ryanpetrello has quit IRC20:14
*** sabeen2 has joined #openstack-infra20:14
bknudsonfound it http://docs.openstack.org/infra/manual/drivers.html20:14
*** ryanpetrello has joined #openstack-infra20:15
clarkblhcheng I am interested to know if setup.py makes sdists with .git20:16
*** tonytan4ever has quit IRC20:16
lhchengclarkb: nope it doesn't.   It seems like the xstatic-* directory was uploaded as tarball instead of just the tar from sdist20:18
*** annegentle has quit IRC20:18
*** sabeen has quit IRC20:18
*** annegentle has joined #openstack-infra20:18
*** SumitNaiksatam has quit IRC20:18
lhchengclarkb: the problem is just for xtatic-magic-search btw.20:18
timrcpabelanger, Will you be at OSCON this year?20:19
openstackgerritMatthew Treinish proposed openstack/requirements: Cap oslo.il8n to allow for transition away from oslo namespace  https://review.openstack.org/19480020:19
mtreinishdims: ^^^ you broke the world20:20
pabelangertimrc, sadly no20:20
*** Sukhdev has joined #openstack-infra20:20
pabelangertimrc, sounds like a few -infra people are going20:20
fungitimrc: i'll be there20:20
timrcfungi, Cool.  Me too.20:20
mtreinishdhellmann, sdague, fungi, lifeless: ^^^ that g-r review should probably be fast tracked20:20
pleia2I'll be around for mon-tues20:20
timrcpleia2, Are you presenting?20:21
lifelessmtreinish: please no20:21
lifelessmtreinish: I've finally gotten my conflicts from hell patch +A'd, don't ruin it :)20:21
pleia2timrc: no, and not strictly oscon-ing either, doing community leadership summit over the weekend + beer track mon-tues20:21
lifelessmtreinish: unless- has that broken the world ?20:21
mtreinishlifeless: heh, well you won't be able to land them20:21
mtreinishbecause that broke the world20:21
*** SumitNaiksatam has joined #openstack-infra20:22
lifelessdhellmann: ^ we need to stop the namespace transition until constraints are in place I think20:22
crinklebeer track \o/20:22
openstackgerritBrant Knudson proposed openstack-infra/project-config: Grant keystone core pushMerge on keystoneauth feature branch  https://review.openstack.org/19480120:22
*** jasondotstar has quit IRC20:22
lifelessdhellmann: because we're going to get more such surprises, statistically speaking20:22
pleia2crinkle: yes, we should gather people and beer track together!20:22
crinklepleia2: ++20:22
lifelessmtreinish: ok so we need a root person to unwedge the world?20:23
mtreinishlifeless: well maybe, hopefully not. The g-r cap should stop pulling in a version without the oslo namespace which will let us fix things20:23
lifelessright but everything currently queued is going to fail ?20:24
mtreinishassuming it's just oslo.il8n which broke things20:24
lifelessqueued-in-the-gate I mean20:24
mtreinishwhich it isn't as jecarey just pointed out in -qa it wasn't20:24
mtreinishlifeless: that ship has already sailed all jobs that get the new versions from pypi that dims just pushed are hosed20:25
mtreinishif it launched already and pulled in the old version it'll still pass20:25
*** jogo has quit IRC20:25
lifelessmessaging, db, middleware i18n are the releases I see with namespace in the changelog20:25
*** jogo has joined #openstack-infra20:26
lifelessyeah, I'm trying to figure if 189250 is in or out :)20:26
mtreinishlifeless: also log I think20:26
openstackgerritSteve Martinelli proposed openstack/requirements: Remove marconiclient from global requirements  https://review.openstack.org/19480420:27
lifelessmtreinish: log only says20:27
lifeless2e2100f Updated from global requirements20:27
lifelessc7f354d Updated from global requirements20:27
lifelessbf0c020 Switch badges from 'pypip.in' to 'shields.io'20:27
lifelessdcd480a Deprecate use-syslog-rfc-format for removal20:27
sdaguemtreinish: well if gerrit didn't keep being "the sad" I might be able to20:28
mtreinishlifeless: yep going through the release notes now t20:28
mtreinishsdague: i need to respin anyway to add a cap to all the new oslo releases20:28
pabelangerreview.o.o issues?20:28
pabelangertiming out here20:29
sdaguemtreinish: so, is there no other way around this? because these caps have made all the integration infrastructure pretty fragile right now20:29
dhellmannmtreinish, lifeless : I can't get gerrit to load that page. What project is actually broken?20:29
sdagueyou basically can't run grenade outside of the gate20:29
*** rguillebert has joined #openstack-infra20:29
sdagueand devstack outside of the gate is getting dicey as well20:29
mtreinishdhellmann: well right now oslo.il8n, because there a bunch of things still using the namespace20:30
mtreinishlike glance_store: http://git.openstack.org/cgit/openstack/glance_store/tree/glance_store/i18n.py20:30
dhellmannmtreinish: right, what is still using it, because I have been doing a lot of testing20:30
*** rossella_s has quit IRC20:30
mtreinishpatrickeast: said there were some other clients too20:30
jeblairfungi, pleia2: seem to be having another gerrit partial-outage; i can't dig in right now20:30
mtreinishdhellmann: I don't have the full list20:30
dhellmannmtreinish: ok, maybe they didn't land the patch I gave them like I thought they did20:30
dhellmannmtreinish: ok, I'll look again after my meetings are over, but go forth with capping for now20:31
lifelessdhellmann: from the message (I know you can't get it) - glance_store and a bunch of the clients usage of oslo.il8n.20:31
openstackgerritMerged openstack-infra/elastic-recheck: add query for py27 bug 1468087  https://review.openstack.org/19479720:31
openstackbug 1468087 in OpenStack Compute (nova) "py27 job failing on test_save_objectfield_missing_instance_row" [High,Confirmed] https://launchpad.net/bugs/146808720:31
*** rossella_s has joined #openstack-infra20:31
dhellmannlifeless: I submitted patches to all those things. Maybe we just need to release them.20:31
mtreinishdhellmann: https://review.openstack.org/#/c/194800/ is the cap for il8n, I'm trying to figure out if I need to respin to add the other namespace removal releases20:31
*** baoli has quit IRC20:31
mtreinishI think it might be just il8n20:32
*** dontalton has quit IRC20:32
fungijeblair: i'm looking20:32
mtreinishsdague: yeah we have to cap, or push a new oslo.i8ln release to add the namespace back20:32
*** dontalton has joined #openstack-infra20:33
*** greghaynes has joined #openstack-infra20:33
dhellmannmtreinish: cap, please20:33
mtreinishif it's more than 1 project that is still using the namespace20:33
*** greghaynes has quit IRC20:33
fungiwe're getting tcp/syn packets destined for port 443, but not any more than usual20:33
pleia2fungi: seems to be going in and out, of being accessible (I can get to it now)20:33
pleia2looking at cacti to see if unusual load20:33
*** claudiub has quit IRC20:34
*** greghaynes has joined #openstack-infra20:34
fungiyeah, for the moment it's responding to me just fine20:34
pleia2yeah, tcp seems to be the only thing that spiked, maybe some networking issues where it lives?20:35
fungiyou know what? the same crawler i was seeing tank our paste server also looks like the main source of new https connections to gerrit20:35
openstackgerritMichael Krotscheck proposed openstack-infra/project-config: Added ironic-webclient to OpenStack  https://review.openstack.org/19442420:35
fungidouble-checking that now since i haven't memorized the ip address20:35
fungiyep! blocking now20:36
pleia2nice work20:36
fungiit was varying between 15-25% of all new connections per tcpdump20:37
mtreinishnow I'm not seeing what I was looking at before. Maybe I just misread something20:37
mtreinishI'm not sure we need the cap anymore20:37
fungino idea if this is the cause of our recent pain, but worth blocking regardless20:37
* pleia2 nods20:37
dhellmannmtreinish: glance_store is not using the oslo namespace in the file you linked to -- do you have a link to a log with a failure?20:39
*** esker has joined #openstack-infra20:39
lifelessdhellmann: but has it been released not using it ?20:39
*** rwsu has joined #openstack-infra20:39
dhellmannlifeless: that's not clear, yet, that's why I want to see the actual traceback20:40
mtreinishdhellmann: yeah I just noticed that, I swear I saw from oslo import il8n somewhere in a glance_store repo, but I already marked the patch as wip20:40
mtreinishI might have jumped the gun20:40
*** Rockyg has quit IRC20:40
mtreinishdhellmann: all i have was this: http://paste.openstack.org/show/317614/20:40
mtreinishwhich patrickeast reported in -qa20:40
dhellmannlifeless: version 0.4 should have had the fix20:41
mtreinishdhellmann: yeah I think it's just patrickeast's env20:41
sdaguemtreinish: yeh, I thought we were pretty on top of this20:41
*** Rockyg has joined #openstack-infra20:41
lifelesspatrickeast: do you have the full log ?20:41
dhellmannmtreinish: is this from a grenade job where something didn't get upgraded?20:41
*** DericHorn-HP has quit IRC20:41
lifelessit could be a downgrade20:41
patrickeastyea one sec20:41
*** dustins_ has quit IRC20:41
lifelessbecause we still have lots of caps all over the place that trigger random downgrades20:41
dhellmannmtreinish: are we actually seeing gate failures?20:41
sdaguemtreinish: yeh, I expect that's probably related to the pip caching issue20:41
lifelesssdague: pip caching issue ?20:42
sdaguelifeless: something is sticking in environments20:42
*** rm_work is now known as rm_work|away20:42
sdaguewe've had a few reports of it, it hit me, but I blew away some things to get past it20:42
*** DericHorn-HP has joined #openstack-infra20:42
lifelessI'm happy to dig in if someone can loop me in20:43
lifelesssdague: virtualenvs I presume you mean?20:43
*** dkranz has quit IRC20:44
dhellmannlifeless: fwiw, I've been holding oslo releases until all of the other libraries I've patched have had releases, so it's possible I missed one but I've spent hours and hours running tests over and over on these changes20:44
ianwjhesketh: http://i.imgur.com/4LygAIy.png <- are these weird chars in the log page my problem or remote config?20:44
*** claudiub has joined #openstack-infra20:44
*** ddieterly has joined #openstack-infra20:45
mtreinishdhellmann: yeah I'm pretty sure I just jumped the gun because I misread something and that pastebin20:45
mtreinishI don't see any gate failures20:45
dhellmannmtreinish: ok, good20:45
mtreinishso I owe dims an apology for blaming him for breaking the world when nothing is broken:)20:45
dhellmannmtreinish: :-)20:45
lifelessdhellmann: totally - I was assuming a coverage gap, not an effort gap20:45
* dhellmann takes off his fire helmet and goes back to the tc meeting20:46
patrickeast:o sry for the disruption20:46
*** dimtruck is now known as zz_dimtruck20:46
patrickeastthe whole log is here if anyone is still itnerested http://ec2-54-67-51-189.us-west-1.compute.amazonaws.com/err.log20:46
dhellmannlifeless: I have a script that checks out all of openstack/*, figures out what uses what libs, and runs all of their unit tests.20:46
dhellmannlifeless: it takes forever to run :-)(20:46
lifelessdhellmann: yeah, you need map-reduce20:46
lifelessI have a thing20:46
*** puranamr has quit IRC20:46
dhellmannlifeless: yeah, I need to spin up more vms to do that or something, but I've been focusing on one lib at a time so it's mostly working -- run job, go to bed, wake up, read logs20:47
sdaguelifeless: no, it was system level20:47
lifelesspatrickeast: 2015-06-23 19:55:33.363 | Collecting glance-store>=0.3.0 (from glance==
lifeless2015-06-23 19:55:33.404 |   Using cached glance_store-0.6.0-py2.py3-none-any.whl20:47
*** Rockyg has quit IRC20:47
*** EricGonczer_ has quit IRC20:48
lifelesspatrickeast: https://pypi.python.org/pypi/glance_store is up to 0.6.020:48
*** Rockyg has joined #openstack-infra20:48
lifelesspatrickeast: so you have something constraining your available versions of some bits (glance_store) but not others (oslo.i18n)20:48
*** puranamr has joined #openstack-infra20:48
lifelesspatrickeast: thats a local foot-gun of some sort20:48
bknudsonI'll need https://review.openstack.org/#/c/194801/ in order to merge up our keystoneclient feature branch20:49
patrickeastlifeless: thanks!20:49
bknudsonthere was a merge conflict :(20:50
*** flepied has quit IRC20:50
clarkbbknudson reviewed20:52
*** zz_dimtruck is now known as dimtruck20:52
*** shardy has quit IRC20:52
*** shardy has joined #openstack-infra20:53
*** HeOS has joined #openstack-infra20:53
*** timothyb89 has joined #openstack-infra20:54
openstackgerritBrant Knudson proposed openstack-infra/project-config: Grant keystone core pushMerge on keystoneclient feature branches  https://review.openstack.org/19480120:54
*** barnaby has joined #openstack-infra20:56
*** ddieterly has quit IRC20:56
bknudson"There are 1 projects not normalized." -- wonder what that means.20:56
*** changbl_ has quit IRC20:56
*** alexpilotti has quit IRC20:57
*** barnaby has quit IRC20:57
clarkbthere is a test to lint our acls20:58
clarkbit should tell you what specifically failes20:58
openstackgerritBrant Knudson proposed openstack-infra/project-config: Grant keystone core pushMerge on keystoneclient feature branches  https://review.openstack.org/19480120:59
*** DericHorn-HP has quit IRC20:59
*** ociuhandu has joined #openstack-infra20:59
*** peristeri has quit IRC21:00
fungibknudson: basically the sections and options get sorted, whitespace removed, et cetera21:00
dimsmtreinish: no worries :)21:00
fungibknudson: makes them easier to de-dupe and to provide mass cleanup patches across them when they follow a consistent order21:01
*** bdemers has quit IRC21:01
openstackgerritPaul Belanger proposed openstack-infra/system-config: WIP: Add trystack.o.o into -infra  https://review.openstack.org/19481621:01
bknudsonfungi: I hadn't sorted correctly. I think I fixed it now since it passed locally.21:02
pabelangerjeblair, would be interested in your thoughts^21:02
*** annegentle has quit IRC21:03
*** chen12 has quit IRC21:04
*** chen12 has joined #openstack-infra21:04
mordredpabelanger: I think that's a great idea - but I think the contents need to move from github to gerrit21:04
fungipabelanger: hogepodge will be interested in that one21:04
pabelangermordred, ya, I was thinking about that too21:04
mordredpabelanger: I'm +2 if we can make that happen21:04
pabelangerwanted to make sure current owner was cool with moving it under -infra too.  then get a governance patchset up21:05
fungithis'll probably be similar to when i moved the devstack.org documentation out of github pages21:05
jeblairpabelanger: i left the same comment, and also another one (if it's static. we can host it on static.o.o as a vhost)21:05
pabelangerjeblair, Ya, I was initially going to do static.o.o.  So, moving it there seems good to me21:06
pabelangerjeblair, mordred: do you think we need a spec?21:06
mordredpabelanger: nah21:07
mordredwell, ignore me though21:07
mordredI'm not the PTL :)21:07
mordredbut honestly, it's a static site - spec seems overkill to me21:07
*** tvanderwiel_ has joined #openstack-infra21:08
*** puranamr has quit IRC21:08
*** tvanderwiel_ has quit IRC21:08
*** eharney has quit IRC21:09
SergeyLukjanovjeblair, fungi, mordred, sorry about delay with stackalytics, they said to me that they are talking directly with Jonathan Bryce on that issue21:09
*** puranamr has joined #openstack-infra21:09
*** tvanderwiel has quit IRC21:09
*** tvanderwiel has joined #openstack-infra21:09
fungioh! fun. i guess jbryce can update us once that happens. thanks21:09
jeblairpabelanger: it's borderline, but if you could do a quick spec i think it'd be good.  since there's a new hostname, etc.21:09
openstackgerritMerged openstack-infra/devstack-gate: Add new oslo libraries  https://review.openstack.org/19436721:09
dhellmannmtreinish: now that the tc meeting is done, I would like to suggest that we have a general policy that we wait for a real gate job to fail before raising the alarm about releases. I *know* they do break things, but let's make sure things are really broken?21:10
pabelangerjeblair, okay, I can do that.21:10
dimsdhellmann: ++21:10
SergeyLukjanovjeblair, fungi, mordred, I think if we'll have written prove of the agreement to drop other tools - it would help a lot21:10
jeblairpabelanger: would also be good to mention that, for the moment at least, operating trystack itself is out of scope (which we can revisit after infra-cloud exists)21:10
pabelangerjeblair, right21:10
openstackgerritMaru Newby proposed openstack-infra/project-config: Enhance run-tox.sh to support functional testing  https://review.openstack.org/16145621:10
mordredpabelanger, jeblair: ++21:10
SergeyLukjanovI'm sorry about deplay - it's really difficult to communicate to marketing team, heh21:10
*** aysyanne has quit IRC21:10
mordredpabelanger: I actually thinking operating trystack could be _fun_ ... once we get infra-cloud going21:10
mordredSergeyLukjanov: no worries! I maen ... marketing people ;)21:10
jeblairmordred: yes, _after_.  :)21:11
pabelangermordred, Ya, really interested to see how the infra-cloud plays out21:11
fungiSergeyLukjanov: i married a marketing person. i know first hand how difficult communicating with them can be ;)21:11
mordredpabelanger: you and me both21:11
SergeyLukjanovfungi, :)21:11
jeblairfungi: this is also not a scalable solution to our problem21:13
*** rlandy has quit IRC21:13
mordredjeblair: marrying marketing people?21:13
fungijeblair: i guess not. at least not while staying legal21:14
mordredyeah, I mean, last I checked DinaBelova is not a marketing person, so SergeyLukjanov can't really help any of us with that - it's all on fungi I guess21:14
jeblairSergeyLukjanov, pabelanger: so for the moment, we should wait to hear more from mirantis marketing/foundation?21:14
*** ldnunes has quit IRC21:15
*** rm_work|away is now known as rm_work21:15
*** bhunter71 has quit IRC21:15
*** annegentle has joined #openstack-infra21:16
*** dannywil_ has quit IRC21:16
*** soren has quit IRC21:16
*** ramashri has quit IRC21:17
*** fawadkhaliq has joined #openstack-infra21:17
*** ramashri has joined #openstack-infra21:17
*** tonytan4ever has joined #openstack-infra21:20
*** fawadkhaliq has quit IRC21:21
*** erikmwilson has quit IRC21:22
*** esmute__ has quit IRC21:23
bknudsonCan I get rid of a feature branch? http://git.openstack.org/cgit/openstack/python-keystoneclient/log/?h=feature/keystone-v321:27
*** esmute__ has joined #openstack-infra21:27
*** e0ne is now known as e0ne_21:28
*** ghostpl has quit IRC21:28
*** abitha has joined #openstack-infra21:28
*** krtaylor has quit IRC21:29
clarkbbknudson: ya I can take care of that21:29
*** e0ne_ has quit IRC21:29
bknudsonclarkb: thanks!21:29
clarkb62c55bc2f0f13630db6d9962ae0edf468f495276 is the old feature/keystone-v3 HEAD (now its logged in irc logs)21:30
clarkband done21:30
fungii too have been using irc logs as my official record of shas when i delete branches21:30
fungiit feels like the old days when i would comtinually stream important syslog facilities to a serial printer21:31
openstackgerritVictor Stinner proposed openstack-infra/project-config: Make gate-nova-python34 voting and add py34 gate  https://review.openstack.org/19412821:31
fungicontinually too21:31
*** esmute__ has quit IRC21:33
*** radez is now known as radez_g0n321:33
*** annegentle has quit IRC21:34
jeblairfungi: i'm just glad my dot-matrix irc printout has finally quieted down21:34
*** annegentle has joined #openstack-infra21:34
*** e0ne has joined #openstack-infra21:34
nibalizerthat would be quite a thing to see21:35
jeblairand hear21:35
clarkboh I meant to mention that yes puppet was disabled on etherpad* to have my patch applied21:36
nibalizerand feed, it would be so hungry21:36
clarkbI need to follow up with upstream about my patch but it doesn't pass tests and from what I can tell that is because HEAD doesn't pass tests21:36
nibalizerjeblair: so am i core on system-config too ?21:36
clarkbwhich is :/21:36
jeblairnibalizer: no, or at least, not yet21:36
jeblairnibalizer: think might make things interesting21:36
nibalizerokay, just getting clear21:37
electricalhiya gents21:37
jeblairnibalizer: this is new, let us know how it works out?  :)21:37
nibalizerjeblair: will do o721:38
*** jrniemijr has quit IRC21:38
*** tlian2 has joined #openstack-infra21:38
*** rguillebert has quit IRC21:38
clarkbhttps://github.com/ether/etherpad-lite/pull/2675 is where we are getting my change upstreamed so we can consume it directly21:39
openstackgerritMerged openstack/requirements: Mark as Python2.7 only the dependencies that are.  https://review.openstack.org/18925021:39
*** austin81 has joined #openstack-infra21:40
fungilifeless: ^ unblocked21:40
*** erikmwilson_ is now known as erikmwilson21:40
electricaland girls obviously :-)21:40
openstackgerritMerged openstack/requirements: Add convenience tox environments  https://review.openstack.org/19430521:41
*** tlian has quit IRC21:41
fungielectrical: i assumed "gents" was just short for both "gentlemen" and "gentlewomen"21:41
electricalfungi: hehe.21:42
lifelessfungi: WOOOOOO21:42
*** hodos has joined #openstack-infra21:42
*** e0ne is now known as e0ne_21:42
lifelessfungi: now, I think we need an experimental job on requirements to run devstack with constraints21:43
*** tvanderwiel has quit IRC21:44
fungilifeless: would that do something like swap out update for generate-constraints in devstack?21:44
fungimight be able to test that out initially by just wip-patching devstack to see what happens21:44
clarkbmtreinish: fungi did the stable periodic jobs get sorted out yesterday?21:44
nibalizeraustin81: timothyb89 hello! welcome!21:44
*** e0ne_ is now known as e0ne21:45
openstackgerritMerged openstack/requirements: Add yaql  https://review.openstack.org/19318021:45
mtreinishclarkb: not quite yet, I've got https://review.openstack.org/194302 up to fix the failure we're seeing21:45
fungiclarkb: https://review.openstack.org/194302 needs reviews21:45
fungiyeah, what mtreinish said21:45
mtreinishit should address the issue once it lands21:45
tonybjeblair: Thanks for the link21:45
clarkbmtreinish: fungi and thats necessary because we are set -e now and need to guard against exit >0 ?21:45
timothyb89hello all!21:46
fungiaustin81: timothyb89: new recruits!21:46
lifelessfungi: no, I have the constraints in the stack21:46
openstackgerritlifeless proposed openstack/requirements: Teach the constraints generator about excludes.  https://review.openstack.org/19431421:47
openstackgerritlifeless proposed openstack/requirements: Show not-in-global errors again.  https://review.openstack.org/19434921:47
openstackgerritlifeless proposed openstack/requirements: Add upper-constraints.txt.  https://review.openstack.org/18925121:47
openstackgerritlifeless proposed openstack/requirements: Manual update run to get things aligned.  https://review.openstack.org/19435121:47
openstackgerritlifeless proposed openstack/requirements: Remove unused test requirement toposort.  https://review.openstack.org/19435021:47
lifelessfungi: mordred: ^ all the +2's please.21:47
lifelessI dont' care about the A yet, once it passes check I'll pester someone to add the A, but the 2's would be nice to make that discussion easy21:48
*** sabeen2 has quit IRC21:48
_odyIs there anyplace where the status of all jobs ran across the sevenish jenkinsi are rolled up to?  Trying to identify a possible way to contruct a CI status radiator for our project.21:50
clarkbmtreinish: fungi reviewed, please check my bash :)21:50
clarkb_ody: http://status.openstack.org/zuul/21:51
clarkb_ody: you can also get the raw json from the zuul endpoint21:51
clarkbwhich is how that page is rendered21:51
SergeyLukjanovjeblair, yup, correct, should wait some more time21:51
fungiclarkb: oh! i believe you're right21:51
fungiclarkb: since test -z requires a parameter21:51
*** esmute__ has joined #openstack-infra21:51
clarkbfungi: yup and non existant var results in no param21:51
fungiclarkb: no, wait, it doesn't21:52
clarkbfungi: it doesn't?21:52
_odyclarkb: Isn't that only as long as it is the zuul queue?21:52
fungiclarkb: you can run `test -z` and the return value is 021:52
clarkb_ody: yes, you want historical data?21:52
lifelessfungi: https://review.openstack.org/#/c/194314/ is the next one specifically needing a +221:52
_odyclarkb: I want to be able to tell that something failed an our ago.21:52
clarkb_ody: for historical data you will want to look at http://graphite.openstack.org21:52
*** rguillebert has joined #openstack-infra21:53
*** thedodd has quit IRC21:53
* greghaynes waves to _ody21:53
clarkbfungi: fun my testing says it does work as does -n21:53
*** esker has quit IRC21:53
clarkbfungi: [ too21:54
fungiyep. a pleasant surprise21:54
clarkbwill flip my vote21:54
*** reed has quit IRC21:54
mtreinishclarkb: yeah I normally write bash stuff by trial and error21:55
mtreinishit worked so I used it :)21:55
mtreinishI'm sure there is a better way to do it all21:55
greghaynesclarkb: it doesnt matter on that patch because its using [[, not [21:55
mtreinishs/better/more bash-like21:55
_odyAfternoon greghaynes.21:55
clarkbgreghaynes: even [ works21:55
*** e0ne has quit IRC21:55
greghaynesclarkb: orly21:55
clarkbat least as a bash builtin, let me try real [21:55
*** DericHorn-HP has joined #openstack-infra21:56
clarkbyup real [ works too21:56
greghaynesah, I guess unaries DTRT21:56
clarkbya I think its multi param tests that fail21:56
greghaynesanyhow, I definitely tested that line because I was afraid of the same ;)21:56
*** reed has joined #openstack-infra21:57
yolandahi, sorry but i couldn't attend infra meeting today and i missed the good news21:58
yolandai'm so happy to be core on puppet, thx for that21:58
greghaynesyolanda: congrats21:58
pleia2yolanda: yay :)21:59
*** marun has quit IRC21:59
yolandathx, will try to do a good job there and work hard on it21:59
fungithank you yolanda!22:00
clarkbsdague: before you go on vacation next week can we make some progress on https://review.openstack.org/#/c/180552/ ? I may behind on things but I really hate to see that work continue to stall because we need to add new features to devstack22:00
*** chen12 has quit IRC22:00
fungiclarkb: what new features, specifically?22:02
clarkbfungi: see sdague's -122:02
fungior what exactly is breaking with soft updates when running the loganalyze plugin22:02
clarkbfungi: basically allow plugins to control requirements22:03
*** yamamoto has joined #openstack-infra22:03
clarkbfungi: I have no idea22:03
fungiyeah, i ask this after reading his reply there22:03
fungii was pretty sure other openstack plugins were working without having all their requirements enforced22:03
clarkboh I see there is a new comment there since I last read the comments22:03
fungibut i could be wrong22:03
clarkbthe first comment was add feature to devstack22:03
*** doug-fish has left #openstack-infra22:04
*** hodos has quit IRC22:04
*** otter768 has joined #openstack-infra22:04
*** hashar has quit IRC22:04
*** claudiub has quit IRC22:05
openstackgerritJames E. Blair proposed openstack-infra/system-config: Add a tool to update gerrit groups  https://review.openstack.org/18659622:05
openstackgerritJames E. Blair proposed openstack-infra/system-config: Add a script to configure the infra group structure  https://review.openstack.org/18686122:05
openstackgerritJames E. Blair proposed openstack-infra/system-config: Add infra-puppet-core and release groups  https://review.openstack.org/19484222:05
clarkbwe would need the plugin to be able to affect the enforcement from within the plugin itself, I am not familiar enough to know if that is possible today22:05
clarkbmaybe that is the missing mechanism mentioned in the first -1 comment?22:05
mgagneI see Jenkins slaves are still configured with Puppet. Any plan or change in progress to move to DIB or ansible?22:06
jeblairmgagne: yes, to dib; at least most of the configuration22:06
mgagnejeblair: any change in flight or mostly in etherpad or ideas floating around?22:07
jeblairmgagne: i'm not sure if we will ever completely remove puppet (we might use it to install our keys)22:07
*** yamamoto has quit IRC22:07
jeblairmgagne: right now we have some images configured partially by dib22:07
mgagneyes, I saw the elements22:08
*** Rockyg has quit IRC22:08
jeblairyeah, and there's a puppet element to pull in what we're doing now22:08
mgagneI'm mostly curious as to how it will be organized as Puppet abstract a lot of stuff already and there is some logic in there22:08
*** annegent_ has joined #openstack-infra22:09
*** otter768 has quit IRC22:09
jeblairi think we want to get almost all of what puppet does for those nodes into dib.  maybe even all of it.  we're not actively moving right now because we're still moving to dib from snapshot images.  but i think the plan is:22:09
*** bknudson has quit IRC22:09
jeblairfinish moving from snapshot to dib22:09
jeblairremove the dependency installation from puppet by using bindep22:10
mgagneshould there only be one big element called single-use-jenkins-slave (or whatever) with a gazillion scripts in there which would mainly translate Puppet into Bash?22:10
jeblairmgagne: oh goodness no, the elements should get simpler, not more complex22:10
*** radez_g0n3 is now known as radez22:11
*** signed8bit has quit IRC22:11
jeblairmgagne: i should also say: it's dib, so you can use whatever is appropriate in there; puppet, chef, ansible, etc.22:11
*** signed8bit has joined #openstack-infra22:11
*** tlian2 has quit IRC22:11
jeblairmgagne: but i think what we want is a set of simple elements that set up the most basic dependencies on our build nodes, like zuul-cloner, bindep, etc.  another set to pre-cache things we might need (git repos, binary packages for bindep, cirros images, etc).  we may or may not have puppet run to set up our keys.22:12
jeblairmgagne: then the jobs themselves install whatever dependencies they need using bindep (and hopefully most of them should be cached)22:12
mgagneI'll read more about bindep then22:13
*** annegentle has quit IRC22:13
*** chen12 has joined #openstack-infra22:15
mgagnejeblair: thanks!22:15
openstackgerritMerged openstack-infra/elastic-recheck: Update tox environment list  https://review.openstack.org/19060922:16
jeblairmgagne: the really short version of how we want to use it is to allow projects to specify their binary dependencies in tree so we can install them automatically22:16
jeblairyolanda, nibalizer, greghaynes: you're all set up, let me know if something doesn't look right22:17
greghaynesjeblair: awesome, ty22:17
fungiyep, i'm about to upload the change that will iterate through all the branches of all the cached git repos and concatenate all the bindep other-requirements.txt lists it finds, then predownload the packages for them22:17
fungithough i need to finish cooking and eating dinner before the nova vulnerability management review starts in 40 minutes22:18
jeblairso that's the part we would run in an element22:18
mgagnejeblair: I agree with this vision. I'm trying to deal with what is available now =)22:18
fungijeblair: i was adding it to the nodepool-base element, but can make it a separate element if desired22:18
jeblairmgagne: right now, we are doing _all of the things_ since just about everything is mid-transition :)22:18
jeblairfungi: i wonder if a separate element might make it more reusable for folks like mgagne?22:19
jeblair(for that matter, 'cache all the git repos' is probably another good reusable unit of work)22:19
clarkbit would have to depend on the nodepool element for the git repos right?22:19
mgagneI did partially imported nodepool-base due to nodepool requiring /etc/nodepool to exists for whatever reason22:19
*** Sukhdev has quit IRC22:19
*** dontalton2 has joined #openstack-infra22:19
clarkbmgagne: thats where nodepool writes things, you do need it22:20
clarkb(for nodepool)22:20
mgagneclarkb: sure, what I meant is: why can't it create it itself?22:20
*** sabeen has joined #openstack-infra22:20
*** mriedem has quit IRC22:20
*** husanu has quit IRC22:20
jeblairmgagne: it doesn't necessarily have access at that point22:20
mgagneclarkb: nodepool assumes that your image has /etc/nodepool precreated which forces me to add it in DIB22:20
clarkbmgagne: es for the reason jeblair just stated22:20
mgagnejeblair: what is "it" ?22:21
clarkbthe nodepool user may not be allowed to write to /etc at VM boot time22:21
timrcMy fingers like typing opensack-infra.  Bad fingers.22:21
jeblairmgagne: (nodepool logging in as the test user which may or may not have sudo access)22:21
mgagnevery good point22:21
*** jgrimm is now known as zz_jgrimm22:21
*** rfolco has quit IRC22:21
fungiclarkb: yep, it needs to drag in the git repo caching element. luckily the source repos caching happens in the extra-data phase before anything in the install phase where package caching runs, so priority juggling won't be an issue22:21
*** husanu has joined #openstack-infra22:22
fungiclarkb: it also needs to depend on the element which installs bindep into the iname22:22
*** annegent_ has quit IRC22:22
*** annegentle has joined #openstack-infra22:22
*** gyee has quit IRC22:23
openstackgerritMerged openstack-infra/irc-meetings: Add Mercador project meeting to IRC.  https://review.openstack.org/19473822:23
*** yamahata has quit IRC22:23
*** puranamr has quit IRC22:24
fungibut if i'm putting the distro package caching into a separate element, maybe i'll move the bindep installation into the same element22:24
*** dizquierdo has quit IRC22:24
fungiwill make it easier to keep the numbers correct relative to each other22:24
fungithe tricky bit is that it has to happen after puppet installs pip22:25
*** sigmavirus24 is now known as sigmavirus24_awa22:26
*** xyang1 has quit IRC22:26
*** gyee has joined #openstack-infra22:27
*** puranamr has joined #openstack-infra22:28
*** ghostpl has joined #openstack-infra22:29
*** signed8bit has quit IRC22:30
*** austin81 has quit IRC22:31
*** amitgandhinz has quit IRC22:31
*** dboik has quit IRC22:32
*** puranamr has quit IRC22:32
*** puranamr has joined #openstack-infra22:32
*** esmute__ has quit IRC22:33
*** ghostpl has quit IRC22:34
*** puranamr has quit IRC22:36
*** puranamr has joined #openstack-infra22:36
_odyclarkb: found kibana, that is the most promising for what I was looking for...except the old version of kibana.22:38
*** esmute__ has joined #openstack-infra22:40
*** linkedinyou has joined #openstack-infra22:40
*** puranamr has quit IRC22:40
pleia2jesusaurus is doing some work to support kibana3 in our infrastructure, so I suspect an upgrade is in the future at some point22:40
*** angela-s has quit IRC22:41
*** puranamr has joined #openstack-infra22:41
*** maxandersen has left #openstack-infra22:41
pleia2open reviews for it: https://review.openstack.org/#/q/status:open+topic:kibana3,n,z22:41
openstackgerritMorgan Fainberg proposed openstack-infra/project-config: Do not run requirements check on keystonemiddleware kilo branch  https://review.openstack.org/19484922:42
pleia2_ody: ^^22:42
rluciohey guys, i've starting getting CI failures from a heat update complaining about needing anew version of qpid-python, but i cant find a new version from pip22:42
rlucioanyone else seen this?  qpid-python;python_version=='2.7'22:42
clarkb_ody its old because es didnt work with kibana 3 in read only mode22:44
pleia2clarkb: ah, good to know22:44
openstackgerritJoshua Harlow proposed openstack-infra/release-tools: Optional include pypi package url link in release notes  https://review.openstack.org/19485022:44
clarkb_ody that may have changed and anyone can test against http://logstash.openstack.org/elasticsearch22:44
clarkbif it works we can deploy new thing but wheb I checked forever ago it didnt work as it needed to write to an index22:45
*** rwsu has quit IRC22:47
clarkbapparently acls are in the paid version of es22:47
*** puranamr has quit IRC22:49
*** puranamr has joined #openstack-infra22:49
*** rlucio has quit IRC22:49
lifelessfungi: can has +A ? https://review.openstack.org/#/c/194314/22:52
fungiwow. rlucio didn't stick around long enough to get an answer22:52
*** tim_o has quit IRC22:52
fungilifeless: ahh, yep. i had already reviewed that but not voted on it22:53
lifelesshttps://review.openstack.org/#/c/189251/ is in the same state22:53
fungiyeah, i followed the needed-by link22:54
*** weshay has quit IRC22:54
*** prad_ has joined #openstack-infra22:55
*** prad has quit IRC22:56
*** prad_ is now known as prad22:56
lifelessfungi: if you go up the stack22:56
lifelessfungi: there are 5 things that I think dhellmann has +2'd all of22:56
fungii just finished reviewing them. all lgtm22:56
lifelessfungi: if you were to +2+A, we'd be onto the iterate-on-jobs buts22:56
fungiand with that, i need to shift gears to nova vulnerability review meeting22:57
*** puranamr has quit IRC22:57
fungiand hopefully find time to get back to our nodepool bindep prep script/dib element integration22:57
*** baoli has joined #openstack-infra22:58
lifelessfungi: Can we setup an experimental job to run devtack in constraints mode first?22:58
lifelessfungi: because I don't think I can do that via Depends-On22:58
lifelessfungi: I linked the WIP devstack patch earlier22:59
*** Guest15326 has quit IRC22:59
*** baoli has quit IRC22:59
clarkblifeless just have a devstack change that does constraints depend on the constraint code?22:59
fungilifeless: yeah, didn't have time to look at it yet. adding an experimental job for that is likely straightforward if the feature exists22:59
lifelessclarkb: REQUIREMENTS_MODE is passed in from zuul AIUI23:00
clarkbjust set it in devstack23:00
*** baoli has joined #openstack-infra23:00
fungiif devstack lacks a non-default constraints mode, then that's probably the next order of business23:00
lifelessclarkb: I need it to be different everywhere23:00
*** r-daneel has quit IRC23:00
clarkbright but23:00
clarkbto test it just set it in a chabge23:00
lifelessclarkb: I guess I have too vague an idea on the entry points here23:01
lifelessclarkb: I'll poke around23:01
lifeless[but I think we're ready for the jobs anyway, so I'd like to get that happening]23:01
clarkbI dont think you want new devstack jobs, just jobs for bumping testibg then proposing conatraints right?23:01
lifelesswe wanted a full set of parallel jobs23:02
*** rlucio has joined #openstack-infra23:02
clarkbthen existing devstack jobs set the enforcemebt as appropriate23:02
lifelessstarting experimental23:02
*** bknudson has joined #openstack-infra23:02
lifelessthen nonvoting23:02
clarkbwat ok23:02
lifelessthen finally a flag day23:02
*** dimtruck is now known as zz_dimtruck23:02
clarkbI would just do the chabge like I describe the flip switch when it works23:02
clarkbmuch easuer23:02
clarkbparallel jobs get tough to juggle when you have >123:02
lifelessso I'll happily follow guidance and variation from the spec23:03
lifelessOTOH, if there's a long tail here23:03
lifelessI don't want to be chasing regressions in one part to land a working thing in a another part23:03
*** annegentle has quit IRC23:03
clarkbthe place where new jobs will be helpful is if we expect changes yo many repos to be nexessary23:03
lifelessso the idea was that we'd be able to move them voting step by step23:04
lifelesswe're expecting to have to change the tox.ini on everyones project, one you and I and Nakato have figured out what a constraints-using tox.ini looks like23:04
clarkbis there a long tail though?23:04
clarkbthey either work or dont23:04
clarkbthey should not race etc23:04
*** annegentle has joined #openstack-infra23:05
*** zz_dimtruck is now known as dimtruck23:05
*** puranamr has joined #openstack-infra23:05
clarkb(since we are making the tests deterministic)23:05
*** mriedem has joined #openstack-infra23:05
*** baoli has quit IRC23:06
lifelessSo look23:06
lifelessI'm piggy in the middle here23:06
lifelesswe came up with a plan23:06
lifelessI'm executing.23:06
*** sdake has joined #openstack-infra23:07
lifelessI don't particularly care about the steps. If they should be different, I'm not going to argue.23:07
clarkbok I am just suggesting that we take advatage of self testing gated prohects as that is why the tests exist as they do. we can do it the other way, I thibk it just complicated and works arou d the checks we already have in place23:07
*** tonytan4ever has quit IRC23:08
lifelesswe don't CI changes to the zuul config though23:08
clarkbbut many people dont frel confortable flipping switches if results come back green (eg pass the gate)23:08
lifelessif we did we could do this as a single switch with confidence23:08
clarkblifeless we shouldnt br changing zuul config for the dwitch, its devstack and tox.ini23:08
lifelessbut as we don't, I don't understand how we'd do that23:08
fungiyeah, i too balked at the idea that we're going to have a way to run every job for every repo on a single change right now. that might become easier with zuul v3, but starting with a basic set of integration jobs run on the constraints update (assuming that's what you're talking about) should be better than what we have23:09
clarkbdevstack starts enforcing everywhere is self tested by devstack,tox.ini enforcing is self tested by project unittests23:09
*** prometheanfire has left #openstack-infra23:09
NakatoYes, started looking at that yesterday.  lifeless to confirm what I'm thinking the idea is, constraints will run in addition to the current pyXX tests for each python right?23:09
lifelessNakato: the goal is to have the unit tests in the gate use the upper-constraints.txt file that is being added to global-requirements23:10
*** lhcheng has quit IRC23:10
*** annegentle has quit IRC23:10
fungiwhich our generate-constraints proposal job will attempt to keep in sync to the projects, though slightly asynchronously23:10
lifelessclarkb: so its devstack-vm-gate-wrap.sh that needs to change23:10
Nakatolifeless: Just the gate, not on developer systems?23:10
lifeless./devstack-vm-gate-wrap.sh:export REQUIREMENTS_MODE=${REQUIREMENTS_MODE:-strict}23:10
clarkblifeless ok, that eould be a self tested change as well23:11
*** timothyb89 has quit IRC23:11
lifelessok, that I can do23:11
lifelessNakato: we want to make it possible for developers to do that too23:11
lifelessNakato: but we don't want to break developers existing workflows23:11
clarkbright so devstack default is old way, gate default is bew way, maybe they converge later23:12
*** dboik has joined #openstack-infra23:12
clarkbtox is trickier but can switch on flags too I think23:13
lifelessclarkb: we kind of handwaved a bit about tox at the summit23:13
lifelessbut TBH since unit tests are not global wedges23:13
clarkbya unittests are less of a priority23:13
lifelessI'm entirely happy if we bring that in more slowly or even opt-in23:13
docaedoHi .. I have a question about enabling CORS for app catalog site. Make that two.23:13
lifelesslike 'if you want this, come and get it'23:13
*** dboik has quit IRC23:14
docaedoFirst, are there any security concerns with adding that header (as noted in https://bugs.launchpad.net/app-catalog/+bug/1467699)?23:14
openstackLaunchpad bug 1467699 in Community App Catalog "CORS not enabled" [Undecided,In progress] - Assigned to Christopher Aedo (docaedo)23:14
*** Hal has joined #openstack-infra23:15
*** esmute__ has quit IRC23:15
fungidocaedo: is there an api there currently which will benefit from cors implementation? though if it's just a tweak to the apache config we can add that to the puppet module pretty easily yeah?23:15
*** Hal is now known as Guest7449323:15
docaedofungi: this is to enable work we're doing on a horizon panel, to integrate catalog contents directly in horizon (hopefully for a one-click add to an environment)23:16
*** dontalton2 has quit IRC23:16
*** dontalton has quit IRC23:16
fungidocaedo: ok, cool. krotscheck: do you have any input on that plan?23:16
docaedofungi: think it's a very easy add, but needs mod_headers module to be enabled, and I am not sure I can do that only via the app-catalog repo23:16
* krotscheck is reading scrollback23:17
fungidocaedo: yeah, the apache puppet module lets us specify mods to load23:17
_odyclarkb: running kibana3 from my laptop pointed at http://logstash.openstack.org/elasticsearch working so far...will hack around with it and see if it is even going to do what I want.23:17
_odyThanks for the URL.23:17
krotscheckdocaedo: Is the app catalog a python wsgi app?23:18
fungidocaedo: i don't recall the exact syntax off the top of my head, but it'll be just a few extra lines in the puppet-apps_site manifest23:18
docaedokrotscheck: um, app catalog is not right now, but kfox1111 was working on what I think was a horizon plugin that would fetch things from the catalog23:20
*** erlon has quit IRC23:20
*** hichihara has joined #openstack-infra23:20
krotscheckdocaedo: Ah-hah. What language is it built in?23:20
docaedofungi: should be as easy as "include apache::mod::headers" .. except headers is not an enabled module by default23:21
krotscheckdocaedo: I'm just checking because we have a thing in OpenStack that does CORS now, and if you use that thing it'd be maintained for you, but if you're not using python then it doesn't really make sense.23:21
docaedokrotscheck: just JS, catalog contents are flat YAML files23:21
anteayafungi: I don't know if I noticed it before but reading the text log of the neutron meeting has not single quotes in positions where folks seem to be making contractions23:21
openstackgerritJoshua Harlow proposed openstack-infra/project-config: Remove py26 gating for automaton  https://review.openstack.org/19486023:21
anteayaI'm feeling confident the single quote is the intention23:22
docaedokrotscheck: thanks for checking :) we're talking about best ways to integrate catalog stuff with other environments (making it as painless as possible)23:22
krotscheckdocaedo: Nice. Yeah, that'd be a mod header thing. The only security concern I have is that using the '*' allowed_domain is a crazy security hole if you actually allow modifications to the service, but if it's all static then I don't have a problem with it.23:22
openstackgerritMichael Still proposed openstack/requirements: Nova requires concurrency 2.1.0 or better  https://review.openstack.org/19486123:22
openstackgerritIan Wienand proposed openstack/diskimage-builder: Update default Fedora to 22  https://review.openstack.org/18584323:23
openstackgerritIan Wienand proposed openstack/diskimage-builder: Work around yum/dnf differences  https://review.openstack.org/18584223:23
openstackgerritIan Wienand proposed openstack/diskimage-builder: Add YUM variable to for Fedora >= 22  https://review.openstack.org/18584123:23
openstackgerritIan Wienand proposed openstack/diskimage-builder: Cleanup yum downloading  https://review.openstack.org/18662023:23
*** sdake_ has joined #openstack-infra23:23
krotscheckdocaedo: I expect you want this to work for any horizon install out there, so you'd want to wildcard the domain.23:23
*** jasondotstar has joined #openstack-infra23:23
docaedokrotscheck: yes, agree would be a security concern if it wasn't static, but I wanted to check with you folks to make sure there wasn't a bigger concern I was missing23:23
ianwgreghaynes: ^ i think that f22 stack has addresssed all comments23:23
docaedokrotscheck: and exactly, would want this to work from anyones horizon23:23
fungidocaedo: it seems fine to me. mainly wanted to drag in krotscheck as our resident cors whipping post23:24
ianwi also think puppet 3 isn't supporting f22, which throws another spanner in the works ... need to research what's going on23:24
jeblairdhellmann: i think i had an epiphany all over your change here: https://review.openstack.org/191193  let me know if i now understand23:24
pleia2anteaya: looks like armax is using some utf8 character or something for ' (mestery's single quotes are fine)23:24
docaedofungi: more eyes and opinions is always good with me23:24
*** yamamoto has joined #openstack-infra23:24
anteayapleia2: ah okay thanks for checking, I'll talk to armax and see what he is using23:25
*** otter768 has joined #openstack-infra23:25
jeblairdocaedo: we enable cors for zuul23:25
jeblairdocaedo: so the puppet to do that should be floating around already23:26
jeblairfungi: ^23:26
krotscheckdocaedo: I don't have security concerns as long as it's GET only. Other alternatives could include hosting a 1px iframe with javascript on the app catalog, and then using iframe messages to pass data back and forth.23:26
docaedojeblair: excellent, I'll see if I can find it23:26
*** sdake has quit IRC23:26
fungijeblair: oh, awesome. i forgot we needed that to be able to embed its status.json23:26
openstackgerritMatthew Treinish proposed openstack/os-testr: Disable printing percent change on run time by default  https://review.openstack.org/19486523:26
fungidocaedo: puppet-zuul repo i believe23:26
krotscheckBut that's annoying and complicated.23:26
jeblairyeah, it's a read-only api though, so the security review for that was pretty quick.  :)23:27
fungiwell, apps.o.o is also read-only at the moment23:27
* krotscheck wanders off again to go beat his head against internal HP things.23:27
fungithanks krotscheck!23:27
anteayapleia2: xgerman as well23:27
jeblairkrotscheck: wear a helmet!23:27
fungiif only he had a holocaust cloak23:27
pleia2anteaya: yeah, maybe they're using the same client?23:27
mtreinishjeblair: if you get a sec want to +A https://review.openstack.org/194302 ? It'll be good to land before the next round of periodic jobs start23:28
pleia2anteaya: likely fix is we could look into adding "AddDefaultCharset utf-8" to the apache config for logs, but I haven't actually looked at our current config23:29
*** stevemar has quit IRC23:29
anteayapleia2: I too have not looked at the current config23:29
*** hdd has quit IRC23:30
anteayapleia2: but adding something might be worth considering23:30
* pleia2 nods23:30
*** ghostpl has joined #openstack-infra23:30
anteayawas going to finish reading the meeting logs for all the meetings I missed first23:30
*** hdd has joined #openstack-infra23:30
openstackgerritMatthew Treinish proposed openstack/os-testr: Dogfood things for unit tests  https://review.openstack.org/19487023:34
jeblairmtreinish: the same icehouse jobs failed twice; are you sure they are benign?23:34
*** ZZelle has quit IRC23:35
*** ghostpl has quit IRC23:35
*** rguillebert has quit IRC23:35
jeblairit looks like ianw does23:35
jeblairit's just that we're changing branch selection and all, so when we lose an entire branch, it's worth a double check :)23:36
mtreinishjeblair: I assumed it was nova-api wasn't starting because of an oslo thing: http://logs.openstack.org/02/194302/4/check/check-tempest-dsvm-neutron-icehouse-nv/ca4b680/logs/screen-n-api.txt.gz23:36
mtreinishI didn't think my change could have caused that23:36
*** dimtruck is now known as zz_dimtruck23:37
mtreinishit also looks like icehouse is generally failing for that too: http://logs.openstack.org/24/194724/4/check/check-tempest-dsvm-full-icehouse/ee2f41a/logs/screen-n-api.txt.gz23:37
mtreinishwe also just pushed the last icehouse release right? So we can just eol everything now23:38
*** mriedem has quit IRC23:38
jeblairmtreinish: i have an inline question23:38
mtreinishdhellmann, dims: btw, the il8n thing from before is hitting us, just on icehouse ^^^23:38
krotscheckjeblair: Where was that page with the "Here's the structure of build things" again that I need to update for javascript things?23:39
fungimtreinish: icehouse is being eol-tagged/deleted later this week, yeah23:39
*** mriedem has joined #openstack-infra23:39
jeblairkrotscheck: http://governance.openstack.org/reference/project-testing-interface.html23:39
fungiwe usually give it 7 days for emergency brown-bag releases after the last stable point release23:39
krotscheckjeblair: Thanks23:40
openstackgerritMerged openstack-infra/project-config: Fix fuel-docs acls to allow merge commit  https://review.openstack.org/19471723:40
openstackgerritChristopher Aedo proposed openstack-infra/puppet-apps_site: Enable CORS for app-catalog  https://review.openstack.org/19487523:40
*** sdake_ has quit IRC23:40
*** yamamoto has quit IRC23:41
krotscheckSo on a related note, I'd like to create a project as a bellweather for publishing from infra to NPM.23:42
krotscheckAnd, well, I think a good target would be a hacking-like project for javascript style guidelines.23:42
krotscheckWhat namespace should that live under?23:42
mtreinishjeblair: responded, but basically I just assumed using the fallback upfront was safer when the override ref was undefined23:42
lifelesskrotscheck: so, openstack-dev hosts hacking, but AIUI we're steering things towards openstack/ these days23:43
jeblairmtreinish: i believe previous successful runs had both variables empty; i actually think that even with this you will be setting it to empty because fallback will be23:43
mtreinishfungi: is there anyway we could accelerate that? because I don't think we can land any tempest changes until we either eol it or fix the cap23:43
jeblairmtreinish: get_fetch_at_ref has a special case for ''23:44
mtreinishfungi: actually you know what I'm just going to premptively remove the tempest icehouse jobs23:44
jeblairmtreinish: http://logs.openstack.org/periodic-qa/periodic-tempest-dsvm-all-master/92543ab/logs/devstack-gate-setup-workspace-new.txt.gz  (a previous successful run)23:44
mtreinishjeblair: oh, ok. I probably should have looked at the def for get_fetch_at_ref23:44
fungimtreinish: yes, i think dropping the backward-compat jobs for icehouse is fine at this stage23:44
mtreinishjeblair: I can respin and remove that bit, one sec23:45
fungioh, i didn't realize we had the null string special cased there23:45
jeblairmtreinish: great, thanks.  i think that is worth doing so that it doesn't become voodoo when we look at it next.23:45
*** DericHorn-HP has quit IRC23:45
*** ramashri has quit IRC23:46
dimsmtreinish: icehouse. wow23:46
*** yamamoto has joined #openstack-infra23:47
openstackgerritJoshua Harlow proposed openstack/requirements: Enable 'python_version=='2.6' for futures and enum34  https://review.openstack.org/19487723:48
openstackgerritMatthew Treinish proposed openstack-infra/devstack-gate: Fix the setup_workspace to handle missing ZUUL_BRANCH  https://review.openstack.org/19430223:49
mtreinishfungi, jeblair: ^^^23:49
openstackgerritMatthew Treinish proposed openstack-infra/project-config: Remove icehouse compat jobs from tempest  https://review.openstack.org/19487823:50
mtreinishfungi, jeblair: also ^^^23:50
jeblairso we _just_ released the last icehouse and now it's failing?23:50
*** Swami has quit IRC23:51
mtreinishjeblair: yep23:51
jeblairah, i have read commit msg23:51
mtreinishit was the oslo releases from this afternoon23:51
mtreinishprobably missing a cap on something somewhere23:51
mtreinishon icehouse23:51
jeblairi'm assuming this falls into the category of "should be fixed by ongoing requirements work"23:51
openstackgerritJoshua Harlow proposed openstack-infra/project-config: Remove py26 check/gating for oslo.messaging  https://review.openstack.org/19487923:52
mtreinishjeblair: this is actually an improvement earlier stable releases (from feb or jan) was pushed when the gate was completely fubar and nothing could land23:52
mtreinishat least this one was pushed before things were broke23:52
mtreinishjeblair: yeah, hopefully this kinda thing will all be sorted by the requirements work23:53
*** isq has quit IRC23:54
anteayafungi: I too am sick to death of fireworks23:55
*** dims has quit IRC23:56
anteayano idea how some folks can think 7am on a Sunday is the right time for fireworks23:56
Clintthat's special23:56
anteayaI am hoping to figure out how it can be less special23:56
anteayaphoning the municipality is not the solution unfortunately23:57
*** ajmiller has quit IRC23:58
ianwjeblair: 00000aa0  6c 61 72 6d 2d 6e 6f 74  69 66 69 65 72 2e 74 78  |larm-notifier.tx| <- that "f" shows up funny in my firefox, but i can't see that it's encoded in any funny way23:58
fungianteaya: here its every single night now. and i think will be until septemberish23:58
*** chlong has joined #openstack-infra23:58
anteayafungi: you beat me then23:58
mtreinishanteaya: heh, I would call the mounties :)23:58
lifelessfireworks are illegal here except for a couple nights of the year23:58
fungiianw: probably fi glyph munging23:58
anteayawe have the occasional monday and wednesday when it is just birds and frogs23:58
anteayaand loons23:58
fungilifeless: here too23:58
fungilifeless: doesn't stop anyone23:59
lifelesshah, does here :)23:59
anteayamtreinish: please do, tell them to fix this23:59
lifelessturns out people don't like loud bangs at night23:59
anteayalifeless: ensure it stays that way23:59
anteayafireworks used to be illegal here too23:59
anteayaexcept by special permit23:59
*** hdd has quit IRC23:59
lifelessthen a fireworks company sued?23:59
anteayaturns out we were standing in the way of people's fun23:59

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