| r1chardj0n3s | jeblair: I'd be tempted to read and maybe pdb/instrument /usr/lib/python2.7/sysconfig.py to see what it's actually doing | 00:00 |
|---|---|---|
| r1chardj0n3s | 'cos my money is on a broken something in sysconfig.py | 00:00 |
| r1chardj0n3s | I | 00:00 |
| r1chardj0n3s | I'm firing up my (14.04) ubuntu to have a look | 00:00 |
| *** gokrokve_ has quit IRC | 00:00 | |
| *** amcrn has quit IRC | 00:00 | |
| dstufft | reminder there are two sysconfigs | 00:00 |
| jeblair | lifeless: i suppose it's unlikely to be pbr, because i think it's happening while trying to satisfy setup_requires (which is pbr) | 00:00 |
| dstufft | there is sysconfig and distutils.sysconfig | 00:01 |
| dstufft | because reasons | 00:01 |
| *** david-lyle has quit IRC | 00:01 | |
| jeblair | lifeless: so it's the "install pbr" step which is the first thing that happens, and we don't get that far | 00:01 |
| *** bradjones has quit IRC | 00:01 | |
| r1chardj0n3s | dstufft: sure, but the traceback is in one of them :) | 00:01 |
| *** pballand has quit IRC | 00:01 | |
| dstufft | r1chardj0n3s: I just wanted to wtf about the fact there is two different sysconfigs | 00:02 |
| dstufft | one isn't an alias for the other | 00:02 |
| dstufft | they are different | 00:02 |
| r1chardj0n3s | dstufft: oh sure, "distutils" | 00:02 |
| r1chardj0n3s | :) | 00:02 |
| jeblair | return os.path.join(get_path('platstdlib').replace("/usr/local","/usr",1), "config" + (sys.pydebug and "_d" or ""), "Makefile") | 00:03 |
| fungi | jeblair: confirmed unreproducible on ubuntu/trusty as well | 00:03 |
| jeblair | so that happens a few calls before the traceback | 00:03 |
| *** ramashri has quit IRC | 00:03 | |
| *** esker has quit IRC | 00:03 | |
| r1chardj0n3s | jeblair: I think you have a winner :/ | 00:04 |
| *** esker has joined #openstack-infra | 00:04 | |
| lifeless | just going to prove, there are more ways to mess things up than one can ever imagine | 00:04 |
| r1chardj0n3s | jeblair: and as far as I can tell, that's part of the debuntu /local/ modifications | 00:05 |
| *** erlon has joined #openstack-infra | 00:05 | |
| *** Shrews_ is now known as Shrews | 00:05 | |
| dstufft | r1chardj0n3s: you know how ubuntu is an african word for humankind? distutils is actually an african word for the most terrible thing ever | 00:06 |
| jeblair | so is easy_install normally the thing that satisfies the setup_requires? | 00:06 |
| dstufft | yes | 00:06 |
| fungi | people. distutils is made from people | 00:06 |
| wenlock | clarkb got to play with r10k , i do like it better, the behavior is different than librarian because it tries to purge the modules folder you managing but it seems like it has 3x more features | 00:06 |
| lifeless | easy_install is never normal | 00:06 |
| dstufft | distutils is soylet green | 00:06 |
| *** mmaglana has quit IRC | 00:06 | |
| jeblair | okay, just making sure i haven't done something locally to cause that to happen | 00:06 |
| r1chardj0n3s | dstufft: yeah, but in this case it's debuntu's fault jeblair is having an issue ;) | 00:06 |
| wenlock | clarkb i didn't get the puppet gem install for librarian though, so i think they must have removed that issue | 00:06 |
| dstufft | oh well, downstream distributors are the bane of my life too | 00:07 |
| fungi | fwiw, it looks like more recent debian/ubuntu releases have solved whatever this specific point of stupidity happens to stem from | 00:07 |
| dstufft | I think they changed their patch to have giant "IF IN VIRTUALENV" checks all over the place | 00:07 |
| wenlock | clarkb i think though based on what I see in install_module.sh the only thing I'd want would be support for an external data option. | 00:07 |
| jeblair | so we'd expect this to happen for any project with a setup_requires being installed into an ubuntu precise virtualenv located under /usr/local/ | 00:07 |
| r1chardj0n3s | fungi: cool, so jeblair just needs to upgrade :) | 00:07 |
| jeblair | r1chardj0n3s: unfortunately, this is on our test nodes and we need to keep precise around for a year | 00:08 |
| fungi | sounds like we need to not make virtualenvs in /usr/local for at least a year | 00:08 |
| dstufft | /opt is cooler than /usr/local, use that! | 00:08 |
| *** esker has quit IRC | 00:09 | |
| jeblair | dstufft: well, we use it for everything else; i kind of want to keep this away from there though | 00:09 |
| r1chardj0n3s | yeah /opt is *edgy* | 00:09 |
| jeblair | i'll probably just put it in /usr with a nice long comment explaining my sin against the fhs | 00:10 |
| fungi | we should totally use /srv instead, since it's of equal levels of vagueness, if perhaps lacking in handlebar mustaches | 00:10 |
| jeblair | fungi: srv used to be cool | 00:10 |
| *** ramashri has joined #openstack-infra | 00:10 | |
| dstufft | srv is old school | 00:10 |
| fungi | yeah, like striped socks | 00:10 |
| *** wenlock has quit IRC | 00:10 | |
| dstufft | (You can tell we are nerds, we are discussing the relative coolness of fielsystem directories) | 00:10 |
| lifeless | I want /{{templates}} now | 00:11 |
| r1chardj0n3s | woah woah I was totally being ironic | 00:11 |
| * anteaya is taking notes | 00:11 | |
| fungi | oh, don't get me wrong, i used the heck out of some /opt back when i ran sco. it was the hip and with it place to install all your crufty crap | 00:11 |
| fungi | clearly by the transitive property of hipsterishness, /opt is now as cool as sco | 00:12 |
| pleia2 | aw, I still <3 srv | 00:12 |
| jeblair | https://bugs.launchpad.net/ubuntu/+source/python2.7/+bug/839588 | 00:12 |
| uvirtbot` | Launchpad bug 839588 in python2.7 "Python 2.7 unable to find it's header files in virtualenv" [Medium,Fix released] | 00:12 |
| jesusaurus | dont forget about /var/lib/, thats been making a comeback recently | 00:13 |
| lifeless | russellb: perhaps you can point me at reproduction of https://bugs.launchpad.net/testrepository/+bug/1331140 ? | 00:14 |
| uvirtbot` | Launchpad bug 1331140 in testrepository "testr reports double failure counts" [Undecided,Incomplete] | 00:14 |
| fungi | jesusaurus: the modern hipstertrend is actually to just pretend nobody is going to use multiple filesystems any longer, symlink /usr/* and /var/* into / and then tell everyone else to gfy | 00:14 |
| r1chardj0n3s | had some weirness yesterday and found one of my scripts had the #! of /opt/usr/local/bin/python2.7 no idea wtf it thought it was doing | 00:14 |
| lifeless | fungi: ITYM GTFOML | 00:14 |
| fungi | lifeless: oh, right. debian-devel | 00:15 |
| jeblair | how do i add precise to that bug? | 00:15 |
| anteaya | r1chardj0n3s: nice | 00:15 |
| lifeless | jeblair: propose somethingorother | 00:15 |
| lifeless | jeblair: url? | 00:15 |
| jeblair | lifeless: https://bugs.launchpad.net/ubuntu/+source/python2.7/+bug/839588 | 00:15 |
| uvirtbot` | Launchpad bug 839588 in python2.7 "Python 2.7 unable to find it's header files in virtualenv" [Medium,Fix released] | 00:15 |
| dstufft | just mount swift as / and then you have like an unlimited size hard drive, why do yoy even need more than one filesystem!! | 00:15 |
| jesusaurus | fungi: sounds like a docker-ism, so it must be hip | 00:16 |
| * fungi waits for the inevitable afs reference | 00:16 | |
| lifeless | jeblair: done | 00:16 |
| jeblair | fungi: you just made it :) | 00:16 |
| lifeless | jeblair: it was 'nominate for series' | 00:16 |
| *** pcm has quit IRC | 00:16 | |
| jeblair | lifeless: ah thx. left a comment there | 00:18 |
| *** pabelanger_ is now known as pabelanger | 00:20 | |
| *** pabelanger has joined #openstack-infra | 00:20 | |
| *** SumitNaiksatam has quit IRC | 00:23 | |
| *** gargola has quit IRC | 00:23 | |
| *** jlk_clone is now known as jlk | 00:24 | |
| *** jlk has joined #openstack-infra | 00:24 | |
| *** chuckC has joined #openstack-infra | 00:27 | |
| anteaya | can I ask for an opinion on this contributor's reason for using [submit] action = cherry pick in the acl file for this stackforge project: https://review.openstack.org/#/c/115250/4/modules/openstack_project/files/gerrit/acls/stackforge/swiftonfile.config | 00:27 |
| *** armax has left #openstack-infra | 00:30 | |
| *** bhuvan_ has quit IRC | 00:31 | |
| *** ZZelle has quit IRC | 00:38 | |
| *** rcarrillocruz has quit IRC | 00:40 | |
| lifeless | klindgren_: fixed | 00:43 |
| *** rcarrillocruz has joined #openstack-infra | 00:44 | |
| klindgren_ | lifeless, thanks! | 00:46 |
| lifeless | yw | 00:46 |
| lifeless | sorry about that | 00:47 |
| clarkb | anteaya you can -1 for that | 00:48 |
| clarkb | cherry pick is problematic | 00:49 |
| *** yaguang has joined #openstack-infra | 00:49 | |
| *** flaviof is now known as flaviof_zzz | 00:49 | |
| jeblair | agreed | 00:49 |
| anteaya | okay thanks | 00:55 |
| *** alexpilotti has quit IRC | 00:56 | |
| *** rick__ has joined #openstack-infra | 00:57 | |
| *** gokrokve has joined #openstack-infra | 00:57 | |
| rick__ | Hi Team, How can I re-enable my service account? | 00:58 |
| *** tomoe_ has joined #openstack-infra | 00:58 | |
| *** sflanigan_ is now known as sflanigan | 00:59 | |
| *** todd_dsm has joined #openstack-infra | 01:00 | |
| *** rcarrillocruz has quit IRC | 01:01 | |
| *** alexpilotti has joined #openstack-infra | 01:02 | |
| *** dims has joined #openstack-infra | 01:02 | |
| *** signed8b_ has quit IRC | 01:03 | |
| *** yamahata has joined #openstack-infra | 01:03 | |
| *** TheJulia_ is now known as TheJulia | 01:04 | |
| *** bdpayne has quit IRC | 01:05 | |
| *** dims_ has joined #openstack-infra | 01:05 | |
| *** marcoemorais has quit IRC | 01:05 | |
| *** rcarrillocruz has joined #openstack-infra | 01:05 | |
| *** alexpilotti has quit IRC | 01:06 | |
| *** otherwiseguy has joined #openstack-infra | 01:06 | |
| anteaya | rick__: hello | 01:08 |
| anteaya | so have you tested any patches on the sandbox repository? | 01:09 |
| *** dims has quit IRC | 01:09 | |
| *** ramashri has quit IRC | 01:09 | |
| *** unicell1 has quit IRC | 01:10 | |
| rick__ | anteaya: Yes , this is my test patch https://review.openstack.org/#/c/115886/ | 01:10 |
| *** ramashri has joined #openstack-infra | 01:10 | |
| rick__ | anteaya: I cannot use prophetstor-ci, because it disabled. I use my account to verify . | 01:10 |
| rick__ | anteaya: http://download.prophetstor.com/prophetstor_ci/115886/18/check/noop-check-communication/61/ | 01:11 |
| rick__ | anteaya: Does I still miss any thing? | 01:11 |
| *** renlt has joined #openstack-infra | 01:12 | |
| anteaya | yes, remove the comment that says starting check jobs | 01:12 |
| *** morganfainberg is now known as morganfainberg_Z | 01:13 | |
| rick__ | anteaya: OK | 01:14 |
| *** tomoe_ has quit IRC | 01:14 | |
| *** camunoz has quit IRC | 01:18 | |
| rick__ | anteaya: Done. | 01:18 |
| *** dims_ has quit IRC | 01:19 | |
| rick__ | anteaya: Does it ok? | 01:19 |
| anteaya | rick__: let's see another successful test on the sandbox patch | 01:19 |
| bknudson | looks like there's a new release of python-subunit, but it might be missing something. | 01:19 |
| bknudson | keystone py26 jobs are failing, and it also didn't work in my env | 01:19 |
| *** otherwiseguy has quit IRC | 01:20 | |
| *** terryw has joined #openstack-infra | 01:20 | |
| anteaya | fungi clarkb jeblair ^^ | 01:20 |
| anteaya | bknudson: do you have a link to a log file? | 01:21 |
| bknudson | http://logs.openstack.org/78/111578/8/check/gate-keystone-python26/c8ae14c/console.html#_2014-08-27_01_03_03_950 | 01:21 |
| * anteaya clicks | 01:21 | |
| bknudson | I get the same error with `.tox/py27/bin/pip install -U "python-subunit>=0.0.20"` | 01:21 |
| bknudson | 0.0.19 works for me | 01:22 |
| anteaya | 6 failures in the gate :( | 01:22 |
| *** trinaths has joined #openstack-infra | 01:22 | |
| clarkb | you want lifeless | 01:22 |
| clarkb | also my glasses exploded so I am temporarily blind | 01:22 |
| anteaya | clarkb: you didn't get hurt did you? | 01:23 |
| *** sflanigan has quit IRC | 01:23 | |
| clarkb | no lense popped out safely | 01:23 |
| *** RicoLin has joined #openstack-infra | 01:23 | |
| bknudson | I hope that doesn't happen to me | 01:23 |
| anteaya | are you without glasses or injured | 01:23 |
| clarkb | *lens | 01:23 |
| anteaya | clarkb: okay, whew | 01:23 |
| anteaya | clarkb: what repo should I offer a patch to pin subunit? | 01:24 |
| anteaya | and our deleting group just expanded | 01:24 |
| clarkb | We can probably get lifeless to fix it | 01:25 |
| anteaya | the running group is very pinched right now | 01:25 |
| bknudson | I opened bug 1361924 for the rechecks | 01:27 |
| uvirtbot` | Launchpad bug 1361924 in keystone "python-subunit 0.0.20 failing to install causing gate failure" [Undecided,New] https://launchpad.net/bugs/1361924 | 01:27 |
| anteaya | thanks bknudson | 01:27 |
| SlickNik | bknudson / clarkb / anteaya: FYI multiple trove py26 jobs are hitting the same issue with subunit | 01:28 |
| bknudson | does py26 use a different pip mirror? | 01:28 |
| anteaya | py26 is tested on centos6 | 01:29 |
| anteaya | py27 is tested on trusty | 01:29 |
| lifeless | hai | 01:29 |
| lifeless | whats up ? | 01:29 |
| anteaya | subunit is causing issues | 01:29 |
| anteaya | bknudson: has a log above | 01:29 |
| lifeless | gnar, stab | 01:30 |
| lifeless | fixing | 01:30 |
| *** rcarrillocruz has quit IRC | 01:31 | |
| anteaya | great thanks | 01:31 |
| *** rcarrillocruz has joined #openstack-infra | 01:31 | |
| *** paulrad has joined #openstack-infra | 01:32 | |
| rick__ | anteaya: https://review.openstack.org/#/c/115886/ this is another patchset trigger. | 01:33 |
| rick__ | anteaya: Our service account is "prophetstor-ci". | 01:33 |
| *** morganfainberg_Z is now known as morganfainberg | 01:33 | |
| lifeless | subunit 0.0.21 up | 01:34 |
| bknudson | lifeless: that worked for me. | 01:34 |
| anteaya | clarkb: how long will it take for our system to use subunit 0.0.21? | 01:34 |
| *** sthillma has quit IRC | 01:35 | |
| anteaya | rick__: you need to ensure your system does not post this message: Patch Set 20: | 01:35 |
| anteaya | Starting check jobs. http://127.0.0.1/zuul/status | 01:35 |
| anteaya | rick__: you can't post Starting check jobs | 01:35 |
| anteaya | only the comment with the build status and logs | 01:36 |
| anteaya | one comment per patchset for third party ci systems | 01:36 |
| rick__ | anteaya: So I need how to do..? | 01:36 |
| anteaya | I think asselin knows | 01:36 |
| anteaya | rick__: can you ask him in the #openstack-cinder channel | 01:37 |
| anteaya | perhaps tomorrow, he is probably asleep now | 01:37 |
| rick__ | anteaya: Does any sample let me reference..? | 01:37 |
| anteaya | that is what I am saying, I don't have a sample, you need to talk to someone who runs a third party ci who has this set properly | 01:38 |
| *** unicell has joined #openstack-infra | 01:38 | |
| clarkb | anteaya: ~5 minutes is all that it will take | 01:38 |
| anteaya | clarkb: great thanks | 01:38 |
| rick__ | anteaya: OK, Thanks. | 01:38 |
| anteaya | clarkb: can bknudson just recheck now then? | 01:38 |
| anteaya | rick__: np, you are close | 01:38 |
| anteaya | rick__: you could grep your code for "Starting check jobs" | 01:39 |
| anteaya | rick__: that would be a start | 01:39 |
| *** weshay has quit IRC | 01:39 | |
| *** sflanigan has joined #openstack-infra | 01:41 | |
| *** sflanigan has joined #openstack-infra | 01:41 | |
| *** baohua has joined #openstack-infra | 01:41 | |
| rick__ | anteaya: OK! | 01:41 |
| *** hogepodge has quit IRC | 01:47 | |
| rick__ | anteaya: Does like this https://review.openstack.org/#/c/115886/? I already remove "http://127.0.0.1/zuul/status". | 01:49 |
| *** amotoki has joined #openstack-infra | 01:50 | |
| *** r1chardj0n3s is now known as r1chardj0n3s_afk | 01:52 | |
| *** todd_dsm has quit IRC | 01:52 | |
| anteaya | you are still posting: Patch Set 21: | 01:54 |
| anteaya | Starting check jobs. | 01:54 |
| anteaya | you can't comment at all when you start the jobs | 01:54 |
| *** todd_dsm has joined #openstack-infra | 01:55 | |
| clarkb | new subunit is at http://pypi.openstack.org/simple/python-subunit/ | 01:55 |
| clarkb | r1chardj0n3s_afk: is now good? or are you really afk? | 01:56 |
| *** nosnos has joined #openstack-infra | 01:56 | |
| *** pcrews has quit IRC | 01:57 | |
| *** mmaglana has joined #openstack-infra | 01:57 | |
| *** Ryan_Lane has quit IRC | 01:58 | |
| rick__ | OK, I try to disable it. | 01:58 |
| anteaya | rick__: okay | 01:59 |
| anteaya | I am going offline now for the night | 01:59 |
| anteaya | rick__: ask in #openstack-cinder tomorrow to find someone to help you if you can't figure it out yourself | 01:59 |
| rick__ | anteaya: OK, I got | 01:59 |
| *** david-lyle has joined #openstack-infra | 02:02 | |
| *** dims has joined #openstack-infra | 02:02 | |
| *** mriedem has joined #openstack-infra | 02:05 | |
| *** david-lyle has quit IRC | 02:07 | |
| *** camunoz has joined #openstack-infra | 02:07 | |
| *** rcarrillocruz has quit IRC | 02:08 | |
| *** Guest61572 is now known as dansmith | 02:09 | |
| *** sflanigan has quit IRC | 02:11 | |
| *** rcarrillocruz has joined #openstack-infra | 02:12 | |
| *** dims has quit IRC | 02:13 | |
| *** todd_dsm has quit IRC | 02:15 | |
| *** david-lyle has joined #openstack-infra | 02:16 | |
| *** homeless has quit IRC | 02:17 | |
| *** zehicle_at_dell has joined #openstack-infra | 02:17 | |
| *** ramashri has quit IRC | 02:18 | |
| *** stevemar has joined #openstack-infra | 02:18 | |
| *** todd_dsm has joined #openstack-infra | 02:19 | |
| *** todd_dsm has quit IRC | 02:20 | |
| *** david-lyle has quit IRC | 02:21 | |
| *** fabo has quit IRC | 02:22 | |
| *** sflanigan has joined #openstack-infra | 02:23 | |
| *** sflanigan has joined #openstack-infra | 02:23 | |
| *** r1chardj0n3s_afk is now known as r1chardj0n3s | 02:26 | |
| *** renlt has quit IRC | 02:27 | |
| r1chardj0n3s | hey clarkb, sorry was out getting some exercise | 02:27 |
| *** mriedem has quit IRC | 02:28 | |
| clarkb | r1chardj0n3s: can you point me at the nova change that added the test so that I can get an idea of exactly what we want to run? | 02:28 |
| r1chardj0n3s | https://review.openstack.org/#/c/114122/ | 02:28 |
| *** fabo has joined #openstack-infra | 02:29 | |
| r1chardj0n3s | so ignore the requirements.txt change ;) | 02:29 |
| r1chardj0n3s | hm | 02:29 |
| r1chardj0n3s | so you can see that the tox command is customised for nova | 02:29 |
| r1chardj0n3s | as we want to ignore requirements in tests | 02:29 |
| *** rcarrillocruz has quit IRC | 02:29 | |
| r1chardj0n3s | & also it has to specify the application source location ("nova") | 02:30 |
| clarkb | so maybe the thing to do is have the requirements job run tox -epip-missing-reqs | 02:30 |
| clarkb | then individual projects can customize for their specific needs | 02:30 |
| * clarkb whips that up | 02:31 | |
| r1chardj0n3s | yep, that is AFAIK what jhesketh did but not in the requirements job | 02:31 |
| *** rcarrillocruz has joined #openstack-infra | 02:33 | |
| *** xchu has joined #openstack-infra | 02:34 | |
| *** praneshp has quit IRC | 02:35 | |
| *** camunoz has quit IRC | 02:35 | |
| *** sflanigan has quit IRC | 02:35 | |
| *** ianw has quit IRC | 02:35 | |
| *** sflanigan_ has joined #openstack-infra | 02:35 | |
| *** camunoz has joined #openstack-infra | 02:35 | |
| *** adalbas has joined #openstack-infra | 02:35 | |
| *** mrrrgn has joined #openstack-infra | 02:36 | |
| *** sflanigan_ has quit IRC | 02:36 | |
| *** sflanigan_ has joined #openstack-infra | 02:36 | |
| *** sflanigan_ has joined #openstack-infra | 02:36 | |
| *** camunoz_ has joined #openstack-infra | 02:36 | |
| *** tgohad has quit IRC | 02:36 | |
| clarkb | r1chardj0n3s: https://review.openstack.org/117072 I think that will do it | 02:36 |
| clarkb | though we will have to update projects to include that tox target | 02:36 |
| *** camunoz_ has quit IRC | 02:36 | |
| *** sflanigan_ has quit IRC | 02:36 | |
| *** camunoz_ has joined #openstack-infra | 02:37 | |
| *** tsg has quit IRC | 02:37 | |
| *** sflanigan_ has joined #openstack-infra | 02:37 | |
| *** sflanigan_ has joined #openstack-infra | 02:37 | |
| *** paulrad has quit IRC | 02:37 | |
| *** jamielennox|away has quit IRC | 02:37 | |
| *** camunoz_ has quit IRC | 02:37 | |
| *** sflanigan_ has quit IRC | 02:37 | |
| *** sflanigan_ has joined #openstack-infra | 02:37 | |
| *** camunoz_ has joined #openstack-infra | 02:37 | |
| *** jamielennox|away has joined #openstack-infra | 02:37 | |
| r1chardj0n3s | clarkb: yep, that's part of my job now ;) | 02:38 |
| *** sflanigan_ has quit IRC | 02:38 | |
| *** sflanigan_ has joined #openstack-infra | 02:38 | |
| *** adalbas has quit IRC | 02:38 | |
| *** sflanigan_ has quit IRC | 02:38 | |
| *** sflanigan_ has joined #openstack-infra | 02:38 | |
| *** sflanigan_ has quit IRC | 02:38 | |
| *** sflanigan_ has joined #openstack-infra | 02:38 | |
| *** camunoz__ has joined #openstack-infra | 02:38 | |
| clarkb | r1chardj0n3s: we should make sure the other infra core are on board with the way I shuffled the jobs but it is a simple change so I don't expect much opposition | 02:38 |
| *** ianw has joined #openstack-infra | 02:38 | |
| r1chardj0n3s | clarkb: yep cool. also jhesketh is on leave at the moment, but I don't expect him to be too upset ;) | 02:39 |
| *** sflanigan_ has quit IRC | 02:39 | |
| *** sflanigan_ has joined #openstack-infra | 02:39 | |
| *** sflanigan_ has joined #openstack-infra | 02:39 | |
| *** camunoz has quit IRC | 02:39 | |
| *** ianw has quit IRC | 02:39 | |
| *** sflanigan_ has quit IRC | 02:40 | |
| *** sflanigan__ has joined #openstack-infra | 02:40 | |
| r1chardj0n3s | clarkb: once the check is in place I'll force a check for nova to make sure it's all peachy | 02:40 |
| *** camunoz has joined #openstack-infra | 02:40 | |
| *** ianw has joined #openstack-infra | 02:40 | |
| *** camunoz has quit IRC | 02:40 | |
| *** sflanigan__ has quit IRC | 02:40 | |
| *** camunoz has joined #openstack-infra | 02:40 | |
| *** sflanigan__ has joined #openstack-infra | 02:40 | |
| *** ianw has quit IRC | 02:40 | |
| *** camunoz has quit IRC | 02:41 | |
| *** camunoz has joined #openstack-infra | 02:41 | |
| *** sflanigan__ has quit IRC | 02:41 | |
| *** sflanigan_ has joined #openstack-infra | 02:41 | |
| *** sflanigan_ has joined #openstack-infra | 02:41 | |
| *** ianw has joined #openstack-infra | 02:41 | |
| *** laxity has joined #openstack-infra | 02:41 | |
| *** camunoz_ has quit IRC | 02:42 | |
| *** sflanigan_ has quit IRC | 02:42 | |
| *** sflanigan_ has joined #openstack-infra | 02:42 | |
| *** sflanigan_ has joined #openstack-infra | 02:42 | |
| *** camunoz_ has joined #openstack-infra | 02:42 | |
| *** jamielennox|away has quit IRC | 02:42 | |
| *** sflanigan__ has joined #openstack-infra | 02:42 | |
| *** jamielennox|away has joined #openstack-infra | 02:42 | |
| *** camunoz__ has quit IRC | 02:43 | |
| *** trinaths has quit IRC | 02:43 | |
| *** camunoz__ has joined #openstack-infra | 02:43 | |
| *** harlowja is now known as harlowja_away | 02:44 | |
| *** ianw has quit IRC | 02:45 | |
| *** camunoz has quit IRC | 02:46 | |
| *** camunoz_ has quit IRC | 02:46 | |
| *** ianw has joined #openstack-infra | 02:46 | |
| *** sflanigan_ has quit IRC | 02:46 | |
| *** dims has joined #openstack-infra | 02:54 | |
| *** bdpayne has joined #openstack-infra | 02:54 | |
| *** sflanigan_ has joined #openstack-infra | 02:58 | |
| *** sflanigan_ has joined #openstack-infra | 02:58 | |
| *** jamielennox|away has quit IRC | 02:59 | |
| *** sflanigan__ has quit IRC | 02:59 | |
| *** ianw has quit IRC | 02:59 | |
| *** sflanigan__ has joined #openstack-infra | 02:59 | |
| *** ianw has joined #openstack-infra | 02:59 | |
| *** jamielennox|away has joined #openstack-infra | 03:00 | |
| *** mrrrgn has quit IRC | 03:00 | |
| *** camunoz has joined #openstack-infra | 03:00 | |
| *** bdpayne has quit IRC | 03:01 | |
| *** rcarrillocruz has quit IRC | 03:02 | |
| *** camunoz__ has quit IRC | 03:02 | |
| *** rcarrillocruz has joined #openstack-infra | 03:02 | |
| *** sflanigan_ has quit IRC | 03:03 | |
| *** ianw has quit IRC | 03:04 | |
| *** KanagarajM has joined #openstack-infra | 03:04 | |
| *** ianw has joined #openstack-infra | 03:05 | |
| *** annegentle has joined #openstack-infra | 03:10 | |
| *** xchu has quit IRC | 03:11 | |
| *** changbl has joined #openstack-infra | 03:12 | |
| *** dims has quit IRC | 03:16 | |
| *** gyee has quit IRC | 03:17 | |
| *** david-lyle has joined #openstack-infra | 03:17 | |
| *** sdake_ has quit IRC | 03:20 | |
| *** zhiyuan has joined #openstack-infra | 03:21 | |
| *** david-lyle has quit IRC | 03:22 | |
| *** nosnos has quit IRC | 03:23 | |
| *** rcarrillocruz has quit IRC | 03:24 | |
| *** nosnos has joined #openstack-infra | 03:24 | |
| *** rcarrillocruz has joined #openstack-infra | 03:25 | |
| *** nosnos has quit IRC | 03:28 | |
| *** rick_ has joined #openstack-infra | 03:28 | |
| *** nkinder has joined #openstack-infra | 03:28 | |
| rick_ | anteaya: Hi Anteaya, I remove the "Start checking jobs." https://review.openstack.org/#/c/115886/ | 03:29 |
| *** harlowja_away has quit IRC | 03:29 | |
| *** ianw has quit IRC | 03:38 | |
| *** sflanigan__ has quit IRC | 03:38 | |
| *** jamielennox|away has quit IRC | 03:38 | |
| *** ianw has joined #openstack-infra | 03:38 | |
| *** sflanigan has joined #openstack-infra | 03:38 | |
| *** sflanigan has joined #openstack-infra | 03:38 | |
| *** camunoz_ has joined #openstack-infra | 03:38 | |
| *** jamielennox|away has joined #openstack-infra | 03:39 | |
| *** camunoz has quit IRC | 03:41 | |
| *** rcarrillocruz has quit IRC | 03:45 | |
| *** rcarrillocruz has joined #openstack-infra | 03:49 | |
| *** nosnos has joined #openstack-infra | 03:53 | |
| *** cnesa7 has quit IRC | 03:54 | |
| *** harlowja_at_home has joined #openstack-infra | 03:55 | |
| *** praneshp has joined #openstack-infra | 03:56 | |
| *** gokrokve has quit IRC | 03:59 | |
| *** gokrokve has joined #openstack-infra | 03:59 | |
| *** praneshp_ has joined #openstack-infra | 04:00 | |
| *** oomichi has joined #openstack-infra | 04:01 | |
| *** praneshp has quit IRC | 04:02 | |
| *** praneshp_ is now known as praneshp | 04:03 | |
| *** bdpayne has joined #openstack-infra | 04:03 | |
| *** hdd has quit IRC | 04:05 | |
| *** garyh has quit IRC | 04:09 | |
| *** dane_leblanc has quit IRC | 04:09 | |
| *** rcarrillocruz has quit IRC | 04:11 | |
| *** daya_k has joined #openstack-infra | 04:13 | |
| *** rcarrillocruz has joined #openstack-infra | 04:15 | |
| *** primeministerp has quit IRC | 04:19 | |
| *** trinaths has joined #openstack-infra | 04:23 | |
| *** todd_dsm has joined #openstack-infra | 04:26 | |
| *** rick__ has quit IRC | 04:29 | |
| *** lcheng_ has joined #openstack-infra | 04:34 | |
| *** koolhead17 has joined #openstack-infra | 04:41 | |
| *** rcarrillocruz has quit IRC | 04:43 | |
| *** ominakov has quit IRC | 04:43 | |
| *** Ryan_Lane has joined #openstack-infra | 04:44 | |
| *** annegentle has quit IRC | 04:44 | |
| *** annegentle has joined #openstack-infra | 04:46 | |
| *** annegentle has quit IRC | 04:46 | |
| *** daya_k has quit IRC | 04:47 | |
| *** rcarrillocruz has joined #openstack-infra | 04:47 | |
| *** chuck_ has quit IRC | 04:49 | |
| *** ppai has joined #openstack-infra | 04:52 | |
| *** yfried_ has quit IRC | 04:53 | |
| *** primeministerp has joined #openstack-infra | 04:53 | |
| ppai | anteaya, ping | 04:53 |
| ppai | hi, does anyone know why "action = cherry pick" mode is discouraged for stackforge projects ? | 04:55 |
| *** tomoe_ has joined #openstack-infra | 04:56 | |
| *** yaguang is now known as yaguang_lunch | 04:57 | |
| *** ildikov_ has quit IRC | 04:57 | |
| *** yamahata has quit IRC | 04:58 | |
| *** yamahata has joined #openstack-infra | 04:58 | |
| *** tomoe_ has quit IRC | 05:00 | |
| clarkb | because it doesnt work well | 05:07 |
| clarkb | gerrit rewrites commits and updates commit messages | 05:08 |
| *** rcarrillocruz has quit IRC | 05:08 | |
| clarkb | tooling doesnt handle that well | 05:08 |
| clarkb | we tried it with the config repo for a while | 05:09 |
| clarkb | ppai ^ | 05:09 |
| ppai | clarkb, okay.. gerrit appends commit msgs with certain additional info..is that causing the problem ? | 05:10 |
| clarkb | yes its part of the problem | 05:10 |
| StevenK | ppai: If the commit message changes, the sha1 hash of the commit changes too | 05:10 |
| ppai | StevenK, yes | 05:10 |
| StevenK | Which can also cause problems for developer workflow | 05:11 |
| ppai | is there a way to avoid the additional merge commit entries in git log ? | 05:11 |
| *** rfolco has quit IRC | 05:11 | |
| clarkb | not today | 05:11 |
| clarkb | they arent that problematic | 05:12 |
| *** rcarrillocruz has joined #openstack-infra | 05:12 | |
| ppai | the only down side of cherry pick i see is the repo maintainer must merge "chained patchsets" in the right order | 05:12 |
| clarkb | ppai it makes all of the tooling cranky | 05:13 |
| clarkb | zuul, git, and so on | 05:13 |
| *** rfolco has joined #openstack-infra | 05:14 | |
| ppai | clarkb, Stevenk, alright, thanks for the info...will change it to "merge if necessary" :) | 05:14 |
| jogo | I take it you guys have noticed http://logs.openstack.org/26/115526/4/check/check-tempest-dsvm-full/81389e2/logs/devstacklog.txt.gz#_2014-08-27_01_13_56_103 | 05:14 |
| jogo | file '/tmp/pip-build.uznj8/python-subunit/filters/subunit2cvs' does not exist | 05:14 |
| StevenK | If the patchset's parent is current master HEAD, it will apply without a merge commit | 05:14 |
| StevenK | jogo: That was pointed out to lifeless a few hours ago, not sure what happened with it | 05:15 |
| clarkb | jogo its fixed | 05:16 |
| jogo | clarkb: thanks, is there a bug for it so I can file a e-r patch | 05:16 |
| jogo | unless someone else has already | 05:16 |
| StevenK | There is a bug for it, not sure about an e-r change | 05:16 |
| ppai | glusterfs is one project where i've seen cherrypick work really well...and the git log looks so clean and commits ordered..BUT they don;t have zuul | 05:16 |
| jesusaurus | clarkb: with cherry-picking does the zuul gate queue do the Right Thing for testing? | 05:17 |
| *** harlowja_at_home has quit IRC | 05:18 | |
| jogo | clarkb: I don't see any infra bug | 05:18 |
| clarkb | jesusaurus iirc it can but not mixed with projects that do merges | 05:18 |
| jesusaurus | ahh | 05:18 |
| clarkb | jogo its not an infra bug | 05:18 |
| jogo | clarkb: subunit2cvs has 0 hits on launchpad for openstack | 05:18 |
| jogo | clarkb: not sure where it would be filed | 05:18 |
| jogo | or where to file one | 05:18 |
| clarkb | jogo its a subunit bug which was fixed | 05:19 |
| jogo | clarkb: nothin in launchpad subunit either | 05:19 |
| jogo | clarkb: I just need a bug number so I can add a e-r patch | 05:19 |
| clarkb | jogo no e-r patch | 05:19 |
| clarkb | its fixed... | 05:20 |
| jogo | clarkb: so we can get the unclassifiedpage to be accurate | 05:20 |
| StevenK | jogo: bug 1361924 | 05:20 |
| uvirtbot` | Launchpad bug 1361924 in subunit "python-subunit 0.0.20 failing to install causing gate failure" [Critical,Fix released] https://launchpad.net/bugs/1361924 | 05:20 |
| jogo | StevenK: thans | 05:21 |
| *** isaacb has joined #openstack-infra | 05:21 | |
| *** penguinRaider has quit IRC | 05:24 | |
| jogo | StevenK clarkb: https://review.openstack.org/117093 | 05:25 |
| *** markmc has joined #openstack-infra | 05:32 | |
| *** terryw has quit IRC | 05:32 | |
| *** melwitt1 has quit IRC | 05:34 | |
| *** bdpayne has quit IRC | 05:36 | |
| *** SumitNaiksatam has joined #openstack-infra | 05:40 | |
| *** rlandy has joined #openstack-infra | 05:43 | |
| *** yfried_ has joined #openstack-infra | 05:43 | |
| *** rcarrillocruz has quit IRC | 05:45 | |
| yfried_ | anyone knows how I can push something to gerrit on behalf of a colleague? | 05:46 |
| *** harlowja_at_home has joined #openstack-infra | 05:46 | |
| *** rcarrillocruz has joined #openstack-infra | 05:46 | |
| trinaths | I get this error when I run devstack-gate. fatal: http://zuul.openstack.org/p/openstack/neutron/info/refs not found | 05:47 |
| trinaths | can any one seen this before. Kindly help me resolve this issue. | 05:47 |
| *** bdpayne has joined #openstack-infra | 05:48 | |
| *** garyh has joined #openstack-infra | 05:48 | |
| *** todd_dsm has quit IRC | 05:51 | |
| trinaths | it is The URL from which to fetch ZUUL references | 05:51 |
| trinaths | when I browse this URL it give me an empty page | 05:51 |
| *** k4n0 has joined #openstack-infra | 05:59 | |
| *** ildikov_ has joined #openstack-infra | 05:59 | |
| *** sflanigan has quit IRC | 06:08 | |
| *** ianw has quit IRC | 06:08 | |
| *** sflanigan has joined #openstack-infra | 06:08 | |
| *** camunoz has joined #openstack-infra | 06:08 | |
| *** ianw has joined #openstack-infra | 06:08 | |
| *** jamielennox|away has quit IRC | 06:10 | |
| *** jamielennox|away has joined #openstack-infra | 06:11 | |
| *** camunoz_ has quit IRC | 06:11 | |
| *** harlowja_at_home has quit IRC | 06:11 | |
| *** loki184 has joined #openstack-infra | 06:12 | |
| *** chandankumar has joined #openstack-infra | 06:14 | |
| *** flaper87|afk is now known as flaper87 | 06:15 | |
| *** yaguang_lunch is now known as yaguang | 06:15 | |
| *** isaacb has quit IRC | 06:16 | |
| *** bdpayne has quit IRC | 06:18 | |
| *** rcarrillocruz has quit IRC | 06:19 | |
| *** Longgeek has joined #openstack-infra | 06:20 | |
| *** frank_zhou has joined #openstack-infra | 06:20 | |
| *** rcarrillocruz has joined #openstack-infra | 06:22 | |
| *** skraynev has quit IRC | 06:23 | |
| *** skraynev has joined #openstack-infra | 06:25 | |
| *** nelsnelson has quit IRC | 06:29 | |
| *** nelsnelson has joined #openstack-infra | 06:33 | |
| *** xchu has joined #openstack-infra | 06:36 | |
| *** jcoufal has joined #openstack-infra | 06:38 | |
| *** mpaolino has joined #openstack-infra | 06:40 | |
| *** rcarrillocruz has quit IRC | 06:41 | |
| *** dkehn_ has quit IRC | 06:43 | |
| *** dkehnx has quit IRC | 06:43 | |
| *** rcarrillocruz has joined #openstack-infra | 06:44 | |
| *** dkehn has joined #openstack-infra | 06:44 | |
| *** dkehn_ has joined #openstack-infra | 06:45 | |
| *** isaacb has joined #openstack-infra | 06:51 | |
| *** tnurlygayanov has joined #openstack-infra | 06:56 | |
| *** doude has joined #openstack-infra | 06:57 | |
| *** daya_k has joined #openstack-infra | 06:58 | |
| *** afazekas has joined #openstack-infra | 06:59 | |
| *** rcarrillocruz has quit IRC | 07:02 | |
| *** nosnos has quit IRC | 07:04 | |
| *** nosnos_ has joined #openstack-infra | 07:04 | |
| *** rcarrillocruz has joined #openstack-infra | 07:05 | |
| flaper87 | SergeyLukjanov: fungi et all: Good morning guys. This is happening in Glance https://review.openstack.org/#/c/113566/ | 07:07 |
| flaper87 | before that patch lands, we thought that it would be nice to double check that it won't break infra/gate/put_here_whatever_could_be_broken | 07:07 |
| *** HeOS has joined #openstack-infra | 07:08 | |
| flaper87 | Do you guys think moving the tests out of glance could break something? If so, what can we do to fix/workaround that? We need this patch for glance.store to succeed | 07:08 |
| *** praneshp_ has joined #openstack-infra | 07:09 | |
| *** freyes has quit IRC | 07:09 | |
| *** praneshp has quit IRC | 07:10 | |
| *** praneshp_ is now known as praneshp | 07:10 | |
| *** freyes has joined #openstack-infra | 07:10 | |
| *** HeOS has quit IRC | 07:14 | |
| *** hanzhang has joined #openstack-infra | 07:15 | |
| *** hanzhang has left #openstack-infra | 07:16 | |
| *** gema has quit IRC | 07:16 | |
| *** hanzhang has joined #openstack-infra | 07:17 | |
| *** dizquierdo has joined #openstack-infra | 07:17 | |
| hanzhang | hi,all | 07:17 |
| *** jgallard has joined #openstack-infra | 07:18 | |
| *** andreykurilin has joined #openstack-infra | 07:19 | |
| hanzhang | I want to install ipset on devstack if q-agt service is available, who can help me? | 07:19 |
| *** e0ne has joined #openstack-infra | 07:20 | |
| hanzhang | now I put the ipset name into files/*/q-agt, but the jekins failed | 07:21 |
| *** isaacb has quit IRC | 07:22 | |
| *** hanzhang has quit IRC | 07:23 | |
| *** jgallard has quit IRC | 07:24 | |
| *** Ryan_Lane has quit IRC | 07:24 | |
| *** jgallard has joined #openstack-infra | 07:25 | |
| *** r1chardj0n3s is now known as r1chardj0n3s_afk | 07:25 | |
| *** hanzhang has joined #openstack-infra | 07:25 | |
| hanzhang | ping jgallard | 07:25 |
| hanzhang | I want to install ipset on devstack if q-agt service is available,now I put the ipset name into files/*/q-agt, but the jekins failed | 07:26 |
| hanzhang | ,who can help me? | 07:26 |
| jgallard | hi hanzhang | 07:28 |
| jgallard | sorry, I don't know | 07:28 |
| hanzhang | thanks anywaythanks all the same | 07:29 |
| *** e0ne has quit IRC | 07:31 | |
| *** dkehn_ has quit IRC | 07:31 | |
| gus | hrm, perhaps we should make new changes to our pypi mirror also pass zuul first :/ | 07:32 |
| *** dkehn has quit IRC | 07:32 | |
| *** Hal_ has joined #openstack-infra | 07:33 | |
| *** linkid has joined #openstack-infra | 07:34 | |
| *** isaacb has joined #openstack-infra | 07:35 | |
| *** e0ne has joined #openstack-infra | 07:36 | |
| *** jgallard has quit IRC | 07:36 | |
| *** HeOS has joined #openstack-infra | 07:37 | |
| *** e0ne has quit IRC | 07:38 | |
| *** jamielennox|away has quit IRC | 07:38 | |
| *** ianw has quit IRC | 07:38 | |
| *** camunoz_ has joined #openstack-infra | 07:38 | |
| *** sflanigan has quit IRC | 07:38 | |
| *** ianw has joined #openstack-infra | 07:38 | |
| *** sflanigan has joined #openstack-infra | 07:39 | |
| *** sflanigan has quit IRC | 07:39 | |
| *** sflanigan has joined #openstack-infra | 07:39 | |
| *** jamielennox|away has joined #openstack-infra | 07:39 | |
| *** gokrokve_ has joined #openstack-infra | 07:40 | |
| *** yolanda has joined #openstack-infra | 07:41 | |
| *** camunoz has quit IRC | 07:41 | |
| *** yolanda_ has quit IRC | 07:42 | |
| *** camunoz_ has quit IRC | 07:42 | |
| *** e0ne has joined #openstack-infra | 07:42 | |
| *** ppai has quit IRC | 07:43 | |
| *** amuller has joined #openstack-infra | 07:43 | |
| *** gokrokve has quit IRC | 07:43 | |
| *** ociuhandu has joined #openstack-infra | 07:44 | |
| *** HeOS has quit IRC | 07:50 | |
| *** isaacb has quit IRC | 07:50 | |
| *** rcarrillocruz has quit IRC | 07:51 | |
| *** fbo_away has joined #openstack-infra | 07:52 | |
| *** fbo_away is now known as fbo | 07:52 | |
| *** fbo has quit IRC | 07:55 | |
| *** e0ne has quit IRC | 07:55 | |
| *** rcarrillocruz has joined #openstack-infra | 07:56 | |
| *** e0ne has joined #openstack-infra | 07:59 | |
| *** jpich has joined #openstack-infra | 07:59 | |
| *** oanufriev has quit IRC | 07:59 | |
| *** salv-orlando_ has joined #openstack-infra | 08:00 | |
| *** oanufriev has joined #openstack-infra | 08:00 | |
| *** jistr has joined #openstack-infra | 08:00 | |
| *** funzo has joined #openstack-infra | 08:00 | |
| *** _buzz has joined #openstack-infra | 08:01 | |
| *** andreykurilin has quit IRC | 08:01 | |
| *** michchap_ has joined #openstack-infra | 08:01 | |
| *** dguerri_ has joined #openstack-infra | 08:01 | |
| *** jgallard has joined #openstack-infra | 08:02 | |
| *** jlibosva has joined #openstack-infra | 08:02 | |
| *** electrichead has joined #openstack-infra | 08:02 | |
| *** HeOS has joined #openstack-infra | 08:02 | |
| *** BadCub has joined #openstack-infra | 08:03 | |
| *** praneshp has quit IRC | 08:03 | |
| *** TheJulia_ has joined #openstack-infra | 08:04 | |
| *** gpocente1 has joined #openstack-infra | 08:05 | |
| *** frank_zhou has quit IRC | 08:05 | |
| *** gilliard_ has joined #openstack-infra | 08:05 | |
| *** Hefeweiz1n has joined #openstack-infra | 08:05 | |
| *** pblaho has joined #openstack-infra | 08:05 | |
| *** e0ne_ has joined #openstack-infra | 08:05 | |
| *** _nadya_ has joined #openstack-infra | 08:05 | |
| *** dmakogon_ has joined #openstack-infra | 08:06 | |
| *** Daviey_ has joined #openstack-infra | 08:08 | |
| *** jd__` has joined #openstack-infra | 08:08 | |
| *** dhellmann_ has joined #openstack-infra | 08:08 | |
| *** portante_ has joined #openstack-infra | 08:08 | |
| *** SergeyLukjanov2 has joined #openstack-infra | 08:09 | |
| *** NobodyCa1 has joined #openstack-infra | 08:09 | |
| *** ruhe2 has joined #openstack-infra | 08:09 | |
| *** loquacity has joined #openstack-infra | 08:09 | |
| *** e0ne has quit IRC | 08:10 | |
| *** hanzhang has quit IRC | 08:10 | |
| *** mattymo has quit IRC | 08:10 | |
| *** asadoughi has quit IRC | 08:10 | |
| *** jd__ has quit IRC | 08:10 | |
| *** NobodyCam has quit IRC | 08:10 | |
| *** TheJulia has quit IRC | 08:10 | |
| *** _buzz_ has quit IRC | 08:10 | |
| *** Daviey has quit IRC | 08:10 | |
| *** dkliban has quit IRC | 08:10 | |
| *** redrobot has quit IRC | 08:10 | |
| *** funzo_ has quit IRC | 08:10 | |
| *** gpocentek has quit IRC | 08:10 | |
| *** michchap has quit IRC | 08:10 | |
| *** BadCub_ has quit IRC | 08:10 | |
| *** jgriffith has quit IRC | 08:10 | |
| *** gilliard has quit IRC | 08:10 | |
| *** ruhe has quit IRC | 08:10 | |
| *** loki184 has quit IRC | 08:10 | |
| *** fabo has quit IRC | 08:10 | |
| *** RicoLin has quit IRC | 08:10 | |
| *** dguerri has quit IRC | 08:10 | |
| *** Hefeweizen has quit IRC | 08:10 | |
| *** denis_makogon has quit IRC | 08:10 | |
| *** salv-orlando has quit IRC | 08:10 | |
| *** loquacities has quit IRC | 08:10 | |
| *** dhellmann has quit IRC | 08:10 | |
| *** portante has quit IRC | 08:10 | |
| *** bradm has quit IRC | 08:10 | |
| *** SergeyLukjanov has quit IRC | 08:10 | |
| *** SergeyLukjanov2 is now known as SergeyLukjanov | 08:10 | |
| *** salv-orlando_ is now known as salv-orlando | 08:10 | |
| *** Daviey_ is now known as Daviey | 08:10 | |
| *** ruhe2 is now known as ruhe | 08:10 | |
| *** loki184 has joined #openstack-infra | 08:10 | |
| *** isaacb has joined #openstack-infra | 08:10 | |
| *** dhellmann_ has quit IRC | 08:12 | |
| *** jd__` is now known as jd__ | 08:12 | |
| *** mattymo has joined #openstack-infra | 08:12 | |
| *** dhellmann has joined #openstack-infra | 08:12 | |
| *** hanzhang has joined #openstack-infra | 08:12 | |
| *** p5ntangle has joined #openstack-infra | 08:12 | |
| *** fabo has joined #openstack-infra | 08:12 | |
| *** jgriffith has joined #openstack-infra | 08:12 | |
| *** stevemar has quit IRC | 08:14 | |
| *** e0ne_ has quit IRC | 08:14 | |
| *** derekh has joined #openstack-infra | 08:14 | |
| *** RicoLin has joined #openstack-infra | 08:15 | |
| *** RicoLin has quit IRC | 08:15 | |
| *** RicoLin has joined #openstack-infra | 08:16 | |
| *** lcheng_ has quit IRC | 08:16 | |
| *** tomoe_ has joined #openstack-infra | 08:16 | |
| *** dkliban has joined #openstack-infra | 08:17 | |
| *** asadoughi has joined #openstack-infra | 08:17 | |
| *** flaper87 is now known as flaper87|afk | 08:17 | |
| *** bradm has joined #openstack-infra | 08:18 | |
| Hal_ | Hi ! I have an issue with the gate-nova-docs, the sphinx setup includes the mathpng extension but the gate does not have dvipng installed, so math formula png generation fails. Do you think this setup can be changed or the use of mathpng must be avoided ? | 08:18 |
| lifeless | Hal_: you can submit patches to fix it to the openstack-infra project | 08:20 |
| *** Longgeek has quit IRC | 08:20 | |
| *** Longgeek has joined #openstack-infra | 08:21 | |
| *** hashar has joined #openstack-infra | 08:23 | |
| Hal_ | lifeless: wich one ? | 08:24 |
| *** alexpilotti has joined #openstack-infra | 08:25 | |
| *** Longgeek has quit IRC | 08:25 | |
| lifeless | Hal_: not 100% offhand - binary deps for tempesty stuff are in devstack, and unittest setup is in the nodepool config (openstack-infra/config) - I haven't looked to see what node type -docs runs on | 08:25 |
| *** rcarrillocruz has quit IRC | 08:27 | |
| *** gpocente1 is now known as gpocentek | 08:28 | |
| *** gpocentek is now known as Guest32108 | 08:28 | |
| *** Guest32108 is now known as gpocentek | 08:30 | |
| *** tomoe_ has quit IRC | 08:30 | |
| *** gpocentek has quit IRC | 08:30 | |
| *** gpocentek has joined #openstack-infra | 08:30 | |
| *** Hal_ has quit IRC | 08:30 | |
| *** rcarrillocruz has joined #openstack-infra | 08:31 | |
| *** ociuhandu has quit IRC | 08:33 | |
| *** _nadya_ has quit IRC | 08:37 | |
| *** AlecTaylor has joined #openstack-infra | 08:38 | |
| *** KanagarajM has quit IRC | 08:41 | |
| *** lcheng_ has joined #openstack-infra | 08:41 | |
| *** hashar_ has joined #openstack-infra | 08:42 | |
| *** dkehn has joined #openstack-infra | 08:42 | |
| *** dizquierdo has quit IRC | 08:43 | |
| *** yfried__ has joined #openstack-infra | 08:43 | |
| *** KanagarajM has joined #openstack-infra | 08:43 | |
| *** Longgeek has joined #openstack-infra | 08:43 | |
| *** yfried_ has quit IRC | 08:44 | |
| *** Hal_ has joined #openstack-infra | 08:45 | |
| *** nosnos_ has quit IRC | 08:45 | |
| *** dkehn_ has joined #openstack-infra | 08:45 | |
| *** hashar has quit IRC | 08:46 | |
| *** nosnos has joined #openstack-infra | 08:46 | |
| *** nosnos has quit IRC | 08:50 | |
| *** isviridov|away is now known as isviridov | 08:51 | |
| *** rcarrillocruz has quit IRC | 08:53 | |
| *** rcarrillocruz has joined #openstack-infra | 08:57 | |
| *** ppai has joined #openstack-infra | 08:58 | |
| *** rick_ has quit IRC | 08:58 | |
| *** lukego has joined #openstack-infra | 09:00 | |
| *** p5ntangle has quit IRC | 09:03 | |
| *** p5ntangle has joined #openstack-infra | 09:03 | |
| *** e0ne has joined #openstack-infra | 09:04 | |
| *** jgallard has quit IRC | 09:04 | |
| *** pelix1 has joined #openstack-infra | 09:05 | |
| *** yamahata has quit IRC | 09:06 | |
| *** jgallard has joined #openstack-infra | 09:07 | |
| *** gokrokve has joined #openstack-infra | 09:09 | |
| *** oomichi has quit IRC | 09:09 | |
| *** nosnos has joined #openstack-infra | 09:11 | |
| *** plars has quit IRC | 09:11 | |
| *** luqas has joined #openstack-infra | 09:12 | |
| *** plars has joined #openstack-infra | 09:12 | |
| *** gokrokve_ has quit IRC | 09:13 | |
| *** gokrokve has quit IRC | 09:13 | |
| *** yamahata has joined #openstack-infra | 09:17 | |
| *** luqas has quit IRC | 09:17 | |
| *** luqas has joined #openstack-infra | 09:19 | |
| *** dizquierdo has joined #openstack-infra | 09:19 | |
| *** dtantsur is now known as dtantsur|brb | 09:20 | |
| *** skolekonov has joined #openstack-infra | 09:20 | |
| *** lcheng_ has quit IRC | 09:22 | |
| *** marun has joined #openstack-infra | 09:23 | |
| *** portante_ has quit IRC | 09:29 | |
| *** AlecTaylor has quit IRC | 09:31 | |
| *** rcarrillocruz has quit IRC | 09:31 | |
| *** soren_ has quit IRC | 09:32 | |
| *** soren_ has joined #openstack-infra | 09:32 | |
| *** soren_ is now known as soren | 09:32 | |
| *** chandankumar has quit IRC | 09:32 | |
| *** krtaylor has quit IRC | 09:32 | |
| *** dkliban has quit IRC | 09:33 | |
| *** sflanigan_ has joined #openstack-infra | 09:33 | |
| *** sflanigan_ has joined #openstack-infra | 09:33 | |
| *** nelsnelson has quit IRC | 09:33 | |
| *** jamielenz has joined #openstack-infra | 09:33 | |
| *** dizquierdo has quit IRC | 09:34 | |
| *** bcrochet has quit IRC | 09:34 | |
| *** plars has quit IRC | 09:34 | |
| *** eharney_ has quit IRC | 09:34 | |
| *** rcarrillocruz has joined #openstack-infra | 09:35 | |
| *** jamielennox|away has quit IRC | 09:36 | |
| *** ianw has quit IRC | 09:37 | |
| *** sflanigan has quit IRC | 09:37 | |
| *** ianw has joined #openstack-infra | 09:37 | |
| *** timrc is now known as timrc-afk | 09:37 | |
| *** yfried has joined #openstack-infra | 09:37 | |
| *** jp_at_hp has joined #openstack-infra | 09:38 | |
| marun | clarkb, fungi, jeblair: The hook scripts have merged to Neutron, so the neutron functional change is now ready to be merged: https://review.openstack.org/#/c/116458/ | 09:39 |
| *** gokrokve has joined #openstack-infra | 09:39 | |
| *** yfried__ has quit IRC | 09:39 | |
| *** MmikeM is now known as Mmike | 09:39 | |
| *** Mmike has quit IRC | 09:39 | |
| *** Mmike has joined #openstack-infra | 09:39 | |
| *** gokrokve has quit IRC | 09:40 | |
| *** chandankumar has joined #openstack-infra | 09:40 | |
| *** gokrokve has joined #openstack-infra | 09:41 | |
| marun | sdague: I'm still curious as to what you think of this attempt to support minimal service installation for the neutron functional job: https://review.openstack.org/#/c/116892 | 09:41 |
| *** cnesa7 has joined #openstack-infra | 09:42 | |
| *** linkid has left #openstack-infra | 09:43 | |
| *** luqas has quit IRC | 09:43 | |
| *** portante has joined #openstack-infra | 09:43 | |
| *** KanagarajM2 has joined #openstack-infra | 09:44 | |
| *** e0ne has quit IRC | 09:44 | |
| *** e0ne has joined #openstack-infra | 09:44 | |
| *** KanagarajM has quit IRC | 09:45 | |
| *** cnesa8 has joined #openstack-infra | 09:45 | |
| *** gokrokve has quit IRC | 09:45 | |
| *** dizquierdo has joined #openstack-infra | 09:46 | |
| Hal_ | lifeless: I tried to add dvipng to test-requirements.txt but it did not work because global-requirements is not contain "dvipng" package. I totally do not know how to handle this issue :-/ | 09:46 |
| *** cnesa7 has quit IRC | 09:47 | |
| *** bcrochet has joined #openstack-infra | 09:47 | |
| *** hashar_ is now known as hashar | 09:47 | |
| *** luqas has joined #openstack-infra | 09:48 | |
| StevenK | Hal_: global-requirements.txt is in the openstack/requirements repository. Clone it and push up a patch for it first. | 09:48 |
| *** pblaho is now known as pblaho|afk | 09:50 | |
| Hal_ | StevenK: is it possible to add non python req in this file ? Like the dvipng program (needed by sphinx mathpng extension) ? | 09:50 |
| *** Longgeek has quit IRC | 09:50 | |
| StevenK | Hal_: requirements.txt is also for Python requirements. | 09:51 |
| *** dkliban has joined #openstack-infra | 09:52 | |
| StevenK | So I'm not sure. If your goal is for docs that are generated using OpenStack's CI, you should look at openstack-infra/devstack-gate. | 09:52 |
| Hal_ | StevenK: yes it is | 09:52 |
| StevenK | Hal_: You could also chat with fungi or jeblair when they're around in a few hours. | 09:53 |
| Hal_ | StevenK: Ok, thanks I will contact them. | 09:53 |
| *** HeOS has quit IRC | 09:58 | |
| *** xianghuihui has joined #openstack-infra | 10:01 | |
| *** luqas has quit IRC | 10:03 | |
| *** AlecTaylor has joined #openstack-infra | 10:05 | |
| *** AlecTaylor has quit IRC | 10:05 | |
| *** AlecTaylor has joined #openstack-infra | 10:05 | |
| *** pcm_ has joined #openstack-infra | 10:06 | |
| *** sflanigan__ has joined #openstack-infra | 10:07 | |
| *** zz_johnthetubagu is now known as johnthetubaguy | 10:07 | |
| *** jamielennox|away has joined #openstack-infra | 10:07 | |
| *** HeOS has joined #openstack-infra | 10:08 | |
| *** pcm_ has quit IRC | 10:08 | |
| *** ianw has quit IRC | 10:09 | |
| *** ianw has joined #openstack-infra | 10:09 | |
| *** gokrokve has joined #openstack-infra | 10:09 | |
| *** pcm_ has joined #openstack-infra | 10:09 | |
| *** jamielenz has quit IRC | 10:09 | |
| *** sflanigan_ has quit IRC | 10:10 | |
| *** Hal_ has quit IRC | 10:10 | |
| *** che-arne has joined #openstack-infra | 10:11 | |
| *** tomoe_ has joined #openstack-infra | 10:11 | |
| *** hashar has quit IRC | 10:13 | |
| *** gokrokve has quit IRC | 10:14 | |
| *** linkid has joined #openstack-infra | 10:15 | |
| *** Longgeek has joined #openstack-infra | 10:16 | |
| *** xchu has quit IRC | 10:16 | |
| *** HeOS has quit IRC | 10:17 | |
| *** rcarrillocruz has quit IRC | 10:17 | |
| *** amuller is now known as amuller_lunch | 10:19 | |
| *** rcarrillocruz has joined #openstack-infra | 10:21 | |
| *** jab has joined #openstack-infra | 10:23 | |
| *** jab has joined #openstack-infra | 10:23 | |
| daya_k | hello infra, had a q on whether zuul should be starting its own http server for the 3rd party ci system | 10:24 |
| daya_k | i dont see on in my setup, only the jenkins http server, and my slave node cannot do a git fetch of the zuul change for the foll url - git fetch http://<servername>/p/openstack/neutron | 10:25 |
| daya_k | the only httpd i have on this node is the one jenkins uses | 10:25 |
| *** marun is now known as marun_afk | 10:25 | |
| *** chandankumar has quit IRC | 10:26 | |
| *** tomoe_ has quit IRC | 10:26 | |
| *** chandankumar has joined #openstack-infra | 10:27 | |
| *** tsufiev_ is now known as tsufiev | 10:29 | |
| *** Longgeek has quit IRC | 10:32 | |
| *** jgallard has quit IRC | 10:35 | |
| *** flaviof_zzz is now known as flaviof | 10:35 | |
| *** jab has quit IRC | 10:35 | |
| *** oanufriev has quit IRC | 10:35 | |
| *** oanufriev has joined #openstack-infra | 10:37 | |
| *** pblaho|afk is now known as pblaho | 10:37 | |
| skolekonov | daya_k, hello. If you installed Zuul manually, you should also configure your Apache server for Zuul (git repo to fetch changes and stats page optionally) | 10:38 |
| daya_k | thanks for responding skolekonov: i used puppet to install all the modules, i am trying a re-install of the scripts | 10:39 |
| ttx | jogo: been looking into categorization for unit tests fails, but in most cases they seem to come from the same change being tested over and over (i.e. same faulty patchset generated 6 fails, gets fixed and all is good) | 10:39 |
| daya_k | skolekonov: trying to understand what to look out for, so i should see a separate httpd owned by zuul in the process list? | 10:39 |
| *** AlecTaylor has quit IRC | 10:41 | |
| skolekonov | No, I think you shouldn't. Check that you have separate file for Zuul in Apache's configuration directories | 10:42 |
| daya_k | ok, let me check | 10:42 |
| *** zul has joined #openstack-infra | 10:43 | |
| skolekonov | I think it should look like sites-enabled/xx-<server-fqdn>.conf | 10:44 |
| daya_k | what dir do i check? /var/cache/apache2 does not have a zuul subdir | 10:45 |
| daya_k | oh yeah, under sites/enabled i see a 50-zuul.conf | 10:45 |
| *** marun_afk has quit IRC | 10:46 | |
| pcm_ | Hi all! Is there an issue with Jenkins? My latest patchset for 114345, failed ALL Jenkins tests. | 10:46 |
| *** flaper87|afk is now known as flaper87 | 10:47 | |
| pcm_ | Looking at some logs, I see this error: | 10:47 |
| pcm_ | error: file '/home/jenkins/workspace/gate-neutron-python27/.tox/py27/build/python-subunit/filters/subunit2cvs' does not exist | 10:47 |
| pcm_ | see this on each test failure | 10:47 |
| pcm_ | I can do a recheck, but didn't want to make things worse, if there is something munged up. | 10:48 |
| *** mpaolino has quit IRC | 10:51 | |
| *** zul has quit IRC | 10:51 | |
| *** HeOS has joined #openstack-infra | 10:51 | |
| *** pblaho has quit IRC | 10:53 | |
| daya_k | skolekonov : in that case, zuul git should be accessible if the zuul server is running? if i just type <server_ip>:80/p in my browser, it errors out, the actual repo is under /var/lib/zuul/git. do you know how this p directory is created, and where it should be located? | 10:53 |
| *** jab has joined #openstack-infra | 10:54 | |
| *** jab has joined #openstack-infra | 10:54 | |
| daya_k | GIT_PROJECT_ROOT is set in the file, but not clear to me where the /p directory comes from | 10:54 |
| daya_k | looking thru the 50-zuul.conf file, looks like its an alias mapping targets to /var/lib/zuul/git | 10:55 |
| *** timrc-afk is now known as timrc | 10:56 | |
| *** mmaglana has quit IRC | 10:57 | |
| *** nelsnelson has joined #openstack-infra | 10:57 | |
| *** plars has joined #openstack-infra | 10:57 | |
| *** krtaylor has joined #openstack-infra | 10:57 | |
| skolekonov | yes, zuul git repo should be available independently of Zuul server, because it's just a directory served by Apache. It has no Web UI, but you can try to clone repo from your server | 10:57 |
| *** ppai has quit IRC | 10:59 | |
| *** Longgeek has joined #openstack-infra | 11:00 | |
| *** hdd has joined #openstack-infra | 11:02 | |
| *** yamahata has quit IRC | 11:04 | |
| *** zhiyuan has quit IRC | 11:04 | |
| *** zul has joined #openstack-infra | 11:04 | |
| *** dmakogon_ is now known as denis_makogon | 11:05 | |
| *** rcarrillocruz has quit IRC | 11:06 | |
| *** zhiyuan has joined #openstack-infra | 11:07 | |
| *** hashar has joined #openstack-infra | 11:07 | |
| *** amuller_lunch is now known as amuller | 11:08 | |
| *** ZZelle has joined #openstack-infra | 11:09 | |
| *** gokrokve has joined #openstack-infra | 11:09 | |
| *** rcarrillocruz has joined #openstack-infra | 11:10 | |
| *** dhellmann has quit IRC | 11:10 | |
| *** zul has quit IRC | 11:11 | |
| daya_k | skolekonov: i see the git repo under this path, but the web page does not load it | 11:11 |
| skolekonov | the web page shouldn't load anything, it's ok | 11:12 |
| *** unicell1 has joined #openstack-infra | 11:13 | |
| daya_k | ok, let me try the git fetch again, i have done a re-install of puppet modules | 11:13 |
| *** dhellmann has joined #openstack-infra | 11:13 | |
| *** gokrokve has quit IRC | 11:14 | |
| *** zul has joined #openstack-infra | 11:14 | |
| *** flaper87 is now known as flaper87|afk | 11:14 | |
| *** ppai has joined #openstack-infra | 11:15 | |
| *** unicell has quit IRC | 11:16 | |
| *** hdd has quit IRC | 11:19 | |
| *** nosnos has quit IRC | 11:20 | |
| *** nosnos has joined #openstack-infra | 11:20 | |
| daya_k | skolekonov: git fetch http://sdnblrm48jm.in.ibm.com/p/openstack/neutron still gives me an error indicating repo not found | 11:21 |
| *** KanagarajM2 has quit IRC | 11:22 | |
| skolekonov | daya_k, check that repo exists on your server in /var/lib/zuul/git | 11:23 |
| *** flaper87|afk is now known as flaper87 | 11:23 | |
| daya_k | skolekonov: yes, its there. i see the entire neutron repo under openstack/neutron in this path | 11:24 |
| skolekonov | does git clone also fail? | 11:24 |
| *** marun_afk has joined #openstack-infra | 11:25 | |
| *** nosnos has quit IRC | 11:25 | |
| daya_k | acually, zuul updates this repo for all the checkins, i see the patch getting updated in this dir. but the slave node cannot fetch it while updating its own repo | 11:25 |
| *** loki184 has quit IRC | 11:25 | |
| daya_k | i checked a few patches while the builds were happening, to make sure they were there | 11:26 |
| daya_k | i tried a clone on the slave, fails | 11:29 |
| skolekonov | have you checked firewall settings? | 11:29 |
| daya_k | i can ping the master from the slave | 11:29 |
| *** yaguang has quit IRC | 11:30 | |
| daya_k | and zuul overall is working, so it triggers jobs, and logs are updated for the patches, so i dont think commu is a problem, the error is - fatal: repository 'http://sdnblrm48jm.in.ibm.com/p/openstack/neutron/' not found | 11:30 |
| daya_k | hey, should env show the GIT_PROJECT_ROOT variable set, i dont see it | 11:32 |
| skolekonov | it's strange, do you see anything in Apache's logs? | 11:32 |
| *** rcarrillocruz has quit IRC | 11:32 | |
| daya_k | so, wondering if the module was run at all or not. | 11:32 |
| *** ppai has quit IRC | 11:32 | |
| daya_k | hmm, i see this other_vhosts_access.log - sdnblrm48jm.in.ibm.com:80 9.126.153.159 - - [27/Aug/2014:11:29:11 +0000] "GET /p/openstack/neutron/info/refs?service=git-upload-pack HTTP/1.1" 404 465 "-" "git/1.9.3" | 11:36 |
| *** rcarrillocruz has joined #openstack-infra | 11:36 | |
| *** ociuhandu has joined #openstack-infra | 11:36 | |
| daya_k | the .159 machins is my slave node from where i do the git | 11:36 |
| *** dims has joined #openstack-infra | 11:36 | |
| daya_k | should the GET show the whole path? | 11:36 |
| *** ZZelle_ has joined #openstack-infra | 11:38 | |
| skolekonov | I think no. It seems like the site root directory is incorrectly set. | 11:38 |
| daya_k | do you see GIT_PROJECT_ROOT is set in your shell env, or is it set dynamically when the git command is received i wonder. ok, let me check | 11:39 |
| *** dims has quit IRC | 11:40 | |
| daya_k | so, the 50-zuul.conf file has this - SetEnv GIT_PROJECT_ROOT /var/lib/zuul/git/ which is the right dir, but i dont know when it gets executed, because in my shell, i dont see this env var. | 11:40 |
| *** ZZelle_ has quit IRC | 11:40 | |
| *** dims has joined #openstack-infra | 11:41 | |
| *** dims has quit IRC | 11:41 | |
| *** dims has joined #openstack-infra | 11:42 | |
| *** flaper87 is now known as flaper87|afk | 11:42 | |
| skolekonov | does sudo apache2ctl -S show that zuul site is enabled? | 11:43 |
| *** dims_ has joined #openstack-infra | 11:44 | |
| *** flaper87|afk is now known as flaper87 | 11:44 | |
| *** weshay has joined #openstack-infra | 11:45 | |
| daya_k | no, getting some errors, let me pastebin them, hang on. | 11:45 |
| *** ppai has joined #openstack-infra | 11:45 | |
| *** dtantsur|brb is now known as dtantsur | 11:45 | |
| daya_k | skolekonov: http://paste.openstack.org/show/100990/ | 11:46 |
| daya_k | is this expected? | 11:47 |
| skolekonov | No. Please try to enable the zuul site with a2ensite <site_name> | 11:47 |
| *** dims has quit IRC | 11:47 | |
| skolekonov | And also it would be good if you setup your Jenkins also using Apache, it's documented on the Jenkins'es web site | 11:48 |
| daya_k | can u please calrify the command for me, i am not too familiar with this | 11:48 |
| daya_k | ok, will do, | 11:48 |
| skolekonov | I think it should be a2ensite 50-zuul | 11:49 |
| daya_k | i tried a2ensite zuul - get 'ERROR: Site zuul does not exist!' | 11:49 |
| daya_k | oh ok, | 11:49 |
| daya_k | hmm, same error | 11:49 |
| skolekonov | try to restart apache with server apache2 restart | 11:50 |
| skolekonov | sorry, service apache2 restart | 11:50 |
| daya_k | y, tried that earlier, but go the same errors as i pasted | 11:50 |
| daya_k | but i also got it for jenkins, and that is working? | 11:51 |
| skolekonov | jenkins behind apache? | 11:51 |
| daya_k | here;s what it gives - http://paste.openstack.org/show/100991/ | 11:51 |
| *** dkranz has joined #openstack-infra | 11:52 | |
| daya_k | so, jenkins site is listed under apache, yes. | 11:52 |
| dtantsur | Hi everyone, quick question please! Is there a way to stop review that was approved by mistake from being merged? Gate job has already started :( | 11:52 |
| *** lukego has quit IRC | 11:52 | |
| dtantsur | I mean, in addition to revoking W+1 | 11:54 |
| daya_k | skolekonov : if it says, the site does not exist, is the path incorrect, or its looking somewhere else also? | 11:55 |
| Kiall | dtantsur: removing the W+1 will prevent it merging.. | 11:56 |
| Kiall | No need to do anything else | 11:56 |
| AJaeger | dtantsur, just send a new patch ;) | 11:56 |
| dtantsur | Kiall, oh thanks | 11:56 |
| AJaeger | any update, e.g. change of commit message, will reset the +1 | 11:56 |
| Kiall | or ^ will cancel the in-progress builds, but they don't do any harm etc | 11:56 |
| skolekonov | daya_k, it's looking somewhere else. Try to disable it first with a2dissite | 11:58 |
| daya_k | skolekonov : i see this entry in 50-hjenkins.conf ' RewriteRule /zuul/status http://127.0.0.1:8001/status [P]' | 11:58 |
| skolekonov | It's for status page | 11:58 |
| daya_k | skolekonov: if i give sitename, its givign not found error again | 11:59 |
| daya_k | i tried wildcard, and it says the file does not exist, so path is wrong - 'ERROR: Site 50-jenkins.conf does not exist!' | 11:59 |
| skolekonov | hm | 12:00 |
| marun_afk | is there anyone here who can help troubleshoot a bug that looks to be related to pbr? | 12:00 |
| *** marun_afk is now known as marun | 12:00 | |
| marun | https://bugs.launchpad.net/grenade/+bug/1361931 | 12:00 |
| uvirtbot` | Launchpad bug 1361931 in python-swiftclient "grenade fails due to ImportError in python-swiftclient" [Undecided,New] | 12:00 |
| marun | It's failing the icehouse grenade jobs consistently | 12:00 |
| marun | mordred: ^^ | 12:00 |
| marun | sdague: ^^ | 12:00 |
| marun | this swiftclient change added a dependency on futures: https://review.openstack.org/#/c/85453/30 | 12:02 |
| daya_k | skolekonov : should i have these sites in sites-available, i only see default and default-ssl | 12:02 |
| *** dims_ has quit IRC | 12:02 | |
| *** mwagner_lap has quit IRC | 12:02 | |
| *** dkranz has quit IRC | 12:03 | |
| russellb | lifeless: updated the bug | 12:03 |
| marun | but looking at the log for a failing job it doesn't appear that pbr is picking up the dependency: http://logs.openstack.org/71/116771/2/check/check-grenade-dsvm-icehouse/429e66a/logs/grenade.sh.txt.gz#_2014-08-27_10_19_16_657 | 12:03 |
| russellb | lifeless: i can't reproduce either, so i closed it | 12:03 |
| *** ppai has quit IRC | 12:03 | |
| sdague | marun: right, so this is the issue where futures is now required for icehouse clouds if you use new swift client | 12:03 |
| sdague | the fix is to propose requirements add in icehouse | 12:04 |
| skolekonov | daya_k, if i'm not mistaken, you shouldn't | 12:05 |
| ttx | sdague: been looking at unit test fails categorization, but so far all pointed to issues in specific patches, which got solved and merged since the fail | 12:05 |
| ttx | so i don't think we have that many uncategorized sticky issues. | 12:06 |
| ttx | (on unit tests side) | 12:06 |
| sdague | ttx cool | 12:06 |
| daya_k | ok, because the man page, for a2ensite seems to indicate otherwise - 'Directory with links to the files in sites-available for enabled sites' | 12:06 |
| *** hashar has quit IRC | 12:06 | |
| marun | sdague: is that already proposed? | 12:06 |
| *** hashar has joined #openstack-infra | 12:07 | |
| ttx | there is a recent surge in nova fail but it doesn't show up in uncategorized page yet | 12:07 |
| sdague | marun: actually, it's that swiftclient now actually requires futures>=2.1.6 | 12:07 |
| sdague | concurrent.futures is not available in 2.1.3, which is what our min version is | 12:08 |
| marun | sdague: arg | 12:08 |
| sdague | and swiftclient apparently has no tests that tickle that | 12:09 |
| *** gokrokve has joined #openstack-infra | 12:09 | |
| marun | sdague: yuck | 12:09 |
| sdague | though it passed grenade when the patch merged | 12:09 |
| marun | sdague: but not grenade icehouse, right? | 12:10 |
| marun | sdague: since that's the failing job | 12:10 |
| sdague | oh, right | 12:10 |
| *** kgiusti has joined #openstack-infra | 12:10 | |
| sdague | because we actually need futures in havana? | 12:10 |
| marun | sdague: I'm getting pretty confused... | 12:10 |
| *** gokrokve has quit IRC | 12:11 | |
| marun | sdague: I think so | 12:11 |
| marun | sdague: this is only to support grenade, right? | 12:11 |
| *** gokrokve has joined #openstack-infra | 12:11 | |
| daya_k | skolekonov: how about if i move these files to sites-available and run a2ensite | 12:11 |
| skolekonov | daya_k, makes sense | 12:12 |
| *** koolhead17 has quit IRC | 12:12 | |
| daya_k | ok, let me try | 12:12 |
| sdague | marun: well it's to support anyone that wants to use new swiftclient on a havana cloud | 12:13 |
| marun | sdague: ah, ok | 12:14 |
| *** flaper87 is now known as flaper87|afk | 12:14 | |
| marun | sdague: I was just trying to assure myself that there would be no implications for packaging | 12:14 |
| sdague | if you release new swiftclient in your havana channel, you'll explode | 12:15 |
| daya_k | skolekonov : tried 'a2ensite /etc/apache2/sites-available/50-jenkins.conf' - get ' ERROR: No site found matching /etc/apache2/sites-available/50-jenkins.conf' | 12:15 |
| marun | sdague: I'm assuming we wouldn't do that... | 12:15 |
| sdague | well, clients don't have stable releases | 12:15 |
| *** p5ntangle has quit IRC | 12:15 | |
| marun | sdague: if someone wanted the new swiftclient to run against havana they'd have to get it from somewhere other than the havana channel | 12:16 |
| *** gokrokve has quit IRC | 12:16 | |
| *** dprince has joined #openstack-infra | 12:16 | |
| sdague | but glance needs swiftclient, for instance | 12:16 |
| sdague | and swiftclient doesn't maintain a stable/havana, so you only get bug fixes by rolling forward | 12:16 |
| isaacb | anyone has experience with zuul-merger? | 12:16 |
| *** dkranz has joined #openstack-infra | 12:16 | |
| marun | sdague: (or by backporting them) | 12:16 |
| *** lukego has joined #openstack-infra | 12:17 | |
| marun | sdague: in any case, is the fix to add the required version of futures to the requirements repo's stable/havana branch? | 12:17 |
| sdague | yeh | 12:17 |
| *** jgallard has joined #openstack-infra | 12:17 | |
| sdague | and also futures should be bumped to 2.1.6 across the board, because swiftclient is now using that | 12:18 |
| marun | sdague: is that already being done? Or could I make sure it happens? | 12:18 |
| *** flaper87|afk is now known as flaper87 | 12:18 | |
| *** mpaolino has joined #openstack-infra | 12:18 | |
| marun | sdague: right | 12:18 |
| sdague | marun: don't know, on first cup of coffee. If you want to propose I can +2 it | 12:18 |
| marun | sdague: ok :) | 12:18 |
| skolekonov | daya_k, i think this command requires only name | 12:18 |
| skolekonov | not full path | 12:18 |
| daya_k | ok | 12:19 |
| isaacb | zuul-merger service fails. anyone faced this problem? | 12:20 |
| daya_k | skolekonov: some success, a2ensite worked, but get some errors in reloading apache, let me paste | 12:20 |
| *** aysyd has joined #openstack-infra | 12:20 | |
| daya_k | skolekonov : http://paste.openstack.org/show/100999/ | 12:21 |
| *** chandankumar has quit IRC | 12:21 | |
| daya_k | still getting the git repo error. | 12:22 |
| skolekonov | I think you should resolve the problems with name resolution, it seems to prevent configuration from being applied. http://httpd.apache.org/docs/2.2/vhosts/name-based.html | 12:25 |
| daya_k | ok, i'll take a look, thanks for your help. | 12:26 |
| jlibosva | sdague: hi, I sent a patch for "futures" global requirements - https://review.openstack.org/#/c/117216/ | 12:30 |
| *** Hal_ has joined #openstack-infra | 12:32 | |
| sdague | jlibosva: ok, so that needs proposed backports to stable/icehouse and stable/havana | 12:39 |
| *** todd_dsm has joined #openstack-infra | 12:39 | |
| sdague | marun: also... isn't havana eol? | 12:39 |
| jlibosva | sdague: done - https://review.openstack.org/#/q/I3161f5bd552b600d2a4483630bed6697e87433e9,n,z | 12:39 |
| sdague | or soon to be eol | 12:39 |
| *** juzuluag has joined #openstack-infra | 12:39 | |
| marun | sdague: from an upstream perspective, sure :) | 12:39 |
| sdague | marun: right, but running grenade in upstream is upstream | 12:40 |
| marun | sdague: I think we have to support for much longer. Although, with Nova and Neutron only being able to coordinate around vm boot since icehouse, we're seeing a need to upgrade people when they run into trouble around that. | 12:40 |
| marun | sdague: ah, right | 12:40 |
| marun | sdague: So can we just kill the job? | 12:40 |
| *** rcarrillocruz has quit IRC | 12:41 | |
| sdague | ttx: what's the clock on havana? | 12:41 |
| *** mmedvede has joined #openstack-infra | 12:44 | |
| *** rcarrillocruz has joined #openstack-infra | 12:44 | |
| *** yamahata has joined #openstack-infra | 12:44 | |
| *** trinaths has quit IRC | 12:45 | |
| *** adalbas has joined #openstack-infra | 12:46 | |
| *** sballe has joined #openstack-infra | 12:47 | |
| HeOS | fungi, clarkb few changes in jeepyb: https://review.openstack.org/#/c/116936/, https://review.openstack.org/#/c/117199/ and gerritlib: https://review.openstack.org/#/c/117225/ | 12:48 |
| sdague | jlibosva: anyway, I fast approved the havana change, and +2ed the others | 12:48 |
| sdague | so hopefully havana will be back on track | 12:48 |
| sdague | in a couple of hours | 12:49 |
| * fungi catches up on the massive scrollback which accumulateth whilst he did slumber | 12:49 | |
| *** p5ntangle has joined #openstack-infra | 12:49 | |
| jlibosva | sdague: cool, thanks | 12:50 |
| marun | sdague: is this change ready to merge then? https://review.openstack.org/#/c/116892/1 | 12:50 |
| *** amuller has quit IRC | 12:50 | |
| marun | sdague: It might be a tiny improvement in gate performance but it would be improvement all the same. | 12:50 |
| rcarrillocruz | hey guys | 12:50 |
| rcarrillocruz | if i wanted to inject two different properties files | 12:51 |
| rcarrillocruz | how to do that? do I put a comma and put the second file? | 12:51 |
| rcarrillocruz | something like: | 12:51 |
| rcarrillocruz | builders: | 12:51 |
| rcarrillocruz | - inject: | 12:51 |
| rcarrillocruz | properties-file: example.prop, foo.prop | 12:51 |
| rcarrillocruz | ? | 12:51 |
| rcarrillocruz | or would I need to put those files between brackets? | 12:51 |
| rcarrillocruz | properties-file: [example.prop, foo.prop] | 12:52 |
| rcarrillocruz | > | 12:52 |
| rcarrillocruz | ? | 12:52 |
| sdague | marun: well it still has a jenkins -1, which might be structural in that approach | 12:52 |
| sdague | marun: have you figured out what that issue is? | 12:52 |
| *** p5ntangle has quit IRC | 12:52 | |
| marun | sdague: It can't be my change, but I'll take a look. | 12:52 |
| sdague | it definitely can be your change | 12:53 |
| *** p5ntangle has joined #openstack-infra | 12:53 | |
| ttx | sdague: "OpenStack Havana stable release will reach EOL at Juno (N+2) milestone-3 and stable releases starting from Icehouse will reach EOL 15 months after their release (approximately N+3 milestone-2)." | 12:53 |
| marun | sdague: it looks like a temporal tempest issue | 12:53 |
| marun | sdague: temporal -> random | 12:53 |
| ttx | sdague: "2013.2.4 - last planned, mid Sep 2014" | 12:53 |
| marun | sdague: the change adds a configuration option that nobody uses (yet), and everything but the grenade job passes. | 12:53 |
| marun | sdague: I'll recheck | 12:54 |
| sdague | ttx: ok, we end up putting so much energy into stable branches the last month of their existance | 12:54 |
| sdague | would *really* like some actual stable main people debugging these issues instead of it randomly falling back to whoever is alive in infra | 12:55 |
| HeOS | fungi, clarkb I fixed a problem when jeepyb system groups created in the database. | 12:55 |
| marun | sdague: does stable/main have a channel? ;) | 12:55 |
| sdague | marun: :P | 12:56 |
| marcusvrn | Hi guys! I'm finishing my CI system and now I'm configuring the logs server...can anyone help me? To setup the server on Jenkins Master, I just need to put my 'site' into site label in macros.yaml? Do you guys have some suggestion about a free server (I saw that someone uses dropbox)? | 12:56 |
| ttx | marun: https://bugs.launchpad.net/openstack-ci/+bug/1360324 | 12:56 |
| uvirtbot` | Launchpad bug 1360324 in openstack-ci "Move #openstack-packaging IRC channel to #openstack-stable" [Medium,In progress] | 12:56 |
| ttx | https://review.openstack.org/#/c/116527/ | 12:56 |
| ttx | :) | 12:56 |
| marun | ttx: huh, shows how much I know | 12:56 |
| *** jgallard has quit IRC | 12:57 | |
| ttx | sdague: One issue is that stable-maint is not really devs, it's more packagers with an interest in common backporting | 12:57 |
| *** julim has joined #openstack-infra | 12:57 | |
| *** jgallard has joined #openstack-infra | 12:57 | |
| ttx | sdague: so the intersection with real devs taht would dive into those issues is limited | 12:57 |
| ttx | I'm almost ready to propose we give stable maint duties to projects | 12:57 |
| ttx | since some of them have been asking for it | 12:58 |
| ttx | and stable-maint would just to release management | 12:58 |
| ttx | and rules enforcement | 12:58 |
| ttx | because the current setup is a bit broken imho | 12:58 |
| sdague | well, I'm not sure that one will make it better | 12:59 |
| marun | I think that sounds reasonable enough. | 12:59 |
| ttx | sdague: not sure iether. It might just make it someone else's problem* | 12:59 |
| marun | Any RH person who is a stable maint focuses on one or more projects anyway | 12:59 |
| sdague | because honestly, what my proposed fix to this would be is stop testing havana | 12:59 |
| *** baohua has quit IRC | 12:59 | |
| sdague | because I don't want to be on the hook for havana fails, and clearly no one else does either :) | 12:59 |
| marun | sdague: +1 | 13:00 |
| ttx | sdague: I wish the stable/* stuff would not interact that much with master stuff, so that when it's broken, it's still stable problem | 13:00 |
| sdague | ttx: ok, so make the clients part of the integrated release | 13:00 |
| ttx | but their failure just show up in the middle of others failures, and in the middle of others pipelines | 13:00 |
| sdague | that's really the big problem towards the end of the stable cycle | 13:01 |
| ttx | so it becomes your problem rather than a self-sustaining group | 13:01 |
| sdague | also, you can't have stable not interact at all with master, otherwise, no upgrade | 13:01 |
| ttx | sdague: the clients used to be, but mordred convinced everyone with his "one client to bind them all" single-channel release approach | 13:02 |
| sdague | yeh, so I think we've got enough evidence now that that doesn't really work | 13:02 |
| sdague | because the clients are used in 2 ways | 13:02 |
| sdague | from the user | 13:02 |
| sdague | and between services | 13:02 |
| marun | sdague: maybe mordred wants to be on call to fix the problems that result? ;) | 13:02 |
| ttx | re: upgrade testing, we could freeze what we upgrade from. | 13:03 |
| *** marun has quit IRC | 13:03 | |
| *** marun has joined #openstack-infra | 13:03 | |
| *** gokrokve has joined #openstack-infra | 13:03 | |
| ttx | there is no way to predict what users would upgrade from anyway | 13:03 |
| sdague | sure... by introducing stable branches to clients .... :) | 13:03 |
| ttx | so "release" is as good as a starting point as anything | 13:03 |
| *** eharney has joined #openstack-infra | 13:03 | |
| *** nkinder has quit IRC | 13:03 | |
| ttx | sdague: would that stable branch approach also apply to oslo libraries? | 13:04 |
| ttx | they are a bit in between currently | 13:04 |
| sdague | maybe | 13:05 |
| sdague | I haven't thought through those implications | 13:05 |
| fungi | flaper87: 113566 should be fine... it self-tested so as long as you analyze the logs from the test runs for any signs that it didn't do what you want (ran too few tests, didn't collect debugging details, et cetera) then it ought to be safe to approve | 13:05 |
| *** jcoufal has quit IRC | 13:05 | |
| sdague | I have all kinds of other issues with the way oslo is currently packaging that causes massive firedrills at milestone 2 every cycle | 13:06 |
| flaper87 | fungi: thanks for checking, would you mind commenting on that bug? Otherwise, I'll just copy paste your message. | 13:06 |
| *** linkid has left #openstack-infra | 13:06 | |
| flaper87 | fungi: s/bug/patch/ | 13:06 |
| *** jcoufal has joined #openstack-infra | 13:08 | |
| sdague | a friend recently made a comment about 'optimizing to reduce fear', so basically thinking about changes that we could make which mean the chances of any future action having catostrophic impacts going down | 13:08 |
| sdague | and I think right now, the unbounded python clients actually have giant fear implications, as seen by this morning's explode around it (which happens quite a bit) | 13:09 |
| *** jgallard has quit IRC | 13:09 | |
| *** jgallard has joined #openstack-infra | 13:09 | |
| *** annegentle has joined #openstack-infra | 13:12 | |
| *** mwagner_lap has joined #openstack-infra | 13:15 | |
| *** yaguang has joined #openstack-infra | 13:19 | |
| mestery | sdague: I heard you spoke with cdub about running tempest tests with ODL in the OpenStack gate, and that you were amendable to this. | 13:20 |
| *** mpaolino has quit IRC | 13:20 | |
| fungi | pcm_: that's fixed now. bad subunit release, which was soon after superseded by a fixed one | 13:20 |
| *** e0ne is now known as e0ne_ | 13:21 | |
| pcm_ | fungi: Yeah, salv-orlando showed me how to look at logstash and we saw that there was a 45 min period with that failing. Someone already did a recheck for me on my commit and it now passed. | 13:22 |
| sdague | mestery: yeh, so that whole post merge + experimental model that jeblair posted out there | 13:22 |
| *** bknudson has quit IRC | 13:22 | |
| sdague | that we worked out in germany | 13:22 |
| * mestery tries to dig for that mail ... | 13:22 | |
| mestery | sdague: Ah, ok, it was in germany | 13:22 |
| sdague | I think any opensource components fit into that model | 13:22 |
| *** ominakov has joined #openstack-infra | 13:22 | |
| mestery | sdague: The reason I ask is that the ODL side is struggling to move things to RAX in a timely manner, though they may have had a break through last night. | 13:23 |
| sdague | docker, ceph, opendaylight | 13:23 |
| mestery | Cool | 13:23 |
| sdague | to be clear though, it won't be pre-merge gating | 13:23 |
| mestery | OK, I think that's the part that cdub was confused on, but it makes sense. | 13:23 |
| sdague | anyway, jeblair and I worked out the model sanely | 13:23 |
| mestery | Nice | 13:24 |
| fungi | sdague: according to https://wiki.openstack.org/wiki/StableBranchRelease#Planned_stable.2Fhavana_releases "2013.2.4 - last planned, mid Sep 2014" so we'll probably eol the branch shortly before that | 13:24 |
| sdague | and he's got a pretty detailed email about it | 13:24 |
| *** mfer has joined #openstack-infra | 13:24 | |
| mestery | Thanks for the info sdague! | 13:24 |
| *** mriedem has joined #openstack-infra | 13:25 | |
| *** p5ntangle has quit IRC | 13:25 | |
| *** balles has quit IRC | 13:25 | |
| fungi | HeOS: for gerrit 2.9? | 13:25 |
| HeOS | fungi, yes. | 13:26 |
| *** zhiyuan has quit IRC | 13:26 | |
| HeOS | fungi, i think, that it will work with 2.10 version. | 13:27 |
| *** zhiyuan has joined #openstack-infra | 13:27 | |
| *** _nadya_ has joined #openstack-infra | 13:28 | |
| fungi | ttx: sdague: ultimately we need the servers to stop depending on the clients as libraries, or else start integration testing all client changes against all supported stable branches and upgrades | 13:28 |
| sdague | fungi: or cut stable releases on them | 13:28 |
| *** yaguang has quit IRC | 13:28 | |
| fungi | though someone has probably already said that, i've apparently still got several screens of scrollback to go | 13:29 |
| sdague | under the assumption that the version of novaclient on a production cloud isn't going to be randomly upgraded | 13:29 |
| *** yaguang has joined #openstack-infra | 13:29 | |
| *** salv-orlando has quit IRC | 13:31 | |
| fungi | sdague: ttx: so with client stable branches, i would need different versions of, say, novaclient imported into my cloud-consuming service simultaneously if i want it to operate cross-provider between one who runs havana and one who runs icehouse? that doesn't seem likely to be a good answer either | 13:32 |
| sdague | fungi: well the current issue is we just made havana clouds have to add dependencies | 13:33 |
| sdague | like major dependencies | 13:33 |
| sdague | fungi: because you assume the clients are a super light wrapper around the openstack API | 13:34 |
| sdague | which they very much are not | 13:34 |
| sdague | the part that broke was not the API | 13:34 |
| *** mpaolino has joined #openstack-infra | 13:34 | |
| *** _nadya_ has quit IRC | 13:34 | |
| sdague | it was all the layers above requiring new dependencies as it refactored | 13:34 |
| fungi | sdague: yeah, i think the answer is to stop having the servers load the clients as libraries. we can't have our cake and eat it too here | 13:34 |
| fungi | servers need stable libraries. users need clients | 13:34 |
| *** mpaolino_ has joined #openstack-infra | 13:35 | |
| sdague | yeh, well now we are boiling the ocean again :) | 13:35 |
| fungi | also the libraries that application developers use to interact with a cloud api should not be the libraries that the cloud servers load to talk to each other. different use cases | 13:35 |
| *** dane_leblanc has joined #openstack-infra | 13:35 | |
| fungi | or we can give up on stable releases | 13:35 |
| sdague | or we cap client requirements on stable release | 13:36 |
| sdague | which would also work | 13:36 |
| fungi | as long as we either install two versions of the clients when used in integration testing (in separate venvs) or stop using them as clients when running integration tests (which seems problematic for other reasons) | 13:37 |
| sdague | so we're removing them from integration testing now | 13:37 |
| sdague | at least from tempest side | 13:37 |
| fungi | i assume this means we're also stopping testing whether the clients work with an actual (well, devstack) cloud? | 13:38 |
| *** pdmars has joined #openstack-infra | 13:38 | |
| sdague | tempest won't | 13:38 |
| mtreinish | sdague: well only partially, the cli tests will still be there | 13:38 |
| sdague | that's where functional testing of clients should come into play | 13:38 |
| mtreinish | at least for a bit | 13:38 |
| fungi | yeah, so the clients need some sort of functional tests at least | 13:38 |
| sdague | mtreinish: only until we can jestison it | 13:38 |
| EmilienM | hey infra, puppet gate is (still) blocked. We have a fix and would need a quick review if possible: https://review.openstack.org/#/c/116915 | 13:38 |
| *** mpaolino has quit IRC | 13:38 | |
| mtreinish | sdague: I plan to push the first lib version at the end of this week. So that will hopefully begin soon. | 13:39 |
| sdague | mtreinish: cool | 13:39 |
| fungi | anyhoo, if servers use clients as dependencies for communicating with one another, then capping them also makes for a very bad security fix story | 13:40 |
| *** laxity has quit IRC | 13:41 | |
| *** bknudson has joined #openstack-infra | 13:41 | |
| fungi | because now we tell havana cloud operators that to not be vulnerable to some particular security hole they need to update swiftclient to a version which havana glance doesn't support, and thus a combination we don't test...? | 13:41 |
| *** mpaolino_ is now known as mpaolino | 13:41 | |
| fungi | or we tell the distros and other downstream consumers to backport and test that themselves and run a hotfixed-not-released version of swiftclient, which in the end just recreates the argument for stable branches of the libs | 13:42 |
| *** e0ne_ has quit IRC | 13:43 | |
| * fungi needs to learn to just declare bankruptcy on scrollback each morning. forgot to have any coffee yet | 13:43 | |
| ttx | I declare bankrupcy on scrollback every 15 min | 13:44 |
| *** signed8bit has joined #openstack-infra | 13:48 | |
| *** paulrad has joined #openstack-infra | 13:50 | |
| *** e0ne has joined #openstack-infra | 13:50 | |
| *** chandankumar has joined #openstack-infra | 13:51 | |
| *** rcarrillocruz has quit IRC | 13:51 | |
| *** rcarrillocruz has joined #openstack-infra | 13:52 | |
| *** pcrews has joined #openstack-infra | 13:52 | |
| *** amotoki has quit IRC | 13:52 | |
| *** nkinder has joined #openstack-infra | 13:53 | |
| *** laxity has joined #openstack-infra | 13:54 | |
| *** homeless has joined #openstack-infra | 13:55 | |
| *** nelsnelson has quit IRC | 14:02 | |
| *** nelsnelson has joined #openstack-infra | 14:03 | |
| daya_k | skolekonov: still around? | 14:04 |
| daya_k | its working now, at least the git fetch part, thought i would update you | 14:04 |
| daya_k | skolekonov : i updated 50-jenkins.conf and 50-zuul.conf a bit, to make the virtualhost work, replaced <VirtualHost zuul:80> with <VirtualHost *:80> | 14:07 |
| *** annegentle has quit IRC | 14:08 | |
| daya_k | thanks for all your help, i will post details on the mailing list, just in case the puppet install scripts need to be updated for this change. | 14:08 |
| *** ildikov_ has quit IRC | 14:09 | |
| *** jbjohnso__ has joined #openstack-infra | 14:10 | |
| Hal_ | fungi: Hi, I am trying to add dvipng software to the gate used for doc generation, devstack I presume. StevenK redirected me to you or jeblair for this matter. I have cloned openstack-infra/devstack repo, but I do not know where to add this dependency. Have you the time to help me ? | 14:13 |
| *** annegentle has joined #openstack-infra | 14:13 | |
| *** baohua has joined #openstack-infra | 14:16 | |
| *** rcarrillocruz has quit IRC | 14:16 | |
| phschwartz | Morning infra | 14:16 |
| *** markmcclain has joined #openstack-infra | 14:17 | |
| fungi | Hal_: yeah, saw the discussion in scrollback. the project docs jobs run from tox (like unit tests) on non-devstack workers. for system (not python) dependencies, we set the distro-specific package names in https://git.openstack.org/cgit/openstack-infra/config/tree/modules/openstack_project/manifests/jenkins_params.pp and then install them in | 14:19 |
| fungi | https://git.openstack.org/cgit/openstack-infra/config/tree/modules/openstack_project/manifests/thick_slave.pp | 14:19 |
| fungi | phschwartz: morning! looks like ops didn't find time to remove those stuck error+deleting instances in dfw yet | 14:19 |
| marun | sdague, fungi, clarkb: ready for merge - https://review.openstack.org/#/c/116892/ | 14:20 |
| *** rcarrillocruz has joined #openstack-infra | 14:20 | |
| *** otherwiseguy has joined #openstack-infra | 14:21 | |
| *** zz_jgrimm is now known as jgrimm | 14:22 | |
| *** p5ntangle has joined #openstack-infra | 14:22 | |
| *** yfried has quit IRC | 14:23 | |
| *** stevemar has joined #openstack-infra | 14:24 | |
| *** p5ntangle has quit IRC | 14:24 | |
| *** KanagarajM has joined #openstack-infra | 14:24 | |
| *** chandankumar has quit IRC | 14:24 | |
| *** david-lyle has joined #openstack-infra | 14:26 | |
| phschwartz | fungi: They promised me they would get to them. let me harash them more | 14:27 |
| *** trinaths has joined #openstack-infra | 14:30 | |
| *** NobodyCa1 is now known as NobodyCam | 14:30 | |
| marun | fungi: So, I have broken the neutron functional job. | 14:30 |
| marun | fungi: given how bad a time this is to happen, do you think making it non-voting until it's fixed would be ok? | 14:31 |
| *** dangers_away is now known as dangers | 14:31 | |
| *** mfer has quit IRC | 14:32 | |
| fungi | marun: how did you break it? | 14:32 |
| fungi | (and did you keep all the pieces?) | 14:32 |
| marun | fungi: I switched to using in-tree hook scripts. | 14:32 |
| fungi | do you have a link to a good example failure? | 14:32 |
| marun | fungi: In retrospect I tried to do too much at once. It's a permissions prolem. | 14:32 |
| *** yaguang has quit IRC | 14:32 | |
| marun | fungi: yes | 14:32 |
| marun | fungi: https://jenkins05.openstack.org/job/check-neutron-dsvm-functional/1762/console | 14:33 |
| fungi | trying to identify if it's something which needs fixing in the job definition or in neutron | 14:33 |
| *** trinaths_afk has joined #openstack-infra | 14:33 | |
| marun | fungi: 2014-08-27 14:26:52.762 | 14:33 |
| marun | fungi: it's in neutron | 14:33 |
| *** yaguang has joined #openstack-infra | 14:33 | |
| marun | I'm just worried at the time it will take to merge a fix given the gate backup | 14:33 |
| *** mfer has joined #openstack-infra | 14:33 | |
| fungi | but i agree that setting it non-voting while you work through major changes to the job is reasonable as long as it's kept brief enough so as to avoid slipping in other changes which break the test in new and exciting ways | 14:33 |
| marun | fungi: if making the job non-voting would reduce the impact, it might be worth doing while the fix to neutron waits to merge | 14:34 |
| marun | fungi: it will be brief | 14:34 |
| marun | fungi: and at least with the hook scripts I can validate the changes in neutron only | 14:34 |
| marun | fungi: I'll submit a patch. | 14:34 |
| fungi | okay, i'll be ready to usher it through rapidly | 14:35 |
| *** dustins has joined #openstack-infra | 14:35 | |
| jeblair | fungi, ttx, sdague: i agree with fungi -- asking users to use different clients for different clouds is a non-starter and will be the death of openstack (if nothing else beats it to it). the end goal should be that we have simple client(s) that users use to talk to any openstack cloud, and we cross-test that in the gate. how openstack handles intra-service communication within a release, i don't care. | 14:35 |
| *** trinaths has quit IRC | 14:36 | |
| *** trinaths_afk is now known as trinaths | 14:36 | |
| jeblair | fungi, ttx, sdague: maybe we need a new client project for that. | 14:36 |
| *** baohua has quit IRC | 14:37 | |
| Hal_ | fungi: thanks, I am on it ;-) | 14:38 |
| *** koolhead17 has joined #openstack-infra | 14:38 | |
| *** sdake_ has joined #openstack-infra | 14:38 | |
| *** medieval1_ has quit IRC | 14:39 | |
| *** tonytan4ever has joined #openstack-infra | 14:40 | |
| *** hdd has joined #openstack-infra | 14:40 | |
| marun | fungi: https://review.openstack.org/#/c/117261/ | 14:40 |
| fungi | marun: taken as a learning experience, probably this should have been set up in parallel as a new in-tree functional job set non-voting initially while the kinks were worked out, and then swapped in for the previous voting job once stabilized? | 14:41 |
| fungi | just curious whether there were concerns about following that pattern at the time | 14:41 |
| *** chandankumar has joined #openstack-infra | 14:42 | |
| marun | fungi: yes. or at a minimum not tried to add to what was already in the job when externalizing to hook scripts. | 14:42 |
| *** amuller has joined #openstack-infra | 14:42 | |
| marun | fungi: I presumed that sudo would still be allowed in the gate_hook, but that clearly is not the case. | 14:43 |
| fungi | jeblair: ttx: sdague: dtroyer: if only there were a unified openstack client users and application developers could run (and then the old per-project clients get relegated to inter-process communication and switched to a different scope and possibly release style) | 14:43 |
| *** afazekas has quit IRC | 14:43 | |
| fungi | marun: right, we normally revoke it in devstack-gate after setting up the environment but before running tests (mainly so that we could ensure tempest tests didn't try to sudo anything) | 14:44 |
| marun | fungi: :/ | 14:44 |
| fungi | or, maybe not so much revoke as run the tests under a different account than the setup | 14:44 |
| marun | fungi: It's not the tests that need special privileges as the test setup. | 14:44 |
| marun | fungi: https://github.com/openstack/neutron/blob/master/neutron/tests/functional/contrib/gate_hook.sh | 14:44 |
| marun | fungi: the unstack and creation of a test-only rootwrap filter | 14:45 |
| fungi | marun: i wonder if that could be split into two hooks in that case | 14:45 |
| fungi | can't remember whether the pre-test hook has sudo access | 14:45 |
| marun | fungi: so add another hook that is run pre-sudo removal? | 14:45 |
| dtroyer | fungi: wouldn't that be great? Hopefully someone has the time to spare to do that someday | 14:46 |
| fungi | dtroyer: i know. seriously | 14:46 |
| *** unicell1 has quit IRC | 14:46 | |
| *** daya_k has quit IRC | 14:46 | |
| sdague | well, we could just flip all the projects back to noop testing for merge, and go off and solve the issue | 14:46 |
| *** che-arne has quit IRC | 14:47 | |
| *** che-arne|2 has joined #openstack-infra | 14:47 | |
| marun | dtroyer, fungi: I'm happy to do that work if you are not opposed. | 14:48 |
| anteaya | if someone could re-enable 12780prophetstor-ciProphetStor cinerama prophetstor.ci@prophetstor.com I'd be grateful | 14:48 |
| marun | I've been nibbling at this for months and it would be nice to finally have it working. | 14:48 |
| Hal_ | fungi: something like that : https://review.openstack.org/117263 ? | 14:48 |
| jeblair | clarkb, fungi: after 2 days of work, i've finally completed my 4 line patch to the nodepool scripts: https://review.openstack.org/117262 | 14:49 |
| fungi | anteaya: done | 14:49 |
| marun | it's pretty embarrassing how slow the progress has been. | 14:49 |
| fungi | dtroyer: marun: honestly, for my (albeit minimal) personal use cases, openstackclient has worked fine. and when it hasn't, it's been because of service providers making compromising choices in their api | 14:50 |
| *** HenryG has joined #openstack-infra | 14:51 | |
| sdague | fungi: openstackclient is still importing the other client libs right? | 14:52 |
| dtroyer | sdague: mostly, yes | 14:52 |
| sdague | so that doesn't actually solve this | 14:52 |
| dtroyer | but finishing the independent sdk does | 14:52 |
| fungi | yeah, at the moment it's still partly a shim | 14:52 |
| dtroyer | swiw, OSC doesn't use swiftclient, and we could dump^H^H^H^Hreplace at least one other pretty easily… | 14:53 |
| fungi | Hal_: having not double-checked the package names myself, that looks like all you need yes (and then we have to wait a day after it merges for it to appear in the next nodepool image update cycle) | 14:53 |
| *** _nadya_ has joined #openstack-infra | 14:53 | |
| *** hemna_ is now known as hemna | 14:53 | |
| *** todd_dsm has quit IRC | 14:54 | |
| Hal_ | fungi: Thanks for your help and feedback. I will now wait for reviews. | 14:54 |
| *** openstackgerrit has joined #openstack-infra | 14:54 | |
| jeblair | marun: for the future, the best way to work out the kinks in a new job like this is to run it locally. devstack-gate has instructions on how to do that | 14:54 |
| *** xianghuihui has quit IRC | 14:55 | |
| *** medieval1 has joined #openstack-infra | 14:55 | |
| marun | jeblair: my goal was to sidestep that by being able to have the jobs run pre-merge against any changes | 14:55 |
| *** flaper87 is now known as flaper87|afk | 14:56 | |
| *** flaper87|afk is now known as flaper87 | 14:56 | |
| marun | jeblair: where I fell down was adding something to the hooks scripts that wasn't in the job. | 14:56 |
| *** Longgeek has quit IRC | 14:56 | |
| marun | jeblair: I should have waited until the job was passing :( | 14:56 |
| marun | jeblair: in any case, it looks like I need to add a new hook that runs after devstack completes and before sudo privileges are removed | 14:58 |
| jeblair | marun: that sounds great | 14:58 |
| marun | jeblair: Ok, I'll post a patch shortly | 14:58 |
| *** isaacb has quit IRC | 14:59 | |
| *** gargola has joined #openstack-infra | 14:59 | |
| jeblair | marun: does pre_test_hook not work? | 14:59 |
| marun | jeblair: it runs before the gate hook, so devstack hasn't run yet | 15:00 |
| *** dizquierdo has quit IRC | 15:00 | |
| marun | jeblair: I can't unstack, and I would have to manually create the target rootwrap dir rather than simply adding a file | 15:00 |
| jeblair | marun: what if you set your gate hook to run the gate script, then your setup script, then your tests? | 15:01 |
| marun | jeblair: That's what I tried to do. But sudo privileges are removed in the gate script and then the setup script fails. | 15:01 |
| marun | jeblair: https://github.com/openstack-infra/devstack-gate/blob/master/devstack-vm-gate.sh#L347 | 15:02 |
| marun | jeblair: https://github.com/openstack/neutron/blob/master/neutron/tests/functional/contrib/gate_hook.sh | 15:02 |
| jeblair | marun: sudo privs are removed for devstack, but not for jenkins | 15:02 |
| AJaeger | jeblair, we missed many docutils pins and I've send patches to all *-specs products - could you give a comment on https://review.openstack.org/117180, please? | 15:02 |
| jeblair | marun: huh, that didn't work? :) | 15:03 |
| marun | jeblair: hmmm... so I should be able to 'sudo -u jenkins'? | 15:03 |
| jeblair | marun: oh i see | 15:03 |
| jeblair | marun: i think you just need to sudo with that sed line | 15:03 |
| openstackgerrit | Matthew Treinish proposed a change to openstack-infra/config: Add subunit2sql gearman workers https://review.openstack.org/108003 | 15:03 |
| jeblair | marun: that script should be running as jenkins | 15:03 |
| jeblair | marun: which has passwordless sudo | 15:03 |
| jeblair | marun: it's just thet /etc/neutron/rootwrap.d/functional.filters is root-owned | 15:04 |
| marun | jeblair: so prefix lines 10 and 16 with 'sudo'? | 15:04 |
| *** mikal has quit IRC | 15:05 | |
| *** loquacity has quit IRC | 15:05 | |
| jeblair | marun: definitely 16; never run unstack myself, but possibly? not really sure how it's supposed to be run | 15:05 |
| marun | jeblair: it fails at present due to permission problems, so I'm guessing sudo is required. | 15:05 |
| sdake_ | unstack is run as the devstack uid - it calls sudo internally iirc | 15:05 |
| jeblair | marun: i can say that prefixing it with sudo will cause it to run as root :) | 15:05 |
| sdake_ | jeblair ;-) | 15:05 |
| anteaya | fungi: thanks, just finishing up scrollback | 15:05 |
| marun | sdake_: you mean that unstack explicitly uses the devstack uid? | 15:06 |
| *** mikal has joined #openstack-infra | 15:06 | |
| *** zz_gondoi is now known as gondoi | 15:06 | |
| jeblair | okay, so having revoked sudo from it could be a problem if it needs to run as stack and also needs sudo. hopefully it's okay to just run it as root | 15:06 |
| AJaeger | fungi, thanks! | 15:06 |
| sdake_ | marun yup pretty sure it does so it can connect to screen - maybe it doesn't sudo | 15:06 |
| jeblair | hrm | 15:07 |
| jeblair | it has an explicit check that it doesn't run as root | 15:07 |
| jeblair | dtroyer: i think we have an awkward situation here | 15:07 |
| sdake_ | best solution is probably to modify unstack.sh imo | 15:08 |
| fungi | should it run as stack? (sudo -u stack unstack.sh)? | 15:08 |
| sdake_ | allow it to sudo internally to root when necessary | 15:08 |
| sdake_ | (not sure when that is) | 15:08 |
| dtroyer | jeblair: catching up…unstack as root? | 15:08 |
| jeblair | okay, so the issue is that we do the following in the gate: | 15:09 |
| jeblair | 1) run devstack, 2) remove sudo access for stack user 3) run tempest | 15:09 |
| fungi | aha. so if unstack is called as the stack user, it needs to happen between 1 and 2 | 15:10 |
| *** skolekonov has quit IRC | 15:11 | |
| *** klindgren_ has quit IRC | 15:11 | |
| *** dprince has quit IRC | 15:11 | |
| *** AJaeger has quit IRC | 15:12 | |
| *** dkehn has quit IRC | 15:16 | |
| *** jesusaurus has quit IRC | 15:16 | |
| sdake_ | unstack attempts to stop systemwide services | 15:17 |
| jeblair | we do that because we want to actually test the rootwrap system | 15:17 |
| sdake_ | mysql and others | 15:17 |
| sdake_ | maybe the mysql is breaking tihngs | 15:17 |
| *** KanagarajM has quit IRC | 15:17 | |
| jeblair | it's pointless to have a rootwrap system if, when you test it, you actually have full sudo access | 15:17 |
| jeblair | what marun wants to do is replace (3) with: run unstack, run some tests | 15:18 |
| jeblair | but unstack assumes it ran with the same privileges that devstack had | 15:18 |
| mtreinish | sdague: is the reason that os-loganalyze isn't working on the keystone logs because of the apache move? | 15:18 |
| mtreinish | sdague: for example: http://logs.openstack.org/25/98725/2/check/check-tempest-dsvm-neutron-full/555ff66/logs/screen-key.txt.gz | 15:18 |
| sdake_ | jeblair would shutting down mysql confuse the node? | 15:18 |
| sdake_ | would/wouldn't? | 15:18 |
| *** KanagarajM has joined #openstack-infra | 15:18 | |
| marun | sdake_: unstack leaves rabbit and mysql alone | 15:18 |
| *** klindgren__ has joined #openstack-infra | 15:18 | |
| dtroyer | jeblair: yes, which in most cases it does. I think it might be ok to run unstack.sh as root, I think that check was as much to remove the whole iddea from random dev minds…although, I'd need to see what it needs root for if UNSTACK_ALL isn't set | 15:18 |
| sdake_ | marun your right, that is only the unstack-all op | 15:18 |
| sdague | mtreinish: that sounds probable | 15:18 |
| sdake_ | eg --all | 15:18 |
| jeblair | i think our options are: (a) if it's physically possible, allow unstack to be run as root (because we can "sudo ./unstack.sh") or (b) modify devstack-gate to not remove stack's sudo access | 15:18 |
| *** che-arne|2 is now known as che-arne | 15:18 | |
| dtroyer | off the top of my head, the iscsi bits in cinder is one... | 15:18 |
| marun | sdake_: ah, ok. I'm not using that, I want mysql and rabbit to persist. | 15:18 |
| *** ildikov has joined #openstack-infra | 15:18 | |
| sdake_ | marun i'd try sudo -u stack unstack.sh | 15:18 |
| sdague | unstack as root should be fine in the non screen case | 15:18 |
| marun | dtroyer, jeblair: I think it is safe because shortly it will only be neutron that will be unstacked | 15:18 |
| sdague | in the screen case... it's going to be goofy I think | 15:18 |
| marun | https://review.openstack.org/#/c/116892/ | 15:18 |
| *** rlandy has quit IRC | 15:18 | |
| *** IvanBerezovskiy has left #openstack-infra | 15:18 | |
| dtroyer | I suspect it would be safe to remove the don't-run-as-root check…safe in the sense of not creating problems down the road from devs… | 15:18 |
| sdague | though.. unstack in the non root case doesn't actually stop nova... | 15:18 |
| sdague | sorry, in the non screen case | 15:18 |
| marun | sdake_: ok, will try | 15:18 |
| dtroyer | sdague: at first thought I don't have anything goofy coming to mind for screen... | 15:18 |
| jeblair | marun, dtroyer, sdague: we could add a devstack-gate option that does unstack before removing stack sudo | 15:18 |
| marun | jeblair: +1 | 15:18 |
| sdague | dtroyer: if you try to run screen commands as not the same tty user things I thought you got tty issues | 15:18 |
| jeblair | so it handles this pattern internally | 15:18 |
| sdague | even if that's root | 15:18 |
| sdague | jeblair: so, there is actually an outstanding bug that will mean you don't quite get expected results | 15:18 |
| sdague | because..... it turns out if you run devstack without screen (which we do in the gate for everything except grenade) the service kills don't actually work | 15:18 |
| sdague | because the pid capture is incorrect | 15:18 |
| sdague | I figured this out on a plane to portland before vacation | 15:18 |
| sdague | dtroyer: did you ever dive into that one further? | 15:18 |
| *** oomichi has joined #openstack-infra | 15:18 | |
| jeblair | marun: do you need service kills, or are you counting on the minimal service start due to that patch to features.yaml? | 15:18 |
| *** hashar_ has joined #openstack-infra | 15:18 | |
| marun | jeblair: I would like to see mysql and rabbit stay running | 15:18 |
| dtroyer | I did, and I'm trying to remember…for most of the situations I was able to get the process leader bits to work, but I think there is a case where that doesn't happen because one of our services daemonizes itself incorectly... | 15:18 |
| marun | jeblair: everything else should be killed | 15:18 |
| fungi | wow. freenode is having some serious server-to-server lag. i see nothing and then conversation going at 10x normal pace | 15:18 |
| *** Daviey has quit IRC | 15:18 | |
| *** dkehn__ has joined #openstack-infra | 15:19 | |
| jeblair | marun: right, but your features.yaml patch suggests that there won't be anything other than those services to kill | 15:19 |
| *** spiffxp- has quit IRC | 15:19 | |
| jeblair | marun: i'm trying to figure out if sdague's bug will affect you | 15:19 |
| *** yaguang has quit IRC | 15:20 | |
| marun | jeblair: I need to enable neutron via DEVSTACK_GATE_NEUTRON to ensure the necessary dependencies are installed | 15:22 |
| marun | jeblair: which bug? | 15:22 |
| morganfainberg | mtreinish, sdague, i'm revisiting the fix to get keystone's logs into shape under apache 2.4 in a few minutes today | 15:22 |
| morganfainberg | mtreinish, sdague, unfortunately got caught up in some FF stuff for keystone | 15:22 |
| yolanda | jeblair, fungi, i did some changes for dib-nodepol config patches, but there are things that i'm not sure about, maybe clarkb or mordred know more about it | 15:22 |
| jeblair | marun: he described it 4 minuse ago ^ | 15:22 |
| yolanda | for example the /tmp/local.pp issue, and the docker changes | 15:22 |
| *** trinaths has quit IRC | 15:22 | |
| *** otherwiseguy has quit IRC | 15:22 | |
| *** nkinder has quit IRC | 15:22 | |
| *** mpaolino has quit IRC | 15:22 | |
| *** kgiusti has quit IRC | 15:22 | |
| *** hashar has quit IRC | 15:22 | |
| *** mattymo has quit IRC | 15:22 | |
| *** ruhe has quit IRC | 15:22 | |
| *** gpocentek has quit IRC | 15:22 | |
| *** dguerri_ has quit IRC | 15:22 | |
| *** vladan has quit IRC | 15:22 | |
| *** proffalk1n has quit IRC | 15:22 | |
| *** dansmith has quit IRC | 15:22 | |
| *** s0nea has quit IRC | 15:22 | |
| *** vhoward- has quit IRC | 15:22 | |
| *** sdake has quit IRC | 15:22 | |
| *** yongli has quit IRC | 15:22 | |
| *** tteggel has quit IRC | 15:22 | |
| *** froek has quit IRC | 15:22 | |
| *** pfallenop has quit IRC | 15:22 | |
| *** gmurphy has quit IRC | 15:22 | |
| *** joa has quit IRC | 15:22 | |
| *** jogo has quit IRC | 15:22 | |
| *** hashar__ has joined #openstack-infra | 15:22 | |
| marun | jeblair: what do you think of overloading the DEVSTACK_GATE_NO_SERVICES flag to indicate the need for an unstack? | 15:22 |
| *** spiffxp_ has joined #openstack-infra | 15:22 | |
| jeblair | yolanda: look at the history for that patch, you'll see where local.pp comes from | 15:22 |
| *** dkehn__ is now known as dkehnx | 15:22 | |
| *** juzuluag has quit IRC | 15:22 | |
| *** Longgeek has joined #openstack-infra | 15:22 | |
| *** trinaths has joined #openstack-infra | 15:22 | |
| *** otherwiseguy has joined #openstack-infra | 15:22 | |
| *** nkinder has joined #openstack-infra | 15:22 | |
| *** mpaolino has joined #openstack-infra | 15:22 | |
| *** kgiusti has joined #openstack-infra | 15:22 | |
| *** mattymo has joined #openstack-infra | 15:22 | |
| *** ruhe has joined #openstack-infra | 15:22 | |
| *** gpocentek has joined #openstack-infra | 15:22 | |
| *** dguerri_ has joined #openstack-infra | 15:22 | |
| *** vladan has joined #openstack-infra | 15:22 | |
| *** dansmith has joined #openstack-infra | 15:22 | |
| *** proffalk1n has joined #openstack-infra | 15:22 | |
| *** s0nea has joined #openstack-infra | 15:22 | |
| *** vhoward- has joined #openstack-infra | 15:22 | |
| *** sdake has joined #openstack-infra | 15:22 | |
| *** yongli has joined #openstack-infra | 15:22 | |
| *** tteggel has joined #openstack-infra | 15:22 | |
| *** froek has joined #openstack-infra | 15:22 | |
| *** pfallenop has joined #openstack-infra | 15:22 | |
| *** gmurphy has joined #openstack-infra | 15:22 | |
| *** joa has joined #openstack-infra | 15:22 | |
| *** jogo has joined #openstack-infra | 15:22 | |
| *** Longgeek has quit IRC | 15:22 | |
| *** jesusaurus has joined #openstack-infra | 15:22 | |
| jeblair | marun: so even with your change https://review.openstack.org/#/c/116892/1 neutron will be running? | 15:23 |
| marun | jeblair: ah, crap | 15:23 |
| yolanda | jeblair, so these changes were removed, looks as this puppet apply /tmp/local.pp just needs to be removed? | 15:23 |
| marun | jeblair: yes. the goal of the patch was to minimize service installation but I still need neutron installed so things like ovs are available | 15:23 |
| jeblair | yolanda: probably? | 15:23 |
| jeblair | marun: how does neutron get in there? | 15:23 |
| openstackgerrit | Romain Chantereau proposed a change to openstack-infra/config: Be able to generate math formulae with Sphinx mathpng base extension. https://review.openstack.org/117263 | 15:23 |
| marun | jeblair: (until I get around to allowing devstack to install dependencies and not start services) | 15:23 |
| *** hashar_ has quit IRC | 15:23 | |
| marun | jeblair: DEVSTACK_GATE_NEUTRON=1 | 15:23 |
| *** jergerber has joined #openstack-infra | 15:24 | |
| marun | jeblair: the reason for the patch in question is that previously the set of service definitions was additive only from the default | 15:24 |
| *** markmcclain has quit IRC | 15:24 | |
| *** jergerber has quit IRC | 15:24 | |
| *** jergerber has joined #openstack-infra | 15:24 | |
| jeblair | marun: ah, so this part at least is a bit of a special case for neutron :) | 15:25 |
| marun | jeblair: the workaround was to allow a minimal set of services to be selected via that flag instead of the default set | 15:25 |
| jeblair | marun: until you make that change to devstack | 15:25 |
| marun | jeblair: correct | 15:25 |
| dtroyer | marun: so I am just now getting an idea of what you are after…basically do the install_prereqs.sh, git clone and setup.py on the repos, and that's it? maybe build config files? | 15:26 |
| *** electrichead is now known as redrobot | 15:26 | |
| anteaya | so as a heads up, there is a conversation that took place in -neutron that will be brought up at the next third party meeting (probably) floating the idea that third party ci systems (at least in Neutron) only vote prior to release | 15:26 |
| jeblair | marun: okay. so yeah, i think the way forward is to add an option to unstack in https://github.com/openstack-infra/devstack-gate/blob/master/devstack-vm-gate.sh#L346 | 15:26 |
| marun | dtroyer: pretty much, yes. | 15:26 |
| anteaya | for those who have thoughts on this idea and would like to attend | 15:26 |
| *** tonytan4ever has quit IRC | 15:26 | |
| fungi | anteaya: only vote prior to release meaning not on stable branches? | 15:26 |
| jeblair | marun: if you want to combine it with the no_services option, i'm okay with that, but i'd prefer a different name so the unstack may come as less of a surprise :) | 15:26 |
| anteaya | fungi: unknown, but a good question | 15:27 |
| marun | jeblair: No problem, I can add a new one | 15:27 |
| fungi | anteaya: just trying to understand what is deemed "before release" since technically every patch is proposed before some release | 15:27 |
| sdague | dtroyer: I *know* none of the nova services stop if you run without screen | 15:27 |
| fungi | anteaya: on master it's before an integrated release, on stable/.* it's before a stable point release (until eol) | 15:28 |
| anteaya | I think the proposal as an example would be plugins/drivers for juno would vote prior to j3 | 15:28 |
| jeblair | marun: and then that patch you're talking about to devstack would be great | 15:28 |
| dtroyer | sdague: currently, yes. I got past most of that, but not all | 15:28 |
| *** atiwari has joined #openstack-infra | 15:28 | |
| *** wenlock has joined #openstack-infra | 15:28 | |
| jeblair | anteaya: some people run openstack in continuous deployment | 15:28 |
| jeblair | anteaya: every commit is a release for them | 15:28 |
| sdague | dtroyer: http://logs.openstack.org/29/100229/1/check/check-grenade-dsvm/0362268/logs/grenade.sh.txt.gz | 15:28 |
| openstackgerrit | Mark Vanderwiel proposed a change to openstack-infra/config: Fix Berkshelf 3.x GeCode 3.x dependency https://review.openstack.org/117274 | 15:28 |
| anteaya | jeblair: I think that is a point that needs to be raised at the meeting, yes | 15:28 |
| sdague | as of a couple weeks ago, nova, cinder, ceilometer at a minimum | 15:28 |
| jeblair | anteaya: i will not be there to raise that point | 15:28 |
| *** Daviey has joined #openstack-infra | 15:28 | |
| marun | jeblair: and those people are crazy ;) | 15:28 |
| anteaya | jeblair: which I can do, but would be grateful for support | 15:28 |
| anteaya | jeblair: thanks for letting me know, I will do my best | 15:29 |
| sdague | dtroyer: you got patches somewhere for that? | 15:29 |
| fungi | ugh. freenode server syncs are still really bursty to the point of making conversation very hard to follow. it's only a matter of time before another massive netsplit i wager | 15:29 |
| jeblair | marun: yeah, but they run the clouds we use in infra, so i have a soft spot for them :) | 15:30 |
| dtroyer | sdague: I have code, yes, but I don't think it's in usable form yet…need to dig up the notes | 15:30 |
| marun | jeblair: I didn't say they were unlikable :) | 15:30 |
| jeblair | though it's more like a commit every few weeks for them | 15:30 |
| *** rcarrillocruz has left #openstack-infra | 15:30 | |
| *** juice has quit IRC | 15:30 | |
| jeblair | fungi: i'm on wilhelm.freenode.net and things seem generally not bursty | 15:30 |
| *** esker has joined #openstack-infra | 15:31 | |
| *** oomichi has quit IRC | 15:31 | |
| fungi | jeblair: yeah, i suspect something's happening to the network on dickson (where i'm connected) | 15:31 |
| openstackgerrit | yolanda.robla proposed a change to openstack-infra/config: Align prepare_node.sh with prepare-node in dib https://review.openstack.org/105931 | 15:32 |
| *** mmaglana has joined #openstack-infra | 15:32 | |
| sdague | dtroyer: so that would solve a bunch of things | 15:32 |
| sdague | it would fix some grenade screen flakiness | 15:32 |
| sdague | it would mean that we could make d-g optionally unstack at the end for these functional jobs | 15:32 |
| sdague | and we could also even test unstack at the end of a run to make sure services were all down | 15:33 |
| *** fungi has quit IRC | 15:34 | |
| *** jcoufal has quit IRC | 15:35 | |
| *** dragondm has quit IRC | 15:37 | |
| *** openstack has joined #openstack-infra | 16:19 | |
| openstackgerrit | A change was merged to openstack/requirements: Update bootstrap-datepicker to 1.3.1.0 https://review.openstack.org/116889 | 16:20 |
| clarkb | marun: jeblair: does unstack output anything? do we need to capture that output? | 16:20 |
| *** sandywalsh has joined #openstack-infra | 16:20 | |
| marun | clarkb: it does output, does it need to go elsewhere than the console? | 16:20 |
| *** dkranz has joined #openstack-infra | 16:20 | |
| *** bdpayne has joined #openstack-infra | 16:21 | |
| *** dustins has quit IRC | 16:21 | |
| clarkb | the stack.sh run uses stdbuf for things | 16:21 |
| clarkb | which is what prompted my question | 16:21 |
| marun | clarkb: it wouldn't be more than 20-30 lines | 16:21 |
| clarkb | marun: ok its probably fine as is for now and we can clean it up if we decide we need to | 16:22 |
| *** _nadya_ has quit IRC | 16:22 | |
| ryanpetrello | anybody in here know how I would go about setting my avatar on https://openstacksummitnovember2014paris.sched.org/speaker/ryan_petrello.1sq7etww#.U_4F5kvMpS8 ? | 16:23 |
| ryanpetrello | I’ve got an account/login w/ an avatar set, but it doesn’t seem to show up here :\ | 16:23 |
| ryanpetrello | ah | 16:23 |
| ryanpetrello | I guess this is sched, not a system you all maintain | 16:23 |
| fungi | ryanpetrello: yeah, we don't manage sched.org | 16:24 |
| ryanpetrello | *digs around some* | 16:24 |
| ryanpetrello | yea, just saw somebody linking to this and saw my absense of a face | 16:24 |
| ryanpetrello | just realized it’s sched.org :) | 16:24 |
| *** marcoemorais has joined #openstack-infra | 16:25 | |
| marun | clarkb: sounds good | 16:25 |
| *** dprince has joined #openstack-infra | 16:25 | |
| *** mmaglana has joined #openstack-infra | 16:25 | |
| * jeblair has no face. all of his avatars are eerily accurate. | 16:25 | |
| *** bknudson has quit IRC | 16:26 | |
| *** flaviof is now known as flaviof_zzz | 16:27 | |
| openstackgerrit | Andreas Jaeger proposed a change to openstack-infra/config: Fix stackforge reference https://review.openstack.org/117297 | 16:27 |
| openstackgerrit | Davanum Srinivas (dims) proposed a change to openstack-infra/config: Create repo for oslo.log https://review.openstack.org/116934 | 16:28 |
| *** yamahata has quit IRC | 16:28 | |
| dmsimard | Is there a process/procedure in place to get a channel added to https://wiki.openstack.org/wiki/IRC (along with IRC and the log bot) ? | 16:30 |
| dmsimard | er… along with chanserv and the log bot | 16:30 |
| *** ildikov_ has joined #openstack-infra | 16:30 | |
| *** dims has joined #openstack-infra | 16:30 | |
| fungi | dmsimard: http://ci.openstack.org/irc.html is some good light reading | 16:31 |
| jeblair | oh, needs updating for accessbot :( | 16:32 |
| jeblair | modules/openstack_project/files/accessbot/channels.yaml should be included in there | 16:32 |
| *** Hefeweiz1n is now known as Hefeweizin | 16:32 | |
| *** ildikov has quit IRC | 16:33 | |
| *** jpich has quit IRC | 16:33 | |
| *** sandywalsh has quit IRC | 16:33 | |
| *** sdake_ has quit IRC | 16:34 | |
| BobBall | oh... perhaps I should have asked in here. Can we set arbitrary variables in nova.conf in the gate? My understanding (hopefully old) is that we need a local.conf rather than localrc but the gate only works on localrc ATM? | 16:34 |
| openstackgerrit | A change was merged to openstack-infra/config: Test oslo.db against all supported SA branches https://review.openstack.org/104134 | 16:34 |
| *** bhuvan has joined #openstack-infra | 16:34 | |
| dmsimard | fungi, jeblair: thanks | 16:35 |
| *** markmc has quit IRC | 16:36 | |
| *** nkinder has quit IRC | 16:37 | |
| *** derekh has quit IRC | 16:38 | |
| openstackgerrit | Andreas Jaeger proposed a change to openstack-infra/config: Fix doc gate https://review.openstack.org/117297 | 16:38 |
| *** Ajaeger has joined #openstack-infra | 16:38 | |
| *** nkinder has joined #openstack-infra | 16:39 | |
| * Ajaeger hopes that his patch for https://review.openstack.org/117297 works fine in the gate as well, reviews are welcome to unbreak the gate | 16:39 | |
| * Ajaeger wonders why gate-ci-docs is not failing for everybody | 16:40 | |
| *** marcoemorais has quit IRC | 16:42 | |
| *** marcoemorais has joined #openstack-infra | 16:42 | |
| openstackgerrit | Davanum Srinivas (dims) proposed a change to openstack/requirements: Add oslo.serialization under Oslo program https://review.openstack.org/117299 | 16:42 |
| clarkb | BobBall: I don't think so. usually for things like restricting nova api workers we want to change those defautls in nova or devstack | 16:42 |
| jeblair | Ajaeger: i don't understand | 16:42 |
| jeblair | Ajaeger: what's broken? | 16:42 |
| *** marcoemorais has quit IRC | 16:43 | |
| *** marcoemorais has joined #openstack-infra | 16:43 | |
| *** marcoemorais has quit IRC | 16:43 | |
| *** marcoemorais has joined #openstack-infra | 16:44 | |
| *** paulrad has quit IRC | 16:44 | |
| BobBall | clarkb: yeah - figured as much... ATM I'm struggling with RAM in the XenServer CI - and a big culprit is the number of workers. Nova-api has 18 (6 CPUs, 3 services) but tempest can only use 3 (since it halves the number of CPUs to get the number of threads) so I figured reducing to TEMPEST_CONCURRENCY was a trivial way to reclaim memory | 16:44 |
| *** alexpilotti has quit IRC | 16:44 | |
| clarkb | BobBall: are you using 8GB nodes? | 16:45 |
| *** bdpayne has quit IRC | 16:46 | |
| Ajaeger | https://review.openstack.org/#/c/117172/ - gate-ci-docs | 16:46 |
| clarkb | BobBall: we restrict to 8GB of ram on our test nodes so curious if you are finding that isn't enough | 16:46 |
| BobBall | ... 'yes' but because XenServer can't allocate VM memory from that 8G and has a dom0 that uses 700M I've only got 4G allocated to the DomU running devstack | 16:46 |
| Ajaeger | jeblair: ^ - and it failed locally for me without patches as well | 16:46 |
| *** todd_dsm has joined #openstack-infra | 16:46 | |
| BobBall | well I think that 8G isn't enough for neutron jobs (dhellmann must get annoyed with me pinging his hilight today) - he found oomkiller killing mysql in the gate | 16:47 |
| *** krtaylor has quit IRC | 16:47 | |
| clarkb | BobBall: I think that is an exception though | 16:47 |
| clarkb | otherwise we would see a ton of OOMKiller | 16:47 |
| BobBall | Devstack's domU will routinely use 6-7GB without considering VMs | 16:48 |
| jeblair | Ajaeger: so it seems like that version of docutils maybe had a behavior change? | 16:48 |
| *** koolhead17 has quit IRC | 16:48 | |
| mtreinish | BobBall: yeah that's an exception, it's a neutron bug | 16:48 |
| BobBall | ah ok | 16:48 |
| BobBall | well - still - the base usage in devstack is over 6G ATM | 16:49 |
| Ajaeger | jeblair: yeah, maybe | 16:49 |
| BobBall | which doesn't leave much space left - and that's made much worse by XenServer's required architecture | 16:49 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/zuul: Cloner: add project-specific branch overrides https://review.openstack.org/116125 | 16:50 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/zuul: Cloner: add upgrade test https://review.openstack.org/116105 | 16:50 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/zuul: Add cache-dir argument to cloner https://review.openstack.org/116740 | 16:50 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/zuul: Cloner: default to ZUUL_BRANCH https://review.openstack.org/116098 | 16:50 |
| Ajaeger | jeblair: yeah, 0.12 is somehow broken - I need to investigate a bit more ;( | 16:50 |
| Ajaeger | the changes I did are fine but not enough | 16:51 |
| * Ajaeger digs further... | 16:51 | |
| *** chuckC has joined #openstack-infra | 16:51 | |
| *** doude has quit IRC | 16:51 | |
| *** tonytan4ever has quit IRC | 16:52 | |
| *** daxatlas has joined #openstack-infra | 16:53 | |
| *** BobBall is now known as BobBall_Away | 16:54 | |
| *** marun has quit IRC | 16:55 | |
| *** mpaolino has quit IRC | 16:56 | |
| jogo | ttx: thanks for looking, that makes sense actually. if a patch with broken unit tests make it into the gate queue that would happen | 16:58 |
| clarkb | zaro: hashar https://review.openstack.org/#/c/117041/ should interest you. I rechecked it to test again now that subunit is fixed | 16:58 |
| jogo | sdague: ^ | 16:58 |
| *** signed8bit has quit IRC | 16:59 | |
| *** markmcclain has joined #openstack-infra | 16:59 | |
| clarkb | jogo: did a patch with broken unittests make it through? that info is probably worth keeping because we got rid of clean check | 17:00 |
| jogo | clarkb: that is what ttx is implying with what he said to me earlier | 17:00 |
| *** gokrokve_ has joined #openstack-infra | 17:00 | |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/zuul: Cloner: add project-specific branch overrides https://review.openstack.org/116125 | 17:00 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/zuul: Add cache-dir argument to cloner https://review.openstack.org/116740 | 17:00 |
| *** gyee has joined #openstack-infra | 17:00 | |
| jogo | clarkb: agreed this is worth noting | 17:00 |
| jeblair | jogo: was it glance? | 17:01 |
| *** unicell has joined #openstack-infra | 17:01 | |
| jeblair | because if so, it's likely the known instance where they kept rechecking after the tox change merged | 17:01 |
| jeblair | i consider that to be an exceptional situation and not worth reinstituting the freshness check | 17:02 |
| *** amotoki has quit IRC | 17:02 | |
| jeblair | clarkb: we did not get rid of clean check | 17:02 |
| clarkb | jeblair: sorry we got rid of the timer on it | 17:02 |
| jeblair | clarkb: we called that the freshness check | 17:02 |
| jogo | jeblair: not sure which jobs as ttx did the analysis | 17:02 |
| jogo | jeblair: and I agree there are other factors at play right now so it is hard to figure out cause and effect | 17:02 |
| *** e0ne has quit IRC | 17:03 | |
| jeblair | i do want to know if it's having an impact, i'm not trying to discourage that. but we need to know the whole story, we can't just do a partial analysis | 17:03 |
| *** Ryan_Lane has joined #openstack-infra | 17:03 | |
| fungi | clarkb: in fact, worth noting that the bizarre corner case rpodolyaka pointed out with 104134 and its out-of-order approval is actually due to the clean check enforcement | 17:03 |
| *** slagle has joined #openstack-infra | 17:03 | |
| *** gokrokve has quit IRC | 17:04 | |
| fungi | er, not so much enforcement as triggering, but regardless | 17:04 |
| *** harlowja has joined #openstack-infra | 17:04 | |
| jogo | jeblair: here is an example: http://logs.openstack.org/53/112853/2/gate/gate-neutron-python26/3157891/console.html#_2014-08-26_07_43_11_267 | 17:04 |
| fungi | the fact that zuul updating a +1 verify with another +1 verify doesn't enqueue into the gate because the comment doesn't mention the +1 vote | 17:04 |
| jogo | http://logs.openstack.org/95/109895/2/gate/gate-neutron-python26/22639db/console.html#_2014-08-26_07_09_24_871 | 17:04 |
| jogo | jeblair: look like there are 3 hits of that in past 2 days | 17:05 |
| jogo | jeblair: I agree that we do not have enough data to draw any conclusions | 17:06 |
| jeblair | jogo: er that's not quite what i said :) | 17:06 |
| jeblair | jogo: so far, the only instances we have found where the freshness check might have caught something are related to the tox change | 17:07 |
| jogo | whee the tox wall of fail: http://logs.openstack.org/34/115634/2/gate/gate-ceilometer-python27/8585cb4/console.html#_2014-08-26_13_22_32_685 | 17:07 |
| jeblair | jogo: and even then, those likely would have been broken for a full 24 hours | 17:07 |
| fungi | with our current setup, it's entirely possible for two changes to enter the gate after each being fine on the branch tip, but then conflicting with one another logically once they're pipelined in the gate | 17:07 |
| fungi | and that was possible even before recent changes | 17:07 |
| jogo | fungi: good point | 17:07 |
| trinaths | Hey, are CI's being added to some mailing list to mail disabled status | 17:08 |
| anteaya | trinaths: hey did you actually read the email? | 17:08 |
| jeblair | jogo: so i'm not saying we don't have enough data to draw conclusions -- my current hypothesis is that the freshness check has little practical impact | 17:09 |
| jeblair | jogo: what we should endeavor to do is to find data that negates that hypothesis :) | 17:09 |
| trinaths | anteaya,: afraid of some spam, so asked here | 17:09 |
| anteaya | trinaths: not spam | 17:09 |
| trinaths | okay. added | 17:09 |
| trinaths | anteaya: done.. added my CI | 17:09 |
| jeblair | jogo: but that means if we find some counter examples, we need to fully understand them | 17:09 |
| *** praneshp has joined #openstack-infra | 17:10 | |
| jogo | jeblair: agreed | 17:10 |
| jeblair | jogo: so looking at that neutron change you pointed at | 17:10 |
| jeblair | https://review.openstack.org/#/c/109895/ | 17:10 |
| jeblair | jogo: it looks like it's related to the hashseed | 17:10 |
| *** signed8bit has joined #openstack-infra | 17:11 | |
| *** signed8bit has quit IRC | 17:11 | |
| jeblair | did neutron set it to 0? | 17:11 |
| *** bdpayne has joined #openstack-infra | 17:11 | |
| *** KanagarajM has quit IRC | 17:11 | |
| *** signed8bit has joined #openstack-infra | 17:11 | |
| *** signed8bit has quit IRC | 17:11 | |
| jogo | jeblair: hmm not sure what the did but, that may be it | 17:12 |
| fungi | they did eventually. i think prior to yesterday when that job ran, but i'll check the commit log | 17:12 |
| *** pradk has joined #openstack-infra | 17:12 | |
| *** signed8bit has joined #openstack-infra | 17:12 | |
| openstackgerrit | Davanum Srinivas (dims) proposed a change to openstack-infra/devstack-gate: Add oslo.utils and oslo.concurrency https://review.openstack.org/117307 | 17:12 |
| jogo | so strange query: message:"No distributions matching the version for oslotest>=1.1.0.0a1" | 17:12 |
| jogo | 39 hits in 7 days all fails | 17:12 |
| jeblair | fungi: it looks like it merged in july https://review.openstack.org/#/c/109729/ | 17:13 |
| jogo | noticed a few of those in the uncategorized page | 17:13 |
| fungi | ahh, yup. it was in place a month prior | 17:13 |
| *** paulrad has joined #openstack-infra | 17:14 | |
| jogo | hmm 9 hits for message:"No distributions matching the version for" AND build_queue:"gate" | 17:15 |
| *** tonytan4ever has joined #openstack-infra | 17:15 | |
| *** trinaths has quit IRC | 17:15 | |
| jogo | was there a py33 issue with oslotest or something? | 17:16 |
| *** trinaths has joined #openstack-infra | 17:16 | |
| fungi | jogo: i wonder if that neutron issue was fixed by https://review.openstack.org/109895 | 17:16 |
| *** tonytan4ever has quit IRC | 17:17 | |
| jogo | fungi: sounds plausible | 17:17 |
| fungi | hrm, no it claims to only surface with a random hash seed | 17:17 |
| *** tonytan4ever has joined #openstack-infra | 17:17 | |
| *** loki184 has quit IRC | 17:17 | |
| fungi | and anyway in the failing log you can see that PYTHONHASHSEED='0' | 17:21 |
| *** paulrad has quit IRC | 17:22 | |
| jogo | fungi: it can still be the case that you mentioned | 17:22 |
| jogo | two things break eachother that both pass current master | 17:23 |
| fungi | jogo: sure, just curious what the other thing was which was being tested ahead of it to cause this... should be able to fetch the zuul ref mentioned in the log and inspect the commit ancestry for it | 17:23 |
| openstackgerrit | Davanum Srinivas (dims) proposed a change to openstack-infra/config: Add missing oslo projects to gerritbot config https://review.openstack.org/117309 | 17:24 |
| phschwartz | fungi: should be cleared up now. close to 120 instances removed | 17:25 |
| fungi | phschwartz: yep--confirmed. thanks a ton!!! | 17:25 |
| phschwartz | fungi: any time | 17:25 |
| fungi | this should hopefully improve our capacity by another 15% or so | 17:25 |
| *** arborism has joined #openstack-infra | 17:26 | |
| phschwartz | now to find a hotel in London and one in either Ireland or Scotland for Nov 10-14 and Nov 15-18 | 17:26 |
| *** arborism has quit IRC | 17:26 | |
| jogo | psedlak: I take it you still haven't been able to land my patch to allow deleting instances in error state? | 17:26 |
| jogo | phschwartz: ^ | 17:27 |
| *** amcrn has joined #openstack-infra | 17:27 | |
| jogo | (typo) | 17:27 |
| *** marcoemorais has quit IRC | 17:27 | |
| *** marcoemorais has joined #openstack-infra | 17:27 | |
| *** marcoemorais has quit IRC | 17:27 | |
| phschwartz | jogo: We are in the process of deploying a recent version to our regions that has it included. Slowed down on the normal deployments for the ironic offering | 17:27 |
| *** marcoemorais has joined #openstack-infra | 17:28 | |
| *** sandywalsh has joined #openstack-infra | 17:28 | |
| *** marcoemorais has quit IRC | 17:28 | |
| *** joools has joined #openstack-infra | 17:28 | |
| jogo | phschwartz: cool, hopefully that patch will mean we don't have to keep asking for instances to be deleted, we can just do it ourselves | 17:28 |
| *** marcoemorais has joined #openstack-infra | 17:28 | |
| fungi | jogo: yeah, it will mean that nodepool just deletes them like normal, at worst when it periodicallt retries | 17:29 |
| fungi | periodically | 17:29 |
| phschwartz | jogo: Correct, and we are investigating the issue in neutron that causes the failure to start with. Looks like an issue in the upstream neutron code that causes at times a query of networks to fail and by default it tries to return all networks which timesout due to our volume. | 17:30 |
| *** p5ntangle has quit IRC | 17:30 | |
| jogo | phschwartz: oh neutron | 17:30 |
| phschwartz | jogo: :) | 17:30 |
| jogo | pleia2: ping - when your ready want to learn how to do a e-r fingerprint | 17:31 |
| jogo | pleia2: and help me write something up about it for others | 17:31 |
| pleia2 | jogo: now is good | 17:31 |
| *** jab has quit IRC | 17:34 | |
| *** ominakov has quit IRC | 17:34 | |
| openstackgerrit | Andreas Jaeger proposed a change to openstack-infra/config: Fix doc problems https://review.openstack.org/117297 | 17:34 |
| *** jab has joined #openstack-infra | 17:35 | |
| *** jab has joined #openstack-infra | 17:35 | |
| *** sdake_ has joined #openstack-infra | 17:36 | |
| *** Sukhdev has quit IRC | 17:36 | |
| *** Sincler has quit IRC | 17:38 | |
| *** Sincler has joined #openstack-infra | 17:39 | |
| *** _nadya_ has joined #openstack-infra | 17:40 | |
| Ajaeger | jeblair: the changes above are independ of docutils, we get errors/warnings already with current docutils... | 17:41 |
| *** _nadya_ has quit IRC | 17:42 | |
| *** zz_dimtruck is now known as dimtruck | 17:44 | |
| Ajaeger | is there any docutils or sphinx expert around that can help figuring out why docutils 0.12 fails with the config repo? See https://review.openstack.org/#/c/117172/ | 17:44 |
| *** dprince has quit IRC | 17:47 | |
| * Ajaeger found that changing the conf.py helps - now figuring out what exactly... | 17:47 | |
| clarkb | Ajaeger: I think dhellmann may know stuff | 17:48 |
| Ajaeger | clarkb: thanks - will ping him once I'm totally lost | 17:48 |
| *** p5ntangle has joined #openstack-infra | 17:49 | |
| openstackgerrit | David Moreau Simard proposed a change to openstack-infra/config: Add #openstack-operators to the eavesdrop bot https://review.openstack.org/117312 | 17:50 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/zuul: Add cache-dir argument to cloner https://review.openstack.org/116740 | 17:51 |
| *** pelix1 has quit IRC | 17:53 | |
| *** zhiyuan has quit IRC | 17:54 | |
| *** zhiyuan has joined #openstack-infra | 17:54 | |
| * clarkb rereviews 99990 | 17:55 | |
| *** jamesmcarthur has joined #openstack-infra | 17:55 | |
| pleia2 | core eyeballs on the 3rd party post would be helpful too: https://review.openstack.org/#/c/116989/ | 17:55 |
| jamesmcarthur | fungi: are you around? | 17:55 |
| pleia2 | once this merges+post queue I can send out the announcement | 17:56 |
| clarkb | pleia2: noted | 17:56 |
| sdague | jeblair: so what's happening when merge checking happens in zuul? | 17:56 |
| *** joools has quit IRC | 17:56 | |
| *** HeOS has quit IRC | 17:57 | |
| mrodden | so i just found a rogue build node that might have been reporting puppet reports to the openstack puppetmaster by accident. i apologize for any issues that might have caused... | 17:57 |
| *** _nadya_ has joined #openstack-infra | 17:58 | |
| clarkb | mrodden: it happens :P and it shouldn't cause any issues on our side | 17:58 |
| pleia2 | mrodden: been there, done that :) | 17:58 |
| mrodden | lol | 17:59 |
| clarkb | we will just have to lcean out the cert request at some point | 17:59 |
| clarkb | sdague: it runs the noop job in a special pipeline for all open changes against the branch that just had a change merged. This runs the intenral zuul merge job which will return failure if mergin fails | 17:59 |
| sdague | ok | 18:00 |
| fungi | jamesmcarthur: yep, i'm here--what's up? | 18:00 |
| jamesmcarthur | hey there | 18:00 |
| sdague | there is a nova change that in the gate that's going to 100% fail because of a bad merge, but I guess it didn't trigger a merge conflict | 18:00 |
| sdague | which I'm surprised by | 18:00 |
| *** mikal has quit IRC | 18:00 | |
| jamesmcarthur | fungi: we are about to make the DNS switch from CloudSites to Cloud Servers | 18:01 |
| fungi | jamesmcarthur: for the www site? | 18:01 |
| sdague | I wanted to figure out if everything worked, and there are just edge cases that aren't catchable (entirely possible) | 18:01 |
| sdague | or if something didn't work | 18:01 |
| jamesmcarthur | Yes sir. We’ve double checked the openstack.org endpoint for gerrit, but wanted to just give you guys a heads up. | 18:01 |
| jamesmcarthur | In the event you had any problems. | 18:01 |
| sdague | clarkb: do you have a moment to help figure out a thing for it? | 18:01 |
| clarkb | sdague: sure | 18:01 |
| fungi | jamesmcarthur: you probably already know, but obviously lower the ttl as low as you can for a while beforehand so you can switch back quickly if you have to ;) | 18:01 |
| sdague | clarkb: 114431,6 | 18:02 |
| clarkb | jamesmcarthur: thank you for the heads up | 18:02 |
| jamesmcarthur | yes sir :) | 18:02 |
| sdague | it's about 13 deep in the gate | 18:02 |
| jamesmcarthur | clarkb: no problem | 18:02 |
| *** dprince has joined #openstack-infra | 18:02 | |
| jamesmcarthur | this should really open up some doors for the infra team and their involvement with the site | 18:02 |
| clarkb | sdague: is that the one that should merge fail? | 18:02 |
| fungi | jamesmcarthur: yep, looks like you dropped it to 5 minutes, so should be all set | 18:02 |
| sdague | clarkb: well I don't know | 18:03 |
| clarkb | sdague: I think there is some potential for racing if a change is approved before the merge check job is run against it | 18:03 |
| clarkb | actually no | 18:03 |
| clarkb | because it will fail to merge in the gate when zuul merges for the gate | 18:03 |
| sdague | so it does rebase clean | 18:03 |
| sdague | however it fails pep8 afterwards | 18:03 |
| clarkb | and it did fail in the gate | 18:04 |
| *** _nadya_ has quit IRC | 18:04 | |
| clarkb | so this may be a case of something getting by the freshness check | 18:04 |
| jamesmcarthur | fungi: if all goes well, we should ultimately have you guys point gerrit at http://192.237.193.83/ to run your queries againsst member data | 18:04 |
| sdague | I looked | 18:04 |
| jamesmcarthur | So that your’e going outside of CloudFlare. However, we’re not launching with CloudFlare. | 18:05 |
| *** mikal has joined #openstack-infra | 18:05 | |
| sdague | it passed check 17 hrs ago | 18:05 |
| fungi | jamesmcarthur: okay, great. i'll make a note of that | 18:05 |
| sdague | so the freshness wouldn't have helped | 18:05 |
| *** p5ntangle has quit IRC | 18:05 | |
| jamesmcarthur | I’ll send a follow up with details before we make that switch. But just a heads up. | 18:05 |
| fungi | jamesmcarthur: thanks! | 18:05 |
| clarkb | sdague: so this is a preexisting corner case brought on by git merges | 18:05 |
| clarkb | sdague: cool! | 18:05 |
| fungi | sdague: one of my nits about git (git nits?) is that it's written for c where whitespace is merely cosmetic most of the time... sigh | 18:06 |
| sdague | ok, I'm going to try to understand this at least a bit more to figure out what's going on | 18:06 |
| sdague | in case it's something we can learn from | 18:06 |
| *** paulrad has joined #openstack-infra | 18:06 | |
| fungi | though i suppose it's cosmetic here too if unit tests are passing, it's more that we're enforcing whitespace-related style which doesn't actually break semantically | 18:06 |
| dhellmann | Ajaeger: do we care about building manpages for openstack-infra/config? | 18:07 |
| dhellmann | Ajaeger: that looks like some sort of limitation of that builder | 18:07 |
| Ajaeger | dhellmann: Just came to the same conclusion ;) Thanks a lot! | 18:07 |
| Ajaeger | Let's ask jeblair, fungi and clarkb | 18:07 |
| Ajaeger | Patch coming for discussion... | 18:07 |
| fungi | Ajaeger: i don't think we have any manpages for that repo. should it need them? | 18:08 |
| clarkb | sdague: I think what happened was git merge resulted in code that could not pass pep8 | 18:08 |
| jroll | jeblair: re-bugging you about https://review.openstack.org/#/c/112143/ :) | 18:08 |
| *** annegentle has quit IRC | 18:08 | |
| *** jp_at_hp has quit IRC | 18:08 | |
| openstackgerrit | Andreas Jaeger proposed a change to openstack-infra/config: Remove docutils pin https://review.openstack.org/117172 | 18:08 |
| clarkb | sdague: so this is a case where the automagic merge isn't so great | 18:08 |
| Ajaeger | fungi: I don't think it needs them - see patch 117172 | 18:08 |
| *** sdake_ has quit IRC | 18:09 | |
| *** ominakov has joined #openstack-infra | 18:09 | |
| Ajaeger | dhellmann: strange that it only failed for config repo. We've removed the docutils pin in 20+ repos and config is the only repository with problems | 18:09 |
| dhellmann | Ajaeger: I think it has something to do with the inlined php code in openstackid.rst | 18:10 |
| dhellmann | Ajaeger: when I turn on warnerrors sphinx falls over on that | 18:10 |
| clarkb | I don't think we need manpages for that document | 18:10 |
| dhellmann | clarkb: ok, I'm not sure how to turn them off in pbr, but I'll look at that | 18:11 |
| Ajaeger | dhellmann: I fixed quite some warnings here: https://review.openstack.org/#/c/117297/ - but not that one... | 18:11 |
| fungi | dhellmann: for the lockfile import ( https://review.openstack.org/101911 ) do you want to get ci access to its pypi registration before we import, or just remember to get it squared away before the next release? | 18:11 |
| Ajaeger | dhellmann: I turned off the manpage in 117172 | 18:11 |
| dhellmann | fungi: oh, let's get that, I assumed jd had done it already | 18:11 |
| dhellmann | fungi: I'll ping him | 18:11 |
| *** lcheng_ has joined #openstack-infra | 18:12 | |
| sdague | clarkb: honestly, I'm seriously confused | 18:12 |
| openstackgerrit | Doug Hellmann proposed a change to openstack-infra/config: Doc cleanup https://review.openstack.org/117319 | 18:12 |
| openstackgerrit | Anita Kuno proposed a change to openstack-infra/elastic-recheck: Expanded fingerprint 1361924 https://review.openstack.org/117320 | 18:12 |
| dhellmann | Ajaeger: I'll let you compare your changes to mine ^^ | 18:12 |
| sdague | because the thing it's freaking out about is a variable I can't find in the file | 18:12 |
| dhellmann | Ajaeger: my patch builds locally, including manpages | 18:12 |
| clarkb | sdague: did the merge algorithm rename a variable or some such? | 18:13 |
| clarkb | I guess I can run the merge locally | 18:13 |
| sdague | clarkb: I don't know | 18:13 |
| sdague | I ran the merge locally | 18:13 |
| *** daxatlas has quit IRC | 18:13 | |
| Ajaeger | sdague and other requirements cores: since we missed all the specs repositories with the docutils pin, could you approve https://review.openstack.org/#/c/117168/ to unbreak the requirements post job? There's still too many repos that need to be fixed. | 18:13 |
| dhellmann | fungi: actually, let's go ahead, I'll make a note about the pypi permission thing and deal with that separately | 18:13 |
| Ajaeger | dhellmann: I'll check, thanks | 18:13 |
| fungi | sdague: you may want to fetch the zuul url for that job in the gate instead | 18:13 |
| sdague | and I got the same results | 18:13 |
| sdague | https://review.openstack.org/#/c/114431/6/nova/tests/api/openstack/compute/plugins/v3/test_flavor_manage.py | 18:13 |
| sdague | fungi: sure, what's the zuul url for that | 18:14 |
| jamesmcarthur | fungi: alright sir - DNS change is made | 18:14 |
| jamesmcarthur | site is back up | 18:14 |
| fungi | sdague: this is the one currently in the gate pipeline still? | 18:14 |
| *** cnesa8 has quit IRC | 18:14 | |
| sdague | yeh | 18:14 |
| fungi | jamesmcarthur: awesome! | 18:14 |
| sdague | and it's running again | 18:14 |
| sdague | because something ahead reset | 18:14 |
| sdague | so it will fail again | 18:14 |
| *** e0ne has joined #openstack-infra | 18:14 | |
| fungi | sdague: oh, i'll need to wait until it gets a worker unless you have the log from the earlier run | 18:15 |
| clarkb | sdague: fungi easiest thing is wait for jobs to start then grab ref from unittest job | 18:15 |
| clarkb | the exact fetch is in the log iirc | 18:15 |
| fungi | yep | 18:15 |
| *** annegent- has joined #openstack-infra | 18:15 | |
| *** daxatlas has joined #openstack-infra | 18:15 | |
| *** annegentle has joined #openstack-infra | 18:15 | |
| fungi | i was going to do that because i'm too lazy to dig it out of the zuul log | 18:15 |
| fungi | <-- very, very lazy | 18:15 |
| jamesmcarthur | funki: curl --data "APPSEC=yEKVvoxEdGPxevE2&email=todd@openstack.org" http://openstack.org/verify/member | 18:15 |
| jamesmcarthur | seems to be working as expected as well | 18:15 |
| * fungi feels very funky now | 18:15 | |
| anteaya | Ajaeger: I'm not certain what is happening with the docutils pin removal revert, can I get a tl;dr? | 18:16 |
| sdague | Ajaeger: sure, I'm fast approving the revert based on the principle that it was a self inflicted recently change, and we should flip back | 18:16 |
| fungi | jamesmcarthur: appreciated. i'll keep an eye out for any reports of oddness and give you and todd a heads up if there is | 18:16 |
| Ajaeger | sdague: thanks! | 18:16 |
| jamesmcarthur | thanks guys | 18:16 |
| anteaya | Ajaeger: but since sdague is approving, you don't need to bother | 18:16 |
| * anteaya moves back to unclassified failures | 18:17 | |
| Ajaeger | anteaya: I followed up the last few days on jeblair's patch to remove the docutils pin (it's not needed anymore since it was fixed). But when jeblair did the patch, I think we had no specs repositories and therefore those were missed. | 18:17 |
| fungi | i'm still unclear on why the specs repos need reqs sync | 18:18 |
| Ajaeger | Now we have in all specs repo a docutils pin (line docutils==0.9.1) and the requirements post job fails. | 18:18 |
| anteaya | Ajaeger: ah | 18:18 |
| Ajaeger | anteaya: so, I've created lots of one liner patches - see https://review.openstack.org/#q,I9107884d964757a9154bdd1feb4ea6df4e9df43e,n,z | 18:18 |
| anteaya | so we need the pin to get the jobs passing then remove the pin for specs repos | 18:18 |
| * anteaya nods | 18:18 | |
| anteaya | got it | 18:18 |
| anteaya | so it isn't necessary except that we said so | 18:19 |
| Ajaeger | not the jobs passing - we need to pin so that the requirements post job works, see: https://jenkins.openstack.org/job/propose-requirements-updates/ | 18:19 |
| *** koolhead17 has joined #openstack-infra | 18:19 | |
| sdague | fungi: I would agree that I don't know why they are in projects.txt ... but that seems like a longer conversation | 18:19 |
| fungi | specs repos don't strike me as something distros are going to package, nor something that is going to be of much interest at all outside of our audience | 18:19 |
| Ajaeger | anteaya: and what's all my patches are in, we can revert the revert ;) | 18:19 |
| fungi | sdague: well, i've questioned it previously when projects asked to add their specs repos to the projects.txt in openstack/requirements | 18:20 |
| Ajaeger | anteaya: the change was too early but we didn't notice | 18:20 |
| JayF | Ajaeger: if you can get https://review.openstack.org/#/c/117180/1 in a mergable state for ironic-specs I'll gladly review it | 18:20 |
| Ajaeger | JayF: give me a minute... | 18:20 |
| JayF | I was just reviewing that, then saw the chat in here :) | 18:21 |
| *** bknudson has joined #openstack-infra | 18:21 | |
| Ajaeger | dhellmann: I have one suggestion and then propose your patch goes in. Will abandon my version | 18:21 |
| Ajaeger | dhellmann: thanks! | 18:21 |
| fungi | sdague: clarkb: https://jenkins01.openstack.org/job/gate-nova-pep8/24989/console fetched refs/zuul/master/Zad90c82288dc40d081a8f896665c67cf | 18:21 |
| sdague | fungi: so I think we should probably end up doing a requirements session at summit in cross project and just nail down some policy | 18:21 |
| *** RicoLin has quit IRC | 18:21 | |
| clarkb | sdague: git fetch http://zm03.openstack.org/p/openstack/nova refs/zuul/master/Zad90c82288dc40d081a8f896665c67cf | 18:21 |
| fungi | sdague: clarkb: git fetch http://zm03.openstack.org/p/openstack/nova refs/zuul/master/Zad90c82288dc40d081a8f896665c67cf | 18:22 |
| clarkb | sdague: don't forget to git checkout FETCH_HEAD | 18:22 |
| dhellmann | Ajaeger: so you mean change the reference to just point to the section later in the same file? | 18:22 |
| *** Sukhdev has joined #openstack-infra | 18:23 | |
| *** dprince has quit IRC | 18:23 | |
| dhellmann | Ajaeger: sorry, you mean the referenced link in the file don't you | 18:23 |
| fungi | sdague: i couldn't agree more. the recent "let's revisit the what and why of reqs" documentation improvement is a nice segue into that session | 18:23 |
| sdague | fungi: yeh, that was part of the plan :) | 18:23 |
| *** dprince has joined #openstack-infra | 18:23 | |
| clarkb | and with bandersnatch the mechanics have changes in substantial ways | 18:24 |
| *** yfried has joined #openstack-infra | 18:24 | |
| Ajaeger | dhellmann: i see two options: Change it as I do in https://review.openstack.org/#/c/117297/3/doc/source/gerrit.rst or remove the line 520. | 18:24 |
| sdague | at least get us closer to the same page, with docs in tree so people can figure it out | 18:24 |
| clarkb | (you don't need to be in global reqs to get reliable pip) | 18:24 |
| dhellmann | Ajaeger: I think those 2 references point to the same place, but the form I'm using doesn't require knowing the base URL of the document | 18:24 |
| openstackgerrit | Doug Hellmann proposed a change to openstack-infra/config: Doc cleanup https://review.openstack.org/117319 | 18:25 |
| dhellmann | Ajaeger: ^^ removes the URL definition since it isn't used/needed | 18:25 |
| Ajaeger | dhellmann: Exactly! Thanks! | 18:25 |
| anteaya | dhellmann: typo repots | 18:26 |
| clarkb | pleia2: anteaya: I am +2 on that doc change | 18:26 |
| *** zhiyuan has quit IRC | 18:26 | |
| *** changbl has quit IRC | 18:26 | |
| clarkb | and now I need to start moving otherwise I will be late | 18:26 |
| * anteaya envisions warnerror repotting | 18:26 | |
| anteaya | clarkb: thank you | 18:27 |
| clarkb | I am also +2 on current 99990 | 18:27 |
| anteaya | did it pass jenkins? | 18:27 |
| *** daxatlas_ has joined #openstack-infra | 18:27 | |
| * anteaya checks | 18:27 | |
| Ajaeger | dhellmann: let me make my patch for docutils pin removal dependend on yours, thanks! | 18:27 |
| dhellmann | Ajaeger: did you spot a typo somewhere? | 18:27 |
| *** zhiyuan has joined #openstack-infra | 18:27 | |
| anteaya | I did | 18:28 |
| Ajaeger | dhellmann: I don't remember any typo | 18:28 |
| anteaya | dhellmann: the commit message: repots | 18:28 |
| *** dims has quit IRC | 18:28 | |
| anteaya | methinks reports is the word you wanted | 18:28 |
| dhellmann | anteaya: ah | 18:28 |
| *** dims has joined #openstack-infra | 18:28 | |
| dhellmann | anteaya: sorry, mixed you up with Ajaeger :-/ | 18:29 |
| openstackgerrit | Andreas Jaeger proposed a change to openstack-infra/config: Remove docutils pin https://review.openstack.org/117172 | 18:29 |
| openstackgerrit | Doug Hellmann proposed a change to openstack-infra/config: Doc cleanup https://review.openstack.org/117319 | 18:29 |
| dhellmann | anteaya: ^ | 18:29 |
| dhellmann | oops, sorry, Ajaeger | 18:29 |
| Ajaeger | dhellmann: no problem | 18:29 |
| openstackgerrit | Andreas Jaeger proposed a change to openstack-infra/config: Remove docutils pin https://review.openstack.org/117172 | 18:29 |
| anteaya | dhellmann: it happens | 18:29 |
| *** gokrokve has joined #openstack-infra | 18:29 | |
| sdague | clarkb: so honestly, I think the merge went funny because a ton of related things were being changed not in a patch series | 18:31 |
| Ajaeger | JayF: https://review.openstack.org/#/c/117180/ rebased | 18:31 |
| sdague | the merge results create a patch which is completely unlike what I'm looking at in review | 18:32 |
| sdague | like there are new functions in it | 18:32 |
| *** changbl has joined #openstack-infra | 18:32 | |
| sdague | clarkb: what is doing the merge? | 18:33 |
| Ajaeger | anteaya: thanks for all your +1s | 18:33 |
| sdague | is it cgit or jgit? | 18:33 |
| fungi | sdague: clarkb it looks like it's picking up a line from https://review.openstack.org/95059 which merged at 14:13 (after it entered the gate ahead of it) | 18:33 |
| *** gokrokve_ has quit IRC | 18:33 | |
| sdague | fungi: that might be it | 18:33 |
| clarkb | cgit does the merge in a way similar to jgit | 18:34 |
| anteaya | Ajaeger: np | 18:34 |
| anteaya | dhellmann: I want to see jenkins finish before I +1, I'd like to see the pages first | 18:34 |
| *** jab has quit IRC | 18:34 | |
| sdague | well, I was getting very different local results, so who knows | 18:35 |
| dhellmann | anteaya: sure thing, and feel free to take that over and make changes if you want -- I was mostly trying to help Ajaeger debug the build failure | 18:35 |
| *** flaviof_zzz is now known as flaviof | 18:36 | |
| *** lcheng_ has quit IRC | 18:36 | |
| Ajaeger | dhellmann: that debugging was highly appreciated! | 18:36 |
| dhellmann | Ajaeger: any time! | 18:36 |
| *** hashar has joined #openstack-infra | 18:36 | |
| *** gokrokve has quit IRC | 18:36 | |
| *** gokrokve has joined #openstack-infra | 18:37 | |
| anteaya | dhellmann: actually I was learning a lot from reviewing your patch so much rst I don't know, not trying to take it over, but between the three of us we hopefully can ensure it is in a mergable state | 18:37 |
| clarkb | sdague locally you have to merge all the things ahead in the fate too | 18:37 |
| clarkb | *gate | 18:37 |
| Ajaeger | sdague, dhellmann : Since you're both in the TC: please cast your vote also for the governance change to remove the docutils pin https://review.openstack.org/#/c/117171/ Thanks! | 18:37 |
| anteaya | clarkb: no, fate sounded right | 18:37 |
| Ajaeger | anteaya: run "doc8" locally, it will show you some of the problems. | 18:37 |
| *** annegentle has quit IRC | 18:37 | |
| *** annegent- is now known as annegentle | 18:38 | |
| dhellmann | Ajaeger: that's probably something ttx can just approve | 18:38 |
| * Ajaeger learned also a bit from dhellmann's changes | 18:38 | |
| sdague | Ajaeger: that just needs ttx | 18:38 |
| sdague | he has the only +A bit | 18:38 |
| Ajaeger | sdague, dhellmann : Ah, ok. | 18:38 |
| *** johnthetubaguy is now known as zz_johnthetubagu | 18:38 | |
| anteaya | Ajaeger: um, what does doc8 use? tox? | 18:38 |
| dhellmann | Ajaeger: we like to vote, but little stuff like that we don't worry about :-) | 18:39 |
| Ajaeger | dhellmann: :) | 18:39 |
| *** hashar has quit IRC | 18:39 | |
| *** gokrokve has quit IRC | 18:39 | |
| Ajaeger | doc8 is a python package, just install and run it like "doc8 doc/source" | 18:39 |
| *** gokrokve has joined #openstack-infra | 18:39 | |
| dhellmann | Ajaeger: does that show things turning on warnerrors doesn't catch? | 18:39 |
| anteaya | Ajaeger: ah, I can try | 18:40 |
| Ajaeger | dhellmann: I didn't know about warnerrors ;( | 18:40 |
| dhellmann | Ajaeger: ah :-) | 18:40 |
| Ajaeger | Let me show the output of doc8... | 18:40 |
| *** trinaths has quit IRC | 18:40 | |
| dhellmann | Ajaeger: I'm running it now | 18:41 |
| Ajaeger | anteaya, dhellmann: http://paste.openstack.org/show/101165/ - on head without dhellmann's changes | 18:41 |
| dhellmann | interesting | 18:41 |
| *** ominakov has quit IRC | 18:41 | |
| openstackgerrit | Doug Hellmann proposed a change to openstack-infra/config: add doc8 venv https://review.openstack.org/117322 | 18:42 |
| *** zigo has quit IRC | 18:43 | |
| dhellmann | Ajaeger: warnerrors doesn't catch long lines or trailing whitespace | 18:43 |
| Ajaeger | dhellmann: yep. do you want to sent a followup patch? | 18:43 |
| openstackgerrit | Anita Kuno proposed a change to openstack-infra/elastic-recheck: Expanded fingerprint 1361924 https://review.openstack.org/117320 | 18:43 |
| anteaya | wow is it unhappy | 18:44 |
| *** dmsimard is now known as dmsimard_away | 18:44 | |
| dhellmann | Ajaeger: I have some other stuff I need to finish before the end of the day | 18:44 |
| *** gokrokve has quit IRC | 18:44 | |
| Ajaeger | dhellmann: that's one you could approve - oslo-specs docutils-pin removal: https://review.openstack.org/#/c/117163/ | 18:44 |
| Ajaeger | dhellmann: then let me send one, my day is nearly over ;) | 18:45 |
| *** gokrokve has joined #openstack-infra | 18:45 | |
| dhellmann | Ajaeger: +2A | 18:45 |
| Ajaeger | dhellmann: thanks! | 18:45 |
| *** dmsimard_away is now known as dmsimard | 18:45 | |
| anteaya | my body is crying for a walk, back in a while | 18:45 |
| pleia2 | anteaya: enjoy | 18:45 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/config: Add integration apply test https://review.openstack.org/117323 | 18:45 |
| jogo | pleia2: how about after lunch? | 18:46 |
| jeblair | clarkb, fungi, nibalize1, jesusaurus: ^ that's the final piece to the puppet integration test | 18:46 |
| *** harlowja has quit IRC | 18:46 | |
| *** zigo has joined #openstack-infra | 18:46 | |
| jeblair | clarkb, fungi, nibalize1, jesusaurus: (the zuul cloner and nodepool image updates being the first 2 pieces) | 18:46 |
| pleia2 | jogo: sure, I'm here until 3 today, then have to head toward the airport (debconf in portland) | 18:46 |
| anteaya | pleia2: thanks | 18:46 |
| *** harlowja has joined #openstack-infra | 18:46 | |
| jogo | pleia2: that should work I think | 18:47 |
| *** lukego has quit IRC | 18:48 | |
| *** _nadya_ has joined #openstack-infra | 18:49 | |
| *** gokrokve has quit IRC | 18:49 | |
| *** ominakov has joined #openstack-infra | 18:51 | |
| marcusvrn | Hi guys! I'm finishing my CI system and now I'm configuring the logs server...can anyone help me? To setup the server on Jenkins Master, I just need to put my 'site' into site label in macros.yaml? Do you guys have some suggestion about a free server (I saw that someone uses dropbox)? | 18:54 |
| *** trinaths has joined #openstack-infra | 18:54 | |
| *** tsg has joined #openstack-infra | 18:55 | |
| jeblair | jroll: +3 | 18:55 |
| *** ociuhandu has quit IRC | 18:55 | |
| *** Sincler has quit IRC | 18:56 | |
| fungi | jeblair: does 117323 depend on your series of updates to zuul-cloner? | 18:58 |
| fungi | jeblair: oh, i'll interpret your comment in here to mean that it does ;) | 18:59 |
| *** atiwari has quit IRC | 18:59 | |
| jeblair | fungi: all three changes are technically independent, it just won't all work until they all land and images are updated | 19:00 |
| jeblair | fungi: but the job is experimental, so landing in any order won't break anything | 19:01 |
| jeblair | so no need to delay/order approval or anything | 19:01 |
| fungi | ahh, right-o. that's fair | 19:01 |
| fungi | i'll still review them together for contextual reasons | 19:01 |
| *** todd_dsm has quit IRC | 19:02 | |
| jeblair | fungi: the viewing order for proper context is: zuul, nodepool prep scripts, integration test job | 19:02 |
| fungi | thanks! | 19:02 |
| *** gokrokve has joined #openstack-infra | 19:02 | |
| jesusaurus | jeblair: im confused by what youre trying to do here: https://review.openstack.org/#/c/117323/1/install_modules.sh | 19:03 |
| *** sdake_ has joined #openstack-infra | 19:03 | |
| sdague | jeblair: question: on your super extra awesome CI results table, would it be possible to always make Jenkins be on top? (instead of alpha sorting) | 19:04 |
| jeblair | sdague: i think that's a good idea, and i expect it to be possible; there's sufficient data fidelity to do that (the results are grouped by system, and the system has a name, and all that is represented in the data structure) | 19:05 |
| jeblair | jesusaurus: the intent is to be able to avoid installing certain modules with that script | 19:05 |
| sdague | it's amazing how much easier it is to review these patches with all that presented all at the top like it is already | 19:05 |
| jeblair | jesusaurus: when doing an integration test with zuul, zuul gets to decide what commits to check out. that happens in the apply-test script | 19:06 |
| jeblair | jesusaurus: but we still want the remainder of the modules to be installed in the usual manner | 19:06 |
| *** plars_ is now known as plars | 19:06 | |
| *** gokrokve has quit IRC | 19:07 | |
| jeblair | sdague: yeah, i dig it | 19:07 |
| jeblair | (i added it to gertty recently too) | 19:07 |
| lifeless | morning | 19:07 |
| lifeless | russellb: thanks | 19:07 |
| *** krtaylor has joined #openstack-infra | 19:07 | |
| *** hashar has joined #openstack-infra | 19:07 | |
| jesusaurus | jeblair: ah, so that list of modules will need to be identical to the list we are giving zuul-cloner in tools/apply-test.sh? | 19:08 |
| jeblair | jesusaurus: yep. ideally that should be automatic, but i don't see an obvious way to do that right now. at any rate, the test will probably fail if we forget to add one :) | 19:09 |
| jesusaurus | jeblair: is there way we can consolidate that list into its own file? | 19:09 |
| hashar | good evening | 19:09 |
| jeblair | flaper87|afk, dhellmann, sdague: i'm happy to do a project rename saturday for glance.store if we want | 19:10 |
| jesusaurus | ah, yeah, i guess the tests failing will be a good enough indicator for now | 19:10 |
| Ajaeger | dhellmann: your patch 117319 does not build for me with docutils 0.12 ;( | 19:10 |
| dhellmann | jeblair, flaper87|afk, sdague : oh, that's great, I assumed you guys would be busy so we were just planning to put that off to a better time | 19:10 |
| Ajaeger | dhellmann: something still missing ;( | 19:11 |
| dhellmann | Ajaeger: testing... | 19:11 |
| dhellmann | Ajaeger: ah, still that fault in the manpage writer | 19:11 |
| jeblair | dhellmann: i expect saturday will still be quiet enough to do that, and i don't have weekend plans :) | 19:11 |
| openstackgerrit | Adam Gandelman proposed a change to openstack-infra/config: Fix DEVSTACK_GATE_GRENADE values for sideways jobs https://review.openstack.org/117333 | 19:12 |
| Ajaeger | dhellmann: yeah - and it's not the php, if I remove that it still fails. Your change is a great improvement but still... | 19:12 |
| *** rhe00 has joined #openstack-infra | 19:12 | |
| dhellmann | jeblair: ok, I wouldn't want you to do it without making sure it's not going to cause issues for flaper87|afk, but it's definitely fine with me | 19:12 |
| jeblair | dhellmann: sending an email too. | 19:13 |
| fungi | jeblair: dhellmann: flaper87|afk: sdague: i too am around saturday and happy to help make it happen quickly if it's decided | 19:13 |
| dhellmann | jeblair: follow-up email to that thread on the ml? | 19:13 |
| dhellmann | heh, yeah | 19:13 |
| * dhellmann checks the list of oslo-related changes for infra... | 19:13 | |
| sdague | jeblair / fungi: cool | 19:13 |
| *** denis_makogon_ has joined #openstack-infra | 19:13 | |
| sdague | I will *not* be around for the weekend :) (also probably off on friday). | 19:14 |
| fungi | i have weekend plans, but can put down the soldering iron for an hour with no serious repercussions (i'll just make sure to turn it off first!) | 19:14 |
| dhellmann | fungi: do you have time to look at https://review.openstack.org/116934 and https://review.openstack.org/117309 ? | 19:14 |
| jeblair | fungi: i have learned that it's also important to pick it up from the correct end when you resume. | 19:14 |
| fungi | jeblair: ouch! | 19:15 |
| dhellmann | jeblair: protip +1 | 19:15 |
| *** gokrokve has joined #openstack-infra | 19:15 | |
| *** paulrad has quit IRC | 19:16 | |
| dhellmann | fungi: also, I think it's fine to go ahead with the pylockfile import in https://review.openstack.org/101911 and I'll get jd to deal with the release permissions issue when he's ready to cut a release | 19:17 |
| *** gokrokve has quit IRC | 19:17 | |
| dhellmann | s/when/before | 19:17 |
| *** gokrokve_ has joined #openstack-infra | 19:17 | |
| *** paulrad has joined #openstack-infra | 19:17 | |
| fungi | dhellmann: okay, approved in that case | 19:17 |
| dhellmann | fungi: ok, I just re-saw clarkb's comment about the release group name, too | 19:17 |
| fungi | dhellmann: i assume there are other modifications needed to it anyway before we could hope to do automated release via ci | 19:18 |
| dhellmann | fungi: to the lib, yes, I'm sure | 19:18 |
| fungi | right, that's what i meant ;) | 19:18 |
| dhellmann | fungi: should I fix that group name? | 19:18 |
| fungi | dhellmann: if you want to real fast, i' | 19:18 |
| dhellmann | sure, sec | 19:18 |
| fungi | ll approve as soon as you push it | 19:18 |
| openstackgerrit | Doug Hellmann proposed a change to openstack-infra/config: Import pylockfile https://review.openstack.org/101911 | 19:19 |
| dhellmann | fungi: better? | 19:20 |
| fungi | dhellmann: perfect | 19:20 |
| sdague | dhellmann: you have +2 on req correct? - https://review.openstack.org/#/c/117218/ is the last piece of cleanup to adjust to the swiftclient changes that needs a +A | 19:20 |
| dhellmann | sdague: I don't know the background on that, is it just a new release? | 19:20 |
| sdague | swiftclient actually integrated a feature that has a new minimum | 19:21 |
| sdague | which is higher than we specified | 19:21 |
| *** homeless has quit IRC | 19:21 | |
| sdague | so https://review.openstack.org/#/c/117218/ makes it reflect reality | 19:21 |
| dhellmann | ok | 19:21 |
| dhellmann | +2a | 19:21 |
| dhellmann | Ajaeger: ok, let me look at the doc build issue again | 19:22 |
| sdague | it was part of the exploding earlier today on havana and icehouse, just wanted to get all those cruising through so we don't get bit again | 19:22 |
| sdague | thanks! | 19:22 |
| Ajaeger | dhellmann: thanks | 19:22 |
| *** homeless has joined #openstack-infra | 19:23 | |
| *** r-daneel has joined #openstack-infra | 19:24 | |
| *** Ryan_Lane has quit IRC | 19:24 | |
| *** Ryan_Lane has joined #openstack-infra | 19:26 | |
| *** HeOS has joined #openstack-infra | 19:27 | |
| *** Ryan_Lane1 has joined #openstack-infra | 19:28 | |
| dhellmann | Ajaeger: I suspect this is a bug in the new docutils, when was it released? | 19:29 |
| *** Ryan_Lane has quit IRC | 19:30 | |
| dhellmann | Ajaeger: you can turn off manpages building, so I'll submit that patch in front of my other one | 19:31 |
| Ajaeger | https://pypi.python.org/pypi/docutils/0.12 - 14th July | 19:31 |
| Ajaeger | yeah, turning it one option - we really don't need the manpage | 19:31 |
| Ajaeger | dhellmann: I can turn manpages of as part of https://review.openstack.org/117172 | 19:32 |
| openstackgerrit | Doug Hellmann proposed a change to openstack-infra/config: add doc8 venv https://review.openstack.org/117322 | 19:33 |
| openstackgerrit | Doug Hellmann proposed a change to openstack-infra/config: Doc cleanup https://review.openstack.org/117319 | 19:33 |
| openstackgerrit | Doug Hellmann proposed a change to openstack-infra/config: Turn off manpage building https://review.openstack.org/117337 | 19:33 |
| *** paulrad has quit IRC | 19:33 | |
| dhellmann | Ajaeger: ^^ | 19:33 |
| *** _nadya_ has quit IRC | 19:33 | |
| *** denis_makogon has quit IRC | 19:33 | |
| lifeless | dhellmann: oh hai | 19:33 |
| *** denis_makogon_ is now known as denis_makogon | 19:33 | |
| *** dmakogon_ has joined #openstack-infra | 19:34 | |
| dhellmann | lifeless: hi | 19:34 |
| dhellmann | lifeless: tell me about your troubles, sir | 19:34 |
| lifeless | dhellmann: they are many and multiple | 19:34 |
| lifeless | dhellmann: but first | 19:35 |
| lifeless | clarkb: I'd really like to unblock swift and glance.store - https://review.openstack.org/#/c/116775/ | 19:35 |
| dhellmann | aha, yes, we should push that one through now that the gate is happy | 19:35 |
| * fungi is gonna go grab some eats, then come back | 19:35 | |
| lifeless | dhellmann: and I want to push https://review.openstack.org/#/c/108269/ through too | 19:35 |
| openstackgerrit | Andreas Jaeger proposed a change to openstack-infra/config: Remove docutils pin https://review.openstack.org/117172 | 19:35 |
| lifeless | dhellmann: the failure was my bad subunit release | 19:36 |
| Ajaeger | dhellmann: yeah, thanks | 19:36 |
| *** Ryan_Lane1 has quit IRC | 19:36 | |
| lifeless | so uhm clarkb or fungi can I ask for a +2 [or -1 :P] on https://review.openstack.org/#/c/108269/ before someone does a new tag in our repos | 19:36 |
| lifeless | dhellmann: I actioned all your review feedback in the whole stack | 19:37 |
| dhellmann | lifeless: does that subunit issue apply to 108269 as well? | 19:37 |
| rhe00 | I am using Jay's setup for my CI. I seem to have a ssh key issue with zuul-merger. Does this sound familiar to anyone? Anyone knows the password for the zuul user in that setup? | 19:37 |
| lifeless | dhellmann: thats the one I was saying it applied to. | 19:38 |
| dhellmann | lifeless: oh, ok, the gate still doesn't like that one but it does like 116775 | 19:38 |
| lifeless | dhellmann: it will like 108269, I just triggered a recheck to prove it :) | 19:39 |
| dhellmann | lifeless: maybe it's a mirror since thing | 19:39 |
| dhellmann | lifeless: gotcha | 19:39 |
| dhellmann | lifeless: I'll work my way through the stack of open changes | 19:39 |
| *** todd_dsm has joined #openstack-infra | 19:40 | |
| lifeless | http://logs.openstack.org/03/114403/4/check/gate-pbr-python33/0d24977/console.html is at the very top but the only expected failure I know of. I just have NFI whats up - once I get this load_tests thing backported via testtools (http://bugs.python.org/issue16662) I'll locally debug it with 3.3. | 19:41 |
| *** andreykurilin has joined #openstack-infra | 19:41 | |
| jroll | jeblair: danke :) | 19:42 |
| *** Mmike has quit IRC | 19:45 | |
| *** Mmike has joined #openstack-infra | 19:45 | |
| *** koolhead17 has quit IRC | 19:47 | |
| *** todd_dsm has quit IRC | 19:48 | |
| openstackgerrit | Andreas Jaeger proposed a change to openstack-infra/config: Fix doc8 issues https://review.openstack.org/117342 | 19:48 |
| *** rlandy has joined #openstack-infra | 19:49 | |
| anteaya | marcusvrn: two things 1) attend the weekly third party meeting: https://wiki.openstack.org/wiki/Meetings/ThirdParty 2) serving logs from dropbox is very slow and you are likely to get a complaint | 19:50 |
| dhellmann | lifeless: I believe I've +2 all of the patches in your series now | 19:52 |
| lifeless | dhellmann: wooo yayayayaya | 19:52 |
| lifeless | dhellmann: now I need to find a second wee viewer | 19:52 |
| lifeless | dhellmann: and we're golden | 19:53 |
| lifeless | [and fix that py33 thing at the very top; later today]. | 19:53 |
| dhellmann | lifeless: I'm prepared to call https://review.openstack.org/#/c/116775/ critical if it's blocking other projects | 19:53 |
| dhellmann | and use my ptl-ness on it | 19:53 |
| dhellmann | lifeless: you mentioned glance and swift? | 19:53 |
| dhellmann | are they actually blocked? | 19:53 |
| lifeless | dhellmann: its was reported when it caused CI jobs in swift and glance.store to fail | 19:54 |
| lifeless | dhellmann: the associated bug has a logstash search to see impact | 19:54 |
| dhellmann | ok, I saw those failures, but I haven't heard any yelling so I wasn't sure if it was having any impact on the projects | 19:54 |
| dhellmann | glance.store retagged with a valid version to work around it, irrc | 19:54 |
| lifeless | its still triggering, lets see | 19:55 |
| lifeless | swift is still affected | 19:56 |
| lifeless | http://logstash.openstack.org/#eyJzZWFyY2giOiJtZXNzYWdlOlwiRXJyb3IgcGFyc2luZ1wiIEFORCBtZXNzYWdlOlwiVmFsdWVFcnJvcjogVW5rbm93biByZW1haW5kZXJcIiBBTkQgdGFnczpcImNvbnNvbGVcIiIsImZpZWxkcyI6W10sIm9mZnNldCI6MCwidGltZWZyYW1lIjoiNjA0ODAwIiwiZ3JhcGhtb2RlIjoiY291bnQiLCJ0aW1lIjp7InVzZXJfaW50ZXJ2YWwiOjB9LCJzdGFtcCI6MTQwOTE2OTM3OTc0Nn0= | 19:56 |
| * dhellmann wonders why notmyname hasn't complained | 19:57 | |
| lifeless | > 10 jobs / hour hitting this | 19:57 |
| notmyname | dhellmann: because I just flew in from the ops meetup this morning and have been away from it. I'm about to grab lunch, then complain :-) | 19:57 |
| *** mmaglana has quit IRC | 19:57 | |
| * notmyname registers a complaint | 19:58 | |
| dhellmann | notmyname: yay! I just approved the fix so I can tell you there's nothing to complain about :-) | 19:58 |
| notmyname | dhellmann: lol | 19:58 |
| dhellmann | notmyname: welcome back :-) | 19:58 |
| *** todd_dsm has joined #openstack-infra | 19:58 | |
| dhellmann | notmyname: after https://review.openstack.org/#/c/116775/ merges your rechecks shouldn't have the same issue any longer | 19:59 |
| notmyname | thanks! | 19:59 |
| *** tsg has quit IRC | 19:59 | |
| *** e0ne has quit IRC | 19:59 | |
| dhellmann | lifeless, notmyname : I'll give it a day or so to make sure that's true, and then cut a release. Since that would be friday, maybe monday morning. | 19:59 |
| *** tsg has joined #openstack-infra | 19:59 | |
| lifeless | dhellmann: ok; I'd obviously like to get more in before then :) | 20:00 |
| notmyname | dhellmann: remember that monday is a US holiday | 20:00 |
| dhellmann | lifeless: releases are easy to make, so we can do more whenver you want | 20:00 |
| dhellmann | notmyname: no kidding? | 20:00 |
| notmyname | labor day | 20:00 |
| dhellmann | indeed, so tuesday then | 20:00 |
| lifeless | dhellmann: it was more a statement about +A's than the release process :) | 20:01 |
| notmyname | dhellmann: fyi, we'll be tagging 2.1.0 final on monday (current schedule) | 20:01 |
| dhellmann | lifeless: yeah | 20:01 |
| lifeless | dhellmann: specifically the preversion-error-if-too-low | 20:01 |
| notmyname | dhellmann: ie I'll send ttx and email this weekend and he'll do it | 20:01 |
| dhellmann | notmyname: ok, this only affects pre-release versions so you should be ok if it's a real release | 20:01 |
| notmyname | ok | 20:01 |
| lifeless | dhellmann: since that is requiring manual effort to keep clean | 20:01 |
| lifeless | dhellmann: until it lands and can enforce it | 20:01 |
| notmyname | dhellmann: even if it's on the same commit? | 20:02 |
| dhellmann | notmyname: the problem is in an unreleased version of pbr, so you're seeing it in the gate where we install from master but it shouldn't be seen "in the wild" | 20:02 |
| dhellmann | notmyname: and yes, the tag sorting should pick up the real release as > the pre-release and use that value | 20:02 |
| dhellmann | lifeless: which patch is that? | 20:03 |
| lifeless | https://review.openstack.org/#/c/108269 | 20:03 |
| * notmyname wonders why the gate is using unreleased support libraries instead of what would be used in prod | 20:03 | |
| dhellmann | notmyname: the gate is our integration test environment for everything | 20:03 |
| lifeless | notmyname: same reason it uses unreleased versions of other API servers. | 20:03 |
| dhellmann | the alternative is to go back to having breaks when untested libs are released | 20:04 |
| *** hogepodge has joined #openstack-infra | 20:04 | |
| dhellmann | at least this way we're only breaking our devs | 20:04 |
| *** Sincler has joined #openstack-infra | 20:05 | |
| *** stevemar has quit IRC | 20:06 | |
| *** jamespd_ is now known as jamespd | 20:08 | |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/zuul: Cloner: add project-specific branch overrides https://review.openstack.org/116125 | 20:09 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/zuul: Add cache-dir argument to cloner https://review.openstack.org/116740 | 20:09 |
| jeblair | hashar: i'm not sure i like the idea of cloner doing hard-links to the cache repos by default | 20:11 |
| jeblair | hashar: that seems dangerous (i don't like that about git itself either :) | 20:11 |
| *** hashar has quit IRC | 20:12 | |
| *** Ajaeger has quit IRC | 20:12 | |
| *** todd_dsm has quit IRC | 20:12 | |
| *** hashar has joined #openstack-infra | 20:12 | |
| *** trinaths has quit IRC | 20:13 | |
| *** kgiusti has left #openstack-infra | 20:13 | |
| hashar | jeblair: yeah I am not sure what it is going to cause : -D | 20:13 |
| clarkb | before I go fix glasses jeblair what prompted the if self.args.project_branch check? does argparse give you None instead of [] when there are no project branches? | 20:13 |
| hashar | jeblair: I have been using git clone --reference /gerrit/replica for a while | 20:14 |
| jeblair | clarkb: yeah, i probably could have changed the default as an alternative | 20:14 |
| clarkb | meh this works | 20:14 |
| lifeless | clarkb: I realise you're busy, but would really love a couple more pbr reviews today | 20:15 |
| clarkb | lifeless: hoping to dive into reviews after these glasses have been reconstructed | 20:15 |
| lifeless | clarkb: cool, thank you! | 20:16 |
| hashar | jeblair: sorry --reference is doing something else. it instructs the cloned repo to fetch from the cached one, but does not cause hardlink. I should dig in a bit more :D | 20:17 |
| hashar | jeblair: clarified. cache dir patch is +1 now. Will do a follow up to support git clone --reference :D | 20:20 |
| jeblair | hashar: we're not using reference, we're doing git clone, which will hardlink if you don't do file:// | 20:20 |
| hashar | jeblair: yeah understood I +1ed the patch | 20:21 |
| hashar | will hack something for my use case, give it a try and propose a patch | 20:21 |
| hashar | or I can just order some more disks :-d | 20:21 |
| jeblair | hashar: i don't think --reference will save space? | 20:21 |
| *** adalbas has quit IRC | 20:22 | |
| *** dprince has quit IRC | 20:22 | |
| *** todd_dsm has joined #openstack-infra | 20:22 | |
| hashar | jeblair: it points to the cached repo objects | 20:22 |
| *** Sukhdev has quit IRC | 20:23 | |
| hashar | so if you have a 400MB pack in the cache, it fetch object from it to populate the working-dir files | 20:23 |
| hashar | and the cloned repo has a tiny .git directory (since everything is in the cached dir) | 20:23 |
| openstackgerrit | Davanum Srinivas (dims) proposed a change to openstack/requirements: Add oslo.log to global requirements https://review.openstack.org/117346 | 20:23 |
| hashar | but yeah all of that is scary and the doc is vastly confusing (there is --local file:// --reference --shared and so on) | 20:24 |
| clarkb | hashar: while you are here did you see https://review.openstack.org/#/c/117041/ ? | 20:24 |
| clarkb | hashar: it does actually pass tests. those failures were related to lib relased yesterady which has been fixed | 20:24 |
| hashar | jeblair: it is not a big deal. Just keep file:// and do copy. I will amend later on for my use case | 20:24 |
| hashar | clarkb: haven't seen it | 20:25 |
| hashar | clarkb: Iguess all JJB patches are failling now? | 20:25 |
| clarkb | hashar: no they pass | 20:25 |
| clarkb | hashar: my change just got unluck.y it is rechecking now and it looks good | 20:26 |
| * clarkb disappears again | 20:27 | |
| *** joools has joined #openstack-infra | 20:27 | |
| *** nkinder has quit IRC | 20:28 | |
| hashar | clarkb: you might want to look at https://review.openstack.org/#/c/64610/ which fixed some unicode issue as well | 20:28 |
| *** hogepodge has quit IRC | 20:29 | |
| *** mmaglana has joined #openstack-infra | 20:29 | |
| hashar | clarkb: commented on change. I am disappearing as well | 20:30 |
| *** rlandy has quit IRC | 20:31 | |
| *** sdake_ has quit IRC | 20:31 | |
| *** Ryan_Lane has joined #openstack-infra | 20:32 | |
| dkranz | clarkb: Is there a simple explanation why if you do a 'git log', some commits have Merge: and authored by Jenkins while others have no Merge: and list real names as authors? | 20:33 |
| lifeless | dkranz: yes. | 20:33 |
| dkranz | :) | 20:33 |
| *** dmsimard is now known as dmsimard_away | 20:33 | |
| *** dmsimard_away is now known as dmsimard | 20:33 | |
| *** rhe00 has left #openstack-infra | 20:33 | |
| lifeless | dkranz: its very simple. git log is showing both the commits people have done, and the merges that gerrit has done. | 20:33 |
| lifeless | dkranz: default will show both always. You can of course change this in git :) | 20:34 |
| dkranz | lifeless: Ah, ok, thanks! | 20:34 |
| lifeless | dkranz: there is a special case where fast forward merges happen, but the only impact that has is that no gerrit 'merge' commit is created. | 20:34 |
| *** julim has quit IRC | 20:36 | |
| *** david-lyle has quit IRC | 20:37 | |
| *** david-lyle has joined #openstack-infra | 20:37 | |
| *** ildikov has joined #openstack-infra | 20:37 | |
| *** pballand has joined #openstack-infra | 20:38 | |
| *** pdmars has quit IRC | 20:39 | |
| *** ildikov_ has quit IRC | 20:41 | |
| *** dizquierdo has joined #openstack-infra | 20:44 | |
| *** nkinder has joined #openstack-infra | 20:45 | |
| *** cnesa has quit IRC | 20:47 | |
| sdague | dkranz: git log --graph --abbrev-commit will show the graph, which might make it easier to grok as well | 20:48 |
| *** Loquacity has joined #openstack-infra | 20:48 | |
| *** stevebaker has quit IRC | 20:49 | |
| *** stevebaker has joined #openstack-infra | 20:49 | |
| *** stevebaker has quit IRC | 20:49 | |
| *** stevebaker has joined #openstack-infra | 20:49 | |
| *** stevebaker has quit IRC | 20:49 | |
| *** stevebaker has joined #openstack-infra | 20:49 | |
| *** stevebaker has quit IRC | 20:49 | |
| *** stevebaker has joined #openstack-infra | 20:49 | |
| *** stevebaker has quit IRC | 20:49 | |
| *** stevebaker has joined #openstack-infra | 20:49 | |
| *** stevebaker has quit IRC | 20:50 | |
| *** stevebaker has joined #openstack-infra | 20:50 | |
| *** stevebaker has quit IRC | 20:50 | |
| *** stevebaker has joined #openstack-infra | 20:50 | |
| *** stevebaker has quit IRC | 20:50 | |
| *** stevebaker has joined #openstack-infra | 20:50 | |
| *** stevebaker has quit IRC | 20:50 | |
| *** stevebaker has joined #openstack-infra | 20:50 | |
| openstackgerrit | Matt Riedemann proposed a change to openstack-infra/elastic-recheck: Add query for horizon not registered bug 1362325 https://review.openstack.org/117350 | 20:51 |
| uvirtbot` | Launchpad bug 1362325 in horizon "Dashboard with slug "sahara" is not registered" [Undecided,New] https://launchpad.net/bugs/1362325 | 20:51 |
| *** dkranz has quit IRC | 20:52 | |
| *** michchap_ has quit IRC | 20:52 | |
| *** michchap has joined #openstack-infra | 20:53 | |
| *** che-arne|2 has joined #openstack-infra | 20:53 | |
| *** juzuluag has quit IRC | 20:54 | |
| *** che-arne has quit IRC | 20:55 | |
| *** melwitt has joined #openstack-infra | 20:56 | |
| *** mfer has quit IRC | 20:56 | |
| *** paulrad has joined #openstack-infra | 20:57 | |
| *** loquacities has joined #openstack-infra | 20:58 | |
| *** trinaths has joined #openstack-infra | 20:58 | |
| *** Loquacity has quit IRC | 20:58 | |
| *** homeless has quit IRC | 20:58 | |
| *** homeless has joined #openstack-infra | 20:59 | |
| jroll | are folks seeing a lot of fails on tempest.scenario.test_dashboard_basic_ops.TestDashboardBasicOps ? | 20:59 |
| jroll | oops, I see https://bugs.launchpad.net/horizon/+bug/1362325 | 21:00 |
| uvirtbot` | Launchpad bug 1362325 in horizon "Dashboard with slug "sahara" is not registered" [Undecided,New] | 21:00 |
| *** melwitt1 has joined #openstack-infra | 21:01 | |
| *** melwitt has quit IRC | 21:03 | |
| *** krtaylor has quit IRC | 21:04 | |
| *** tgohad has joined #openstack-infra | 21:07 | |
| *** marcoemorais has quit IRC | 21:07 | |
| *** marcoemorais has joined #openstack-infra | 21:07 | |
| *** marcoemorais has quit IRC | 21:08 | |
| *** marcoemorais has joined #openstack-infra | 21:08 | |
| *** stevebaker has quit IRC | 21:08 | |
| *** tsg has quit IRC | 21:08 | |
| *** stevebaker has joined #openstack-infra | 21:08 | |
| *** markmcclain has quit IRC | 21:12 | |
| anteaya | so if I am proposing a patch to create a stackforge repo that only has the merge-check template for jobs and check and gate are noop, do I need any entry in jjb/config/projects.yaml? | 21:12 |
| anteaya | the merge-check template appears to be defined in zuul/layout.yaml not from a jjb job | 21:12 |
| *** paulrad has quit IRC | 21:14 | |
| *** e0ne has joined #openstack-infra | 21:15 | |
| *** aysyd has quit IRC | 21:15 | |
| *** mwagner_lap has quit IRC | 21:15 | |
| lifeless | ok I have a frankenmonster | 21:16 |
| lifeless | muhahahaha | 21:16 |
| *** markmcclain has joined #openstack-infra | 21:18 | |
| *** e0ne has quit IRC | 21:18 | |
| zaro | anteaya: if you have no jenkins jobs to run then you do not need an entry in projects.yaml | 21:19 |
| anteaya | zaro: thanks | 21:20 |
| *** r1chardj0n3s_afk is now known as r1chardj0n3s | 21:20 | |
| nibalize1 | wenlock: was it you I was speaking to about hiera and puppet at the last meeting? | 21:21 |
| *** nibalize1 is now known as nibalizer | 21:21 | |
| wenlock | nibalizel yeah, i added comments on the spec, in the review | 21:21 |
| mriedem | what was the old auto-abandon timeout? 10 days? 2 weeks? | 21:22 |
| *** e0ne has joined #openstack-infra | 21:22 | |
| openstackgerrit | Anita Kuno proposed a change to openstack-infra/config: Create a stackforge repo for ci dashboard project - radar https://review.openstack.org/114315 | 21:22 |
| sdague | 2 weeks | 21:22 |
| mriedem | sdague: ok, trying not to be too offensive in my e-r abandons | 21:23 |
| fungi | commence review of all new and exciting scrollback | 21:23 |
| *** ominakov has quit IRC | 21:24 | |
| openstackgerrit | Anita Kuno proposed a change to openstack-infra/config: Create a stackforge repo for ci dashboard project - radar https://review.openstack.org/114315 | 21:24 |
| *** pcm_ has quit IRC | 21:25 | |
| sdague | mriedem: yeh, I've been going through really old reviews that I'm listed on and abandoning with reasons | 21:26 |
| *** ominakov has joined #openstack-infra | 21:26 | |
| * SergeyLukjanov wake up | 21:28 | |
| * fungi finds it amusing that here in america we have a national holiday called "labor day" where we're expect to to be away from work. proof there's something wrong with our grasp of the english language | 21:28 | |
| r1chardj0n3s | clarkb: could you please URL me the change from yesterday you made to support pip-missing-reqs? | 21:28 |
| morganfainberg | fungi, lol | 21:32 |
| morganfainberg | fungi, right long weekend and such. | 21:32 |
| fungi | anteaya: specifically, putting it in jjb's projects.yaml is for when you want to instantiate a job template (possibly via a job group) and pass in parameters. most of the time it's because you have jobs where the project name is part of the job name. if your project is not adding any jobs in zuul's layout.yaml which include the name of the project, then it's likely not going to be needed (also the | 21:33 |
| *** sdake_ has joined #openstack-infra | 21:33 | |
| fungi | layout validation job will complain if it actually was) | 21:33 |
| *** e0ne has quit IRC | 21:33 | |
| anteaya | fungi: good to know | 21:33 |
| anteaya | thanks | 21:34 |
| *** dizquierdo has quit IRC | 21:34 | |
| *** melwitt1 is now known as melwitt | 21:35 | |
| *** bhuvan_ has joined #openstack-infra | 21:35 | |
| openstackgerrit | A change was merged to openstack-infra/elastic-recheck: Expanded fingerprint 1361924 https://review.openstack.org/117320 | 21:35 |
| *** bhuvan has quit IRC | 21:38 | |
| *** che-arne|2 has quit IRC | 21:38 | |
| *** nkinder has quit IRC | 21:39 | |
| *** bhuvan_ has quit IRC | 21:42 | |
| *** bhuvan has joined #openstack-infra | 21:44 | |
| clarkb | r1chardj0n3s: https://review.openstack.org/#/c/117072/ | 21:45 |
| r1chardj0n3s | thanks clarkb, am +1'ing now to keep track :) | 21:46 |
| openstackgerrit | Matthew Treinish proposed a change to openstack-infra/elastic-recheck: Add query for bug 1362347 https://review.openstack.org/117358 | 21:47 |
| uvirtbot` | Launchpad bug 1362347 in neutron "neutron-ns-meta invokes oom-killer during gate runs" [Undecided,New] https://launchpad.net/bugs/1362347 | 21:47 |
| * anteaya admits she doesn't know what oom-killer is | 21:48 | |
| SergeyLukjanov | fungi, jeblair, clarkb, re https://bugs.launchpad.net/horizon/+bug/1362325 - We need to land https://review.openstack.org/#/c/112000 and https://review.openstack.org/#/c/112001 to stop installing sahara-dashboard in the devstack code (we removed the deprecated code from the sahara-dashboard today), it's needed to unbreak the gate | 21:48 |
| uvirtbot` | Launchpad bug 1362325 in horizon "Dashboard with slug "sahara" is not registered" [Undecided,New] | 21:48 |
| jeblair | clarkb, r1chardj0n3s: can that be added to the project-requirements-change.py script? | 21:48 |
| fungi | anteaya: a linux kernel subsystem which identifies the likely biggest culprit for memory consumption out of the process table when virtual memory is exhausted and then kills it | 21:49 |
| jeblair | SergeyLukjanov: how did the gate end up getting broken? | 21:49 |
| adam_g | clarkb, small fix to the sideways jobs @ https://review.openstack.org/#/c/117333/ | 21:49 |
| *** cnesa has joined #openstack-infra | 21:49 | |
| anteaya | fungi: ah | 21:49 |
| clarkb | jeblair: maybe? the thing r1chardj0n3s pointed out yesterday was that different projects may have different needs for that and tox represents that well with a target | 21:49 |
| jeblair | i don't see a pip-missing-reqs env in nova | 21:49 |
| jeblair | where is this running? | 21:49 |
| clarkb | adam_g: gah I don't know how I missed that | 21:50 |
| *** rpodolyaka has quit IRC | 21:50 | |
| clarkb | jeblair: currently its a proposed change against nova | 21:50 |
| clarkb | r1chardj0n3s: ^ where is that change? | 21:50 |
| r1chardj0n3s | jeblair: https://review.openstack.org/#/c/114122/ | 21:50 |
| jeblair | clarkb: we're running silent on a proposed change? | 21:50 |
| jeblair | grr | 21:50 |
| clarkb | I didn't make it silent... | 21:50 |
| clarkb | at leeast I think I didn't | 21:51 |
| fungi | yeah, i think the "correct" way to do that was to merge the change to create the tox env, then create a non-voting check job which ran it demonstrating that it "fails" on the current state of the project, then propose the requiremrnts addition which fixes the project to passing that job | 21:51 |
| jeblair | clarkb: no i'm asking about the hundreds of times/day we're running that job now | 21:51 |
| jeblair | apparently we're collecting data on how reliable it | 21:51 |
| jeblair | is | 21:51 |
| jeblair | it's very reliable. it always fails. | 21:51 |
| *** dane_leblanc has quit IRC | 21:52 | |
| fungi | or s/non-voting check/experimental even/ | 21:52 |
| *** andreykurilin has quit IRC | 21:52 | |
| r1chardj0n3s | fungi the problem with the way clarkb has implemented triggering the check is that it's slightly contorted to actually get it to run (it only runs on requirements.txt changes) | 21:52 |
| *** daxatlas has quit IRC | 21:53 | |
| *** daxatlas_ is now known as daxatlas | 21:53 | |
| r1chardj0n3s | clarkb: I'm starting to think that's a bad idea; the tool picks up problems because of changes to *code* not changes to *requirements* | 21:53 |
| *** otherwiseguy has quit IRC | 21:53 | |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/config: Move pip-missing-reqs to experimental https://review.openstack.org/117360 | 21:53 |
| fungi | r1chardj0n3s: yeah, i also point that out in my review comments i think | 21:53 |
| *** andreykurilin has joined #openstack-infra | 21:53 | |
| jeblair | r1chardj0n3s: what code? | 21:53 |
| r1chardj0n3s | fungi: ok | 21:53 |
| r1chardj0n3s | fungi: yep, your comment does capture my concern | 21:54 |
| fungi | jeblair: any code which starts importing a module which isn't in your requirements list | 21:54 |
| clarkb | right my thinking was that its logically a requirements check though | 21:54 |
| clarkb | and we don't need to run those on every change | 21:55 |
| fungi | jeblair: and probably has a chance of getting merged because it passes your unit tests because it's provided by something which you transitively require rather than explicitly require, so it just happens to be present when you get it tested | 21:55 |
| clarkb | the only people that might get bitten by that are CDers who are no better off today | 21:55 |
| r1chardj0n3s | fungi, jeblair, clarkb: sorry, I gotta be a dad for a while, back in about an hour | 21:55 |
| *** esker has quit IRC | 21:55 | |
| *** dims has quit IRC | 21:55 | |
| jeblair | okay, so it's literally a change to $PROJECT's code that causes the test to fail | 21:55 |
| *** ominakov has quit IRC | 21:56 | |
| SergeyLukjanov | jeblair, it's because of removing old code from the sahara-dashboard repo but it was installed by devstack (installation is going ok - sahara dashboard is an empty package now, but devstack code adds it to the horizon config as dashboard with slug 'sahara' and so horizon fails to start w/ non-existing dashboard :( | 21:56 |
| jeblair | because $PROJECT starts directly using an undeclared dependency? | 21:56 |
| *** laca has quit IRC | 21:56 | |
| *** signed8bit has quit IRC | 21:56 | |
| fungi | jeblair: yeah | 21:56 |
| jeblair | SergeyLukjanov: how did the change to remove the code pass the gate? | 21:56 |
| *** oanufriev has quit IRC | 21:56 | |
| jeblair | fungi: okay that makes sense; it does seem like something we should run on every commit then | 21:57 |
| clarkb | jeblair: yup and the only reason it works is because you happen to have that requirement pulled in via a transitive dependency situation | 21:57 |
| jeblair | how does it actually work? | 21:57 |
| jeblair | what's the run-time like? | 21:57 |
| clarkb | jeblair: let me dig up a run of it | 21:57 |
| *** bhuvan_ has joined #openstack-infra | 21:57 | |
| SergeyLukjanov | jeblair, there were no dsvm job on the sahara-dasboard repo because it was targeted to be deprecated, my fail on it :( | 21:57 |
| fungi | jeblair: right. this actually started out life as a conversation i had with r1chardj0n3s at the mid-cycle because some project's gating (i forget which) spontaneously broke when one of their dependencies released a new version which stopped declaring a dependency on something else the project was using | 21:58 |
| jeblair | neato | 21:58 |
| clarkb | jeblair: http://logs.openstack.org/22/114122/4/experimental/gate-nova-pip-missing-reqs/614f918/console.html.gz looks like 10 minutes for nova | 21:58 |
| jeblair | SergeyLukjanov: okay, so there's nothing to do to prevent this in the future, since when this actually lands, dashboard will be gone, right? | 21:58 |
| *** sandywalsh has quit IRC | 21:58 | |
| fungi | so it's a bit of a time-bomb finder. fast too (i think on the order of seconds though sounds like we may have some actual job logs to measure it) | 21:58 |
| SergeyLukjanov | jeblair, yup | 21:58 |
| fungi | oh, 10 minutes? | 21:59 |
| clarkb | according to that run yup | 21:59 |
| fungi | eek, for some reason i thought it was faster | 21:59 |
| SergeyLukjanov | jeblair, it'll be only in stable/icehouse, but it's not included into the gate for I | 21:59 |
| fungi | i bet that's the "install all your crap into the venv" duration mostly | 21:59 |
| jeblair | yeah, | 21:59 |
| *** bhuvan has quit IRC | 21:59 | |
| jeblair | clarkb, fungi: the actual run after install takes 9 seconds | 21:59 |
| *** bhuvan_ is now known as bhuvan | 21:59 | |
| jeblair | so really, we could shove this into the unit test job | 22:00 |
| *** mriedem has quit IRC | 22:00 | |
| jeblair | oh, but now this is where we don't want test-requirements installed? | 22:00 |
| fungi | i was hoping it was something we might be able to integrate into, say, the pep8 runner or the unit test runners, but its venv expectations might clash with theirs | 22:00 |
| clarkb | jeblair: correct so we would have to do a separate tox and install all that stuff | 22:00 |
| clarkb | we don't gain much by running it in the venv of another job | 22:01 |
| jeblair | clarkb: dumb question: why does it require the requirements actually be installed? | 22:01 |
| fungi | i think because it needs to see if they provide the modules being imported? r1chardj0n3s could probably say for certain though | 22:01 |
| clarkb | jeblair: thats actually a good question. I wonder if pip-missing-reqs is using the interpreter to import stuff | 22:02 |
| *** laca has joined #openstack-infra | 22:02 | |
| * fungi looks at the source | 22:02 | |
| fungi | https://github.com/r1chardj0n3s/pip-missing-reqs | 22:02 |
| clarkb | it uses pkg_resources to build a dataset of what is installed | 22:03 |
| *** ianw has quit IRC | 22:04 | |
| *** ianw has joined #openstack-infra | 22:04 | |
| sdague | clarkb: https://review.openstack.org/#/c/106637/ - long standing thing that should change, been +2ed for 6 weeks | 22:05 |
| sdague | 2 +2ed actually :) | 22:05 |
| *** ociuhandu has joined #openstack-infra | 22:05 | |
| SergeyLukjanov | jeblair, both changes has been approved by devstack folks, but it seems like it will break any tempest-running job, so, probably, I need to promote themto avoid resets? | 22:05 |
| *** otherwiseguy has joined #openstack-infra | 22:06 | |
| clarkb | sdague: fungi beat me to it | 22:06 |
| sdague | ok | 22:06 |
| sdague | I don't remember if we still have to babysit deploys on that | 22:06 |
| fungi | oops, sorry. yeah it was a trivial one | 22:06 |
| sdague | otherwise I'd +A | 22:06 |
| jeblair | clarkb, fungi: i'm not seeing any way more preferable than the current approach for missing-reqs :/ | 22:07 |
| jeblair | clarkb, fungi: i think every code change is the right time to run it, and we don't have a similar virtualenv to use | 22:07 |
| fungi | sdague: i think i just need to double-check that it's importing the updated modules once that lands | 22:07 |
| jeblair | SergeyLukjanov: are all integrated gate changes failing, or just sahara changes? | 22:07 |
| SergeyLukjanov | jeblair, it's about all jobs where both sahara and horizon enabled, so, it's integrated gate for master | 22:08 |
| fungi | jeblair: yeah, also, it really needs to be able to be run and switched from non-voting to voting per project as its phased in, so i didn't think it would be particularly effective condensed into any existing job until it was already in use and gating for all projects | 22:08 |
| clarkb | jeblair: fungi: I can abandon my change and we can keep the old one and move it to check/gate | 22:08 |
| jeblair | SergeyLukjanov: then yeah, you should promote them. have you done that before? | 22:09 |
| openstackgerrit | A change was merged to openstack-infra/os-loganalyze: Use /usr/bin/env python instead of /usr/bin/python https://review.openstack.org/106637 | 22:09 |
| fungi | clarkb: i think we should move it to experimental or non-voting check, yeah (when i reviewed the previous change it was still set for experimental--i didn't see the iteration which put it into the silent pipeline until now or i would have said something sooner) | 22:09 |
| SergeyLukjanov | jeblair, yup, I have enough ninja merge/promotion experience | 22:09 |
| jeblair | SergeyLukjanov: cool. i'd say go for it then. | 22:10 |
| *** bhuvan has quit IRC | 22:10 | |
| *** peluse has quit IRC | 22:11 | |
| mattoliverau | Morning, having fun with pbr I see :) | 22:11 |
| *** bnemec has quit IRC | 22:11 | |
| fungi | mattoliverau: i like your definition of "fun." where did you obtain it? | 22:12 |
| *** bnemec has joined #openstack-infra | 22:12 | |
| *** whoops has joined #openstack-infra | 22:12 | |
| *** pradk has quit IRC | 22:12 | |
| SergeyLukjanov | jeblair, ok, thx | 22:12 |
| SergeyLukjanov | jeblair, here is an example of how it failed on non-sahara stuff - http://logs.openstack.org/92/116892/1/gate/gate-tempest-dsvm-full/8513e10/logs/horizon_error.txt.gz | 22:13 |
| *** harlowja_ has joined #openstack-infra | 22:13 | |
| mattoliverau | fungi: I've discovered the only real way to live is at 100% sarcasm :p | 22:13 |
| *** harlowja_ has quit IRC | 22:15 | |
| *** harlowja_ has joined #openstack-infra | 22:15 | |
| *** jergerber has quit IRC | 22:15 | |
| *** harlowja has quit IRC | 22:16 | |
| *** gondoi is now known as zz_gondoi | 22:16 | |
| *** andreykurilin has quit IRC | 22:16 | |
| fungi | dfw seems to be really unhappy. our quota was full of instances stuck in error+deleting state and so phschwartz got ops to clear those. a few hours later, almost back to all error+deleting there again | 22:17 |
| phschwartz | fungi: yup, and ops is working to clear them right now again :( | 22:17 |
| fungi | phschwartz: oof--thanks! | 22:18 |
| fungi | i was also looking at one of the logstash searches for ubuntu package download errors and the majority of them were spotted in dfw | 22:18 |
| fungi | so i suspect it's got some manner of indigestion at the moment | 22:19 |
| *** Sincler has quit IRC | 22:20 | |
| *** reed has joined #openstack-infra | 22:24 | |
| clarkb | give it tums | 22:28 |
| clarkb | fungi: jeblair I am going to self approve https://review.openstack.org/#/c/116176/ today if you guys don't beat me to it :) | 22:28 |
| *** mmedvede has quit IRC | 22:29 | |
| *** oanufriev has joined #openstack-infra | 22:30 | |
| jeblair | OH! | 22:30 |
| fungi | clarkb: i'll approve it | 22:30 |
| jeblair | http://logs.openstack.org/73/116973/3/check/gate-config-puppet-apply-precise/c2af9bc/console.html | 22:30 |
| jeblair | this is really interesting | 22:30 |
| jeblair | that failed because /etc/puppet/modules/puppetboard already exists before the job runs | 22:31 |
| jeblair | because of the install_modules that happens as part of the node bootstropping | 22:31 |
| fungi | ohhhhhhh | 22:31 |
| jeblair | and install_modules does not roccetly update the commit | 22:31 |
| *** Daviey has quit IRC | 22:31 | |
| fungi | i completely misread the previous failure for that and thought it was complaining that the parameter was invalid for the external version of that module | 22:32 |
| fungi | whoops | 22:32 |
| *** sdake_ has quit IRC | 22:32 | |
| fungi | so this would have magically started working tomorrow | 22:32 |
| jeblair | i will attempt a fix | 22:32 |
| jeblair | fungi: yup | 22:32 |
| jeblair | i mean, if the nodepool image gods willed it | 22:33 |
| fungi | i guess we need to not have the job use /etc/puppet/modules for testing module installation | 22:33 |
| jeblair | fungi: well, that's one option. or the other one would be to correctly update the commit | 22:33 |
| fungi | or we blow away the existing /etc/puppet/modules if we want to just dirty-hack it | 22:33 |
| clarkb | or just overwrite it since we won't use the node again | 22:33 |
| clarkb | jeblair: ya that | 22:33 |
| fungi | aha, yep | 22:33 |
| jeblair | fungi: actually, i kind of like the rm -fr approach | 22:34 |
| jeblair | slightly closer to what would happen on a real deploy | 22:34 |
| *** weshay has quit IRC | 22:34 | |
| fungi | yeah, otherwise we might have lingering modules on the filesystem which we removed from the install list | 22:34 |
| *** sballe_ has joined #openstack-infra | 22:35 | |
| jeblair | Your branch is behind 'origin/master' by 1011 commits, and can be fast-forwarded. | 22:35 |
| jeblair | that just happened in one of my config repos :) | 22:35 |
| fungi | bwahahahaha | 22:35 |
| *** tgohad has quit IRC | 22:36 | |
| *** jgrimm is now known as jgrimm_zz | 22:37 | |
| *** sballe has quit IRC | 22:37 | |
| *** jamesmcarthur has quit IRC | 22:37 | |
| *** markmcclain has quit IRC | 22:38 | |
| clarkb | wow did you go on vacation? | 22:38 |
| fungi | that must have been one _busy_ week | 22:38 |
| nibalizer | sounds like a config repo on a little used host or directory | 22:38 |
| openstackgerrit | A change was merged to openstack-infra/config: Propose new Code Review -1 text https://review.openstack.org/116176 | 22:39 |
| *** dizquierdo has joined #openstack-infra | 22:39 | |
| jeblair | it was my gertty config repo (i don't normally use it because i have lots of symlinks in the one i usually use and i haven't bothered to switch them over yet) | 22:40 |
| *** dizquierdo has quit IRC | 22:40 | |
| jeblair | but this time, i wanted to write that change and then "x" the review i currently had displayed on top of it | 22:40 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/config: Disable notifications in storyboard https://review.openstack.org/116973 | 22:40 |
| *** praneshp has quit IRC | 22:40 | |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/config: Remove /etc/puppet/modules on apply test https://review.openstack.org/117365 | 22:40 |
| *** praneshp_ has joined #openstack-infra | 22:40 | |
| jeblair | like so | 22:40 |
| *** marcoemorais has quit IRC | 22:40 | |
| *** eharney has quit IRC | 22:40 | |
| *** marcoemorais has joined #openstack-infra | 22:40 | |
| *** pballand has quit IRC | 22:41 | |
| *** marcoemorais has quit IRC | 22:41 | |
| *** marcoemorais has joined #openstack-infra | 22:41 | |
| *** denis_makogon has quit IRC | 22:42 | |
| *** Daviey has joined #openstack-infra | 22:42 | |
| fungi | jeblair: lgtm unless puppet sticks stdlib there it expects to find. i guess test results will tell us | 22:42 |
| clarkb | yay tests | 22:43 |
| *** krtaylor has joined #openstack-infra | 22:45 | |
| mrodden | have you guys seen an issue syncing from pypi lately? | 22:50 |
| mrodden | TooManyRedirects: Exceeded 30 redirects. | 22:50 |
| mrodden | on one of the package simple pages | 22:50 |
| clarkb | mrodden: that happens occasionally when a project on pypi renames itself | 22:50 |
| clarkb | mrodden: we usually have to ping dstufft to fix it | 22:51 |
| mrodden | ah | 22:51 |
| mrodden | 2014-08-27 17:44:10,035 ERROR: Error syncing package: instant-rst@1204635 | 22:51 |
| mrodden | guessing he'll catch up on the backlog later and find that | 22:53 |
| *** LandenAuer has joined #openstack-infra | 22:55 | |
| fungi | mrodden: fun, in that case we're probably stuck on it too. checking | 22:55 |
| fungi | yep | 22:55 |
| * fungi sighs | 22:55 | |
| fungi | stuck since 17:05 utc | 22:56 |
| *** dimtruck is now known as zz_dimtruck | 22:57 | |
| fungi | killed the stuck bandersnatch process just now to see if it recovers on its own for the next run or whether it's still looping | 22:58 |
| openstackgerrit | Joe Gordon proposed a change to openstack-infra/elastic-recheck: clean up readme https://review.openstack.org/115011 | 23:00 |
| fungi | huh. actually it's fine now... "Resuming interrupted sync from local todo list. [...] 0 packages to sync." | 23:01 |
| *** whoops has quit IRC | 23:02 | |
| *** bhuvan has joined #openstack-infra | 23:02 | |
| clarkb | fungi: does it get stuck in the loop with no way out then when pypi fixes its fine if it can unloop? | 23:02 |
| fungi | clarkb: i can't actually tell. attaching an strace to the innermost pid just shows it in an endless select polling cycle | 23:02 |
| fungi | clarkb: though it does seem to get stuck indefinitely until restarted | 23:03 |
| clarkb | fungi: maybe we should run it in a half hour timeout for now? | 23:04 |
| openstackgerrit | David Caro proposed a change to openstack-infra/jenkins-job-builder: Added parallelization options https://review.openstack.org/75514 | 23:04 |
| fungi | i was pondering that as a hack to not have to worry about it | 23:04 |
| fungi | though it's being called from within a python script, so we'd presumably want to embed the timeout in that | 23:05 |
| clarkb | we can timeout the whole thing | 23:05 |
| clarkb | the python script is the thing run by cron iirc | 23:05 |
| fungi | we can... it could also stand to be run with a lockfile since right now when it hangs you just get a bunch of python tracebacks from failing to start over and over every 5 minutes after that. makes the log a mess | 23:06 |
| fungi | oh, also i see the "too many redirects" now | 23:07 |
| fungi | i forgot you have to wait for the second run | 23:07 |
| fungi | because the next run after it's interrupted only tries to resume the one action it was interrupted in the middle of | 23:07 |
| openstackgerrit | A change was merged to openstack-infra/elastic-recheck: Add query for horizon not registered bug 1362325 https://review.openstack.org/117350 | 23:08 |
| uvirtbot` | Launchpad bug 1362325 in horizon "Dashboard with slug "sahara" is not registered" [Undecided,New] https://launchpad.net/bugs/1362325 | 23:08 |
| fungi | though this time it didn't hang even though it hit a redirect loop, so no clue what the perfect storm of circumstances is to make that manifest | 23:08 |
| r1chardj0n3s | hullo again fungi, clarkb, jeblair, I gather that the consensus is to move back to triggering pip-missing-reqs as a job in its own right as per jhesketh's patch? | 23:09 |
| fungi | r1chardj0n3s: yeah, but not in the silent pipeline | 23:10 |
| *** jamielennox|away is now known as jamielennox | 23:10 | |
| r1chardj0n3s | not sure how it got into the silent pipeline; jhesketh had it in the experimental so I could trigger it while testing the nova change to make sure it worked | 23:10 |
| fungi | that should have either been added originally in experimental or non-voting in check | 23:10 |
| r1chardj0n3s | yep, it was | 23:11 |
| fungi | and yeah, the last time i had reviewed it, it was targeting experimental | 23:11 |
| *** hdd has quit IRC | 23:11 | |
| r1chardj0n3s | the idea being to move it from experimental once I'd made sure it was working correctly, which it was | 23:11 |
| fungi | huh... yeah https://review.openstack.org/114137 did add it in experimental | 23:13 |
| openstackgerrit | Mark Vanderwiel proposed a change to openstack-infra/config: Fix Berkshelf 3.x GeCode 3.x dependency https://review.openstack.org/117274 | 23:13 |
| *** dangers is now known as dangers_away | 23:14 | |
| *** dmsimard is now known as dmsimard_away | 23:14 | |
| r1chardj0n3s | ok, and my nova patch to demonstrate the check works (https://review.openstack.org/#/c/114122/) should be good to go. I guess the only thing left then is to move the check from experimental to gate non-voting? | 23:15 |
| r1chardj0n3s | or voting. *shrug* I'm not sure how that will interact with projects that don't implement the tox env | 23:16 |
| clarkb | fungi: it was experimental | 23:16 |
| clarkb | then it was moved to silent | 23:16 |
| r1chardj0n3s | as I understand it, the check will FAIL if the tox env doesn't exist. that's suboptimal as it means generating noise of false fails and if the env is then implemented and fails people will just ignore it | 23:17 |
| r1chardj0n3s | ideally the check wouldn't run at all if the env doesn't exist | 23:18 |
| clarkb | r1chardj0n3s: it just means we need to merge your change | 23:18 |
| fungi | r1chardj0n3s: clarkb: git show 8891904 in npva | 23:19 |
| fungi | er, nova | 23:19 |
| fungi | nobody intentionally moved it to silent | 23:19 |
| r1chardj0n3s | yep, I've unblocked that changed | 23:19 |
| r1chardj0n3s | just gonna chase up another nova person to +2 it | 23:20 |
| clarkb | fungi: that ref doesn't exist for me | 23:20 |
| fungi | er, not in nova, in openstack-infra/config sorry | 23:20 |
| fungi | rebase/merge fail when moving the docker job to the silent pipeline | 23:20 |
| openstackgerrit | Joe Gordon proposed a change to openstack-infra/config: Cleanup tempest-logs.html https://review.openstack.org/117373 | 23:21 |
| clarkb | fungi: oh wow | 23:21 |
| clarkb | I was just going to ask about the docker job and why it was related | 23:21 |
| fungi | wow, the parent changes there really should have conflicted. comparing those there's only a one line margin for overlapping context between them | 23:21 |
| fungi | it's definitely the merge that screwed it up | 23:22 |
| fungi | i don't think this was human error at all | 23:22 |
| *** Sincler has joined #openstack-infra | 23:22 | |
| fungi | jeblair: http://git.openstack.org/cgit/openstack-infra/config/commit/?id=8891904 | 23:23 |
| clarkb | fungi: huh, do you want to clean it up? I can approve | 23:23 |
| fungi | yep, on it now | 23:23 |
| fungi | though a little worried, after all, that really shouldn't have happened | 23:23 |
| *** tonytan4ever has quit IRC | 23:24 | |
| openstackgerrit | Jeremy Stanley proposed a change to openstack-infra/config: Correct merge calculation error in layout.yaml https://review.openstack.org/117375 | 23:30 |
| fungi | clarkb: jeblair: r1chardj0n3s: ^ | 23:30 |
| *** tomoe_ has joined #openstack-infra | 23:30 | |
| r1chardj0n3s | thanks fungi, a very odd error that one | 23:31 |
| *** gargola has quit IRC | 23:31 | |
| clarkb | fungi: I am going to just approve it | 23:31 |
| fungi | r1chardj0n3s: yeah, proof that git is not infallible | 23:31 |
| fungi | i've seen this sort of thing happen on blind rebases in files with highly repetitive content, but this is the first time i've seen it on a merge commit in an area of a file with minimal repetition | 23:32 |
| fungi | i'm a little worried that neither cgit (in zuul) nor jgit (in gerrit) felt that conflicted | 23:33 |
| *** tomoe_ has quit IRC | 23:34 | |
| openstackgerrit | Spencer Krum proposed a change to openstack-infra/config: Conditionally manage puppetdb connection https://review.openstack.org/117377 | 23:34 |
| clarkb | lifeless: comments on 108270. I didn't -1 because I am fine with it as is but mostly wanted to spark that converstaion | 23:35 |
| *** oomichi has joined #openstack-infra | 23:39 | |
| lifeless | clarkb: thanks | 23:39 |
| lifeless | clarkb: the spec covers the combination of symbols. I am guessing you want https://review.openstack.org/#/c/108270/18/doc/source/index.rst tweaked to include that? | 23:39 |
| clarkb | lifeless: yeah, since that is the thing that will be published and shown to devs | 23:40 |
| clarkb | lifeless: but I think that can be a follow up change | 23:40 |
| *** marcusvrn has quit IRC | 23:41 | |
| *** david-lyle has quit IRC | 23:42 | |
| lifeless | clarkb: on https://review.openstack.org/#/c/108269/ | 23:42 |
| lifeless | clarkb: I've linked the etherpad where I analysed and fixed the versions | 23:42 |
| *** joools has quit IRC | 23:42 | |
| clarkb | lifeless: thanks | 23:42 |
| clarkb | looks like we have a few outstanding fixed | 23:43 |
| clarkb | er *fixes | 23:43 |
| lifeless | clarkb: none that matter | 23:43 |
| lifeless | clarkb: they can all be landed after this lands | 23:44 |
| lifeless | clarkb: the murano ones are on branches where CI fails for other reasons | 23:44 |
| clarkb | lifeless: so I should approve that change then? | 23:44 |
| lifeless | clarkb: the ospurge folk are AWOL - I mailed them directly | 23:44 |
| *** sdake_ has joined #openstack-infra | 23:45 | |
| lifeless | clarkb: the oslo-incubator one failed landing due to my subunit 0.0.20, rechecking it now (and its not run in grenade so can land later) | 23:45 |
| lifeless | clarkb: I think it would be ok to approve it - its not going to cause mass panic | 23:46 |
| *** RicoLin has joined #openstack-infra | 23:50 | |
| clarkb | lifeless: done | 23:52 |
| lifeless | clarkb: danke | 23:53 |
| *** sdake_ has quit IRC | 23:53 | |
| *** gokrokve_ has quit IRC | 23:54 | |
| lifeless | wow, I really can't get my head around hg's model | 23:54 |
| lifeless | bzr and git are so much closer to each other | 23:54 |
| clarkb | lifeless: ++ so much | 23:54 |
| clarkb | every time I use hg its like visiting a foreign country | 23:55 |
| clarkb | you can get around but it isn't always easy | 23:55 |
| lifeless | clarkb: http://bugs.python.org/issue16662 is what I'm poking at atm | 23:55 |
| lifeless | hg phases set draft $uuid | 23:55 |
| lifeless | I mean wtf | 23:55 |
| fungi | wow | 23:55 |
| *** sdake_ has joined #openstack-infra | 23:55 | |
| *** unicell has quit IRC | 23:55 | |
| *** Sincler has quit IRC | 23:56 | |
| clarkb | are we wtf'ing the bug? | 23:56 |
| lifeless | fungi: hg tracks if you've pushed a commit and if you have makes it's position in your branch immutable | 23:56 |
| * fungi has friends who swore by hg, but even they're mostly just giving in and using git now | 23:56 | |
| *** hdd has joined #openstack-infra | 23:56 | |
| *** unicell has joined #openstack-infra | 23:57 | |
| lifeless | clarkb: I am btw doing this to do the fix for zzzeek properly | 23:57 |
| clarkb | lifeless: thats what I thought | 23:57 |
| openstackgerrit | Steve Baker proposed a change to openstack-infra/config: check-heat-dsvm-functional pre/post hooks https://review.openstack.org/116533 | 23:57 |
| clarkb | so that root tests/__init__.py can load all the resources? | 23:57 |
| lifeless | yes | 23:57 |
| lifeless | I'm tired of workarounds :) | 23:57 |
| clarkb | hg branches are not bzr or git branches | 23:58 |
| clarkb | they are immutable things that are included in any merge commits iirc | 23:58 |
| clarkb | so the branch itself is part of history rather than a pointer into history | 23:59 |
| *** dims has joined #openstack-infra | 23:59 | |
| jeblair | sdague, dtroyer: https://review.openstack.org/#/c/112000 | 23:59 |
| jeblair | sdague, dtroyer: apparently now both of the sahara changes need to go in at once because of the gate breakage | 23:59 |
| anteaya | clarkb fungi is my comment on this patch accurate or incorrect? https://review.openstack.org/#/c/117117/1/modules/openstack_project/files/gerrit/acls/stackforge/requests-mock.config | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!