| *** dims has joined #openstack-infra | 00:02 | |
| *** hemna is now known as hemnafk | 00:04 | |
| *** dkehn_ has joined #openstack-infra | 00:05 | |
| *** dkehn has quit IRC | 00:07 | |
| *** matsuhashi has joined #openstack-infra | 00:09 | |
| *** changbl has joined #openstack-infra | 00:12 | |
| *** changbl has quit IRC | 00:23 | |
| hashar | fungi: are pip commands using an internal openstack mirror of pip ? | 00:30 |
|---|---|---|
| clarkb | hashar: it depends where the job is running, but in many cases yet | 00:30 |
| clarkb | s/yet/yes/ | 00:31 |
| hashar | Jenkins Job Builder test-requires has sphinxcontrib-programoutput as a new addition. Some job is apparently unable to install that python module | 00:31 |
| hashar | so I guess I need to get the python module added to the mirror | 00:31 |
| fungi | i don't think jjb's jobs use the mirror... | 00:32 |
| hashar | http://logs.openstack.org/15/45715/3/check/jenkins-job-builder-compare-xml/9cabd2a/console.html tells me """ Could not find any downloads that satisfy the requirement sphinxcontrib-programoutput """ | 00:32 |
| HenryG | Can anyone help me with this dependency error I am getting with devstack? oslo.messaging vs. pbr ... http://paste.openstack.org/show/46399/ | 00:32 |
| *** marun has quit IRC | 00:32 | |
| clarkb | hashar: http://git.openstack.org/cgit/openstack-infra/config/tree/modules/jenkins/files/slave_scripts/select-mirror.sh#n54 should have it fall back on pypi proper | 00:33 |
| clarkb | fungi: JJB does but it should use pypi proper too | 00:33 |
| fungi | aah | 00:33 |
| clarkb | HenryG: your global version of pbr is too old and setup_requires won't upgrade it | 00:33 |
| clarkb | HenryG: if you manually upgrade pbr gobally it should make the error go away | 00:33 |
| hashar | clarkb: or can I add the missing module to the mirror ? | 00:34 |
| clarkb | hashar: no, the change merging should add it to the mirror | 00:34 |
| clarkb | hashar: this is why we allow it to go to pypi, can you link to the job that failed? | 00:35 |
| hashar | https://review.openstack.org/#/c/45715/ the check on patchset 3 | 00:35 |
| hashar | or did you mean console log? that is http://logs.openstack.org/15/45715/3/check/jenkins-job-builder-compare-xml/9cabd2a/console.html | 00:36 |
| HenryG | clarkb: aha, I thought I had done that. But the upgrade is failing because it can't uninstall the old version. | 00:36 |
| clarkb | hashar: now that that change has merged the new requirement should end up on the mirror once the mirror jobs run | 00:37 |
| clarkb | unless | 00:37 |
| HenryG | clarkb: 'pip uninstall pbr' says "No files were found to uninstall." | 00:37 |
| clarkb | fungi: do we run the job for not the openstack mirror? | 00:37 |
| clarkb | HenryG: you may need to manually remove it, old versions of pip are bad at uninstalling stuff | 00:37 |
| fungi | clarkb: no, that mirror is empty | 00:37 |
| clarkb | fungi: so maybe select mirror needs to stop using that mirror until we can build it properly? | 00:38 |
| fungi | assuming you mean the openstack-infra mirror tree | 00:38 |
| fungi | clarkb: is it causing issues? | 00:38 |
| clarkb | oh we fall back on http://git.openstack.org/cgit/openstack-infra/config/tree/modules/jenkins/files/slave_scripts/select-mirror.sh#n53 | 00:39 |
| clarkb | fungi: see hashar's failed job, I htink pypi is failing us | 00:39 |
| hashar | fungi: http://logs.openstack.org/15/45715/3/check/jenkins-job-builder-compare-xml/9cabd2a/console.html :) | 00:39 |
| hashar | fungi: JJB can't find the module sphinxcontrib-programoutput which has been added to tools/test-requires | 00:39 |
| HenryG | clarkb: thanks. It seems that /opt/stack/pbr was the old version. Deleting that dir exposed the new version installed. | 00:40 |
| clarkb | should install https://pypi.python.org/pypi/sphinxcontrib-programoutput/0.8 but pip isn't able to isntall it | 00:40 |
| *** nosnos has joined #openstack-infra | 00:40 | |
| clarkb | hashar: pip can't find the package on pypi | 00:40 |
| clarkb | hashar: I think JJB itself is fine | 00:40 |
| *** thomasm has joined #openstack-infra | 00:40 | |
| hashar | maybe it has hit an outdated pypi mirror | 00:41 |
| fungi | is pypi.p.o broken? cdn fail? | 00:41 |
| clarkb | fungi: the log isn't verbose enough | 00:41 |
| fungi | :/ | 00:41 |
| hashar | at least I managed to install it manually but I have no idea which mirror I have hit :D | 00:41 |
| clarkb | I think just typicaly pypi derpage | 00:41 |
| fungi | tried to recheck? | 00:42 |
| hashar | nop :/ | 00:42 |
| clarkb | the change has already merged | 00:42 |
| fungi | oh | 00:42 |
| clarkb | leading more weight to once in a while derpage | 00:42 |
| hashar | I will submit a new change to find out what happens | 00:42 |
| clarkb | hashar: I wouldn't worry about it too much | 00:42 |
| clarkb | hashar: but yes, give another change a shot and see if this is consistent | 00:43 |
| fungi | well, yeah, probably a fastly cdn cache miss or similar | 00:43 |
| openstackgerrit | Giulio Fidente proposed a change to openstack-infra/devstack-gate: add c-bak to MY_ENABLED_SERVICES https://review.openstack.org/45797 | 00:43 |
| openstackgerrit | Antoine Musso proposed a change to openstack-infra/jenkins-job-builder: can Jenkins / pip install sphinxcontrib-programoutput ?? https://review.openstack.org/45798 | 00:44 |
| fungi | still not finding any obvious common thread between the three mis-cloned repos on zuul | 00:45 |
| hashar | same :( No distributions at all found for sphinxcontrib-programoutput https://jenkins01.openstack.org/job/jenkins-job-builder-compare-xml/56/console | 00:45 |
| clarkb | hashar: it is weird that none of the other jobs are running into that | 00:45 |
| hashar | maybe something is misconfigured in jjb | 00:46 |
| clarkb | oh, I think I know what the issue is | 00:47 |
| clarkb | the compare job is not running select-mirror so if the last job that ran select mirror was for an openstack/ project it won't be allowed to talk to the world | 00:48 |
| *** mriedem has joined #openstack-infra | 00:48 | |
| fungi | ha. that's awesome | 00:48 |
| fungi | this must be jjb's first requirement not in the openstack/* mirror | 00:49 |
| clarkb | I will push a fix | 00:49 |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/config: Select mirror when diffing JJB XML files. https://review.openstack.org/45801 | 00:54 |
| clarkb | fungi: hashar ^ that should do it | 00:54 |
| *** Ryan_Lane has quit IRC | 00:57 | |
| fungi | derp in the commit message? (not that it especially matters) | 00:57 |
| clarkb | fungi: where? | 00:57 |
| *** zehicle has quit IRC | 00:58 | |
| *** zehicle has joined #openstack-infra | 00:58 | |
| *** thomasm has quit IRC | 00:59 | |
| hashar | clarkb: crazy stuff :-] | 00:59 |
| * mordred did not accomplish as much today as he might have wanted | 01:00 | |
| fungi | clarkb: the filenames | 01:00 |
| clarkb | fungi: gah yes | 01:00 |
| clarkb | I will fix because why not | 01:00 |
| fungi | sure | 01:00 |
| clarkb | mordred: me too, but at least the review queue is much smaller and review backups have been kickstarted | 01:00 |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/config: Select mirror when diffing JJB XML files. https://review.openstack.org/45801 | 01:01 |
| *** giulivo has joined #openstack-infra | 01:02 | |
| mordred | clarkb: yay! | 01:04 |
| mordred | clarkb: my outstanding patches list is less than a browser page!!! | 01:05 |
| hashar | I am off, thanks clarkb! | 01:07 |
| *** dkehn_ has quit IRC | 01:07 | |
| clarkb | hashar: have fun in SF | 01:08 |
| fungi | clarkb: what was the reasoning beyond running it in the job shell fragment instead of the script file? we seem to do far more of the latter (everything except hacking and requirements) | 01:08 |
| hashar | beers.each do |beer| { throw_at('hashar') } | 01:08 |
| fungi | er, s/beyond/behind/ | 01:09 |
| clarkb | fungi: because that is what the old jobs did | 01:09 |
| *** sarob has joined #openstack-infra | 01:09 | |
| clarkb | fungi: and each one is different ... | 01:09 |
| fungi | ah | 01:10 |
| *** changbl has joined #openstack-infra | 01:10 | |
| *** dkehn has joined #openstack-infra | 01:10 | |
| *** hashar has quit IRC | 01:11 | |
| fungi | right, passes the project name | 01:15 |
| *** yaguang has joined #openstack-infra | 01:21 | |
| *** ericw has joined #openstack-infra | 01:32 | |
| *** blamar has quit IRC | 01:32 | |
| *** senk has quit IRC | 01:35 | |
| *** dkehn has quit IRC | 01:38 | |
| *** xchu has joined #openstack-infra | 01:38 | |
| *** sarob has quit IRC | 01:38 | |
| *** sarob has joined #openstack-infra | 01:39 | |
| *** fandikurnia02 has quit IRC | 01:40 | |
| *** adalbas has quit IRC | 01:58 | |
| *** vipul has quit IRC | 02:01 | |
| *** vipul has joined #openstack-infra | 02:02 | |
| *** sarob_ has joined #openstack-infra | 02:02 | |
| *** sarob has quit IRC | 02:02 | |
| *** ericw has quit IRC | 02:04 | |
| *** portante|afk is now known as portante | 02:13 | |
| openstackgerrit | Zhongyue Luo proposed a change to openstack-dev/hacking: Add check for deprecated method assert_ https://review.openstack.org/45807 | 02:17 |
| *** sarob has joined #openstack-infra | 02:19 | |
| *** sarob_ has quit IRC | 02:19 | |
| *** zehicle has quit IRC | 02:20 | |
| *** zehicle has joined #openstack-infra | 02:25 | |
| *** senk has joined #openstack-infra | 02:26 | |
| *** sarob has quit IRC | 02:28 | |
| *** sarob has joined #openstack-infra | 02:29 | |
| *** mriedem has quit IRC | 02:29 | |
| *** dims has quit IRC | 02:31 | |
| *** senk has quit IRC | 02:31 | |
| *** sarob has quit IRC | 02:33 | |
| *** wenlock has quit IRC | 02:35 | |
| *** wenlock has joined #openstack-infra | 02:35 | |
| *** pcrews has joined #openstack-infra | 02:43 | |
| *** mestery_ is now known as mestery | 02:44 | |
| *** anteaya has quit IRC | 02:47 | |
| *** melwitt has quit IRC | 02:50 | |
| *** KLevenstein has joined #openstack-infra | 02:50 | |
| *** ghena1986 has joined #openstack-infra | 02:50 | |
| *** ghena1986 has left #openstack-infra | 02:50 | |
| *** zhiyan has joined #openstack-infra | 02:56 | |
| *** zhiyan has left #openstack-infra | 02:58 | |
| *** wenlock has quit IRC | 03:00 | |
| *** dcramer_ has joined #openstack-infra | 03:12 | |
| *** DennyZhang has joined #openstack-infra | 03:30 | |
| *** nati_ueno has quit IRC | 03:32 | |
| *** nati_ueno has joined #openstack-infra | 03:32 | |
| *** nati_ueno has quit IRC | 03:33 | |
| *** nati_ueno has joined #openstack-infra | 03:33 | |
| *** boris-42 has joined #openstack-infra | 04:03 | |
| *** vogxn has joined #openstack-infra | 04:07 | |
| openstackgerrit | Joe Gordon proposed a change to openstack-infra/config: Merge duplicate bugs in recheckwatch https://review.openstack.org/45511 | 04:10 |
| *** boris-42 has quit IRC | 04:10 | |
| *** dkehn has joined #openstack-infra | 04:20 | |
| *** nati_uen_ has joined #openstack-infra | 04:21 | |
| *** dkliban has quit IRC | 04:22 | |
| *** nati_uen_ has quit IRC | 04:22 | |
| *** nati_uen_ has joined #openstack-infra | 04:23 | |
| *** nati_ueno has quit IRC | 04:25 | |
| openstackgerrit | Joe Gordon proposed a change to openstack-infra/config: Remove closed bugs from recheckwatch https://review.openstack.org/45813 | 04:30 |
| jog0 | some tweaks to recheckwatch | 04:30 |
| jog0 | to make life easier during the stabilization period | 04:34 |
| *** SergeyLukjanov has joined #openstack-infra | 04:38 | |
| *** dkehn has quit IRC | 04:47 | |
| *** Ryan_Lane has joined #openstack-infra | 04:48 | |
| *** boris-42 has joined #openstack-infra | 04:50 | |
| *** UtahDave has quit IRC | 04:52 | |
| *** HenryG has quit IRC | 04:53 | |
| openstackgerrit | Alex Gaynor proposed a change to openstack-infra/config: Run the troveclient tests under Pypy https://review.openstack.org/45814 | 04:56 |
| *** dguitarbite has joined #openstack-infra | 05:04 | |
| *** sjing has joined #openstack-infra | 05:05 | |
| *** dkehn has joined #openstack-infra | 05:13 | |
| *** pblaho has joined #openstack-infra | 05:27 | |
| *** nicedice has quit IRC | 05:33 | |
| *** nosnos has quit IRC | 05:35 | |
| *** nosnos has joined #openstack-infra | 05:36 | |
| *** SergeyLukjanov has quit IRC | 05:55 | |
| *** SergeyLukjanov has joined #openstack-infra | 06:04 | |
| *** notmyname has quit IRC | 06:05 | |
| *** Shrews has quit IRC | 06:06 | |
| *** oubiwann has quit IRC | 06:06 | |
| *** Shrews has joined #openstack-infra | 06:06 | |
| *** Guest66487 has quit IRC | 06:06 | |
| *** annegentle_ has quit IRC | 06:06 | |
| *** retr0h has quit IRC | 06:07 | |
| *** ttx has quit IRC | 06:07 | |
| *** notmyname has joined #openstack-infra | 06:08 | |
| *** retr0h has joined #openstack-infra | 06:08 | |
| *** retr0h has joined #openstack-infra | 06:08 | |
| *** mordred has quit IRC | 06:08 | |
| *** stevebaker has quit IRC | 06:09 | |
| *** annegentle has joined #openstack-infra | 06:09 | |
| *** SergeyLukjanov has quit IRC | 06:10 | |
| *** comstud has quit IRC | 06:10 | |
| *** DennyZhang has quit IRC | 06:11 | |
| *** comstud has joined #openstack-infra | 06:11 | |
| *** jlk has quit IRC | 06:12 | |
| *** russellb has quit IRC | 06:12 | |
| *** stevebaker has joined #openstack-infra | 06:13 | |
| *** xchu has quit IRC | 06:13 | |
| *** jlk has joined #openstack-infra | 06:13 | |
| *** jlk has joined #openstack-infra | 06:13 | |
| *** vishy has quit IRC | 06:13 | |
| *** vishy has joined #openstack-infra | 06:14 | |
| *** lillie has joined #openstack-infra | 06:15 | |
| *** lillie is now known as Guest11685 | 06:16 | |
| *** vogxn has quit IRC | 06:16 | |
| *** nijaba has joined #openstack-infra | 06:20 | |
| *** russellb has joined #openstack-infra | 06:21 | |
| *** jhesketh has quit IRC | 06:22 | |
| *** xchu has joined #openstack-infra | 06:26 | |
| *** vogxn has joined #openstack-infra | 06:29 | |
| *** yaguang has quit IRC | 06:29 | |
| *** odyssey4me has joined #openstack-infra | 06:30 | |
| *** oubiwann has joined #openstack-infra | 06:42 | |
| *** afazekas_zz is now known as afazekas | 06:51 | |
| *** yaguang has joined #openstack-infra | 06:53 | |
| *** ttx has joined #openstack-infra | 07:09 | |
| *** ttx has quit IRC | 07:09 | |
| *** ttx has joined #openstack-infra | 07:09 | |
| *** jhesketh_ has quit IRC | 07:11 | |
| *** flaper87|afk is now known as flaper87 | 07:14 | |
| *** vogxn has quit IRC | 07:21 | |
| *** shardy_afk is now known as shardy | 07:22 | |
| *** afazekas has quit IRC | 07:24 | |
| *** fbo_away is now known as fbo | 07:33 | |
| *** alexpilotti has joined #openstack-infra | 07:39 | |
| *** Bada has joined #openstack-infra | 07:41 | |
| *** Ryan_Lane has quit IRC | 07:44 | |
| *** nosnos has quit IRC | 07:52 | |
| *** xchu has quit IRC | 07:52 | |
| *** tstevenson has quit IRC | 07:55 | |
| *** vogxn has joined #openstack-infra | 07:55 | |
| *** amotoki has joined #openstack-infra | 08:00 | |
| *** afazekas has joined #openstack-infra | 08:01 | |
| *** jpich has joined #openstack-infra | 08:02 | |
| saschpe_ | zul: hi, any update on python-lesscpy packaging? | 08:02 |
| *** mordred has joined #openstack-infra | 08:05 | |
| *** xchu has joined #openstack-infra | 08:06 | |
| *** KLevenstein has quit IRC | 08:08 | |
| ttx | jog0: for testing launchpad bugs you can use http://staging.launchpad.net/ | 08:11 |
| ttx | jog0: can be a bit slow but at least it does not pollute the main db | 08:12 |
| ttx | (context: https://bugs.launchpad.net/openstack-ci/+bug/1223159) | 08:12 |
| uvirtbot | Launchpad bug 1223159 in openstack-ci "fake bug 2" [Undecided,New] | 08:12 |
| *** Ryan_Lane has joined #openstack-infra | 08:14 | |
| *** Ryan_Lane has quit IRC | 08:21 | |
| *** derekh has joined #openstack-infra | 08:33 | |
| *** SergeyLukjanov has joined #openstack-infra | 08:37 | |
| *** sjing has quit IRC | 08:47 | |
| *** odyssey4me has quit IRC | 08:51 | |
| *** mgagne1 has joined #openstack-infra | 09:02 | |
| *** mgagne1 has joined #openstack-infra | 09:02 | |
| *** afazekas_ has joined #openstack-infra | 09:07 | |
| *** jhesketh_ has joined #openstack-infra | 09:11 | |
| *** afazekas has quit IRC | 09:11 | |
| *** amotoki has quit IRC | 09:11 | |
| *** pabelanger has quit IRC | 09:11 | |
| *** mgagne has quit IRC | 09:11 | |
| *** jpeeler has quit IRC | 09:11 | |
| *** mtreinish has quit IRC | 09:11 | |
| *** markmcclain has quit IRC | 09:11 | |
| *** comstud has quit IRC | 09:11 | |
| *** dcramer_ has quit IRC | 09:11 | |
| *** dragondm has quit IRC | 09:11 | |
| *** SpamapS has quit IRC | 09:11 | |
| *** roaet has quit IRC | 09:11 | |
| *** saper has quit IRC | 09:11 | |
| *** carlp has quit IRC | 09:11 | |
| *** EntropyWorks has quit IRC | 09:11 | |
| *** jesusaurus has quit IRC | 09:11 | |
| *** danger_fo_away has quit IRC | 09:11 | |
| *** rainya has quit IRC | 09:11 | |
| *** derekh has quit IRC | 09:11 | |
| *** nijaba has quit IRC | 09:11 | |
| *** pblaho has quit IRC | 09:11 | |
| *** zehicle has quit IRC | 09:11 | |
| *** soren has quit IRC | 09:11 | |
| *** david-lyle has quit IRC | 09:11 | |
| *** ianw has quit IRC | 09:11 | |
| *** LinuxJedi has quit IRC | 09:11 | |
| *** locke105 has quit IRC | 09:11 | |
| *** koolhead17 has quit IRC | 09:11 | |
| *** pleia2 has quit IRC | 09:11 | |
| *** matsuhashi has quit IRC | 09:11 | |
| *** jcooley has quit IRC | 09:11 | |
| *** jeremyb has quit IRC | 09:11 | |
| *** rockstar has quit IRC | 09:11 | |
| *** zaro has quit IRC | 09:11 | |
| *** chmouel has quit IRC | 09:11 | |
| *** yaguang has quit IRC | 09:11 | |
| *** nati_uen_ has quit IRC | 09:11 | |
| *** pcrews has quit IRC | 09:11 | |
| *** mestery has quit IRC | 09:11 | |
| *** salv-orlando has quit IRC | 09:11 | |
| *** sdague has quit IRC | 09:11 | |
| *** saschpe_ has quit IRC | 09:11 | |
| *** niska has quit IRC | 09:11 | |
| *** briancline has quit IRC | 09:11 | |
| *** samalba has quit IRC | 09:11 | |
| *** paul-- has quit IRC | 09:11 | |
| *** uvirtbot has quit IRC | 09:15 | |
| *** uvirtbot has joined #openstack-infra | 09:15 | |
| *** afazekas_ is now known as afazekas | 09:15 | |
| *** jhesketh__ has joined #openstack-infra | 09:16 | |
| *** pabelanger_ has joined #openstack-infra | 09:17 | |
| *** yaguang has joined #openstack-infra | 09:17 | |
| *** nati_uen_ has joined #openstack-infra | 09:17 | |
| *** pcrews has joined #openstack-infra | 09:17 | |
| *** mestery has joined #openstack-infra | 09:17 | |
| *** salv-orlando has joined #openstack-infra | 09:17 | |
| *** sdague has joined #openstack-infra | 09:17 | |
| *** saschpe_ has joined #openstack-infra | 09:17 | |
| *** niska has joined #openstack-infra | 09:17 | |
| *** briancline has joined #openstack-infra | 09:17 | |
| *** samalba has joined #openstack-infra | 09:17 | |
| *** Ryan_Lane has joined #openstack-infra | 09:19 | |
| *** jpeeler has joined #openstack-infra | 09:20 | |
| *** mtreinish has joined #openstack-infra | 09:20 | |
| *** derekh has joined #openstack-infra | 09:20 | |
| *** nijaba has joined #openstack-infra | 09:20 | |
| *** comstud has joined #openstack-infra | 09:20 | |
| *** pblaho has joined #openstack-infra | 09:20 | |
| *** dcramer_ has joined #openstack-infra | 09:20 | |
| *** zehicle has joined #openstack-infra | 09:20 | |
| *** matsuhashi has joined #openstack-infra | 09:20 | |
| *** rainya has joined #openstack-infra | 09:20 | |
| *** danger_fo_away has joined #openstack-infra | 09:20 | |
| *** jesusaurus has joined #openstack-infra | 09:20 | |
| *** EntropyWorks has joined #openstack-infra | 09:20 | |
| *** carlp has joined #openstack-infra | 09:20 | |
| *** saper has joined #openstack-infra | 09:20 | |
| *** roaet has joined #openstack-infra | 09:20 | |
| *** SpamapS has joined #openstack-infra | 09:20 | |
| *** dragondm has joined #openstack-infra | 09:20 | |
| *** pleia2 has joined #openstack-infra | 09:20 | |
| *** koolhead17 has joined #openstack-infra | 09:20 | |
| *** locke105 has joined #openstack-infra | 09:20 | |
| *** LinuxJedi has joined #openstack-infra | 09:20 | |
| *** ianw has joined #openstack-infra | 09:20 | |
| *** david-lyle has joined #openstack-infra | 09:20 | |
| *** soren has joined #openstack-infra | 09:20 | |
| *** chmouel has joined #openstack-infra | 09:20 | |
| *** zaro has joined #openstack-infra | 09:20 | |
| *** rockstar has joined #openstack-infra | 09:20 | |
| *** jeremyb has joined #openstack-infra | 09:20 | |
| *** jcooley has joined #openstack-infra | 09:20 | |
| *** paul-- has joined #openstack-infra | 09:20 | |
| *** amotoki has joined #openstack-infra | 09:20 | |
| openstackgerrit | Thierry Carrez proposed a change to openstack-infra/odsreg: Expose Comments in admin site https://review.openstack.org/45832 | 09:23 |
| openstackgerrit | A change was merged to openstack-infra/odsreg: Expose Comments in admin site https://review.openstack.org/45832 | 09:24 |
| *** mkerrin has joined #openstack-infra | 09:24 | |
| *** Ryan_Lane has quit IRC | 09:24 | |
| *** vogxn has quit IRC | 09:29 | |
| *** matsuhashi has quit IRC | 09:34 | |
| *** matsuhashi has joined #openstack-infra | 09:36 | |
| *** SergeyLukjanov has quit IRC | 09:45 | |
| *** vogxn has joined #openstack-infra | 09:45 | |
| *** Ryan_Lane has joined #openstack-infra | 09:49 | |
| *** dguitarbite has quit IRC | 09:50 | |
| *** Ryan_Lane has quit IRC | 09:54 | |
| *** lifeless has quit IRC | 09:56 | |
| *** xchu has quit IRC | 09:56 | |
| *** pcm_ has joined #openstack-infra | 09:59 | |
| *** pcm_ has quit IRC | 10:01 | |
| *** afazekas is now known as afazekas_food | 10:01 | |
| *** pcm_ has joined #openstack-infra | 10:01 | |
| *** matsuhashi has quit IRC | 10:03 | |
| *** vogxn has quit IRC | 10:12 | |
| *** vogxn has joined #openstack-infra | 10:14 | |
| *** Ryan_Lane has joined #openstack-infra | 10:20 | |
| *** Ryan_Lane has quit IRC | 10:25 | |
| *** salv-orlando has quit IRC | 10:30 | |
| *** dizquierdo has joined #openstack-infra | 10:43 | |
| *** yaguang has quit IRC | 10:46 | |
| *** rfolco has quit IRC | 10:48 | |
| *** SergeyLukjanov has joined #openstack-infra | 10:48 | |
| *** Ryan_Lane has joined #openstack-infra | 10:50 | |
| *** vogxn has quit IRC | 10:54 | |
| *** Ryan_Lane has quit IRC | 10:55 | |
| *** ArxCruz has quit IRC | 10:56 | |
| *** vogxn has joined #openstack-infra | 10:58 | |
| *** lcestari has joined #openstack-infra | 11:03 | |
| openstackgerrit | Zhongyue Luo proposed a change to openstack-dev/hacking: Add check for equality operator use for singleton comparsion https://review.openstack.org/45824 | 11:12 |
| *** DennyZhang has joined #openstack-infra | 11:12 | |
| *** lifeless has joined #openstack-infra | 11:15 | |
| *** pblaho has quit IRC | 11:16 | |
| *** adalbas has joined #openstack-infra | 11:20 | |
| *** tian has quit IRC | 11:21 | |
| *** vogxn has quit IRC | 11:29 | |
| *** pblaho has joined #openstack-infra | 11:35 | |
| *** ArxCruz has joined #openstack-infra | 11:44 | |
| *** weshay has joined #openstack-infra | 11:45 | |
| *** DennyZhang has quit IRC | 11:46 | |
| *** shadower has quit IRC | 11:47 | |
| *** afazekas has joined #openstack-infra | 11:48 | |
| *** markmcclain has joined #openstack-infra | 11:52 | |
| *** rfolco has joined #openstack-infra | 11:58 | |
| *** yaguang has joined #openstack-infra | 12:00 | |
| *** sandywalsh has joined #openstack-infra | 12:13 | |
| *** dcramer_ has quit IRC | 12:15 | |
| *** zeus has joined #openstack-infra | 12:15 | |
| *** thomasm has joined #openstack-infra | 12:23 | |
| *** sridevi has joined #openstack-infra | 12:25 | |
| *** kaushikc has joined #openstack-infra | 12:26 | |
| sridevi | Hi, could someone help me with https://review.openstack.org/#/c/45822/ | 12:26 |
| sridevi | failure | 12:26 |
| *** dcramer_ has joined #openstack-infra | 12:27 | |
| openstackgerrit | Julien Danjou proposed a change to openstack-infra/config: Run Tempest tests against Ceilometer w/ MongoDB https://review.openstack.org/45862 | 12:33 |
| *** salv-orlando has joined #openstack-infra | 12:34 | |
| *** kaushikc has quit IRC | 12:37 | |
| openstackgerrit | Julien Danjou proposed a change to openstack-infra/config: Run Tempest tests against Ceilometer w/ MongoDB https://review.openstack.org/45862 | 12:37 |
| *** sridevi has quit IRC | 12:38 | |
| *** markmcclain has quit IRC | 12:39 | |
| *** weshay has quit IRC | 12:40 | |
| *** yaguang has quit IRC | 12:55 | |
| *** dprince has joined #openstack-infra | 12:55 | |
| *** weshay has joined #openstack-infra | 12:55 | |
| *** anteaya has joined #openstack-infra | 13:03 | |
| *** ericw has joined #openstack-infra | 13:04 | |
| *** whayutin_ has joined #openstack-infra | 13:06 | |
| zul | saschpe_: should be this week | 13:07 |
| saschpe_ | zul: that'd be great, thanks | 13:07 |
| *** KLevenstein has joined #openstack-infra | 13:08 | |
| *** weshay has quit IRC | 13:08 | |
| *** dims has joined #openstack-infra | 13:10 | |
| *** mriedem has joined #openstack-infra | 13:13 | |
| *** boris-42 has quit IRC | 13:18 | |
| *** boris-42 has joined #openstack-infra | 13:19 | |
| *** dkliban has joined #openstack-infra | 13:22 | |
| *** rfolco has quit IRC | 13:23 | |
| *** yaguang has joined #openstack-infra | 13:37 | |
| *** burt has joined #openstack-infra | 13:43 | |
| *** rfolco has joined #openstack-infra | 13:45 | |
| *** cppcabrera has joined #openstack-infra | 13:46 | |
| *** thedodd has joined #openstack-infra | 13:47 | |
| *** prad has joined #openstack-infra | 13:49 | |
| *** vogxn has joined #openstack-infra | 13:51 | |
| *** dkehn has quit IRC | 14:01 | |
| *** dkehn has joined #openstack-infra | 14:01 | |
| *** tstevenson has joined #openstack-infra | 14:02 | |
| *** vogxn has quit IRC | 14:02 | |
| *** zeus has quit IRC | 14:02 | |
| *** nati_uen_ has quit IRC | 14:05 | |
| *** erfanian has quit IRC | 14:08 | |
| *** mrodden has joined #openstack-infra | 14:11 | |
| *** rnirmal has joined #openstack-infra | 14:12 | |
| *** erfanian has joined #openstack-infra | 14:15 | |
| *** markmcclain has joined #openstack-infra | 14:17 | |
| sdague | what would people think about a new recheck tag for freshness of a patch? | 14:24 |
| sdague | given our rate of change, I find that after a week nothing is mergable, so before reviewing code I like to actually make sure the patch is going to still merge in jenkins before doing so | 14:24 |
| sdague | I've been doing a lot of recheck no bug on old patches, but that seems a little overuse of that tag | 14:25 |
| *** kaushikc has joined #openstack-infra | 14:26 | |
| *** dcramer_ has quit IRC | 14:30 | |
| anteaya | sdague: what would the new recheck tag be called? | 14:30 |
| *** jhesketh_ has quit IRC | 14:33 | |
| *** vogxn has joined #openstack-infra | 14:33 | |
| *** kaushikc has quit IRC | 14:35 | |
| fungi | i'm not seeing how that's overuse of "recheck no bug" | 14:37 |
| fungi | you're rechecking, and there's no bug... unless the goal is to make it clear to other reviewers why you're rechecking it | 14:38 |
| fungi | i agree leaving a second comment explaining the non-bug reason for the recheck is extra makework. how about extending the no bug syntax to allow an optional comment after? something like "recheck no bug: just want to see if this is still working" | 14:39 |
| *** marun has joined #openstack-infra | 14:39 | |
| fungi | and not track the comments or anything, just match the no bug pattern on "recheck no bug(:.+)?" | 14:40 |
| fungi | or rather, "^recheck no bug(:.+)?$" | 14:41 |
| *** ttx has quit IRC | 14:42 | |
| *** oubiwann has quit IRC | 14:43 | |
| *** dcramer_ has joined #openstack-infra | 14:43 | |
| jeblair | that's one of the actual reasons for recheck no bug. also, fun zuul project: https://bugs.launchpad.net/zuul/+bug/1079268 | 14:45 |
| uvirtbot | Launchpad bug 1079268 in zuul "periodic recheck of reviews" [Wishlist,Triaged] | 14:45 |
| *** mriedem has quit IRC | 14:45 | |
| sdague | fungi: yeh, optional comment would be nice | 14:47 |
| *** mriedem has joined #openstack-infra | 14:48 | |
| jeblair | part of the reason for keeping it simple (and not allowing, say, a full sentence) was to avoid accidental triggers. if you extend the syntax (which is already confusing enough for people), please keep that in mind. | 14:49 |
| *** oubiwann has joined #openstack-infra | 14:49 | |
| openstackgerrit | Alex Gaynor proposed a change to openstack-infra/config: Run the troveclient tests under PyPy https://review.openstack.org/45814 | 14:50 |
| jeblair | BobBall: can you attend the infra team meeting today (at 1900 utc)? | 14:50 |
| *** dkranz has joined #openstack-infra | 14:51 | |
| *** UtahDave has joined #openstack-infra | 14:52 | |
| anteaya | jeblair: do you want a salt category added to today's meeeting agenda? | 14:55 |
| *** KLevenstein has quit IRC | 14:55 | |
| jeblair | anteaya: feel free | 14:56 |
| anteaya | okay thanks | 14:56 |
| *** nati_ueno has joined #openstack-infra | 14:58 | |
| *** dkranz has quit IRC | 14:58 | |
| *** dcramer_ has quit IRC | 15:01 | |
| *** pycabrera has joined #openstack-infra | 15:02 | |
| anteaya | so is today bug day for -infra? | 15:05 |
| anteaya | if so, I'll do what I can to help, if not I will keep working on puppet-dashboard | 15:06 |
| *** cppcabrera has quit IRC | 15:06 | |
| jeblair | anteaya: yes, at 1700 utc | 15:06 |
| anteaya | ah okay | 15:07 |
| *** pycabrera is now known as cppcabrera | 15:07 | |
| *** mrodden has quit IRC | 15:09 | |
| *** ttx has joined #openstack-infra | 15:10 | |
| *** ttx has quit IRC | 15:10 | |
| *** ttx has joined #openstack-infra | 15:10 | |
| *** vogxn has quit IRC | 15:10 | |
| *** yaguang has quit IRC | 15:14 | |
| *** nati_ueno has quit IRC | 15:14 | |
| *** dkranz has joined #openstack-infra | 15:15 | |
| *** pblaho has quit IRC | 15:16 | |
| fungi | i definitely want to go over the salt stuff too--can we try to get to it earlyish in the agenda? i'm probably going to have to bail at about 1945z so i can get to red hat and find parking (our monthly local opestack meetup landed on a tuesday this time) | 15:17 |
| *** mrodden has joined #openstack-infra | 15:20 | |
| *** marun has quit IRC | 15:20 | |
| *** mrmartin has joined #openstack-infra | 15:22 | |
| *** marun has joined #openstack-infra | 15:25 | |
| *** sridevi has joined #openstack-infra | 15:29 | |
| *** Bada has quit IRC | 15:30 | |
| *** nati_ueno has joined #openstack-infra | 15:31 | |
| *** nati_ueno has quit IRC | 15:31 | |
| *** vogxn has joined #openstack-infra | 15:32 | |
| *** dkehn has quit IRC | 15:34 | |
| *** dcramer_ has joined #openstack-infra | 15:37 | |
| *** KLevenstein has joined #openstack-infra | 15:39 | |
| jd__ | fungi: where should I look to see that the precise-node are using Ubuntu Cloud Archive? | 15:39 |
| *** KLevenstein has quit IRC | 15:40 | |
| fungi | jd__: i'd git grep in openstack-infra/config for the url, as a start | 15:40 |
| *** KLevenstein has joined #openstack-infra | 15:40 | |
| jd__ | fungi: tried that already :) | 15:40 |
| jd__ | I didn't find anything about sources.list handling actually | 15:41 |
| *** jhesketh__ has quit IRC | 15:41 | |
| fungi | jd__: i just checked directly on one of the precise slaves with 'grep -ir cloud /etc/apt' and found nothing | 15:42 |
| *** pabelanger_ has quit IRC | 15:42 | |
| jd__ | fungi: damn, thanks, I thought it was already in use for whatever reason | 15:42 |
| *** pabelanger has joined #openstack-infra | 15:42 | |
| *** dprince has quit IRC | 15:43 | |
| fungi | was pretty sure that never got implemented because originally we were talking about moving jobs to quantal, which didn't have cloud archive support, but should resurrect the cloud archive work now that we decided to stick with precise until the next lts | 15:43 |
| jd__ | makes sense | 15:43 |
| fungi | especially now that we have python 3.3 and pypy solutions on precise | 15:44 |
| jeblair | it shouldn't be too hard to do, i think we already have some 'apt::repo' usage puppet | 15:44 |
| * fungi agrees | 15:44 | |
| *** rfolco has quit IRC | 15:45 | |
| jd__ | I'm looking into it | 15:45 |
| *** sridevi has quit IRC | 15:50 | |
| openstackgerrit | Julien Danjou proposed a change to openstack-infra/config: Use Ubuntu Cloud Archive on precise nodes https://review.openstack.org/45896 | 15:53 |
| openstackgerrit | Julien Danjou proposed a change to openstack-infra/config: Run Tempest tests against Ceilometer w/ MongoDB https://review.openstack.org/45862 | 15:53 |
| jd__ | my try at it :) | 15:54 |
| anteaya | fungi: I had put salt at the end of the agenda just before open discussion since I thought it was etiquette to append new items to the end | 15:55 |
| jeblair | i'm given to understand the chair has a habit of reordering items as warranted. | 15:55 |
| fungi | i have in fact seen the chair do that | 15:56 |
| anteaya | go chair | 15:56 |
| zaro | morning | 15:56 |
| anteaya | morning zaro | 15:56 |
| *** nati_ueno has joined #openstack-infra | 15:57 | |
| anteaya | can puppet-dashboard connect to a puppet client or does it have to connect to a puppet master? | 16:02 |
| * mordred reorders the chair | 16:02 | |
| *** reed has joined #openstack-infra | 16:03 | |
| *** hashar has joined #openstack-infra | 16:04 | |
| *** gyee has joined #openstack-infra | 16:04 | |
| anteaya | morning mordred | 16:04 |
| openstackgerrit | A change was merged to openstack-infra/nodepool: Add image-update command https://review.openstack.org/44920 | 16:04 |
| openstackgerrit | A change was merged to openstack-infra/nodepool: Add alien-list command https://review.openstack.org/45448 | 16:05 |
| *** dkehn has joined #openstack-infra | 16:05 | |
| *** thedodd has quit IRC | 16:05 | |
| openstackgerrit | A change was merged to openstack-infra/config: Update nodepool init script to nodepoold https://review.openstack.org/44557 | 16:06 |
| openstackgerrit | David Boucha proposed a change to openstack-infra/config: Give Jenkins the ability to trigger puppet runs https://review.openstack.org/45898 | 16:06 |
| openstackgerrit | A change was merged to openstack-infra/nodepool: Add alien-image-list command https://review.openstack.org/45449 | 16:07 |
| *** vogxn has quit IRC | 16:07 | |
| openstackgerrit | A change was merged to openstack-infra/nodepool: Add a hold command https://review.openstack.org/45453 | 16:07 |
| openstackgerrit | A change was merged to openstack-infra/nodepool: Add a delete command https://review.openstack.org/45454 | 16:07 |
| openstackgerrit | A change was merged to openstack-infra/nodepool: Add image-delete command https://review.openstack.org/45455 | 16:07 |
| pleia2 | anteaya: I | 16:07 |
| anteaya | morning pleia2 | 16:07 |
| pleia2 | anteaya: I'll take a closer look, but I believe each node reports in to the dashboard (not mater) | 16:07 |
| anteaya | okay the documentation I am reading leads me to believe that dashboard fetches the info, so if the client nodes report it, I would like to understand that | 16:08 |
| zaro | pretty busy meeting agenda today. | 16:08 |
| pleia2 | bug day in an hour too :) | 16:08 |
| * pleia2 prepping etherpad | 16:09 | |
| pleia2 | anteaya: oh ok, maybe it's the other way around then :) | 16:09 |
| zaro | pleia2: is there an alert system for bug day? i can never seem to remember the schedule. | 16:09 |
| pleia2 | dashboard goes to nodes | 16:09 |
| anteaya | okay | 16:09 |
| *** pycabrera has joined #openstack-infra | 16:09 | |
| openstackgerrit | A change was merged to openstack-infra/config: Select mirror when diffing JJB XML files. https://review.openstack.org/45801 | 16:09 |
| pleia2 | zaro: in the weekly meeting announcement | 16:09 |
| pleia2 | I also mentioned it at the meeting last week ;) | 16:10 |
| zaro | pleia2: thx. i'll remember to read it going forward. i thought it was just boiler plate reminder. | 16:10 |
| *** markmcclain has quit IRC | 16:10 | |
| pleia2 | yeah, I probably should have modified the subject too | 16:10 |
| *** vogxn has joined #openstack-infra | 16:10 | |
| zaro | pleia2: missed last meeting. | 16:10 |
| zaro | pleia2: imo, would be better as a seperate email. | 16:11 |
| *** pycabrera is now known as cppcabrera_afk | 16:12 | |
| *** cppcabrera has quit IRC | 16:12 | |
| *** dkehn has quit IRC | 16:12 | |
| pleia2 | zaro: ok :) | 16:12 |
| *** nati_uen_ has joined #openstack-infra | 16:13 | |
| *** SergeyLukjanov has quit IRC | 16:15 | |
| *** dkranz has quit IRC | 16:15 | |
| *** nati_ueno has quit IRC | 16:16 | |
| *** openstackstatus has quit IRC | 16:18 | |
| anteaya | it got buried in the avalanche but here is UtahDave's inital offering for a salt patch: https://review.openstack.org/#/c/45898/1 | 16:18 |
| mordred | I was just looking at that | 16:19 |
| mordred | it seems so simple | 16:19 |
| anteaya | yay | 16:19 |
| *** rev has joined #openstack-infra | 16:19 | |
| anteaya | when you know what button to push, yeah | 16:19 |
| fungi | pinch of salt | 16:21 |
| UtahDave | fungi: I didn't include the change to enable the salt-master and salt-minion services | 16:22 |
| fungi | UtahDave: cool. i can manually test this first once i'm done reading through it | 16:22 |
| UtahDave | fungi: that would be great! Let me know if you have any questions or if you think something could be done in a better way. | 16:23 |
| fungi | UtahDave: reenabling should be discussed on #openstack-meeting between 19:00-20:00 utc, to confirm we're comfortable that the runaway memory utilization and process build-up is ironed out in a later release than we were running | 16:23 |
| *** dachary has joined #openstack-infra | 16:24 | |
| fungi | i went digging in the release notes a little but didn't spot right off which version fixes it | 16:24 |
| *** dkehn has joined #openstack-infra | 16:25 | |
| jeblair | mordred: do you have a few minutes to bring django-openstack-auth up to standards? https://review.openstack.org/#/c/45655/ | 16:28 |
| *** cppcabrera has joined #openstack-infra | 16:28 | |
| openstackgerrit | A change was merged to openstack-infra/config: link to jjb cache documentation in the handbook . https://review.openstack.org/45738 | 16:28 |
| *** dkranz has joined #openstack-infra | 16:28 | |
| *** cppcabrera_afk has quit IRC | 16:28 | |
| jeblair | mordred: (those are non-voting; it's not urgent; just something we should do at some point, and i know it's your favorite thing to make projects conform) | 16:29 |
| *** cppcabrera is now known as cppcabrera_afk | 16:29 | |
| *** vogxn has quit IRC | 16:31 | |
| openstackgerrit | A change was merged to openstack-infra/config: link git sha in commit comments to the gerrit review https://review.openstack.org/45464 | 16:35 |
| *** nhm has joined #openstack-infra | 16:36 | |
| jog0 | how can I make hacking-pypy non-gating? | 16:37 |
| clarkb | morning | 16:37 |
| jog0 | as it doesn't work | 16:37 |
| jog0 | morning clarkb | 16:37 |
| clarkb | jog0: in the layout.yaml file make it non voting under the jobs sections near the top of the file | 16:37 |
| anteaya | morning clarkb | 16:37 |
| clarkb | jog0: should be a lot of examples of that in that section of the file | 16:37 |
| jeblair | jog0: though we should ask Alex_Gaynor about that | 16:37 |
| clarkb | jeblair: jog0 yes he said it was working | 16:38 |
| jog0 | https://review.openstack.org/#/c/45761/ | 16:38 |
| Alex_Gaynor | hacking tests totally passed for me under PyPy, can you link a failign run? | 16:38 |
| jeblair | (we wouldn't have merged it if he hadn't tested it) | 16:38 |
| mordred | jeblair: I'll take a look | 16:38 |
| jog0 | in general I don't like going from no test to gating without a non-voting phase to confirm | 16:38 |
| *** nicedice has joined #openstack-infra | 16:38 | |
| Alex_Gaynor | (looking) | 16:39 |
| *** yolanda has joined #openstack-infra | 16:40 | |
| *** fbo is now known as fbo_away | 16:41 | |
| yolanda | hi, i have a question about zuul/jenkins communication. I'm finding that zuul tries to access jenkins as anonymous user, that can be a problem with security when defining roles. Is there any way to define some user/password to access jenkins with a named user? | 16:41 |
| jeblair | yolanda: what version of zuul? | 16:41 |
| *** SergeyLukjanov has joined #openstack-infra | 16:43 | |
| openstackgerrit | Joe Gordon proposed a change to openstack-infra/config: Make gate-hacking-pypy non-gating https://review.openstack.org/45905 | 16:43 |
| jog0 | Alex_Gaynor clarkb ^ | 16:43 |
| jog0 | I propose we make this non-gating for now until we can resolve it | 16:43 |
| Alex_Gaynor | jog0: so pypy totally passes locally for me :/ | 16:43 |
| yolanda | jeblair, i'm just grabbing the last one from https://github.com/openstack-infra/zuul.git | 16:44 |
| jog0 | even if resolve means it works in 2 hours | 16:44 |
| jog0 | Alex_Gaynor: I hate when that happens | 16:44 |
| jeblair | yolanda: so you're using the gearman plugin? | 16:44 |
| Alex_Gaynor | jog0: maybe try a recheck no bug? | 16:44 |
| yolanda | jeblair, yes | 16:44 |
| jog0 | Alex_Gaynor: it failed 2x in a row | 16:44 |
| jog0 | so don't think that will help | 16:44 |
| *** nati_uen_ has quit IRC | 16:44 | |
| Alex_Gaynor | jog0: grumble, sounds like a thing then :/ | 16:45 |
| clarkb | jog0: lgtm | 16:45 |
| jog0 | Alex_Gaynor: I am running on https://review.openstack.org/#/c/45824/ just to confirm its not my patch somehow | 16:45 |
| jeblair | yolanda: we use the project-based matrix authorization strategy, and they only perms granted to anonymous are "overall-read" and "job-read" | 16:46 |
| jeblair | s/they/the/ | 16:46 |
| openstackgerrit | A change was merged to openstack-infra/config: Make gate-hacking-pypy non-gating https://review.openstack.org/45905 | 16:47 |
| yolanda | jeblair, that's enough to be able to trigger jobs and retrieve results? | 16:47 |
| jog0 | Alex_Gaynor: did you try running the tests locally for https://review.openstack.org/#/c/45761/1 | 16:47 |
| Alex_Gaynor | jog0: no, let me do that | 16:47 |
| jeblair | yolanda: i believe so because the gerrit-plugin bypasses the authorization system | 16:47 |
| jeblair | er | 16:47 |
| jeblair | gearman-plugin | 16:47 |
| clarkb | gearman-plugin escalates itself iirc | 16:47 |
| anteaya | wow, awesome work on the etherpad round-up, pleia2 | 16:47 |
| clarkb | oh yes bug day | 16:48 |
| Alex_Gaynor | jog0: hmm, results in a "VersionConflict: (pyflakes 0.7.3 (/Users/alex_gaynor/.virtualenvs/23e6d2eafccc0dc9/hacking/.tox/pypy/site-packages), Requirement.parse('pyflakes==0.7.2'))" which doesn't really seem relevant | 16:48 |
| * clarkb finds link to etherpad | 16:48 | |
| yolanda | jeblair, i was having some issues with "Connect" permissions denied for anonymous, but i had anonymous access totally locked | 16:48 |
| anteaya | https://etherpad.openstack.org/cibugreview-september2013 | 16:48 |
| yolanda | i don't want anonymous to be able to read, but if there is no way to specify that, we will try to deal with it | 16:48 |
| pleia2 | anteaya: I'm not done yet, it's not 17:00! :) | 16:49 |
| jeblair | yolanda: i don't know if that is required for gearman-plugin -- we have it set because that's what we want (it's public). | 16:49 |
| clarkb | pleia2: I won't look at it yet then :) | 16:49 |
| jeblair | yolanda: can you pastebin the error message you get? | 16:49 |
| anteaya | don't let me distract, pleia2 | 16:49 |
| Alex_Gaynor | jog0: so tests pass on that other one you rechecked, I wonder if it's somehow your patch! | 16:49 |
| jog0 | Alex_Gaynor: :( | 16:49 |
| Alex_Gaynor | jog0: for the life of me I can't see how though | 16:49 |
| *** dkehn_ has joined #openstack-infra | 16:49 | |
| jog0 | Alex_Gaynor: you get it to fail locally | 16:50 |
| *** dkehn has quit IRC | 16:50 | |
| hashar | clarkb: hi :) Seems like the JJB xml comparison job is working again. I have abandoned my test change 45798 and left a message on the puppet change 45801. Thank you very much! | 16:50 |
| pleia2 | clarkb: everyone can start working on it, I'm about halfway through Mediums | 16:50 |
| yolanda | jeblair: INFO: While serving http://xxxx/computer/ciaas-jenkins-slave-0/slave-agent.jnlp: hudson.security.AccessDeniedException2: anonymous is missing the Connect permission | 16:50 |
| yolanda | and also zuul was reporting that the job didn't exist | 16:50 |
| Alex_Gaynor | jog0: yup, it totally reproduces locally on your patch | 16:51 |
| yolanda | removing security worked | 16:51 |
| jog0 | Alex_Gaynor: :( can you do a git bisect o nit | 16:51 |
| jog0 | on it | 16:51 |
| Alex_Gaynor | jog0: how do you mean? | 16:51 |
| jog0 | good news is its reproducable | 16:51 |
| jeblair | yolanda: i'm not a jenkins expert, but that kind of looks like a problem launching the slave, not a gearman-plugin problem; | 16:52 |
| jog0 | find a patch in that branch where things work, then use git-bisect | 16:52 |
| *** hashar has quit IRC | 16:52 | |
| jeblair | yolanda: it looks like you're using jnlp to launch the slave, which requires the slave make a connection to jenkins, which is denied | 16:52 |
| clarkb | jeblair: yolanda: that is what it looks like to me. slave-agent.jnlp is one way slaves talk to the master | 16:52 |
| yolanda | jeblair, not sure if this log was related anyway, but the zuul one was | 16:52 |
| jeblair | yolanda: (it will affect gearman-plugin because without the slave online, the job won't be registered) | 16:52 |
| Alex_Gaynor | jog0: can you try rebasing that patch on master? if I do that it works ok | 16:52 |
| clarkb | yolanda: you can try running a job on the master with zuul and see if that suffers the same problem? | 16:53 |
| jog0 | Alex_Gaynor: that is already in master | 16:53 |
| jeblair | yolanda: (but there's nothing gearman-plugin can do about it) | 16:53 |
| Alex_Gaynor | jog0: no, I mean on https://review.openstack.org/#/c/45761 | 16:53 |
| yolanda | clarkb, i can't try it at the moment, sorry | 16:53 |
| jog0 | Alex_Gaynor: that is cherry-pick to a stable branch | 16:54 |
| jog0 | so can't rebase on master | 16:54 |
| clarkb | oh maybe stable branch doesn't work? we can make pypy run only on master and new stable branches | 16:54 |
| *** nati_ueno has joined #openstack-infra | 16:54 | |
| jog0 | try running pyp on the 0.7.x branch | 16:54 |
| yolanda | jeblair, so gearman-plugin is bypassing all security rules, but if slave-master relation fails by some security issues, jobs won't be registered, right? | 16:54 |
| jog0 | pypy | 16:54 |
| mordred | what clarkb said | 16:54 |
| yolanda | just trying to get some clues | 16:54 |
| jeblair | yolanda: yes | 16:54 |
| *** UtahDave has quit IRC | 16:55 | |
| yolanda | thx | 16:55 |
| jog0 | clarkb: not sure why that would be the case | 16:55 |
| jog0 | this is a new stable branch | 16:55 |
| jog0 | as of yesterday | 16:55 |
| mordred | jog0: we should really make that branch stable/0.7.x next time we get a chance | 16:55 |
| mordred | or maybe we shouldn't | 16:55 |
| mordred | hrm | 16:55 |
| Alex_Gaynor | jog0: ah ah ah, this patch is against stable/0.7.x, not master | 16:55 |
| mordred | don't pay attention to me - I'm still drinking coffee | 16:55 |
| jog0 | mordred: I forgot why we didn't use stable in it but there was a reason | 16:55 |
| clarkb | mordred: fungi https://review.openstack.org/#/c/45788/ getting that in soonish would be awesome | 16:56 |
| jog0 | sdague and I chatted about it | 16:56 |
| mordred | because that means it becomes stable-maint | 16:56 |
| *** UtahDave has joined #openstack-infra | 16:56 | |
| jog0 | here stable just means no new tests | 16:56 |
| mordred | clarkb: done | 16:56 |
| clarkb | mordred: danke | 16:57 |
| sdague | yeh, that's why, the core teams get all weird if we made it stable/ | 16:57 |
| fungi | clarkb: we double-tapped it apparently ;) | 16:57 |
| *** Ryan_Lane has joined #openstack-infra | 16:57 | |
| clarkb | sdague: jog0: we can override that behavior in your acl with exclusive access if you wish | 16:57 |
| *** derekh has quit IRC | 16:57 | |
| sdague | we could, it's not really a big deal though | 16:57 |
| openstackgerrit | A change was merged to openstack-infra/config: Use bup to backup review-dev to an offsite server https://review.openstack.org/45788 | 16:58 |
| openstackgerrit | Tom Fifield proposed a change to openstack-infra/jeepyb: Update doc impact to only make bugs on merge https://review.openstack.org/45907 | 16:58 |
| jog0 | Alex_Gaynor: well for now pypy is non-gating and when we sort this out (today I hope) we can make it gating again | 16:58 |
| *** fifieldt has joined #openstack-infra | 16:58 | |
| jog0 | I am thinking maybe its a funny deps issue | 16:58 |
| sdague | man, today is so full of meetings.... I can't wait to get back to hacking on things tomorrow | 16:58 |
| Alex_Gaynor | jog0: can we make it only non-gating on the 0.7.x branch? | 16:58 |
| Alex_Gaynor | jog0: since it gates fine on master | 16:59 |
| jog0 | Alex_Gaynor: perhaps, lets see if we can make it gate on 0.7.x branch as well though | 16:59 |
| openstackgerrit | Tom Fifield proposed a change to openstack-infra/jeepyb: Update doc impact to only make bugs on merge https://review.openstack.org/45907 | 16:59 |
| jog0 | Alex_Gaynor: but I am all for that, if you put the patch up I will +1 | 16:59 |
| jeblair | pleia2: ready? | 17:00 |
| jog0 | Alex_Gaynor: I wonder how much faster hacking/flake8 is with pypy | 17:00 |
| jog0 | because for nova running flake8 can be slow | 17:01 |
| fungi | one two three etherpad | 17:01 |
| jog0 | I.E. if its faster I want to switch to using pypy hacking on my own bux | 17:01 |
| jog0 | box | 17:01 |
| Alex_Gaynor | jog0: is http://bpaste.net/show/131157/ what I want? | 17:01 |
| jog0 | Alex_Gaynor: I *think* so | 17:02 |
| Alex_Gaynor | clarkb: ^ | 17:02 |
| jog0 | yeah ask them | 17:02 |
| Alex_Gaynor | I'll put it up for review | 17:02 |
| clarkb | Alex_Gaynor: you want to remove the voting false and invert the regex to be negative lookahead on that string | 17:02 |
| Alex_Gaynor | clarkb: ah ok | 17:02 |
| pleia2 | jeblair: yeah I just have some low ones to review | 17:03 |
| clarkb | Alex_Gaynor: asserting that we don't run that job on that branch, but everything else is fair game | 17:03 |
| pleia2 | https://etherpad.openstack.org/cibugreview-september2013 | 17:03 |
| pleia2 | happy bug day :) | 17:03 |
| openstackgerrit | Alex Gaynor proposed a change to openstack-infra/config: Only make hacking-pypy non-voting on stable/0.7.x https://review.openstack.org/45908 | 17:03 |
| *** mrmartin has quit IRC | 17:03 | |
| jog0 | so the non-gating patch didn't work https://review.openstack.org/#/c/45761/1 | 17:03 |
| jog0 | clarkb: ^ | 17:03 |
| clarkb | jog0: patience :) that change is pushed out by puppet which is on a slightly weird interval | 17:04 |
| clarkb | source updates every 15 minutes, puppet runs about every 10 minutes | 17:04 |
| jog0 | clarkb: ohh ahh | 17:04 |
| jeblair | mordred, clarkb, fungi, anyone-else: bug triage time https://etherpad.openstack.org/cibugreview-september2013 | 17:04 |
| * fungi is already digging in | 17:05 | |
| jeblair | clarkb: you should take the backup bug, you seem to just about have that wrapped up | 17:05 |
| clarkb | jeblair: will do | 17:05 |
| *** changbl has quit IRC | 17:05 | |
| clarkb | mordred: the second bug on that list has your name written all over it | 17:05 |
| jeblair | clarkb: did notmyname or chmouel ever respond about https://bugs.launchpad.net/openstack-ci/+bug/1076481 | 17:06 |
| uvirtbot | Launchpad bug 1076481 in openstack-ci "Add Swift functional tests to full test run" [High,In progress] | 17:06 |
| pleia2 | fungi: I think we can close this one https://bugs.launchpad.net/openstack-ci/+bug/1204843 | 17:08 |
| uvirtbot | Launchpad bug 1204843 in openstack-ci "No Content-Type for logs on eavesdrop.openstack.org" [Medium,In progress] | 17:08 |
| clarkb | jeblair: yes, I think we went ahead and made them gating | 17:08 |
| fungi | pleia2: oh, right. closing | 17:08 |
| anteaya | pabelanger: this one is currently assigned to you: https://bugs.launchpad.net/openstack-ci/+bug/933843 is it still on your radar, or what would you like to do? | 17:08 |
| uvirtbot | Launchpad bug 933843 in openstack-ci "Use upstream puppet modules where possible" [Medium,In progress] | 17:08 |
| jeblair | zaro: ping | 17:08 |
| *** kpepple has joined #openstack-infra | 17:09 | |
| *** dprince has joined #openstack-infra | 17:09 | |
| jeblair | i spoke to david yesterday, and told him about the tag-based releases for the clouddocs plugin | 17:10 |
| jeblair | i'll start an email thread and we can coordinate | 17:10 |
| jeblair | updated bug 1082812 | 17:10 |
| uvirtbot | Launchpad bug 1082812 in openstack-ci "Move rackspace/clouddocs-maven-plugin to openstack-dev/clouddocs-maven-plugin" [High,In progress] https://launchpad.net/bugs/1082812 | 17:10 |
| clarkb | jeblair: ty | 17:10 |
| jeblair | pleia2: is 'incomplete' not a closed state? | 17:12 |
| *** cppcabrera_afk is now known as cppcabrera | 17:13 | |
| pleia2 | jeblair: I manually added the list of ones included here, I can add Incomplete, sec | 17:13 |
| pleia2 | hm, actually I have LPSTATUS = ('New', 'Incomplete', 'Incomplete (with response)', | 17:14 |
| jeblair | pleia2: just wondering why https://bugs.launchpad.net/openstack-ci/+bug/1129172 is in the list | 17:14 |
| uvirtbot | Launchpad bug 1129172 in openstack-ci "devstack-gate - Unable to get pack index" [High,Incomplete] | 17:14 |
| pleia2 | jeblair: I guess "High" trumps "Incomplete" | 17:14 |
| pleia2 | let's see what it does if I remove Incomplete | 17:15 |
| *** sarob has joined #openstack-infra | 17:15 | |
| *** sarob has quit IRC | 17:16 | |
| *** sarob has joined #openstack-infra | 17:17 | |
| *** pmoosh has joined #openstack-infra | 17:17 | |
| *** sarob has quit IRC | 17:18 | |
| pleia2 | jeblair: yeah, so I was specifically including incomplete in my list, I can not do that if we want | 17:18 |
| *** sarob has joined #openstack-infra | 17:18 | |
| zaro | jeblair: pong. sorry i didn't realized that i lost my network connection. | 17:18 |
| jeblair | zaro: see https://launchpad.net/bugs/1082812 | 17:18 |
| uvirtbot | Launchpad bug 1082812 in openstack-ci "Move rackspace/clouddocs-maven-plugin to openstack-dev/clouddocs-maven-plugin" [High,In progress] | 17:18 |
| clarkb | jeblair: how does recent update on https://bugs.launchpad.net/openstack-ci/+bug/1010621 look? | 17:19 |
| uvirtbot | Launchpad bug 1010621 in openstack-ci "important servers should have backups" [Critical,In progress] | 17:19 |
| *** dizquierdo has left #openstack-infra | 17:20 | |
| Alex_Gaynor | jog0: so I'm pretty confused, at this point `git diff master` from 0.7.x is tiny, and there's nothing in there that obviously would cause pypy incompat, but it really doesn't pass on 0.7.x and does on master :/ | 17:20 |
| jog0 | Alex_Gaynor any deps that are different? | 17:20 |
| jeblair | clarkb: lgtm | 17:20 |
| mordred | clarkb: yes. the second bug is all me | 17:20 |
| zaro | jeblair: i can take ownership of that one if someone else isn't already working on it. | 17:20 |
| Alex_Gaynor | jog0: six vs. six<1.4.0 | 17:20 |
| *** jcoufal has joined #openstack-infra | 17:21 | |
| jog0 | Alex_Gaynor: that patch is going into 0.7.x next | 17:21 |
| jeblair | zaro: that would be great, thanks | 17:21 |
| Alex_Gaynor | jog0: cool, we'll see if that changes anything I guess, I can't for the life of me imagine it would | 17:21 |
| jog0 | https://review.openstack.org/#/c/45764/ | 17:21 |
| zaro | clarkb: took ownership of https://launchpad.net/bugs/1082812 should be easy to do since infra is already in place. | 17:22 |
| uvirtbot | Launchpad bug 1082812 in openstack-ci "Move rackspace/clouddocs-maven-plugin to openstack-dev/clouddocs-maven-plugin" [High,In progress] | 17:22 |
| clarkb | zaro: its mostly done | 17:23 |
| Alex_Gaynor | jog0: grumble, pypy still fails there, still one less thing for me to look at | 17:23 |
| clarkb | zaro: we ended up putting it in stackforge iirc | 17:23 |
| clarkb | zaro: I haven't had a chance to update the bug since mordred merged the import change :P | 17:23 |
| jog0 | Alex_Gaynor: good luck | 17:23 |
| clarkb | mordred: ^ you pulled the trigger too soon | 17:23 |
| Alex_Gaynor | jog0: I wonder if git will let me bisect between 0.7.x and master in a sane way | 17:24 |
| clarkb | looks like jeblair updated the bug, good | 17:24 |
| mordred | clarkb: sorry | 17:24 |
| clarkb | mordred: its ok, we will just need to force push to update the repo to the latest version before they start using gerrit | 17:24 |
| *** Ryan_Lane has quit IRC | 17:25 | |
| jeblair | he also has a second plugin he wants to move after this one is done | 17:25 |
| jog0 | Alex_Gaynor: looks like it does | 17:25 |
| jog0 | 'git bisect start master 0.7.x' | 17:25 |
| Alex_Gaynor | jog0: ok cool, I'll try that in a minute | 17:25 |
| *** hub_cap has joined #openstack-infra | 17:26 | |
| hub_cap | hey is there a way to reference a dependency on another project in gerrit? besides just a self-review w a message? | 17:26 |
| clarkb | hub_cap: no | 17:26 |
| hub_cap | ok so then is the rule of thumb just to add a comment? | 17:27 |
| hub_cap | w a link | 17:27 |
| clarkb | hub_cap: yeah, documentation | 17:27 |
| fungi | hub_cap: bug 1172399 | 17:27 |
| uvirtbot | Launchpad bug 1172399 in zuul "support cross-repo-dependencies in zuul" [Undecided,New] https://launchpad.net/bugs/1172399 | 17:27 |
| clarkb | hub_cap: you can also potentially express the relationship in your test suite | 17:27 |
| *** matty_dubs has joined #openstack-infra | 17:27 | |
| clarkb | hub_cap: such that you can only merge in one way with the gate, but that is often more trouble than it is worth | 17:27 |
| mordred | ok. I'm off the phone. where are we? | 17:28 |
| hub_cap | yes clarkb i see | 17:28 |
| hub_cap | but i like the fungi link :) | 17:28 |
| * mordred always seems to get stuck with heavy call days on bug review days | 17:28 | |
| jeblair | hub_cap: sorry it's short notice, but can you drop by the infra team meeting today (starts at 1900 utc) ? | 17:28 |
| clarkb | mordred: https://etherpad.openstack.org/cibugreview-september2013 | 17:28 |
| hub_cap | jeblair: yar | 17:28 |
| clarkb | mordred: work down the list :) | 17:28 |
| jeblair | hub_cap: great thanks! | 17:28 |
| hub_cap | can u do me a solid jeblair? and just ping me when it starts? | 17:28 |
| hub_cap | im lurking | 17:28 |
| jeblair | hub_cap: will do | 17:29 |
| hub_cap | <3<3 | 17:29 |
| *** Ryan_Lane has joined #openstack-infra | 17:30 | |
| zaro | ohh. did you want to keep ownership of 1082812 then? | 17:30 |
| jeblair | bug 1082812 | 17:31 |
| uvirtbot | Launchpad bug 1082812 in openstack-ci "Move rackspace/clouddocs-maven-plugin to openstack-dev/clouddocs-maven-plugin" [High,In progress] https://launchpad.net/bugs/1082812 | 17:31 |
| jeblair | zaro: no, you can have it; the only other thing is that a core needs to do a force push. it's not a big deal. | 17:31 |
| jeblair | zaro: but i'll start an email thread | 17:31 |
| fungi | clarkb: jeblair: are we clear to delete the old paste.o.o vm? we've been on the new one a while | 17:31 |
| clarkb | fungi: isn't that still paste.drizzle.org? | 17:32 |
| jeblair | mordred: ^ ? | 17:32 |
| fungi | ahh, right | 17:32 |
| jeblair | honestly, i think we've given them quite enough time. | 17:32 |
| mordred | yeah. oh - brian and mark said they don't care if that disappears | 17:32 |
| fungi | dns for paste.drizzle.o definitely still points to the old vm | 17:33 |
| * mordred double checking | 17:33 | |
| clarkb | jeblair: https://bugs.launchpad.net/openstack-ci/+bug/1093812 should I go ahead and remove openstack-ci from that bug? no progress from nova and unless we want to fix the nova tree ourselves I don't see much reason for tracking it | 17:33 |
| uvirtbot | Launchpad bug 1093812 in openstack-ci "nova API version doc should point to Nova developer doc" [High,Confirmed] | 17:33 |
| mordred | btw - bug 2 in the list is still waiting on this: https://review.openstack.org/#/c/41945/ | 17:33 |
| mordred | clarkb: I don't think that's a thing nova can fix | 17:34 |
| clarkb | mordred: it is, the links are on nova/doc/source | 17:34 |
| clarkb | s/on/in/ | 17:34 |
| mordred | clarkb: the redirects from nova.o.o? | 17:34 |
| mordred | ah - I see | 17:35 |
| mordred | nevermind | 17:35 |
| mordred | clarkb: actually, I think this is a "we need to re-run the docs jobs for the previous tags" | 17:35 |
| zaro | jeblair: is this done? https://bugs.launchpad.net/openstack-ci/+bug/1082803 | 17:36 |
| uvirtbot | Launchpad bug 1082803 in openstack-ci "Manage jenkins global config and plugin list" [Medium,In progress] | 17:36 |
| mordred | clarkb: http://docs.openstack.org/developer/nova/2013.1/ works just fine - but that's because of how we publish docs on tag events, right? | 17:36 |
| jeblair | zaro: partially; mgagne1's summary is the current status | 17:36 |
| jeblair | zaro: it should be left open because there's still more to do | 17:36 |
| fungi | which may not work if the docs jobs are not backward compatible to the old states represented by those tags | 17:36 |
| mordred | fungi: verified. brian aker does not care if paste.drizzle.org ceases to exist | 17:36 |
| *** mgagne1 is now known as mgagne | 17:36 | |
| clarkb | mordred: developer/nova is off of master | 17:36 |
| fungi | mordred: righteous. i'll blow away the old paste.o.o | 17:37 |
| zaro | jeblair: i'm not sure what else is needed. could you please update? | 17:37 |
| mordred | http://docs.openstack.org/developer/nova/2013.1/ is off the 2013.1 tag | 17:37 |
| clarkb | mordred: I think there may be mutliple issues expressed in that bug and it isn't clearly described | 17:37 |
| jeblair | zaro: the global config is not managed | 17:37 |
| mordred | clarkb: THAT I agree with | 17:37 |
| fungi | mordred: clarkb: jeblair: also should be safe to delete the old wiki.o.o vm and the old-wiki.o.o dns entry i guess. any objections? | 17:37 |
| mordred | fungi: I'm fine with it - I think we've been running on new wiki for quite a while now | 17:38 |
| jeblair | fungi: there's a bug assigned to me to move stuff off that server | 17:38 |
| clarkb | mordred: 1. nova doc/source/* have/had hard coded links to places with versions that are old 2. we don't delete old stuff when doing the ftp upload because doing so leaves a small window where stuff will 404 | 17:38 |
| fungi | jeblair: oh, right that | 17:38 |
| fungi | nevermind | 17:38 |
| fungi | i keep forgetting we had other sites on that system | 17:38 |
| jeblair | i'm still pretty bummed that we're going to lose a lot of content when we shut it down | 17:38 |
| jeblair | i had to use it just the other day to find some etherpads :( | 17:39 |
| clarkb | mordred: I think the solution to 2 is to stop hosting the docs wherever they are hosted so that we can be smarter about the file copying | 17:39 |
| mordred | jeblair: blerg | 17:39 |
| mordred | clarkb: they are currently hosted in cloud sites - perhaps that's what the CI bug should be about? | 17:39 |
| jeblair | mordred, clarkb: i don't object, but how does that help? | 17:40 |
| jeblair | my understanding is that many docs jobs are designed around leaving files in certain directories, eg, branches | 17:40 |
| Alex_Gaynor | jog0: this doesn't appear to work, git bisect appears to have gotten itself into a loop of some sort, it never actually makes progress | 17:40 |
| clarkb | mordred: anteaya's example is great http://docs.openstack.org/developer/nova/object.model.html isn't in latest master (it is an orphaned file that refers to old stuff on the irght hand side | 17:40 |
| jeblair | the file copy mechanism doesn't appear to change that. | 17:40 |
| *** pabelanger has quit IRC | 17:40 | |
| jeblair | clarkb: is the bug about "obsolete files are left in place"? | 17:41 |
| mordred | no, the bug is about "links to old tags 404" | 17:41 |
| jeblair | clarkb: because i thought it was about entire missing deployments of docs | 17:41 |
| clarkb | jeblair: the bug should be split in two. obsolete files are left in place and in some parts of nova docs we hard code links to old stuff | 17:41 |
| mordred | clarkb: there is a third thing | 17:41 |
| mordred | it's not about the stale links in the nova docs | 17:41 |
| mordred | it's about the fact that the thing they want to point to does not exist anywhere | 17:41 |
| jog0 | Alex_Gaynor: :( it was worth a short | 17:41 |
| mordred | so the link can't be updated | 17:41 |
| jeblair | thu bug says "Currently nova.openstack.org redirects to docs.openstack.org/developer/nova. | 17:42 |
| mordred | because there is no location | 17:42 |
| jeblair | But the /versionnumber sites (http://nova.openstack.org/2011.2 ), points 404." | 17:42 |
| jeblair | that seems pretty clear | 17:42 |
| fungi | ttx: do you still need the old summit.o.o vm at 50.57.121.48 or can it be deleted? | 17:42 |
| jog0 | I assume after running each test you say git bisect good or bad | 17:42 |
| jeblair | i don't see anything about individual files that stick around | 17:42 |
| clarkb | jeblair: then read the description | 17:42 |
| clarkb | jeblair: it is a very confusing bug and the exmaple doesn't support the title and so on which is why at least a couple different bugs are expressed in it | 17:42 |
| jog0 | Alex_Gaynor: are only a few patches that are missing in 0.7.x so shouldnt' be hard to manually do | 17:42 |
| clarkb | jeblair: anteaya's comment points out the thing about individual files sticking around | 17:43 |
| notmyname | jeblair: did you get an answer about swift functional tests? | 17:43 |
| jeblair | clarkb: are you saying that 'object.model.html' isn't a current file? | 17:43 |
| jeblair | notmyname: yes | 17:43 |
| clarkb | jeblair: yes | 17:43 |
| notmyname | jeblair: kk. anything needed from our side? | 17:43 |
| notmyname | jeblair: or any changes we should know about? | 17:43 |
| mordred | three bugs: orphan files, stale links in nova docs, missing publications of old tags of nova | 17:43 |
| jeblair | notmyname: not that i'm aware of, thanks; we're just cleaning up old bugs | 17:43 |
| clarkb | jeblair: so the bug there isn't that we link to a thing that doesn't exist its that we still host content that is dead | 17:44 |
| *** nati_uen_ has joined #openstack-infra | 17:44 | |
| jeblair | mordred: my evolving understanding of the problem is now that 'stale links in nova docs' doesn't exist, because the stale links are only in files that shouldn't exist, so that should be folded into "orphan files" | 17:44 |
| jeblair | mordred, clarkb: ^ is that wrong? | 17:45 |
| jog0 | Alex_Gaynor: I try running 'pypy /usr/local/bin/flake8' I get no module named pkg_resources | 17:45 |
| *** nati_ueno has quit IRC | 17:45 | |
| mordred | jeblair: yes. I believe I agree with that | 17:45 |
| Alex_Gaynor | jog0: yeah, you need to have pypy installed into its own virtualenv, `virtualenv -p `which pypy`` | 17:45 |
| *** nati_uen_ has quit IRC | 17:46 | |
| jog0 | eww | 17:46 |
| *** atiwari has joined #openstack-infra | 17:46 | |
| jeblair | clarkb: can you take 1168455 as part of your etherpad hatchet job? | 17:48 |
| jeblair | bug 1168455 | 17:48 |
| uvirtbot | Launchpad bug 1168455 in openstack-ci "Etherpad export function always exports another document" [Medium,Triaged] https://launchpad.net/bugs/1168455 | 17:48 |
| jeblair | clarkb: i think the understanding is that we should drop our rewrite rule | 17:48 |
| jeblair | clarkb: (but consider putting in a redirect, if it's not too much trouble) | 17:49 |
| clarkb | jeblair: yes I can take it, and yes export still doesn't work with new etherpad when we have our rewrite rules | 17:49 |
| clarkb | jeblair: import does work however... I tested that as part of the hatchet job | 17:49 |
| Alex_Gaynor | jog0: ce4f1f35 fixes it, I have no clue how why | 17:49 |
| clarkb | jeblair: redirect https://etherpad.openstack.org/\([^p][^/]\) to https://etherpad.openstack.org/p/$1 ? | 17:50 |
| jog0 | Alex_Gaynor: weird | 17:50 |
| *** pycabrera has joined #openstack-infra | 17:50 | |
| jeblair | clarkb: do you need to exclude the etherpad js, css, images and such dependencies? | 17:51 |
| clarkb | jeblair: yes | 17:51 |
| notmyname | clarkb: jeblair: mordred: anything going on right now that would cause networking errors on the swift functests gate? | 17:52 |
| *** cppcabrera has quit IRC | 17:52 | |
| portante | and along with those failing tests are other tempest jobs as well | 17:53 |
| notmyname | portante: ah? it's not just our tests? | 17:53 |
| portante | well | 17:53 |
| portante | the tempest jobs fail often | 17:53 |
| portante | so I can't tell | 17:54 |
| clarkb | notmyname: portante: can you link to examples? | 17:54 |
| portante | http://logs.openstack.org/43/44943/6/check/gate-swift-devstack-vm-functional/9c1f080 | 17:54 |
| notmyname | clarkb: specivic http://logs.openstack.org/97/45897/1/gate/gate-swift-devstack-vm-functional/7cb7029/console.html | 17:54 |
| notmyname | clarkb: but I see 8 patches with it right now | 17:54 |
| portante | clarkb: https://bugs.launchpad.net/swift/+bug/1223466 | 17:54 |
| uvirtbot | Launchpad bug 1223466 in swift "Functional test failures occurring sporadically" [Undecided,New] | 17:54 |
| jeblair | mordred: i think you should close 1192039 | 17:54 |
| jeblair | mordred: i think you should close bug 1192039 | 17:54 |
| uvirtbot | Launchpad bug 1192039 in openstack-ci "Merge release tags back into master" [Medium,In progress] https://launchpad.net/bugs/1192039 | 17:54 |
| zaro | clarkb: is Review in progress at https://review.openstack.org/#/c/43263/ | 17:55 |
| portante | that bug shows a failing functional test on man page changes. :) | 17:55 |
| jeblair | mordred: ha, sorry, you should not close it, nm | 17:55 |
| *** pycabrera is now known as cppcabrera | 17:55 | |
| clarkb | zaro: no, see jeblair's comment | 17:55 |
| zaro | clarkb: is bug 1200215 an upstream bug? | 17:55 |
| uvirtbot | Launchpad bug 1200215 in openstack-ci "Gerrit revert does not include Change-Id in commit message" [Medium,Triaged] https://launchpad.net/bugs/1200215 | 17:55 |
| jeblair | mordred: i'm confused by why that didn't lik | 17:56 |
| clarkb | zaro: no | 17:56 |
| jeblair | link | 17:56 |
| *** melwitt has joined #openstack-infra | 17:56 | |
| clarkb | zaro: that is a quirk in how git hooks work | 17:56 |
| clarkb | fungi: has lots of details | 17:56 |
| mordred | jeblair: I am also confused by that | 17:56 |
| *** sarob has quit IRC | 17:56 | |
| clarkb | portante: notmyname: that is networking failing against localhost | 17:56 |
| clarkb | that should just work (tm_ | 17:56 |
| notmyname | clarkb: thus our confusion ;-) | 17:56 |
| *** sarob has joined #openstack-infra | 17:56 | |
| jog0 | Alex_Gaynor: so pypy version of hacking/flake8 seems take the same amount of time | 17:57 |
| clarkb | notmyname: portante: are those tests using swiftclient? at least a few of them appear to | 17:58 |
| jog0 | oh wait that was cold cache | 17:58 |
| Alex_Gaynor | jog0: find a larger codebase, and see if its gets faster ;) | 17:58 |
| clarkb | notmyname: portante: perhpas we have a non symmetric gate with swift and swiftclient that allowed swiftclient to break swift ? | 17:58 |
| zaro | clarkb: is it even a bug at all? | 17:58 |
| jog0 | Alex_Gaynor: in nova flake8 takes 72 seconds or so | 17:58 |
| jeblair | clarkb: also, why are there no logs for that job? | 17:59 |
| notmyname | clarkb: the last patch merged to python-swiftclient was Aug29 | 17:59 |
| notmyname | clarkb: IOW, no changes recently, and the problem only surfaced since I got on the bus this morning ;-) | 17:59 |
| portante | these functional test failures seemed to start today | 18:00 |
| portante | what he said | 18:00 |
| * portante SF buses are probably not the root cause, though | 18:00 | |
| clarkb | jeblair: I am not sure, probably due to the way the job uses the d-g hooks | 18:00 |
| clarkb | notmyname: portante: are the failures reproducable? | 18:01 |
| clarkb | *reproduceable locally? | 18:01 |
| notmyname | clarkb: no. works for me locally | 18:01 |
| jeblair | so i wonder about a devstack or other dependency change | 18:01 |
| clarkb | jeblair: possibly in the swift(client) dependencies as well | 18:01 |
| Alex_Gaynor | jog0: oh. | 18:02 |
| Alex_Gaynor | jog0: can we just backport this to 0.7.x ;) | 18:02 |
| clarkb | jeblair: actually, the job doesn't have the scp stuff for devstack jobs | 18:02 |
| clarkb | jeblair: so we aren't copying what logs may be there | 18:02 |
| *** pblaho has joined #openstack-infra | 18:02 | |
| notmyname | portante: the bus I got on this morning was surprisingly clean. that plus the gate failures are 2 weird things that have happened already today | 18:03 |
| jog0 | Alex_Gaynor: looks like pypy does result ina 20% speedup afterall | 18:03 |
| portante | ;) | 18:03 |
| jeblair | clarkb: i'll fix the job defn | 18:03 |
| Alex_Gaynor | jog0: oh, cool | 18:03 |
| *** salv-orlando has quit IRC | 18:03 | |
| *** pabelanger has joined #openstack-infra | 18:03 | |
| clarkb | portante: in your local env are you starting from a fresh devstack isntall so that dependencies are installed as in the gate? | 18:04 |
| clarkb | er notmyname ^ | 18:04 |
| *** salv-orlando has joined #openstack-infra | 18:04 | |
| jog0 | Alex_Gaynor: yeah so if you want to backport that change to 0.7.x i will aprove it but it sounds like the wrong way to fix it | 18:04 |
| *** markmcclain has joined #openstack-infra | 18:04 | |
| Alex_Gaynor | jog0: it is totally wrong, but I completely don't understand :/ | 18:04 |
| notmyname | clarkb: not devstack, but I've installed all the deps from test-requirements | 18:04 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/config: Use correct gate hook for swift functional tests https://review.openstack.org/45917 | 18:04 |
| jog0 | Alex_Gaynor: maybe you can file a bug with pypy to deal with later | 18:05 |
| *** cppcabrera has quit IRC | 18:05 | |
| Alex_Gaynor | jog0: ahhh, I think I found a bug maybe | 18:05 |
| clarkb | notmyname: devstack and devstack gate are going to install deps from openstack/requirements/global-requirements.txt to make sure everyone works on the same page however we are asymmetric with this test so that may be related | 18:05 |
| clarkb | mordred: ^ | 18:05 |
| *** cppcabrera has joined #openstack-infra | 18:06 | |
| portante | clarkb: link to that file? | 18:06 |
| clarkb | portante: https://git.openstack.org/cgit/openstack/requirements/tree/global-requirements.txt | 18:06 |
| Alex_Gaynor | jog0: ok I understand why <something> doesn't work, I have no idea how it only affects pypy though | 18:07 |
| anteaya | clarkb: could this patch be playing a role? https://review.openstack.org/#/c/45283/1 | 18:07 |
| clarkb | anteaya: no, that is for gating global-requirements and shouldn't affect anything else | 18:07 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/config: Add swift-functional test to client and reqs https://review.openstack.org/45918 | 18:07 |
| anteaya | clarkb: k | 18:07 |
| jog0 | Alex_Gaynor: haha | 18:08 |
| clarkb | anteaya: actual changes to global-requirements or the unpinned dependencies in global-requirements would potentially be causes | 18:08 |
| *** salv-orlando_ has joined #openstack-infra | 18:08 | |
| notmyname | clarkb: perhaps, but has that changed in the last 2 hours? | 18:08 |
| anteaya | k, I'll look again | 18:08 |
| anteaya | is it just within the last two hours? | 18:08 |
| notmyname | anteaya: ish. it started just a few hours ago. I don't have an exact start time | 18:09 |
| Alex_Gaynor | jog0: yeah, I'm confused :/ I'll put up the cherry-pick | 18:09 |
| Alex_Gaynor | jog0: what's the procedure for cherry-picking a thing | 18:09 |
| jog0 | Alex_Gaynor: kk, when/if yo u find the issue I am interested in hearing what it was | 18:09 |
| jog0 | 'git cherry-pick -ex sha1' | 18:10 |
| notmyname | anteaya: the oldest test failure is at 9:23am (pacific). so just under 2 hours ago | 18:10 |
| jog0 | that keeps the original change-id | 18:10 |
| *** cppcabrera has quit IRC | 18:10 | |
| Alex_Gaynor | jog0: so part of hte issue is that when you have `return` in a generator it return an empty generator, it doesn't actually return None, and moving the thing outside into a different funciton really does make it return None | 18:10 |
| anteaya | notmyname: hmmm okay | 18:10 |
| portante | notmyname, clarkb: certainly swift's requirements.txt needs to be reconciled against the global one | 18:10 |
| *** salv-orlando has quit IRC | 18:11 | |
| *** salv-orlando_ is now known as salv-orlando | 18:11 | |
| clarkb | https://jenkins.openstack.org/job/periodic-mirror-python26/29/console | 18:11 |
| anteaya | notmyname: what was the last passing test prior to that? | 18:11 |
| portante | global is missing swift's requirement of pastedeploy | 18:11 |
| clarkb | keyring was updated | 18:11 |
| anteaya | keyring, good old keyring | 18:11 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/devstack-gate: Add a pip freeze invocation to save with the logs https://review.openstack.org/45919 | 18:11 |
| Alex_Gaynor | jog0: https://review.openstack.org/#/c/45920/ let me know if I scrwed that up | 18:12 |
| portante | swift wants dnspython >= 1.10.0, while global wants 1.9.4 | 18:12 |
| jog0 | Alex_Gaynor: add a little blurb saying why we are backporting | 18:12 |
| Alex_Gaynor | jog0: ok | 18:12 |
| portante | those are the only two requirements that look out of sync | 18:13 |
| portante | notmyname,clarkb: ^ | 18:13 |
| jog0 | Alex_Gaynor: once this merges lets turn pypy gating back on | 18:13 |
| notmyname | anteaya: hard to say. the next oldest patch is from 8:13 am. (but there may have been passing earlier runs in the ones that are marked failed now) | 18:13 |
| anteaya | notmyname: okay | 18:14 |
| clarkb | the changes jeblair has proposed should give us a lot more info | 18:14 |
| Alex_Gaynor | jog0: yup, going to go abandon my other patch | 18:14 |
| clarkb | it may be valuable to reproduce using an env similar to the one used in the gate | 18:15 |
| notmyname | portante: doesn't look like our requirements or test requirements has changed since april | 18:15 |
| mordred | notmyname: I know one of the reasons for dnspython pinning is around ipv6 stuff- it's possible that testnodes are starting to do somethign with ipv6 on localhost or something? | 18:15 |
| *** dcramer_ has quit IRC | 18:15 | |
| notmyname | mordred: again, is that something that changed in the last few hours? | 18:16 |
| clarkb | notmyname: right, but in the devstack jobs we don't use your requirements | 18:16 |
| mordred | notmyname: it's possible - I don't mean in our config | 18:16 |
| mordred | notmyname: I mean in the underlying machiens we're getting from the cloud | 18:16 |
| clarkb | falcon was also semi recently updated https://jenkins.openstack.org/job/post-mirror-python26/172/console | 18:16 |
| Alex_Gaynor | jog0: ok as soon as that's landed I'll put the re-enable gate up | 18:16 |
| *** cppcabrera has joined #openstack-infra | 18:16 | |
| jog0 | Alex_Gaynor: awesome | 18:16 |
| notmyname | clarkb: so related (but perhaps distracting, for now) question: when upstream requirements are updated, how do we get them? like I said, our requirements files haven't changed since april | 18:17 |
| mordred | notmyname: there is a pending patch which will propose changes to your repo when that happens | 18:17 |
| clarkb | notmyname: you update your requirements file to match | 18:17 |
| clarkb | mordred: I thought you decided that you wouldn't do that? | 18:17 |
| jog0 | jeblair: when is it possible to get https://review.openstack.org/#/c/45813/ running? | 18:17 |
| mordred | notmyname: in the mean time, cloning openstack/requirements and using its update.py script is the best bet currently | 18:17 |
| mordred | clarkb: no, I just decided to deprioritize getting it finished | 18:18 |
| notmyname | mordred: clarkb: ok, so there is not outstanding patch for swift to update the requirements. so something in that process is broken if the 2 are out of sync | 18:18 |
| jeblair | sdague: can you look at https://review.openstack.org/#/c/45813/ and jog0's other recheckwatch patch when you have a moment? | 18:18 |
| mordred | since it's not an operational need atm | 18:18 |
| mordred | notmyname: no - the job doesn't exist yet | 18:18 |
| mordred | notmyname: at the moment, it's purely a thing that projects can consume when they choose | 18:18 |
| mordred | well, purely is the wrong word | 18:19 |
| sdague | jeblair: sure | 18:19 |
| jog0 | jeblair: good idea, sdague thanks | 18:19 |
| mordred | consuming it via changes to your repo is at the project's choice of velocity | 18:19 |
| notmyname | mordred: "when they choose" == when something upstream changes and causes gate failures? | 18:19 |
| openstackgerrit | Pierre Rognant proposed a change to openstack-infra/jenkins-job-builder: Fix plot plugin support https://review.openstack.org/45754 | 18:19 |
| mordred | notmyname: often that is a motivating factor, yes | 18:19 |
| *** UtahDave has quit IRC | 18:19 | |
| sdague | so my concern is that closed bugs might actually be fair recheck bugs | 18:20 |
| mordred | thus why I do still want to get the auto-propose things going | 18:20 |
| notmyname | mordred: but perhaps this is distracting. it hasn't yet been established that this is the root cause yet | 18:20 |
| mordred | notmyname: indeed | 18:20 |
| openstackgerrit | A change was merged to openstack-infra/config: Use correct gate hook for swift functional tests https://review.openstack.org/45917 | 18:20 |
| mordred | potentially many moving bits | 18:20 |
| sdague | which is why I left them in, but aged them more | 18:20 |
| clarkb | notmyname: right, hopefully jeblair's changes give us enough info to stab at this less blindly | 18:20 |
| *** UtahDave has joined #openstack-infra | 18:20 | |
| clarkb | jeblair: mordred fungi https://bugs.launchpad.net/openstack-ci/+bug/1129172 safe to mark that something stronger than incomplete with the new git.o.o in place? | 18:21 |
| uvirtbot | Launchpad bug 1129172 in openstack-ci "devstack-gate - Unable to get pack index" [High,Incomplete] | 18:21 |
| mordred | clarkb: looking | 18:22 |
| openstackgerrit | Pierre Rognant proposed a change to openstack-infra/jenkins-job-builder: Fix plot plugin support https://review.openstack.org/45754 | 18:22 |
| jeblair | clarkb: what would you suggest? pleia2 was explicitly including incomplete bugs in her list; i think that was the problem. | 18:22 |
| clarkb | jeblair: hmm, ya I think that is valid | 18:23 |
| mordred | clarkb: the logs are gone, so I don't know enough about that bug to say otherwise | 18:23 |
| clarkb | jeblair: so stick to incomplete and update etherpad populator? | 18:23 |
| mordred | ++ | 18:23 |
| clarkb | pleia2: ^ | 18:23 |
| pleia2 | already patched my script for next time | 18:24 |
| pleia2 | need to push it to the githubs | 18:24 |
| Alex_Gaynor | jog0: so who do I bribe for the second review :) | 18:24 |
| jog0 | Alex_Gaynor: mordred or clarkb | 18:24 |
| mordred | Alex_Gaynor: which review? | 18:24 |
| Alex_Gaynor | mordred: https://review.openstack.org/#/c/45920/ | 18:24 |
| Alex_Gaynor | mordred: backport of a thing to get hacking passing on 0.7.x against pypy | 18:24 |
| openstackgerrit | A change was merged to openstack-infra/config: Add swift-functional test to client and reqs https://review.openstack.org/45918 | 18:24 |
| jog0 | sdague: if a closed bug is called in recheck we can re-open it | 18:25 |
| mordred | done | 18:25 |
| Alex_Gaynor | mordred: thanks! | 18:25 |
| jeblair | mordred: https://bugs.launchpad.net/openstack-ci/+bug/1214366 i'm sure that's fixed; i'm just not sure why; can you resolve that? | 18:25 |
| jog0 | sdague: think the tradeoff of a more concise recheck page is worth this edge case | 18:25 |
| uvirtbot | Launchpad bug 1214366 in openstack-ci "Could not find any downloads that satisfy the requirement markupsafe" [Undecided,New] | 18:25 |
| sdague | jog0: that's fine, I zeroed it as a review | 18:25 |
| mordred | jeblair: got it | 18:26 |
| sdague | plus it will mean I'm not the last person to touch that :) | 18:26 |
| clarkb | notmyname: portante: if you recheck tests in about 15 minutes we should get a lot more info | 18:27 |
| portante | clarkb: okay, thx | 18:27 |
| notmyname | clarkb: ah, ok. I already set one to reverify | 18:27 |
| jog0 | sdague: heh | 18:27 |
| clarkb | notmyname: https://jenkins02.openstack.org/job/gate-swift-devstack-vm-functional/368/ will capture the pip freeze but not the other things jeblair added | 18:28 |
| clarkb | notmyname: still potentially useful | 18:28 |
| notmyname | ok | 18:28 |
| clarkb | though we may not copy it off of the host... beause jenkins | 18:28 |
| jeblair | clarkb: we need the log copy to get the pip freeze | 18:28 |
| jog0 | sdague: also I figure someone will just open a new bug or search the bug list and find it | 18:28 |
| *** UtahDave has left #openstack-infra | 18:28 | |
| jog0 | sdague: there is this one too https://review.openstack.org/#/c/45511/ | 18:28 |
| *** UtahDave has joined #openstack-infra | 18:28 | |
| jeblair | notmyname, portante, clarkb: should we hold that node and try to debug on-host? | 18:29 |
| jog0 | Merge diplciate bugs | 18:29 |
| openstackgerrit | Alex Gaynor proposed a change to openstack-infra/config: Re-enable gating for hacking on PyPy https://review.openstack.org/45923 | 18:29 |
| openstackgerrit | Pierre Rognant proposed a change to openstack-infra/jenkins-job-builder: Fix plot plugin support https://review.openstack.org/45754 | 18:29 |
| Alex_Gaynor | jog0: ^ yay | 18:29 |
| clarkb | jeblair: yes please | 18:29 |
| openstackgerrit | Pierre Rognant proposed a change to openstack-infra/jenkins-job-builder: Fix plot plugin support https://review.openstack.org/45754 | 18:30 |
| clarkb | mordred: https://bugs.launchpad.net/openstack-ci/+bug/1118469 why not simply push a .md5 file? | 18:30 |
| uvirtbot | Launchpad bug 1118469 in openstack-ci "Upload integrity: jenkins should publish expected MD5" [Medium,Triaged] | 18:30 |
| clarkb | mordred: your suggestion seems complicated | 18:30 |
| mordred | clarkb: I think my suggestion is very old | 18:30 |
| jog0 | Alex_Gaynor: now I just need one more review to push out the 0.7.1 tag https://review.openstack.org/#/c/45764/ | 18:31 |
| Alex_Gaynor | jog0: seems like you need to ask clarkb :) | 18:31 |
| jeblair | clarkb: http://paste.openstack.org/show/46509/ | 18:31 |
| clarkb | ok I will update a suggestion for uploading a .md5 file. That seems super simple. Also I think that can be low-hanging-fruit in that it should be straightforward for not us to do | 18:31 |
| mordred | clarkb: we could totally publish an md5 file - however, there is no SSL path to the tarballs.o.o- so it's still tough to verify that you didn't get mitm'd | 18:31 |
| clarkb | mordred: is jenkins https much better? I suppose you only need to make a cert exception once | 18:32 |
| mordred | clarkb: well.... when I made the suggestoin, there was only one jenkins and it had a real cert | 18:32 |
| mordred | clarkb: BUT - I like the md5 file idea | 18:32 |
| clarkb | jeblair: I really liked how cookiecutter the nodepool command additions were | 18:32 |
| mordred | clarkb: I'm just saying we should consider adding an ssl cert on things we tell people to download from | 18:33 |
| notmyname | clarkb: just got to the swift tests part. they failed again | 18:33 |
| mordred | clarkb: you know, for the reasons | 18:33 |
| clarkb | mordred: ++ | 18:33 |
| jeblair | clarkb: yeah, i think i need to set the default config file location; i had to say "nodepool -c /etc/.... hold ..." instead of just "nodepool hold ..." which was the intent. but other than that... | 18:33 |
| mordred | jeblair: yes. I am a fan of those | 18:33 |
| jeblair | mordred: we don't tell people to download from tarballs.o.o | 18:34 |
| jeblair | mordred: we publish releases to launchpad | 18:34 |
| *** dcramer_ has joined #openstack-infra | 18:34 | |
| mordred | jeblair: although, as I was reading them, it made me think "wow, these should be a rest API and a client lib" ... | 18:34 |
| mordred | jeblair: ttx publishes to launchpad after we upload the artifacts to tarballs.o.o | 18:34 |
| jeblair | yes, but where ttx gets the tarball is different than where "people" get them. :) | 18:34 |
| mordred | jeblair: seems like there is a hole in the trust chain there for ttx | 18:35 |
| mordred | jeblair: sure. I misspoke then when I said "people" - I forgot, ttx isn't human ;) | 18:35 |
| jeblair | mordred: indeed there is, but solving that problem for ttx is not the same as solving it for everyone | 18:35 |
| jeblair | mordred: which is, in fact, why you stated your original idea the way you did | 18:35 |
| mordred | also, we do have tarballs.o.o links for oslo.config in our repos mid-cycle | 18:35 |
| mordred | jeblair: but point well taken | 18:36 |
| jeblair | mordred: and we serve git repos over the insecure git:// protocol | 18:36 |
| mordred | also, since the publish jobs run on jenkins.o.o | 18:36 |
| mordred | the original suggestion should still work | 18:36 |
| clarkb | mordred: you shouldn't calculate on publish though | 18:36 |
| clarkb | mordred: you should calculate in the -tarball job | 18:37 |
| mordred | this is true | 18:37 |
| jog0 | clarkb: can you review https://review.openstack.org/#/c/45764/ its one line | 18:37 |
| mordred | also, we can attach md5 files to our pypi uploads as well, iirc | 18:37 |
| jeblair | mordred: yep. i'm not saying i like the original suggestion or that we shouldn't put an https cert on tarballs; i just don't want to lose sight of the actual problem | 18:37 |
| mordred | jeblair: ++ | 18:37 |
| jog0 | and then the 0.7.x branch is ready for hackign | 18:37 |
| clarkb | mordred: we already do md5 stuff with pypi | 18:37 |
| *** MarkAtwood has joined #openstack-infra | 18:38 | |
| clarkb | jog0: approved | 18:38 |
| clarkb | jog0: see comment on the change though | 18:39 |
| adam_g | anyone know sup /w the 'run.py: error: no such option: -t' errors when running troveclient tests? https://jenkins.openstack.org/view/All/job/gate-python-troveclient-python27/52/console | 18:39 |
| clarkb | adam_g: testr does a test discovery pass that involves importing the python modules that it can find. I am willing to bet importing run.py creates that outpt | 18:40 |
| jog0 | clarkb: lets talk about that hacking requirements check later | 18:41 |
| jog0 | I like the idea | 18:41 |
| jog0 | there maybe some gotchas though | 18:41 |
| clarkb | jeblair: mordred: is there a reason that people should consume releases from launchpad? | 18:41 |
| jeblair | clarkb: because that's where our release manager publishes releases | 18:42 |
| clarkb | that seems like an implementation detail of how we do it and not necessarily a requirement (but I may be missing something important like gpg key stuff) | 18:42 |
| mordred | clarkb, adam_g: that is correct | 18:42 |
| *** thomasm has quit IRC | 18:42 | |
| clarkb | adam_g: easy work around is to restrict where testr looks for tests. You do that in .testr.conf | 18:43 |
| jeblair | clarkb: it means that tarballs.o.o is not required to have the sorts of things you would expect for an official release channel for a project | 18:43 |
| clarkb | jeblair: thats fair | 18:43 |
| jeblair | clarkb: we can add those things to tarballs.o.o, and then it could be offered as an option | 18:43 |
| mordred | clarkb, adam_g: or, fix the actual fail in that repo | 18:43 |
| *** dmakogon_ has joined #openstack-infra | 18:43 | |
| *** adalbas has quit IRC | 18:44 | |
| mordred | which is a) importing run shouldn't try to do whatever it's trying to do | 18:44 |
| mordred | and b) see a | 18:44 |
| * fungi just had a thought... an openstack/releases git repo where people can submit detached openpgp signatures for review, which end up being published alongside the tarballs to which they correspond. would be trivial to implement | 18:45 | |
| clarkb | fungi: mordred jeblair I left a comment on the bug that solicits more discussion | 18:46 |
| *** pcrews has quit IRC | 18:47 | |
| fungi | theoretical workflow would involve retrieving the tarball, signing it locally after verifying what needs veryfying, then committing the detached signature to the releases repo, and submitting for review | 18:47 |
| * fungi will mull over that for a bit | 18:47 | |
| clarkb | https://bugs.launchpad.net/openstack-ci/+bug/1129436 seems like that could be left open until we run our own apt mirror | 18:47 |
| uvirtbot | Launchpad bug 1129436 in openstack-ci "devstack failed: dpkg was interrupted" [Medium,Triaged] | 18:47 |
| jeblair | fungi: neat | 18:48 |
| clarkb | maybe make that a wishlist item? | 18:48 |
| fungi | that would also allow anyone who wanted to attest to a tarball to submit a signature for possible addition. this is similar to some of the suggestions which have been batted around for attestation on the next generation of pypi | 18:49 |
| *** dachary has quit IRC | 18:51 | |
| clarkb | jeblair: the devstack slave you held unpings | 18:52 |
| *** thomasm has joined #openstack-infra | 18:52 | |
| clarkb | jeblair: did the completion of testing mark it used or delete and it went away? | 18:52 |
| *** nicedice has quit IRC | 18:53 | |
| openstackgerrit | A change was merged to openstack-infra/devstack-gate: Add a pip freeze invocation to save with the logs https://review.openstack.org/45919 | 18:53 |
| *** nicedice has joined #openstack-infra | 18:53 | |
| notmyname | clarkb: jeblair: https://review.openstack.org/#/c/44570/ just failed, but still no logs | 18:53 |
| *** dachary has joined #openstack-infra | 18:53 | |
| clarkb | notmyname: yeah the job change to grab logs merged after that job started | 18:53 |
| clarkb | notmyname: at this point everything should be in. I would try again | 18:54 |
| notmyname | k | 18:54 |
| jeblair | clarkb: yeah, i think delete must have ignored hold, sorry. :( | 18:54 |
| clarkb | jeblair: oh well | 18:54 |
| jeblair | next run should get logs, yeah? | 18:54 |
| chmouel | clarkb, notmyname: I launched those functional tests on a fresh devstack which update the dependences and I could not reproduce either | 18:55 |
| clarkb | jeblair: yes | 18:55 |
| clarkb | chmouel: thanks good to know | 18:55 |
| chmouel | I don't use localhost tho but the public IP of my vm | 18:55 |
| adam_g | clarkb, mordred thanks | 18:55 |
| *** vipul is now known as vipul-away | 18:56 | |
| * jog0 just pushed the 0.7.1 tag for hacking | 18:57 | |
| *** adalbas has joined #openstack-infra | 19:00 | |
| *** fbo_away is now known as fbo | 19:00 | |
| *** jhesketh has joined #openstack-infra | 19:01 | |
| *** dcramer_ has quit IRC | 19:01 | |
| jeblair | meeting time! | 19:01 |
| notmyname | clarkb: jeblair: perhaps different error, but still no logs https://jenkins02.openstack.org/job/gate-swift-devstack-vm-functional/369/ | 19:03 |
| *** thomasm has quit IRC | 19:04 | |
| clarkb | notmyname: the logs are copied offsite | 19:04 |
| clarkb | notmyname: I think there is a new issue as a result of one of jeblairs changes ... | 19:05 |
| clarkb | .functests: line 6: nosetests: command not found because we are running in the test and not post test portion of devstack-gate | 19:05 |
| clarkb | jeblair: & | 19:05 |
| clarkb | I think we need to revert that particular change | 19:06 |
| *** dkehn_ is now known as dkehn | 19:06 | |
| hub_cap | jeblair: are yall in openstack-meeting or -alt? | 19:07 |
| pleia2 | -meeting | 19:07 |
| hub_cap | n/m i was scrolled up and didnt notice :) | 19:07 |
| hub_cap | thx pleia2 | 19:07 |
| jeblair | clarkb: go for it. it's still the wrong hook to use though. :) i wouldn't revert the log copying though | 19:07 |
| * hub_cap is dumb | 19:07 | |
| clarkb | jeblair: I htink it is the correct hook bceause it needs devstack to actually run | 19:07 |
| jeblair | clarkb: ah, ok. | 19:09 |
| *** vipul-away is now known as vipul | 19:10 | |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/config: Partial revert of 257b10d37405e28f3bf292e1041bf2c3010c02c6 https://review.openstack.org/45928 | 19:10 |
| clarkb | there we go | 19:10 |
| *** salv-orlando has quit IRC | 19:12 | |
| *** dcramer_ has joined #openstack-infra | 19:14 | |
| *** prad has quit IRC | 19:15 | |
| pabelanger | anteaya, Yes, you better re-assign it. Haven't had much free time lately to hack away on puppet modules | 19:16 |
| *** vipul is now known as vipul-away | 19:17 | |
| *** vipul-away is now known as vipul | 19:17 | |
| anteaya | pabelanger: very good thanks, I'll take it then, can you unassign yourself please? | 19:17 |
| *** fbo is now known as fbo_away | 19:21 | |
| *** HenryG has joined #openstack-infra | 19:24 | |
| *** Ryan_Lane has quit IRC | 19:26 | |
| *** gyee has quit IRC | 19:27 | |
| mordred | clarkb: yes. I believe that is the reason I did it originally in post-test | 19:28 |
| clarkb | mordred: can haz approval? | 19:28 |
| jeblair | BobBall: ping | 19:28 |
| *** fbo_away is now known as fbo | 19:30 | |
| *** markmcclain has quit IRC | 19:31 | |
| ttx | jeblair: how is 'where ttx gets the tarball is different than where "people" get them.' ? I download it from tarballs.o.o | 19:32 |
| *** vipul is now known as vipul-away | 19:32 | |
| clarkb | ttx: but then you push to launchpad and that is where people consume them with signatures and SSL | 19:32 |
| *** markmcclain has joined #openstack-infra | 19:32 | |
| ttx | clarkb: right -- although the tarball copy is more of a convenience. We could just publish signatures/ssl | 19:33 |
| ttx | (in a non-LP setup) | 19:33 |
| clarkb | ttx: right jeblair touched on that when I asked him why LP was required | 19:34 |
| clarkb | ttx: and basically it depends on being able to provide the things we need that LP provides | 19:34 |
| *** sandywalsh has quit IRC | 19:34 | |
| ttx | clarkb: agreed | 19:34 |
| notmyname | jeblair: clarkb: when the global requirements are updated, are they gated on all of the gate tests? | 19:36 |
| ttx | clarkb: pypi is another interesting "copy", basically where we push libraries | 19:36 |
| clarkb | notmyname: not all of them, only the devstack tests (we had missed your functional tests previously, but jeblair just added them so we are symmetric now) | 19:37 |
| *** yolanda has quit IRC | 19:37 | |
| *** pmoosh has quit IRC | 19:37 | |
| *** sergmelikyan has quit IRC | 19:42 | |
| *** ruhe has joined #openstack-infra | 19:44 | |
| *** MarkAtwood has quit IRC | 19:46 | |
| *** mdenny has quit IRC | 19:47 | |
| *** mdenny has joined #openstack-infra | 19:48 | |
| *** thomasm has joined #openstack-infra | 19:48 | |
| *** sandywalsh has joined #openstack-infra | 19:49 | |
| *** pblaho has quit IRC | 19:51 | |
| *** annegentle has quit IRC | 19:53 | |
| openstackgerrit | Andreas Jaeger proposed a change to openstack-infra/config: Move admin-guide-cloud to docs.o.o/admin-guide-cloud/content https://review.openstack.org/45938 | 19:53 |
| *** annegentle has joined #openstack-infra | 19:53 | |
| *** toddmorey has joined #openstack-infra | 19:54 | |
| *** MarkAtwood has joined #openstack-infra | 19:55 | |
| *** afazekas has quit IRC | 19:55 | |
| *** dolphm has joined #openstack-infra | 19:57 | |
| openstackgerrit | A change was merged to openstack-infra/config: Partial revert of 257b10d37405e28f3bf292e1041bf2c3010c02c6 https://review.openstack.org/45928 | 19:59 |
| jog0 | jeblair: now that sdague commented on https://review.openstack.org/#/c/45813/ ... | 20:01 |
| jog0 | I am hoping to get the changes into recheck so we can use them during stablization | 20:01 |
| jhesketh | jeblair: good meeting | 20:02 |
| clarkb | zaro: you might just try hacking something together for the other ACL categories | 20:02 |
| notmyname | portante: clarkb: jeblair: failure logs http://logs.openstack.org/97/45897/1/gate/gate-swift-devstack-vm-functional/9207838/ | 20:02 |
| jeblair | jhesketh: thanks! glad you could stop by! | 20:02 |
| jhesketh | jeblair: I was hoping to discuss some log stuff with you during it but didn't quite realise how it would work | 20:02 |
| clarkb | notmyname: 45928 should actually get the tests to run and fail as before | 20:03 |
| jeblair | notmyname: next puppet update should fix that | 20:03 |
| jhesketh | maybe we should discuss it here first before in such a meeting anyway | 20:03 |
| jeblair | jhesketh: no problem; if you ever want to discuss anything, the agenda is on a wiki page that you may feel free to edit https://wiki.openstack.org/wiki/Meetings/InfraTeamMeeting | 20:04 |
| dolphm | :( is it possible for requirements/global-requirements.txt to be bypassed somehow? | 20:04 |
| jeblair | jhesketh: i think it might be best for me to send an email about the log thing | 20:04 |
| jhesketh | yeah, I decided since I hadn't attended before I should get a feel first before adding an item (although I have been reading the logs+minutes for a few meetings) | 20:04 |
| clarkb | dolphm: not supposed to be. what is wrong? | 20:04 |
| dolphm | global-requirements.txt specified six<1.4.0 but keystone is failing checks because 1.4.1 is installed https://bugs.launchpad.net/ceilometer/+bug/1219694/comments/3 | 20:05 |
| uvirtbot | Launchpad bug 1219694 in ceilometer/grizzly "Jenkins broken due to incompatibility between six 1.4.0 and wsme 0.5b2" [Critical,Confirmed] | 20:05 |
| jeblair | jhesketh: it's been meetings all morning for me, and i haven't had a chance to do anything with email yet | 20:05 |
| jhesketh | jeblair: sure, I was meaning to do the same | 20:05 |
| jhesketh | no worries, and definitely no hurry | 20:05 |
| clarkb | jog0: is dolphm's thing related to hacking? | 20:05 |
| jeblair | jhesketh: are you on the openstack-infra@lists.o.o mailing list? | 20:05 |
| clarkb | dolphm: I think that is an issue that jog0 is working to correct | 20:06 |
| clarkb | dolphm: and may even be fixed at this point | 20:06 |
| clarkb | jhesketh: jeblair: log thing? | 20:06 |
| zaro | clarkb: that's going to be interesting. but i don't see another alternative. | 20:07 |
| jhesketh | jeblair: to be honest, I can't remember and I'm on the wrong computer :p | 20:07 |
| jhesketh | but I believe so | 20:07 |
| jhesketh | the archives look familiar | 20:07 |
| jhesketh | clarkb: so there's this vague thing where zuul should handle logs | 20:07 |
| jog0 | Lear it may be, but hacking 0.7.1 is out which fixes that | 20:07 |
| dolphm | clarkb: there's 3 failed runs showing in zuul right now | 20:07 |
| jhesketh | I think jeblair has some more formed thoughts on it but it has been something I've been considering too | 20:08 |
| *** rfolco has joined #openstack-infra | 20:08 | |
| clarkb | http://pypi.openstack.org/openstack/hacking/ new version is there | 20:08 |
| jeblair | clarkb: mikal mentioned to me that he and jhesketh have log storing needs and it sounded like their needs might be very similar to the 'log storing/retrieving' thing (to replace scp-to-static.o.o) | 20:08 |
| jhesketh | specifically because I've been working on a non-jenkins based worker for zuul | 20:08 |
| portante | clarkb: so where are we at with swift functional test failures? | 20:08 |
| jeblair | which we also need for non-jenkins based workers for zuul :) | 20:08 |
| *** melwitt1 has joined #openstack-infra | 20:08 | |
| *** melwitt has quit IRC | 20:08 | |
| clarkb | portante: pushed a fix to one of the fixes :) now waiting for that to get applied so that we get real logs | 20:09 |
| jeblair | jhesketh, clarkb: so i'll send an email with a summary of things we've brainstormed about on the subject to the openstack-infra list | 20:09 |
| clarkb | jeblair: interesting | 20:09 |
| portante | okay, so will that take another 15 minutes? | 20:09 |
| clarkb | portante: you may already be good at this point | 20:09 |
| clarkb | portante: jobs started now I expect to have the new config | 20:09 |
| jeblair | i think the list is a good place for that, as it might be a bit of an evolving design discussion | 20:09 |
| portante | okay | 20:09 |
| clarkb | jeblair: ++ | 20:10 |
| jhesketh | jeblair: sounds good | 20:10 |
| clarkb | apparently we don't do a pip freeze in the pep8 job :/ | 20:11 |
| clarkb | jog0: https://jenkins01.openstack.org/job/gate-keystone-pep8/706/console according to timestamps there and on the mirror that job should've used hacking 0.7.1 | 20:12 |
| clarkb | oh unless maybe old hacking was installed because of a pin? | 20:12 |
| jog0 | clarkb: it shouldn't be we pin between 0.7 and 0.8 | 20:13 |
| jog0 | but not sure if everything is there yet | 20:13 |
| *** vipul-away is now known as vipul | 20:13 | |
| clarkb | portante: notmyname: https://jenkins02.openstack.org/job/gate-swift-devstack-vm-functional/372/console just started. I am watching it | 20:13 |
| *** sarob has quit IRC | 20:14 | |
| mordred | jog0: we should make a gate job for hacking that runs proposed hacking change against trunk of all projects that gate on hacking | 20:14 |
| jog0 | clarkb: the pinning looks right | 20:14 |
| jog0 | mordred: we have | 20:14 |
| mordred | jog0: we have? | 20:14 |
| jog0 | http://logs.openstack.org/24/45824/3/check/gate-hacking-integration-nova/7e3fed0/console.html | 20:15 |
| jog0 | it only does nova right now | 20:15 |
| jog0 | but we can add others | 20:15 |
| jog0 | mordred: it doesn't do any deps checks though | 20:15 |
| clarkb | dolphm: I am trying to reproduce locally to get a better handle on this | 20:16 |
| jog0 | Clark not surwe whats going on | 20:16 |
| clarkb | jog0: yeah I am going to work on reproducing | 20:17 |
| dolphm | clarkb: thank you | 20:17 |
| jog0 | clarkb: 2013-09-10 20:03:17.120 | pkg_resources.VersionConflict: (six 1.4.1 (/home/jenkins/workspace/gate-keystone-pep8/.tox/pep8/lib/python2.7/site-packages), Requirement.parse('six<1.4.0')) | 20:17 |
| jog0 | so yeah we pinned six because ... | 20:18 |
| jog0 | https://review.openstack.org/#/c/44656/ | 20:19 |
| *** sarob has joined #openstack-infra | 20:19 | |
| jog0 | and https://review.openstack.org/#/c/45641/ | 20:20 |
| jog0 | dolphm clarkb ^ | 20:20 |
| *** dprince has quit IRC | 20:22 | |
| *** ArxCruz has quit IRC | 20:23 | |
| * jog0 hates dependency hell | 20:24 | |
| mordred | jog0: you LOVE dependency hell | 20:24 |
| clarkb | jog0: fails that way with new hacking (0.7.1) | 20:25 |
| jog0 | mordred: we should get to the point where hacking is autosynced with global openstack reqs | 20:25 |
| clarkb | something else has unpinned six | 20:25 |
| mordred | jog0: ++ | 20:25 |
| clarkb | jog0: it is easy, just add the job :) | 20:25 |
| * jog0 looks ata mordred to make the magic | 20:25 | |
| *** markmcclain has quit IRC | 20:25 | |
| jog0 | well we need the job for the stable branch | 20:25 |
| * mordred wrote the job already | 20:25 | |
| jog0 | 'stable' | 20:25 |
| clarkb | http://paste.openstack.org/show/46535/ somewhere in that list of stuff we pull in six > 1.40 | 20:26 |
| mordred | https://review.openstack.org/#/c/40068/ | 20:27 |
| mordred | jog0: ++ | 20:27 |
| mordred | jog0: ^^ I mean | 20:27 |
| jog0 | mordred: hacking 0.7.x branch is missing from that | 20:27 |
| clarkb | http://git.openstack.org/cgit/openstack/oslo.config/tree/requirements.txt?id=1.2.0a3 | 20:27 |
| clarkb | I think we have a winner | 20:27 |
| mordred | jog0: yah. I'm just saying, those are the mechanics | 20:27 |
| jog0 | mordred: pretty slick I like it | 20:28 |
| clarkb | I am going to propose something different | 20:28 |
| clarkb | we revert the openstack/requirements six requirement | 20:28 |
| clarkb | and make ceilometer and wsme play nice | 20:28 |
| clarkb | because | 20:28 |
| clarkb | ceilometer + wsme is a small web | 20:28 |
| clarkb | six is a massive web | 20:28 |
| * dhellmann perks up his ears | 20:28 | |
| dhellmann | "play nice"? | 20:29 |
| portante | clarkb: notmyname: job failed, not sure what was supposed to look different in the logs ... | 20:29 |
| portante | https://jenkins02.openstack.org/job/gate-swift-devstack-vm-functional/372/console | 20:29 |
| clarkb | dhellmann: https://review.openstack.org/#/c/44656/ | 20:29 |
| jog0 | clarkb: any solution sounds good to me | 20:29 |
| mordred | dhellmann: we're stubbing our toes on six pinning | 20:29 |
| *** salv-orlando has joined #openstack-infra | 20:29 | |
| clarkb | and I know that getting a new oslo.config release out is going to be hard particularly with markmc busy | 20:29 |
| dhellmann | clarkb: I think we have a fix for that in wsme, and just need to cut another release | 20:29 |
| jeblair | portante: we new have logs: http://logs.openstack.org/44/45944/1/check/gate-swift-devstack-vm-functional/877a4a9/logs/ | 20:30 |
| jeblair | s/new/now/ | 20:30 |
| portante | huh, xattr at 0.6.2 ... | 20:31 |
| *** sdake_ has quit IRC | 20:32 | |
| portante | so we test with the latest python-*client versions (-dev, -master, etc.) | 20:33 |
| clarkb | portante: yes | 20:33 |
| clarkb | portante: so that a release doesn't break everything | 20:33 |
| *** gyee has joined #openstack-infra | 20:34 | |
| *** changbl has joined #openstack-infra | 20:34 | |
| portante | k | 20:34 |
| dims | clarkb, what do you need with oslo.config? | 20:34 |
| *** bswrchrd has joined #openstack-infra | 20:34 | |
| clarkb | dims: possibly to pin six, but I am working with ceilo folks to take a different direction and remove the six pin | 20:35 |
| dims | clarkb, ok. if something is needed in oslo, i can pitch in | 20:36 |
| *** Ryan_Lane has joined #openstack-infra | 20:36 | |
| jog0 | clarkb: is this breaking everything right now? if so lets unpin in hacking right now | 20:36 |
| clarkb | jog0: not everything, just things that have dependencies that install six before a thing with six pinned is installed | 20:37 |
| boris-42 | guys does anybody know about this bug | 20:37 |
| boris-42 | pkg_resources.VersionConflict: (six 1.4.1 (/home/jenkins/workspace/gate-rally-pep8/.tox/pep8/lib/python2.7/site-packages), Requirement.parse('six<1.4.0')) | 20:37 |
| jog0 | clarkb: so is it worth it to get a quick dirty fix hacking wise now | 20:38 |
| jog0 | boris-42: ^ | 20:38 |
| mordred | boris-42: yup. it's the error of the moment and is being worked | 20:38 |
| boris-42 | oh nice thanks guys!! | 20:38 |
| portante | clarkb: what added information is in those logs, the pip freeze? | 20:38 |
| jeblair | portante: also all of the service logs; the entire http://logs.openstack.org/44/45944/1/check/gate-swift-devstack-vm-functional/877a4a9/logs/ directory | 20:39 |
| *** sdake_ has joined #openstack-infra | 20:39 | |
| *** sdake_ has joined #openstack-infra | 20:39 | |
| clarkb | I have sent mail to ceilo folks asking about the feasibility of fixing this on the ceilo side so that we can unpin six everywhere | 20:40 |
| fungi | clarkb: 1700 utc this saturday works fine for me. we'll get 'em moved then | 20:40 |
| clarkb | fungi: great | 20:40 |
| jeblair | fungi: thanks | 20:40 |
| * fungi re-lurks for meetup, already in progress ;) | 20:40 | |
| jog0 | revert six pin in 0.7.x: https://review.openstack.org/45950 | 20:42 |
| jog0 | clarkb mordred ^ once merged will cut 0.7.2 | 20:42 |
| clarkb | jog0: lets hold off on merging until we know if we can fix this in ceilo | 20:43 |
| portante | clarkb: can we get those functional tests runs to happen where nose saves the stdout and stderr of each test and prints it out with the failures? | 20:43 |
| jog0 | clarkb: fine with me | 20:43 |
| clarkb | jog0: otherwise we may need to make a great big pinning everywhere :) | 20:43 |
| jog0 | -1 or 2 the patch | 20:43 |
| portante | clarkb: and can we get nose to use -v there? | 20:43 |
| jog0 | i just wanted it up just in case | 20:43 |
| jog0 | figure break one project instead of >1 | 20:43 |
| clarkb | Alex_Gaynor: ^ can you take a look at the six stuff? any opinions? | 20:44 |
| mordred | portante: we basically just run .functest - so whatever that does | 20:45 |
| clarkb | portante: we should be able to by updating the job http://git.openstack.org/cgit/openstack-infra/config/tree/modules/openstack_project/files/jenkins_job_builder/config/swift-functional.yaml#n22 assuming the functests script knows how to do that | 20:45 |
| clarkb | portante: also we should work on getting you guys on testr once Icehouse opens as we currently do all of that for all of the testr test runs | 20:46 |
| mordred | ++ | 20:47 |
| mordred | we'll also want that so that we can do "zuul-processes-test-run-output-streaming" everywhere | 20:47 |
| clarkb | portante: looks like you will need to update .functests to get that info | 20:48 |
| clarkb | it isn't very configurable | 20:48 |
| clarkb | actually it does use $@ | 20:49 |
| clarkb | so .functests -v | 20:50 |
| clarkb | according to the nose docs stdout capture is enabled by default | 20:50 |
| clarkb | we don't seem to override it in setup.cfg | 20:50 |
| *** sarob has quit IRC | 20:50 | |
| portante | hmm, wonder where that output is coming from then | 20:51 |
| clarkb | portante: stderr? looking at a nose issue they don't capture stderr | 20:52 |
| clarkb | I would link to it but I had to open it in google's cache because the direct link won't open | 20:52 |
| portante | clarkb, mordred: can you update the yaml to add -v to functests | 20:53 |
| clarkb | portante: I think we needt to set --verbosity=3 | 20:54 |
| *** ericw has quit IRC | 20:54 | |
| clarkb | -v is the same as 2 which is set in setup.cfg. I will propose that change | 20:54 |
| portante | k thanks | 20:54 |
| portante | clarkb, in the mean time, I am going to modify our func test client to report the errors that it is encountering | 20:55 |
| *** adalbas has quit IRC | 20:57 | |
| annegentle | hey infra team, we're seeing a ton of FATAL ERROR cannot assign instance of org.apache.fop.fonts.FontTriplet to field org.apache.fop.fonts.FontTriplet.style | 20:57 |
| annegentle | it's causing a lot of rechecks. An example is at http://logs.openstack.org/24/45924/1/gate/gate-openstack-manuals-validate/94d5ee2/console.html | 20:58 |
| openstackgerrit | Khai Do proposed a change to openstack-infra/config: Diff between installed packages and requirements https://review.openstack.org/45329 | 21:00 |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/config: Run swift functional tests verbosely. https://review.openstack.org/45953 | 21:00 |
| clarkb | jeblair: fungi mordred ^ | 21:01 |
| mordred | call for tech is 6 | 21:01 |
| mordred | actor call at 7 | 21:01 |
| mordred | gah | 21:01 |
| mordred | well, you all know when the actor call is now | 21:01 |
| openstackgerrit | Khai Do proposed a change to openstack-infra/config: Diff between installed packages and requirements https://review.openstack.org/45329 | 21:01 |
| *** prad has joined #openstack-infra | 21:01 | |
| clarkb | annegentle: that stacktrace is being generated in the cloud docs plugin | 21:02 |
| clarkb | annegentle: I would check with David | 21:02 |
| annegentle | clarkb: okay will do | 21:02 |
| *** bodepd has quit IRC | 21:02 | |
| clarkb | annegentle: 2013-09-10 18:58:29.951 | at com.rackspace.cloud.api.docs.builders.PDFBuilder.postProcessResult(PDFBuilder.java:245) being the important line in the long | 21:03 |
| clarkb | s/long/log/ | 21:03 |
| *** bodepd has joined #openstack-infra | 21:03 | |
| annegentle | clarkb: oo thanks for spelunking for me | 21:04 |
| jgriffith | anything funky going on, or is this just a reverify no bug: http://logs.openstack.org/68/41868/19/gate/gate-cinder-pep8/b6ad019/console.html | 21:04 |
| clarkb | jgriffith: major funkyness | 21:04 |
| jgriffith | clarkb: can't figure out where the req's conflict came in from the rebase | 21:05 |
| clarkb | jgriffith: long story short we pinned six in openstack/requirements to < 1.4.0 to accomodate ceilometer, but in doing so made the hurt spread elsewhere | 21:05 |
| *** rfolco has quit IRC | 21:05 | |
| jgriffith | Ahh :) | 21:05 |
| clarkb | jgriffith: currently working to unpin six and fix the problems on the ceilo side so that everyone else can continue on as usual | 21:05 |
| jgriffith | clarkb: sounds good, thanks for the details | 21:06 |
| jog0 | clarkb: lets mege this https://review.openstack.org/#/c/45950/ | 21:06 |
| clarkb | jog0: ok it has my +2 I will let you hit the go button | 21:06 |
| *** matty_dubs is now known as matty_dubs|gone | 21:07 | |
| jog0 | clarkb: kk | 21:07 |
| clarkb | portante: https://review.openstack.org/#/c/45953/1/modules/openstack_project/files/jenkins_job_builder/config/swift-functional.yaml your review on that is valuable as well | 21:07 |
| *** dmakogon_ has quit IRC | 21:08 | |
| *** rockyg has joined #openstack-infra | 21:09 | |
| mordred | stevebaker, dstufft: pip mucking with file mode things | 21:09 |
| stevebaker | dstufft: hey | 21:10 |
| *** adalbas has joined #openstack-infra | 21:11 | |
| stevebaker | oh, hang on. It looks like https://github.com/pypa/pip/blob/develop/pip/util.py#L590 already has some mode fixing | 21:11 |
| stevebaker | let me test that | 21:12 |
| stevebaker | ok, the problem is fixed in the develop branch, nice! | 21:13 |
| * stevebaker waits for 1.4.2 | 21:13 | |
| *** dolphm has left #openstack-infra | 21:15 | |
| *** sarob has joined #openstack-infra | 21:15 | |
| notmyname | portante: changes are pending to verbosely run functional tests? is that where we are? | 21:17 |
| jeblair | notmyname: yes https://review.openstack.org/#/c/45953/1 | 21:18 |
| *** vipul is now known as vipul-away | 21:18 | |
| jeblair | notmyname: we would like a swift person to look at that | 21:18 |
| jeblair | portante: may have been working on a similar change to .functests itself, from what i read | 21:19 |
| *** adalbas has quit IRC | 21:19 | |
| notmyname | jeblair: lgtm | 21:20 |
| *** Ryan_Lane has quit IRC | 21:22 | |
| *** Ryan_Lane1 has joined #openstack-infra | 21:22 | |
| * jog0 cut hacking 0.7.2 now with no six pinning | 21:24 | |
| *** Ryan_Lane1 is now known as Ryan_Lane | 21:24 | |
| *** pcm_ has quit IRC | 21:24 | |
| *** Ryan_Lane has quit IRC | 21:24 | |
| *** Ryan_Lane has joined #openstack-infra | 21:24 | |
| notmyname | jeblair: so that needs to land then a test suite rerun? | 21:24 |
| jeblair | notmyname: yes; i approved it; puppet will need to reconfigure the job (so it'll take a bit) | 21:24 |
| notmyname | jeblair: bit=5 minutes? bit=1 hour? | 21:25 |
| jeblair | notmyname: (for immediate results, propose a change to swift that makes that the default; you don't have to merge it, the check run will use the change) | 21:25 |
| jeblair | notmyname: 0-25 minutes | 21:25 |
| notmyname | jeblair: k | 21:26 |
| jeblair | notmyname: some folks are working on making that instantaneous | 21:26 |
| openstackgerrit | A change was merged to openstack-infra/config: Run swift functional tests verbosely. https://review.openstack.org/45953 | 21:26 |
| notmyname | jeblair: I'll wait for the job | 21:26 |
| *** KLevenstein1 has joined #openstack-infra | 21:26 | |
| *** KLevenstein has quit IRC | 21:30 | |
| clarkb | jeblair: I also think some of the log weirdness may be related to devstack-gate logging | 21:31 |
| jeblair | clarkb: log weirdness? | 21:31 |
| clarkb | jeblair: I am trying to read up on how it works to make sure we are doing the right thing for post_test_hook | 21:31 |
| *** sarob has quit IRC | 21:31 | |
| clarkb | jeblair: http://logs.openstack.org/01/44901/4/check/gate-swift-devstack-vm-functional/9d8245c/console.html around 2013-09-10 20:33:24.988 we appear to have overlapping streams | 21:31 |
| clarkb | jeblair: and the actual call to .functest isn't logged | 21:32 |
| clarkb | I don't think this is the reason for the failures, just makes it a little harder to read what is going on | 21:32 |
| *** HenryG has quit IRC | 21:32 | |
| clarkb | jeblair: I think we might want to redirect the output of pre and post hooks and set -oxtrace like we do for setupworkspace | 21:33 |
| jeblair | clarkb: wfm | 21:33 |
| *** jhesketh has quit IRC | 21:33 | |
| *** thomasm has quit IRC | 21:34 | |
| *** flaper87 is now known as flaper87|afk | 21:37 | |
| *** vipul-away is now known as vipul | 21:38 | |
| *** prad has quit IRC | 21:39 | |
| *** mriedem has quit IRC | 21:41 | |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/devstack-gate: Log pre and post test hooks to file and stdout. https://review.openstack.org/45956 | 21:42 |
| clarkb | jeblair: something like ^ | 21:42 |
| clarkb | jeblair: any chance we can get that nodepool hold bug fixed as well so that holding a slave is an optio? | 21:43 |
| jeblair | clarkb: sure, i'll work on that in a sec. | 21:43 |
| jeblair | clarkb: we should gate d-g on the functests run | 21:43 |
| clarkb | jeblair: ++ but I think we should get them passing again (debugging may be hard otherwise) | 21:44 |
| *** ruhe has quit IRC | 21:44 | |
| jeblair | yes; just thinking that it will help exercise this part of the script | 21:44 |
| *** fbo is now known as fbo_away | 21:47 | |
| *** Ryan_Lane has quit IRC | 21:47 | |
| *** Ryan_Lane1 has joined #openstack-infra | 21:47 | |
| *** flaper87|afk is now known as flaper87 | 21:47 | |
| *** flaper87 is now known as flaper87|afk | 21:48 | |
| *** thomasm has joined #openstack-infra | 21:49 | |
| *** Ryan_Lane has joined #openstack-infra | 21:49 | |
| *** Ryan_Lane1 has quit IRC | 21:49 | |
| *** cppcabrera has left #openstack-infra | 21:52 | |
| *** prad has joined #openstack-infra | 21:54 | |
| *** rockyg has quit IRC | 21:55 | |
| *** amotoki_ has joined #openstack-infra | 22:00 | |
| *** amotoki_ has quit IRC | 22:00 | |
| *** amotoki_ has joined #openstack-infra | 22:00 | |
| *** prad has quit IRC | 22:01 | |
| *** krtaylor has joined #openstack-infra | 22:02 | |
| *** burt has quit IRC | 22:02 | |
| *** boris-42 has quit IRC | 22:03 | |
| clarkb | notmyname: portante: I just started https://jenkins02.openstack.org/job/gate-swift-devstack-vm-functional/375/console | 22:03 |
| clarkb | that should include verbose functests | 22:04 |
| *** whayutin_ has quit IRC | 22:04 | |
| *** toddmorey has quit IRC | 22:06 | |
| *** dkranz has quit IRC | 22:06 | |
| *** SergeyLukjanov has quit IRC | 22:06 | |
| notmyname | clarkb: thanks | 22:06 |
| *** kgriffs has joined #openstack-infra | 22:10 | |
| openstackgerrit | Kiall Mac Innes proposed a change to openstack-infra/jenkins-job-builder: Ensure Conditional Builder produces valid class paths https://review.openstack.org/45960 | 22:10 |
| *** sarob has joined #openstack-infra | 22:10 | |
| clarkb | notmyname: the output looks a lot better now | 22:10 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/nodepool: Add default location for config file https://review.openstack.org/45961 | 22:11 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/nodepool: Fix fake provider https://review.openstack.org/45962 | 22:11 |
| notmyname | clarkb: boom | 22:11 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/nodepool: Fix HOLD state https://review.openstack.org/45963 | 22:11 |
| *** jhesketh_ has joined #openstack-infra | 22:11 | |
| *** Ryan_Lane has quit IRC | 22:12 | |
| *** Ryan_Lane has joined #openstack-infra | 22:13 | |
| clarkb | is it trying to talk to keystone for the account info? | 22:13 |
| *** UtahDave has quit IRC | 22:15 | |
| notmyname | perhaps if devstack uses keystone | 22:16 |
| *** sarob has quit IRC | 22:16 | |
| *** lcestari has quit IRC | 22:16 | |
| *** prad has joined #openstack-infra | 22:17 | |
| *** rnirmal has quit IRC | 22:17 | |
| jeblair | this is starting to sound like an integration test | 22:19 |
| *** SergeyLukjanov has joined #openstack-infra | 22:19 | |
| clarkb | jeblair: I am trying to decide if it is, there is code that creates account containers not sure if that is providing user auth through swift directly or not | 22:19 |
| *** SergeyLukjanov has quit IRC | 22:20 | |
| notmyname | jeblair: clarkb: for swift dev environments, we use tempauth for this (that's why we have tempauth--to lower dependency and integration issues). | 22:20 |
| jeblair | clarkb, mordred: if one/both of you could review a few quick changes: https://review.openstack.org/#/q/status:open+project:openstack-infra/nodepool,n,z | 22:20 |
| jeblair | clarkb, mordred: i'll restart nodepool to fix the hold bug | 22:20 |
| notmyname | but, tempauth is not The OpenStack Way (tm) | 22:20 |
| clarkb | jeblair: looking | 22:21 |
| *** prad has quit IRC | 22:21 | |
| *** lcestari has joined #openstack-infra | 22:21 | |
| *** torgomatic has joined #openstack-infra | 22:22 | |
| clarkb | jeblair: lgtm. nice and simple | 22:22 |
| openstackgerrit | A change was merged to openstack-infra/nodepool: Fix fake provider https://review.openstack.org/45962 | 22:25 |
| openstackgerrit | A change was merged to openstack-infra/nodepool: Add default location for config file https://review.openstack.org/45961 | 22:25 |
| openstackgerrit | A change was merged to openstack-infra/nodepool: Fix HOLD state https://review.openstack.org/45963 | 22:25 |
| notmyname | clarkb: even so, it looks like the keystone endpoints it's using (actually, it looks like it got back a good service catalog) are at localhost | 22:26 |
| *** thomasm has quit IRC | 22:26 | |
| clarkb | notmyname: https://review.openstack.org/#/c/45201/ any chance that is related | 22:27 |
| clarkb | there are ceilometer warnigns in the proxy log | 22:27 |
| *** melwitt1 has quit IRC | 22:29 | |
| *** melwitt has joined #openstack-infra | 22:29 | |
| clarkb | I wonder if that extra sed is breaking the config line | 22:31 |
| clarkb | jeblair: can you hold a swift functional test test node once you have nodepool updated? | 22:32 |
| jeblair | clarkb: will do | 22:32 |
| jeblair | clarkb: the important one of those 3 changes has not shown up here: https://git.openstack.org/cgit/openstack-infra/nodepool/log/ | 22:33 |
| jeblair | clarkb: it _has_ appeared here: https://github.com/openstack-infra/nodepool/commits/master | 22:33 |
| notmyname | clarkb: what cm errors are you seeing? in http://logs.openstack.org/01/44901/4/check/gate-swift-devstack-vm-functional/9d8245c/logs/screen-s-proxy.txt.gz there are lots of proxy responses that seem to indicate that good things are happening | 22:34 |
| clarkb | notmyname: its jus ta warning right at the beginning of the log | 22:34 |
| jeblair | (i'm looking at gerrit's error log) | 22:34 |
| *** jpich has quit IRC | 22:36 | |
| *** thomasm has joined #openstack-infra | 22:39 | |
| jhesketh_ | jeblair: at the moment with the logs, do we apply any formatting to any of them? | 22:39 |
| clarkb | jhesketh_: we do | 22:39 |
| *** jcoufal has quit IRC | 22:39 | |
| jeblair | jhesketh_: the on-the-fly htmlization in that mod_python script | 22:39 |
| clarkb | jhesketh_: http://logs.openstack.org/01/44901/4/check/gate-swift-devstack-vm-functional/9d8245c/logs/screen-n-crt.txt.gz for example | 22:39 |
| jhesketh_ | righto, which logs/gates does that apply to | 22:40 |
| notmyname | clarkb: I don't think it's cm | 22:40 |
| notmyname | clarkb: "Sep 10 20:33:59 devstack-precise-hpcloud-az2-245229 rsyslogd-2177: imuxsock lost 4065 messages from pid 21437 due to rate-limiting" is odd (in the syslog) | 22:40 |
| *** sarob has joined #openstack-infra | 22:40 | |
| clarkb | jhesketh_: all of them, the script will bail if it doesn't recognize the format | 22:40 |
| jhesketh_ | ah okay | 22:40 |
| jhesketh_ | what would you say is the largest text file that actually gets formatted? | 22:41 |
| clarkb | notmyname: I think that is syslog complaining that it got ratelimited | 22:41 |
| notmyname | clarkb: right. but it means that we might not have all the data | 22:41 |
| clarkb | jhesketh_: nova cpu, nova api, or nova computer. They can get into the several megabytes | 22:41 |
| clarkb | notmyname: well we don't use syslog | 22:41 |
| jhesketh_ | okay | 22:41 |
| clarkb | notmyname: for this reason | 22:41 |
| jhesketh_ | clarkb: so I | 22:42 |
| jhesketh_ | *eh, fail | 22:42 |
| jhesketh_ | so I have been working on some private CI testing that currently just stores logs on disk | 22:42 |
| jhesketh_ | and serves them as is | 22:42 |
| *** harlowja has quit IRC | 22:42 | |
| jhesketh_ | however then I use javascript to apply formatting using regex | 22:43 |
| *** thomasm has quit IRC | 22:43 | |
| jhesketh_ | this way I don't need to store mutliple copies and I figure javascript is good enough these days to handle it (coupled with fast CPU's) | 22:43 |
| jhesketh_ | they are much much smaller in size though | 22:43 |
| jhesketh_ | but it could be an option up to a certain size before a log server would need to cache html formatted versions | 22:44 |
| clarkb | jhesketh_: I have killed my browser with these log files and that is without javascript doing stuff with them | 22:44 |
| *** dkliban has quit IRC | 22:44 | |
| jhesketh_ | but then agian, I'm not sure we care about disk space | 22:44 |
| clarkb | however js might be smart enough to only transform a window | 22:44 |
| jhesketh_ | heh, fair enough | 22:44 |
| *** changbl has quit IRC | 22:44 | |
| clarkb | so that we aren't doing it for the whole thing | 22:44 |
| jhesketh_ | clarkb: what do you mean? | 22:44 |
| jhesketh_ | oh right, I see | 22:44 |
| *** mrodden has quit IRC | 22:44 | |
| jhesketh_ | that's not a terrible idea | 22:45 |
| clarkb | jhesketh_: yeah I can see us doing that | 22:45 |
| *** kgriffs has left #openstack-infra | 22:46 | |
| *** MarkAtwood2 has joined #openstack-infra | 22:46 | |
| *** MarkAtwood has quit IRC | 22:47 | |
| notmyname | clarkb: portante: I've got about 10 more minutes today to look at this. any ideas on next steps? | 22:51 |
| jeblair | clarkb: http://paste.openstack.org/show/46559/ | 22:51 |
| jeblair | run https://jenkins02.openstack.org/job/gate-swift-devstack-vm-functional/377/console | 22:51 |
| clarkb | notmyname: I am poking at making make_requests log actual useful info | 22:52 |
| *** amotoki_ has quit IRC | 22:52 | |
| openstackgerrit | Steve Baker proposed a change to openstack/requirements: Add diskimage-builder and tripleo-image-elements https://review.openstack.org/45967 | 22:52 |
| *** pcrews has joined #openstack-infra | 22:52 | |
| notmyname | clarkb: once that run finishes and we still have access to the box, then running the functests by hand would be a good exercise | 22:53 |
| clarkb | notmyname: yup | 22:53 |
| *** Ryan_Lane has quit IRC | 22:54 | |
| *** Ryan_Lane has joined #openstack-infra | 22:55 | |
| mikal | jeblair: I just want to confirm... It was linuxcon europe that you and mordred are going to, right? | 22:55 |
| clarkb | notmyname: portante https://review.openstack.org/45970 something like that | 22:57 |
| jeblair | mikal: yes | 22:57 |
| mikal | jeblair: sweet | 22:57 |
| openstackgerrit | Pierre Rognant proposed a change to openstack-infra/jenkins-job-builder: Fix plot plugin support https://review.openstack.org/45754 | 22:58 |
| jeblair | clarkb: 2013-09-10 23:00:57,430 INFO nodepool.NodeCompleteThread: Node id: 245709 is complete but in HOLD state | 23:01 |
| jeblair | should be good to go | 23:01 |
| clarkb | jeblair: woot | 23:01 |
| clarkb | notmyname: did you want to jump on that host? | 23:01 |
| clarkb | portante: ^ | 23:01 |
| stevebaker | hey, could I please have some eyeballs on https://review.openstack.org/#/c/45967/ so I can continue with heat-slow testing? | 23:01 |
| notmyname | clarkb: do you want that upstream? you could patch it locally on the hold node. I'd also recommend adding the try_count and the request itself | 23:02 |
| notmyname | clarkb: I don't think I have creds there | 23:02 |
| clarkb | notmyname: I think it is valuable upstream to help debug future problems (as is the error is basically useless) | 23:02 |
| clarkb | notmyname: not yet, but I can add your ssh key :) | 23:02 |
| notmyname | clarkb: right, but patching locally is a lot faster :-) (actually we can't even merge upstream yet) | 23:03 |
| notmyname | clarkb: pub key https://gist.github.com/notmyname/c5bb54d03a802cff2f9d | 23:04 |
| jhesketh_ | So currently if I want to find a log for a job, is the only way to do that following the links from gerrit? | 23:04 |
| notmyname | clarkb: I can look at it for just a few more minutes | 23:04 |
| clarkb | notmyname: ssh jenkins@15.185.234.173 | 23:05 |
| clarkb | jhesketh_: the url is deterministic, so you can construct it | 23:05 |
| clarkb | jhesketh_: but otherwise yes follwoing links from gerrit is easiest | 23:05 |
| notmyname | clarkb: where is the swift code on there? | 23:05 |
| *** nati_ueno has joined #openstack-infra | 23:05 | |
| jeblair | notmyname: /opt/stack/new/swift | 23:06 |
| jhesketh_ | thanks | 23:07 |
| *** nati_ueno has quit IRC | 23:07 | |
| *** sandywalsh has quit IRC | 23:08 | |
| *** alexpilotti has quit IRC | 23:08 | |
| Alex_Gaynor | jeblair: if you've got a sec would you mind reviewing https://review.openstack.org/#/c/45923/ | 23:10 |
| jeblair | jog0: ^ | 23:11 |
| Alex_Gaynor | thanks | 23:11 |
| Alex_Gaynor | Also did someone in here ping me a while ago? | 23:11 |
| clarkb | Alex_Gaynor: yes wanted feedback on unpinning six | 23:12 |
| Alex_Gaynor | clarkb: that's waiting on a WSME release | 23:13 |
| clarkb | Alex_Gaynor: right, but fixing ceilometer broke everyone else | 23:14 |
| Alex_Gaynor | clarkb: how so? | 23:14 |
| clarkb | Alex_Gaynor: lots of things don't pin six, as soon as they install before pinned six you get a dependency conflict in pip | 23:14 |
| clarkb | Alex_Gaynor: because pip dependency resolution needs dstufft's awesomeness | 23:14 |
| Alex_Gaynor | clarkb: ah, so we needed to send review requests to all sub projects or something? | 23:15 |
| jog0 | jeblair: loos good to me | 23:15 |
| *** w_ has joined #openstack-infra | 23:15 | |
| clarkb | Alex_Gaynor: no, WSME release is supposed to happen rsn | 23:15 |
| clarkb | Alex_Gaynor: then we unpin I think | 23:15 |
| Alex_Gaynor | right | 23:15 |
| Alex_Gaynor | clarkb: I've got a patch up that's waiting on a WSME release | 23:15 |
| clarkb | Alex_Gaynor: I think if we really need to pin it we have to put six first on the dependency list | 23:16 |
| clarkb | that way the pinned version is installed before anything else. but right now I think we can limp by until WSME release | 23:16 |
| Alex_Gaynor | clarkb: in global-requirements.txt? | 23:16 |
| openstackgerrit | A change was merged to openstack-infra/config: Re-enable gating for hacking on PyPy https://review.openstack.org/45923 | 23:17 |
| clarkb | Alex_Gaynor: no in the individual projects | 23:17 |
| Alex_Gaynor | clarkb: ah, yes | 23:17 |
| *** thomasm has joined #openstack-infra | 23:17 | |
| *** olaph has quit IRC | 23:17 | |
| stevebaker | what is needed to get diskimage-builder onto the devstack image /var/cache/pip? Is it enough to just have it in requirements/global-requirements.txt? | 23:19 |
| Alex_Gaynor | stevebaker: global-reqiurements.txt + the nightly job which caches stuff | 23:19 |
| stevebaker | Alex_Gaynor: ok, so it doesn't need to be in some project's requirements.txt too? | 23:20 |
| Alex_Gaynor | stevebaker: you eventually need it in your project's requirements.txt so it gets installed | 23:20 |
| clarkb | stevebaker: it doesn't if you install it in the image build, however we don't like doing that | 23:20 |
| clarkb | stevebaker: we like devstack being able to fully configure itself | 23:20 |
| stevebaker | clarkb: I currently do, see https://review.openstack.org/#/c/45969/ | 23:21 |
| stevebaker | But it is not obvious to me which project should depend on diskimage-builder in this context - most likely to me is tempest | 23:22 |
| *** sandywalsh has joined #openstack-infra | 23:22 | |
| stevebaker | since tempest needs the images to do some tests | 23:23 |
| *** Ryan_Lane has quit IRC | 23:23 | |
| *** Ryan_Lane has joined #openstack-infra | 23:23 | |
| clarkb | stevebaker: is it going to be used for heat testing? I think it would be fine to put it in heat | 23:23 |
| stevebaker | does putting it in test-requirements.txt lead to it being installed? | 23:24 |
| *** Ryan_Lane has quit IRC | 23:26 | |
| *** Ryan_Lane has joined #openstack-infra | 23:26 | |
| clarkb | stevebaker: good question, we are pip installing stuff so I think the answer is yes | 23:26 |
| clarkb | mordred: ^ | 23:26 |
| jeblair | stevebaker: it sounds like tempest is probably a good place, since tempest will be the thing using it | 23:26 |
| clarkb | so the problem with using tempest and maybe it isn't really a problem is that we depend on site packages for it in the gate | 23:27 |
| clarkb | any stragglers not installed already should get installed though | 23:27 |
| jeblair | the alternative, i think would be in devstack | 23:28 |
| jeblair | but it doesn't seem to be big on installing dependencies these day | 23:29 |
| jeblair | s | 23:29 |
| clarkb | jeblair: just dependencies out of apt/yum | 23:29 |
| stevebaker | its good to know the solution is not obvious ;) | 23:31 |
| jeblair | also, as an aside, the pip cache isn't terribly important on devstack slaves atm; i'm not sure there's anything in it, as fetching from the openstack mirror is quite fast. though i wouldn't mind rsyncing the mirror if that ever gets set up. | 23:32 |
| stevebaker | It looks like I can't pip install diskimage-builder without it being in the pip cache http://logs.openstack.org/69/45969/1/check/gate-tempest-devstack-vm-full/f555c61/console.html | 23:33 |
| *** HenryG has joined #openstack-infra | 23:33 | |
| clarkb | stevebaker: I think it is not in the mirror | 23:33 |
| clarkb | stevebaker: so you will need to add it to openstack/requirements/global-requirements.txt | 23:33 |
| stevebaker | clarkb: https://review.openstack.org/#/c/45967/ | 23:34 |
| stevebaker | I'm not sure if that gate-requirements-install fail is due to the change | 23:34 |
| clarkb | stevebaker: it isn't | 23:35 |
| *** thomasm has quit IRC | 23:35 | |
| clarkb | stevebaker: I think your change is fine, the tests are just derpy right now | 23:35 |
| stevebaker | ok | 23:36 |
| notmyname | clarkb: found it | 23:36 |
| notmyname | clarkb: it's the ceilometer middleware in the pipeline. please revert that patch from devstack ASAP | 23:37 |
| notmyname | clarkb: torgomatic is going to look at their code to see where it breaks | 23:37 |
| clarkb | notmyname: great thanks | 23:37 |
| clarkb | notmyname: https://jenkins01.openstack.org/job/gate-swift-devstack-vm-functional/419/console is very verbose :) | 23:37 |
| notmyname | clarkb: tldr, it swallows the error response that the proxy returns | 23:37 |
| clarkb | resulting in a BadStatusLine exception? | 23:38 |
| notmyname | clarkb: because it changes a 404 http resonse into a sero byte reply | 23:38 |
| clarkb | notmyname: yup BadStatusLine("\'\'",) | 23:39 |
| notmyname | clarkb: ok, I gotta go prep for a meetup. I leave you to it. please revert ASAP and tickle the outstanding patches in swift, if possible. torgomatic will figure out why it's bad middleware | 23:39 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/config: Gate devstack and devstack-gate on swift-func https://review.openstack.org/45976 | 23:39 |
| notmyname | clarkb: jeblair: thanks for your help | 23:39 |
| jeblair | and that ^ should fix the asymetric gate that let this in | 23:39 |
| koolhead17 | mordred: are you at openstack day india by any chance? | 23:39 |
| jeblair | notmyname: you're welcome, thanks for finding the actual problem. :) | 23:40 |
| clarkb | jeblair: https://review.openstack.org/45977 | 23:41 |
| *** branen__ has joined #openstack-infra | 23:41 | |
| clarkb | notmyname: ^ and I assume we can destroy the slave we held? | 23:41 |
| *** dims has quit IRC | 23:41 | |
| clarkb | jeblair: do you want to recheck 45977 after 45976 goes in? | 23:41 |
| *** branen__ has quit IRC | 23:42 | |
| *** branen has joined #openstack-infra | 23:42 | |
| jeblair | clarkb: let's do that | 23:42 |
| clarkb | ok 45976 lgtm | 23:42 |
| jeblair | clarkb: i'll push the config update through so we can get started | 23:43 |
| clarkb | sounds good | 23:43 |
| *** sarob has quit IRC | 23:44 | |
| openstackgerrit | A change was merged to openstack-infra/config: Gate devstack and devstack-gate on swift-func https://review.openstack.org/45976 | 23:44 |
| clarkb | I am going to take this as an opportunity to talk home (while we wait for puppet) | 23:45 |
| jeblair | i'm running puppet now. but go on home, i'll cancel and recheck. :) | 23:45 |
| clarkb | jeblair: thanks. | 23:45 |
| *** branen_ has quit IRC | 23:45 | |
| *** dkliban has joined #openstack-infra | 23:46 | |
| jeblair | btw, apparently virtualenv '1.10.1' is about to be installed everywhere | 23:46 |
| jeblair | notice: /Stage[main]/Openstack_project::Base/Package[virtualenv]/ensure: ensure changed '1.10' to '1.10.1' | 23:46 |
| *** MarkAtwood2 has quit IRC | 23:48 | |
| *** jhesketh_ has quit IRC | 23:48 | |
| jeblair | erm, i didn't need to abort anything because the reload automatically lanuched that job. oops. | 23:49 |
| *** dims has joined #openstack-infra | 23:53 | |
| *** atiwari has quit IRC | 23:55 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!