| openstackgerrit | Joe Gordon proposed a change to openstack-infra/config: Move large-ops test to check pipeline https://review.openstack.org/46840 | 00:00 |
|---|---|---|
| *** michchap has quit IRC | 00:01 | |
| *** dims has joined #openstack-infra | 00:04 | |
| *** dcramer__ has joined #openstack-infra | 00:06 | |
| *** senk has joined #openstack-infra | 00:08 | |
| *** hemna is now known as hemnafk | 00:12 | |
| *** dcramer__ has quit IRC | 00:20 | |
| *** Ryan_Lane has quit IRC | 00:23 | |
| *** marun has quit IRC | 00:24 | |
| *** pcrews has joined #openstack-infra | 00:32 | |
| *** lnxnut has joined #openstack-infra | 00:32 | |
| *** colinmcnamara has joined #openstack-infra | 00:32 | |
| *** cyeoh_ has joined #openstack-infra | 00:35 | |
| *** cyeoh_ has quit IRC | 00:35 | |
| *** cyeoh_ has joined #openstack-infra | 00:35 | |
| *** cyeoh_ is now known as cyeoh | 00:35 | |
| *** senk has quit IRC | 00:37 | |
| *** dcramer__ has joined #openstack-infra | 00:39 | |
| *** mrodden has joined #openstack-infra | 00:40 | |
| *** colinmcnamara has quit IRC | 00:41 | |
| *** lnxnut has quit IRC | 00:42 | |
| *** mrodden1 has joined #openstack-infra | 00:42 | |
| *** lnxnut has joined #openstack-infra | 00:42 | |
| *** colinmcnamara has joined #openstack-infra | 00:42 | |
| *** mrodden has quit IRC | 00:44 | |
| *** dprince has quit IRC | 00:45 | |
| *** marun has joined #openstack-infra | 00:45 | |
| *** mrodden has joined #openstack-infra | 00:45 | |
| *** mrodden1 has quit IRC | 00:46 | |
| *** senk has joined #openstack-infra | 00:47 | |
| *** senk has quit IRC | 00:47 | |
| *** senk has joined #openstack-infra | 00:49 | |
| *** colinmcnamara has quit IRC | 00:52 | |
| *** sarob has joined #openstack-infra | 00:52 | |
| *** doctorjnupe has quit IRC | 01:01 | |
| *** cody-somerville has quit IRC | 01:02 | |
| *** senk has quit IRC | 01:06 | |
| *** senk has joined #openstack-infra | 01:08 | |
| *** dcramer__ has quit IRC | 01:10 | |
| *** jergerber has quit IRC | 01:23 | |
| *** zehicle_at_dell has joined #openstack-infra | 01:23 | |
| *** jergerber has joined #openstack-infra | 01:23 | |
| *** svarnau has joined #openstack-infra | 01:26 | |
| *** svarnau has quit IRC | 01:28 | |
| *** nati_ueno has quit IRC | 01:29 | |
| *** hashar has joined #openstack-infra | 01:30 | |
| *** cody-somerville has joined #openstack-infra | 01:30 | |
| *** yaguang has joined #openstack-infra | 01:38 | |
| *** sarob has quit IRC | 01:38 | |
| *** sarob has joined #openstack-infra | 01:38 | |
| *** sarob has quit IRC | 01:42 | |
| *** jergerber has quit IRC | 01:43 | |
| *** nati_ueno has joined #openstack-infra | 01:49 | |
| *** pcrews has quit IRC | 01:53 | |
| *** sarob has joined #openstack-infra | 02:02 | |
| *** pcrews has joined #openstack-infra | 02:12 | |
| *** lcestari has quit IRC | 02:17 | |
| openstackgerrit | Mathew Odden proposed a change to openstack-infra/jenkins-job-builder: Add publisher for Git Publisher support https://review.openstack.org/46417 | 02:22 |
| *** jerryz has quit IRC | 02:22 | |
| *** dcramer__ has joined #openstack-infra | 02:31 | |
| *** UtahDave has joined #openstack-infra | 02:32 | |
| *** cody-somerville has quit IRC | 02:33 | |
| *** CaptTofu has quit IRC | 02:41 | |
| *** UtahDave has quit IRC | 02:44 | |
| *** sarob has quit IRC | 02:54 | |
| *** ryanpetrello has quit IRC | 02:57 | |
| *** lnxnut has quit IRC | 03:10 | |
| *** vipul has quit IRC | 03:16 | |
| *** michchap_ has quit IRC | 03:17 | |
| *** vipul has joined #openstack-infra | 03:17 | |
| *** michchap has joined #openstack-infra | 03:17 | |
| *** michchap_ has joined #openstack-infra | 03:21 | |
| *** michchap has quit IRC | 03:21 | |
| *** senk has quit IRC | 03:24 | |
| *** vipul is now known as vipul-away | 03:35 | |
| *** michchap_ has quit IRC | 03:37 | |
| *** michchap has joined #openstack-infra | 03:38 | |
| *** vipul-away is now known as vipul | 03:39 | |
| *** dkliban has joined #openstack-infra | 03:40 | |
| *** afazekas_zz is now known as afazekas | 03:40 | |
| *** Ryan_Lane has joined #openstack-infra | 03:43 | |
| *** vogxn has joined #openstack-infra | 03:46 | |
| *** sarob has joined #openstack-infra | 03:46 | |
| *** lyz has quit IRC | 03:47 | |
| *** senk has joined #openstack-infra | 03:53 | |
| *** lyz has joined #openstack-infra | 03:59 | |
| *** dims has quit IRC | 04:01 | |
| *** hashar has quit IRC | 04:02 | |
| Alex_Gaynor | How long should the zuul tests take, and should they use 600+ MB of RAM per process? | 04:02 |
| openstackgerrit | Alex Gaynor proposed a change to openstack-infra/zuul: On null changes serialize the id as null https://review.openstack.org/46869 | 04:05 |
| *** amotoki has joined #openstack-infra | 04:07 | |
| openstackgerrit | Alex Gaynor proposed a change to openstack-infra/zuul: On null changes serialize the id as null https://review.openstack.org/46869 | 04:09 |
| *** ArxCruz has quit IRC | 04:10 | |
| *** pcrews has quit IRC | 04:12 | |
| *** hashar has joined #openstack-infra | 04:14 | |
| Alex_Gaynor | Ah, I figured it out: if your select module doesn't have a `poll`, the tests busy loop | 04:17 |
| *** yongli_away is now known as yongli | 04:17 | |
| *** lnxnut has joined #openstack-infra | 04:21 | |
| *** lnxnut has quit IRC | 04:26 | |
| openstackgerrit | Joshua Hesketh proposed a change to openstack-infra/config: Test nova changes against a real database https://review.openstack.org/46871 | 04:28 |
| *** UtahDave has joined #openstack-infra | 04:34 | |
| *** dkliban has quit IRC | 04:36 | |
| *** senk has quit IRC | 04:36 | |
| *** hashar has quit IRC | 04:43 | |
| *** sarob has quit IRC | 04:43 | |
| *** sarob has joined #openstack-infra | 04:44 | |
| *** CaptTofu has joined #openstack-infra | 04:46 | |
| jhesketh__ | jeblair, clarkb: ping | 04:49 |
| *** DinaBelova has joined #openstack-infra | 04:56 | |
| *** afazekas is now known as __afazekas | 05:07 | |
| *** nicedice has quit IRC | 05:20 | |
| *** SergeyLukjanov has joined #openstack-infra | 05:20 | |
| *** dcramer__ has quit IRC | 05:21 | |
| *** sarob has quit IRC | 05:23 | |
| *** afazekas has joined #openstack-infra | 05:32 | |
| *** pblaho has joined #openstack-infra | 05:42 | |
| *** pblaho has quit IRC | 05:42 | |
| *** pblaho has joined #openstack-infra | 05:43 | |
| *** thomasbiege1 has joined #openstack-infra | 05:48 | |
| *** DinaBelova has quit IRC | 05:49 | |
| *** thomasbiege1 has quit IRC | 05:49 | |
| *** yaguang has quit IRC | 05:51 | |
| *** SergeyLukjanov has quit IRC | 05:55 | |
| *** boris-42 has joined #openstack-infra | 05:58 | |
| *** yaguang has joined #openstack-infra | 06:01 | |
| *** branen__ has joined #openstack-infra | 06:02 | |
| *** mkerrin1 has joined #openstack-infra | 06:03 | |
| *** nijaba_ has joined #openstack-infra | 06:03 | |
| *** w_ has joined #openstack-infra | 06:04 | |
| *** lifeless_ has joined #openstack-infra | 06:05 | |
| *** jpeeler has quit IRC | 06:09 | |
| *** mkerrin has quit IRC | 06:09 | |
| *** nijaba has quit IRC | 06:09 | |
| *** lifeless has quit IRC | 06:09 | |
| *** olaph has quit IRC | 06:09 | |
| *** branen_ has quit IRC | 06:09 | |
| *** fungi has quit IRC | 06:09 | |
| *** odyi has quit IRC | 06:09 | |
| *** openstackstatus has quit IRC | 06:13 | |
| *** openstack has joined #openstack-infra | 06:18 | |
| *** sileht has joined #openstack-infra | 06:19 | |
| *** afazekas is now known as afazekas_no_irq | 06:19 | |
| *** sdague_ has joined #openstack-infra | 06:21 | |
| *** prad_ has joined #openstack-infra | 06:23 | |
| *** saper_ has joined #openstack-infra | 06:23 | |
| *** comstud_ has joined #openstack-infra | 06:23 | |
| *** rainya has quit IRC | 06:23 | |
| *** sdague has quit IRC | 06:23 | |
| *** sdague_ is now known as sdague | 06:23 | |
| *** rainya has joined #openstack-infra | 06:23 | |
| *** comstud has quit IRC | 06:24 | |
| *** saper has quit IRC | 06:24 | |
| *** prad has quit IRC | 06:24 | |
| *** dragondm has quit IRC | 06:24 | |
| *** prad_ is now known as prad | 06:24 | |
| *** boris-42 has quit IRC | 06:26 | |
| *** dragondm has joined #openstack-infra | 06:26 | |
| *** saper_ is now known as saper | 06:34 | |
| *** odyssey4me has joined #openstack-infra | 06:36 | |
| *** sarob has quit IRC | 06:46 | |
| *** BobBallAway has joined #openstack-infra | 06:53 | |
| *** thingee has joined #openstack-infra | 06:53 | |
| *** Ng has joined #openstack-infra | 06:53 | |
| *** hashar has joined #openstack-infra | 06:53 | |
| *** branen__ has quit IRC | 06:53 | |
| *** branen__ has joined #openstack-infra | 06:53 | |
| *** mkerrin1 has quit IRC | 06:53 | |
| *** mkerrin1 has joined #openstack-infra | 06:53 | |
| *** dizquierdo has joined #openstack-infra | 06:56 | |
| *** vogxn has quit IRC | 07:04 | |
| *** hashar has quit IRC | 07:07 | |
| *** SergeyLukjanov has joined #openstack-infra | 07:10 | |
| *** SlickNik has quit IRC | 07:14 | |
| *** vipul is now known as vipul-away | 07:14 | |
| *** SlickNik has joined #openstack-infra | 07:14 | |
| *** vipul-away is now known as vipul | 07:16 | |
| *** DinaBelova has joined #openstack-infra | 07:19 | |
| *** yaguang_ has quit IRC | 07:23 | |
| *** yaguang has joined #openstack-infra | 07:24 | |
| * saschpe_ is going to pester zul about python-lesscpy | 07:28 | |
| *** gaelL has quit IRC | 07:30 | |
| *** UtahDave has quit IRC | 07:34 | |
| *** cyeoh has quit IRC | 07:37 | |
| *** cyeoh has joined #openstack-infra | 07:38 | |
| *** thomasbiege1 has joined #openstack-infra | 07:42 | |
| *** enikanorov-w has quit IRC | 07:42 | |
| *** thomasbiege1 has quit IRC | 07:43 | |
| *** enikanorov-w has joined #openstack-infra | 07:43 | |
| *** plomakin has quit IRC | 07:46 | |
| *** plomakin has joined #openstack-infra | 07:46 | |
| *** davidhadas has joined #openstack-infra | 07:47 | |
| davidhadas | hi, I just upgraded and encountered ImportError: No module named pbr | 07:49 |
| davidhadas | Looking for documentation on how to reslve this | 07:50 |
| *** nati_ueno has quit IRC | 07:57 | |
| *** gaelL has joined #openstack-infra | 07:58 | |
| *** tstevenson has quit IRC | 08:00 | |
| *** jpich has joined #openstack-infra | 08:03 | |
| *** Ryan_Lane has quit IRC | 08:03 | |
| *** fbo_away is now known as fbo | 08:07 | |
| *** DinaBelova has quit IRC | 08:08 | |
| *** DinaBelova has joined #openstack-infra | 08:11 | |
| *** davidhadas has quit IRC | 08:11 | |
| *** cyeoh has quit IRC | 08:11 | |
| *** cyeoh has joined #openstack-infra | 08:12 | |
| *** SergeyLukjanov has quit IRC | 08:13 | |
| *** dizquierdo has left #openstack-infra | 08:14 | |
| *** oubiwann has quit IRC | 08:19 | |
| *** SergeyLukjanov has joined #openstack-infra | 08:20 | |
| *** oubiwann has joined #openstack-infra | 08:22 | |
| *** davidhadas has joined #openstack-infra | 08:32 | |
| *** BobBallAway is now known as BobBall | 08:33 | |
| *** vogxn has joined #openstack-infra | 08:46 | |
| *** dizquierdo has joined #openstack-infra | 08:46 | |
| *** flaper87|afk is now known as flaper87 | 09:03 | |
| *** yaguang has quit IRC | 09:05 | |
| *** yongli is now known as yongli_away | 09:10 | |
| *** DinaBelova has quit IRC | 09:28 | |
| openstackgerrit | Marton Kiss proposed a change to openstack-infra/config: Add groups-dev.openstack.org node https://review.openstack.org/42608 | 09:43 |
| *** boris-42 has joined #openstack-infra | 09:44 | |
| openstackgerrit | Victor Sergeyev proposed a change to openstack-dev/hacking: Add H234 check to avoid (x)range() function usage. https://review.openstack.org/46912 | 09:45 |
| *** HenryG has joined #openstack-infra | 10:10 | |
| *** mugsie_ is now known as mugsie | 10:11 | |
| *** mugsie has joined #openstack-infra | 10:11 | |
| openstackgerrit | Roman Podolyaka proposed a change to openstack-infra/config: Use run-tox.sh script for sqlalchemy-migrate jobs https://review.openstack.org/44686 | 10:12 |
| openstackgerrit | Yulia Portnova proposed a change to openstack-infra/config: New project: manila, python-manilaclient https://review.openstack.org/46919 | 10:21 |
| openstackgerrit | Paul Bourke proposed a change to openstack/requirements: Add ordereddict for python 2.6 compatibility https://review.openstack.org/46474 | 10:25 |
| *** pcm_ has joined #openstack-infra | 10:29 | |
| *** pcm_ has quit IRC | 10:30 | |
| *** pcm_ has joined #openstack-infra | 10:31 | |
| openstackgerrit | Yulia Portnova proposed a change to openstack-infra/config: New project: manila, python-manilaclient https://review.openstack.org/46919 | 10:38 |
| *** vogxn has quit IRC | 10:46 | |
| *** thomasbiege1 has joined #openstack-infra | 10:48 | |
| openstackgerrit | Yulia Portnova proposed a change to openstack-infra/config: New project: manila, python-manilaclient https://review.openstack.org/46919 | 10:52 |
| openstackgerrit | Yulia Portnova proposed a change to openstack-infra/config: New project: manila, python-manilaclient https://review.openstack.org/46919 | 10:55 |
| openstackgerrit | Victor Sergeyev proposed a change to openstack-dev/hacking: Add H234 check to avoid (x)range() function usage https://review.openstack.org/46912 | 10:59 |
| openstackgerrit | Yulia Portnova proposed a change to openstack-infra/config: New project: manila, python-manilaclient https://review.openstack.org/46919 | 11:03 |
| *** amotoki has quit IRC | 11:12 | |
| *** crank_ is now known as crank | 11:22 | |
| *** thomasbiege2 has joined #openstack-infra | 11:23 | |
| *** thomasbiege1 has quit IRC | 11:26 | |
| *** thomasbiege2 has quit IRC | 11:26 | |
| *** DinaBelova has joined #openstack-infra | 11:28 | |
| *** jcoufal has joined #openstack-infra | 11:39 | |
| *** prad has quit IRC | 11:46 | |
| *** annegentle has quit IRC | 11:48 | |
| *** SergeyLukjanov has quit IRC | 11:49 | |
| *** SergeyLukjanov has joined #openstack-infra | 11:49 | |
| *** blamar has joined #openstack-infra | 11:50 | |
| *** thomasbiege has joined #openstack-infra | 11:51 | |
| *** thomasbiege has quit IRC | 11:51 | |
| *** lcestari has joined #openstack-infra | 11:54 | |
| *** dcramer__ has joined #openstack-infra | 11:55 | |
| *** ArxCruz has joined #openstack-infra | 12:11 | |
| *** cppcabrera has joined #openstack-infra | 12:15 | |
| *** dims has joined #openstack-infra | 12:16 | |
| *** cppcabrera has left #openstack-infra | 12:17 | |
| *** weshay has joined #openstack-infra | 12:19 | |
| *** jhesketh has quit IRC | 12:33 | |
| *** thomasbiege has joined #openstack-infra | 12:35 | |
| *** annegentle has joined #openstack-infra | 12:42 | |
| *** rfolco has joined #openstack-infra | 12:43 | |
| *** lifeless_ is now known as lifeless | 12:43 | |
| *** hashar has joined #openstack-infra | 12:44 | |
| *** w_ is now known as olaph | 12:47 | |
| *** SergeyLukjanov has quit IRC | 12:49 | |
| *** DinaBelova has quit IRC | 12:50 | |
| *** afazekas_no_irq has quit IRC | 12:51 | |
| *** SergeyLukjanov has joined #openstack-infra | 12:51 | |
| *** thomasm has joined #openstack-infra | 12:52 | |
| *** dcramer__ has quit IRC | 12:58 | |
| *** SergeyLukjanov has quit IRC | 13:07 | |
| *** amotoki has joined #openstack-infra | 13:09 | |
| *** SergeyLukjanov has joined #openstack-infra | 13:11 | |
| *** DinaBelova has joined #openstack-infra | 13:11 | |
| *** jergerber has joined #openstack-infra | 13:13 | |
| *** mriedem has joined #openstack-infra | 13:14 | |
| *** changbl has quit IRC | 13:15 | |
| *** prad has joined #openstack-infra | 13:16 | |
| *** prad has quit IRC | 13:19 | |
| *** blamar_ has joined #openstack-infra | 13:21 | |
| *** blamar has quit IRC | 13:22 | |
| *** blamar_ is now known as blamar | 13:22 | |
| *** kpavel has joined #openstack-infra | 13:25 | |
| kpavel | Hi, what is the trigger for the opened bug to be updated with "Fix proposed to branch: XXX" info? Because it did create a link to the bug inside the change, but for some reason didn't create a message on launchpad: https://review.openstack.org/#/c/46649/ | 13:26 |
| *** dims has quit IRC | 13:30 | |
| *** dims has joined #openstack-infra | 13:30 | |
| mordred | kpavel: you want to look at https://wiki.openstack.org/wiki/GitCommitMessages at the Including external references section | 13:32 |
| *** blamar has quit IRC | 13:32 | |
| *** SergeyLukjanov has quit IRC | 13:33 | |
| *** matty_dubs|gone is now known as matty_dubs | 13:34 | |
| *** zehicle_at_dell has quit IRC | 13:38 | |
| kpavel | mordred: just changed it to "Closes-Bug: #XXX" and it doesn't look like it helped. | 13:39 |
| *** wenlock has joined #openstack-infra | 13:40 | |
| *** blamar has joined #openstack-infra | 13:41 | |
| *** mkerrin1 has quit IRC | 13:44 | |
| *** CaptTofu has quit IRC | 13:48 | |
| *** prad has joined #openstack-infra | 13:48 | |
| *** dkliban has joined #openstack-infra | 13:48 | |
| *** wenlock has quit IRC | 13:50 | |
| openstackgerrit | Sascha Peilicke proposed a change to openstack/requirements: Require pyparsing >= 2.0.1 https://review.openstack.org/46949 | 13:51 |
| *** julim has joined #openstack-infra | 13:52 | |
| *** mkerrin has joined #openstack-infra | 13:53 | |
| *** michchap has quit IRC | 13:53 | |
| *** michchap has joined #openstack-infra | 13:54 | |
| *** dcramer__ has joined #openstack-infra | 13:56 | |
| *** blamar has quit IRC | 13:57 | |
| *** blamar has joined #openstack-infra | 13:58 | |
| *** michchap has quit IRC | 14:03 | |
| *** tstevenson has joined #openstack-infra | 14:04 | |
| *** pblaho has quit IRC | 14:05 | |
| *** mestery has joined #openstack-infra | 14:07 | |
| *** DinaBelova has quit IRC | 14:07 | |
| *** ArxCruz has quit IRC | 14:08 | |
| *** SergeyLukjanov has joined #openstack-infra | 14:08 | |
| *** mestery has quit IRC | 14:09 | |
| *** lnxnut has joined #openstack-infra | 14:09 | |
| *** mestery has joined #openstack-infra | 14:09 | |
| *** DinaBelova has joined #openstack-infra | 14:12 | |
| *** CaptTofu has joined #openstack-infra | 14:13 | |
| *** pentameter has joined #openstack-infra | 14:14 | |
| *** dkranz has joined #openstack-infra | 14:16 | |
| *** rnirmal has joined #openstack-infra | 14:21 | |
| fungi | kpavel: it only posts a link comment in the bug if it's the first patchset for that change, to avoid spamming the bug on every new patchset for a change (it has no way of knowing whether it's added a fix-proposed comment already) | 14:24 |
| *** CaptTofu has quit IRC | 14:25 | |
| fungi | could probably be fixed by giving the trigger scrape the bug comments looking for traces of its past activity or adding some sort of persistent storage (it's entirely stateless at the moment), but that's a deal of additional complexity | 14:26 |
| fungi | s/giving/having/ | 14:26 |
| fungi | davidhadas: what did you just upgrade, that you're seeing a pbr importerror exception from? can you throw the whole python traceback onto http://paste.openstack.org/ and give us the link? | 14:28 |
| *** kgriffs has joined #openstack-infra | 14:29 | |
| *** kgriffs has left #openstack-infra | 14:30 | |
| *** NobodyCa1 is now known as NobodyCam | 14:33 | |
| jeblair | jhesketh__: i'm at linuxcon this week with spotty availability, sorry | 14:33 |
| *** markmcclain has joined #openstack-infra | 14:34 | |
| *** sarob has joined #openstack-infra | 14:39 | |
| *** thedodd has joined #openstack-infra | 14:40 | |
| Alex_Gaynor | dhellmann: no quotes around fixed. It's just fixed. | 14:45 |
| dhellmann | Alex_Gaynor: fair enough :-) | 14:45 |
| *** sarob has quit IRC | 14:47 | |
| *** sarob has joined #openstack-infra | 14:47 | |
| *** hashar has quit IRC | 14:50 | |
| *** sarob has quit IRC | 14:52 | |
| *** boris-42 has quit IRC | 14:53 | |
| openstackgerrit | Ryan Petrello proposed a change to openstack-infra/config: Provide a more generic run-tox.sh. https://review.openstack.org/43145 | 14:53 |
| *** DinaBelova has quit IRC | 14:57 | |
| mordred | davidhadas: ping me - I'd lke to know more about your pbr problem - and yes, also, pasetebin traceback please | 14:59 |
| *** ryanpetrello has joined #openstack-infra | 14:59 | |
| ryanpetrello | meh, made a slight change and rebase to https://review.openstack.org/#/c/43145/ and now I'm seeing a unicode issue for gate-ci-docs: http://logs.openstack.org/45/43145/9/check/gate-ci-docs/4c75536/console.html | 15:00 |
| ryanpetrello | anybody know anything about this? | 15:00 |
| *** dkranz has quit IRC | 15:01 | |
| *** danger_fo_away is now known as danger_fo | 15:02 | |
| *** SergeyLukjanov has quit IRC | 15:02 | |
| *** DinaBelova has joined #openstack-infra | 15:06 | |
| fungi | ryanpetrello: well, i grok the exception. you can't unicodedata.normalize('NFC', newpath) | 15:06 |
| *** ArxCruz has joined #openstack-infra | 15:06 | |
| ryanpetrello | right | 15:07 |
| fungi | since newpath needs to be a unicode type rather than string | 15:07 |
| ryanpetrello | what's odd is that all I did here was change one line of my review and rebased | 15:07 |
| fungi | looking to see where newpath is getting assigned in hopes of ferreting it out | 15:07 |
| ryanpetrello | so I'm not sure where this new failure is coming from :\ | 15:07 |
| fungi | i'm suspecting something may have changed external to your patch, like maybe a new sphinx package version or the like. i'll try to reproduce here | 15:10 |
| fungi | looks like the code it's tripping over was added in sphinx 1.2 | 15:11 |
| fungi | ayep... "Running Sphinx v1.2b2" | 15:12 |
| ryanpetrello | oy | 15:13 |
| fungi | landed on pypi today | 15:13 |
| fungi | so, um, yeah | 15:13 |
| fungi | lemme see what we have nailed in global-requirements.txt to possibly prevent this elsewhere. might need to do the same in your repo | 15:14 |
| fungi | er, in the config repo, ahead of your change i mean | 15:14 |
| fungi | though it's possible we need to pin it in the sphinx wrapper project dhellmann's got instead | 15:15 |
| * fungi goes looking for it | 15:15 | |
| dhellmann | fungi: wrapper? | 15:15 |
| fungi | dhellmann: wasn't there some project which was previously auto-sphinxification inside of pbr or oslo or something, and then got separated out? | 15:16 |
| *** thomasbiege has quit IRC | 15:16 | |
| dhellmann | fungi: oh, yeah, I think that's in the theme package | 15:16 |
| * fungi doesn't want to interrupt your conferencing so feel free to ignore. i'll find where we need the pin | 15:16 | |
| dhellmann | https://github.com/openstack/oslo.sphinx | 15:16 |
| *** dkranz has joined #openstack-infra | 15:17 | |
| fungi | or https://git.openstack.org/cgit/openstack/oslo.sphinx right | 15:17 |
| fungi | thanks! | 15:17 |
| dhellmann | I'm listening to a guy from Wired talk about how the internet is like a brain. So I'm not really listening that closely. :-) | 15:17 |
| jd__ | a heavily damaged brain maybe :] | 15:18 |
| fungi | someone at wired has seen a brain before? didn't think they had a point of reference for that | 15:18 |
| *** hashar has joined #openstack-infra | 15:18 | |
| fungi | the internet is a lot like my brain, yes. damaged and demented | 15:18 |
| *** pcrews has joined #openstack-infra | 15:21 | |
| fungi | ryanpetrello: so it looks like we simply dep on sphinx with no version spec in openstack-infra/config tox.ini | 15:22 |
| fungi | now, as to why pip is pulling in a clearly beta-versioned sphinx is my question... (too-old pip?) | 15:22 |
| fungi | i'm going to upload a cap of <1.2 and you can rebase on it to see if that helps | 15:23 |
| ryanpetrello | okay, sounds good | 15:24 |
| mordred | fungi: pip pulls beta versioned things if you do not have pip 1.4 | 15:28 |
| fungi | mordred: yeah. this ran in a venv on precise32, so... | 15:29 |
| mordred | fungi: do we not have pip 1.4 on our precise nodes yet? | 15:29 |
| fungi | i was about to check. i don't think that's likely to be the problem | 15:29 |
| mordred | mordred@precise32:~$ pip --version | 15:29 |
| mordred | pip 1.0 from /usr/lib/python2.7/dist-packages (python 2.7) | 15:30 |
| fungi | gah | 15:30 |
| mordred | I thought we upgraded that globally in puppet - or did we remove that upgrade | 15:30 |
| mordred | ? | 15:30 |
| fungi | yeah, it's pip 1.0 on other precise slaves i've checked too | 15:30 |
| fungi | did we work around it a different way in some jobs by maybe having pip install a new version of itself into the virtualenv first? | 15:31 |
| *** jcoufal has quit IRC | 15:32 | |
| *** jcoufal has joined #openstack-infra | 15:32 | |
| fungi | we have pip>=1.4 in openstack/requirements global-requirements.txt, which of course has no bearing on openstack-infra/config | 15:33 |
| *** marun has quit IRC | 15:33 | |
| *** marun has joined #openstack-infra | 15:34 | |
| * fungi checks a random nova job to see how it gets tackled | 15:34 | |
| *** markmcclain has quit IRC | 15:36 | |
| fungi | yeah, we're ending up running sphinx 1.1.3 elsewhere | 15:36 |
| *** marun has quit IRC | 15:37 | |
| fungi | it's too bad pip freeze doesn't report what version it's running | 15:37 |
| dstufft | mordred: https://mail.python.org/pipermail/python-dev/2013-September/128723.html | 15:37 |
| zul | saschpe_: whats the review url for lesscpy in openstack-ci please | 15:37 |
| *** changbl has joined #openstack-infra | 15:37 | |
| mordred | dstufft: ooh | 15:37 |
| dstufft | mordred: ^ Round 3 of the "include pip with python" PEP, this time it moves from distutils-sig to python-dev | 15:37 |
| dstufft | mordred: Oh, and it includes backporting pip inclusion to 2.7 | 15:37 |
| dstufft | and 3.3 | 15:37 |
| dstufft | so if it lands as it is now, ``pip`` should start to be available by default in python 2.7+ and 3.3+ | 15:38 |
| saschpe_ | zul: https://review.openstack.org/#/c/40218/ | 15:38 |
| zul | thanks | 15:38 |
| *** pcrews has quit IRC | 15:39 | |
| *** julim has quit IRC | 15:39 | |
| *** odyssey4me has quit IRC | 15:40 | |
| *** senk has joined #openstack-infra | 15:40 | |
| *** markmcclain has joined #openstack-infra | 15:40 | |
| mordred | dstufft: that would be stellar | 15:41 |
| *** __afazekas is now known as afazekas | 15:42 | |
| mordred | dstufft: and it would be pip as currently? like 1.4 ish? | 15:42 |
| *** julim has joined #openstack-infra | 15:43 | |
| fungi | and then we'll have one more reason to push red hat to switch rhel to python 2.7 | 15:43 |
| BobBall | Can anyone in here shed any light on six in nova? requirements.txt has six<1.4 but it seems that jsonpointer>=1.0 pulls in six1.4 - as does test-requirements.txt/hacking... I think/hope this is causing my run_tests.sh to die horribly | 15:43 |
| dstufft | mordred: basically it includes a python -m getpip which will have a $LATEST_AT_TIME_PYTHON_X.Y.Z_WAS MADE bundled, and when the windows and OSX installers run they'll execute python -m getpip which will attempt to hit pypi to get an even later version and falls back to the bundled | 15:43 |
| dstufft | mordred: additionally it goes to the Linux distros and says "Hey, plz 2 be making pip and python depend on each other" | 15:43 |
| dstufft | mordred: which Fedora has said "ok" so far | 15:44 |
| *** svarnau has joined #openstack-infra | 15:44 | |
| dstufft | so in 2.7+ you'll always be able to install pip with python -m getpip, and for Windows/OSX from Python.org it'll install pip automatically, and for every thing else we're going to try to apply social pressure to make it install as it makes sense for their system | 15:45 |
| *** zehicle_at_dell has joined #openstack-infra | 15:46 | |
| mordred | BobBall: I believe there was a recent evil around six but I don't remember the details - dhellmann do you? | 15:47 |
| *** flaper87 is now known as flaper87|afk | 15:47 | |
| *** SergeyLukjanov has joined #openstack-infra | 15:47 | |
| mordred | dstufft: if there is any way I can help support that, please let me know | 15:48 |
| mordred | dstufft: like, if it would help to say "OpenStack REALLY needs this" | 15:48 |
| mordred | dstufft: or, if I can help by keeping my mouth shut about OpenStack :) | 15:50 |
| BobBall | recent as in still current? :) I've updated my nova to the very latest... | 15:50 |
| dstufft | mordred: if you want to comment on the python-dev thread that'd be cool :) So far the response has been primarily positive in the previous threads on distutils-sig, and the guy who makes the final decision has already said he agrees in principle but has some issues that he wants addressed | 15:50 |
| BobBall | I'd hope with using a venv that it should just work... unless of course it's using system packages too | 15:50 |
| dstufft | I suspect the longer battle will be convincing all the various linux distros | 15:51 |
| dstufft | luckily we already got Fedora on board, and I think it won't take much convincing to get debian and ubuntu | 15:51 |
| *** wenlock has joined #openstack-infra | 15:54 | |
| *** kpavel has quit IRC | 15:55 | |
| *** flaper87|afk is now known as flaper87 | 15:55 | |
| *** doctorjnupe has joined #openstack-infra | 15:55 | |
| *** boris-42 has joined #openstack-infra | 15:56 | |
| mordred | BobBall: nova _should_ just work in a venv - although it does pull in some system libraries too (because of libvirt) | 15:58 |
| mordred | BobBall: let me dig in a little bit | 15:58 |
| mordred | dstufft: I'm pretty sure we should be able to get ubuntu on board | 15:59 |
| *** mrodden has quit IRC | 15:59 | |
| BobBall | thanks mordred | 16:01 |
| BobBall | I bet it just works for you though :) | 16:01 |
| mordred | BobBall: probably | 16:02 |
| *** mestery has quit IRC | 16:02 | |
| *** marun has joined #openstack-infra | 16:04 | |
| mordred | BobBall: what's pulling in jsonpointer? I donm't see it in nova? | 16:05 |
| BobBall | Requirement already up-to-date: jsonpointer>=1.0 in ./.venv/lib/python2.7/site-packages (from jsonpatch>=0.10,<2->warlock>=1.0.1,<2->python-glanceclient>=0.9.0->-r /usr/workspace/scratch/openstack/nova/requirements.txt (line 26)) | 16:05 |
| *** alexpilotti has joined #openstack-infra | 16:05 | |
| *** julim has quit IRC | 16:06 | |
| mordred | BobBall: k. thanks | 16:06 |
| *** alexpilotti has quit IRC | 16:06 | |
| BobBall | http://paste.openstack.org/show/47157/ | 16:06 |
| *** alexpilotti has joined #openstack-infra | 16:07 | |
| *** matty_dubs is now known as matty_dubs|lunch | 16:07 | |
| *** nati_ueno has joined #openstack-infra | 16:07 | |
| BobBall | can we not get rid of per-project requirements entirely? :) | 16:07 |
| dhellmann | mordred, BobBall: no, but Alex_Gaynor may | 16:07 |
| Alex_Gaynor | dhellmann: err, what? | 16:08 |
| dhellmann | Alex_Gaynor: do you remember any recent issues with six that caused us to pin it? | 16:08 |
| *** julim has joined #openstack-infra | 16:08 | |
| Alex_Gaynor | dhellmann: Yes, I do. | 16:08 |
| Alex_Gaynor | dhellmann: it changed the MRO in the case of with_metaclass, and that somehow broke WSME | 16:09 |
| dhellmann | :-/ | 16:09 |
| dhellmann | is there a wsme bug for that? | 16:09 |
| * dhellmann looks | 16:09 | |
| Alex_Gaynor | dhellmann: I assume so, I thought we were waiting on a thing so we could do a release? | 16:09 |
| mordred | BobBall: nope. totally impossible | 16:09 |
| BobBall | shame :D | 16:10 |
| mordred | BobBall: but - also, to annoy you, it works locally for me | 16:10 |
| Alex_Gaynor | dhellmann: that is to say, WSME is fixed, but not released (AFAIK) | 16:10 |
| mordred | BobBall: can you tell me how you're installing? | 16:10 |
| dhellmann | Alex_Gaynor: I've lost track of all the things fixed in WSME that we were waiting to be able to release. cdevienne did a new beta release 0.5b5 yesterday | 16:11 |
| dhellmann | or earlier today | 16:11 |
| BobBall | I'm running "./run_tests.sh" :) Actually to be a little clearer I'm doing ./run_tests.sh -u -V nova.tests.integrated.test_api_samples.ServerUsageSampleJsonTests | 16:11 |
| *** GheRiver1 is now known as GheRivero | 16:11 | |
| BobBall | because that test is one of the ones that fails | 16:11 |
| mordred | BobBall: ok. let me try runnign that | 16:12 |
| mordred | BobBall: I ran/run tox | 16:12 |
| BobBall | six may be unrelated to that failure - but was the most obvious issue when trying to figure it out... | 16:12 |
| BobBall | http://paste.openstack.org/show/47162/ is the full error I'm getting btw - you can see loads of errors relating to six from line 328 | 16:18 |
| BobBall | ah - I've now convinced myself from the log that it is 100% because of six... :) | 16:19 |
| *** senk has quit IRC | 16:20 | |
| *** nati_ueno has quit IRC | 16:20 | |
| BobBall | if it works for you mordred can you pastebin your pip log so I can try and figure out why it's installing the wrong version of six for me but not for you? | 16:22 |
| BobBall | or, more likely a pip freeze | 16:22 |
| BobBall | also - were you using a fresh env? I wonder if something like I'm picking up a newer version of warlock which doesn't depend on jsonpointer but your version is still within the range... | 16:24 |
| *** senk has joined #openstack-infra | 16:25 | |
| mordred | BobBall: ah! run_tests.sh does, indeed, fail for me too | 16:27 |
| mordred | this is one of the reaons I do not like having divergent test running envs | 16:28 |
| mordred | let me see what run_tests.sh is doing wrong | 16:28 |
| *** senk has quit IRC | 16:29 | |
| BobBall | does tox install requirements from test-requirements.txt? | 16:29 |
| mordred | yes | 16:30 |
| mordred | they _should_ be doing the exact same thing | 16:30 |
| *** senk has joined #openstack-infra | 16:30 | |
| BobBall | oh great! | 16:30 |
| *** hashar has quit IRC | 16:31 | |
| *** markmcclain has quit IRC | 16:31 | |
| mordred | but should doesn't mean are | 16:31 |
| *** CaptTofu has joined #openstack-infra | 16:33 | |
| *** julim has quit IRC | 16:35 | |
| *** jog0_ is now known as jog0 | 16:36 | |
| openstackgerrit | Jeremy Stanley proposed a change to openstack-infra/config: Make sure pip upgrades itself in tox https://review.openstack.org/46984 | 16:39 |
| fungi | ryanpetrello: ^ rebase on that and it should help | 16:40 |
| fungi | mordred: seems that was the issue & | 16:40 |
| *** dkranz has quit IRC | 16:40 | |
| *** markmcclain has joined #openstack-infra | 16:40 | |
| fungi | the commit message is probably a little wrong, since it's presumably upgrading more than just pip with that change | 16:41 |
| mordred | fungi: are you sure that does what you think? | 16:41 |
| mordred | fungi: the way we get newer pip in tox venvs is to have newer tox | 16:42 |
| ryanpetrello | fungi: thanks, will do shortly | 16:42 |
| fungi | mordred: mordred so are you saying tox is too old on our precise saves then? | 16:42 |
| fungi | i pulled that option from nova, which seemed to be causing us to use latest pip in the virtualenv once added | 16:43 |
| *** mrmartin has joined #openstack-infra | 16:43 | |
| fungi | and caused the sphinx version pulled in by the build to be a non-prerelease version as a result | 16:43 |
| mordred | fungi: great! | 16:44 |
| fungi | but it very well may still be an ordering issue there (are we guaranteed that -U will upgrade pip before using older pip to install sphinx?) | 16:45 |
| *** BobBall is now known as BobBallAway | 16:46 | |
| mordred | fungi: it goes in the order listed in requirements :) | 16:47 |
| fungi | well, i have to assume pip is an implicit requirement since it does get installed into the virtualenv but is not listed in the tox deps | 16:48 |
| openstackgerrit | Joe Gordon proposed a change to openstack-infra/config: Move large-ops test to check pipeline https://review.openstack.org/46840 | 16:48 |
| *** flaper87 is now known as flaper87|afk | 16:49 | |
| mordred | BobBallAway: http://paste.openstack.org/show/47169/ <-- there is run_tests.sh on my machine | 16:49 |
| BobBallAway | That's what I get - thanks for confirming the same mordred | 16:50 |
| mordred | BobBallAway: here's the logs from my tox run: http://paste.openstack.org/show/47171 | 16:50 |
| *** sarob has joined #openstack-infra | 16:50 | |
| BobBallAway | mordred: might be something to do with python-glanceclient... tox uses the one already in your environment but run_tests downloads a new one | 16:54 |
| mordred | BobBallAway: oh. fascinating | 16:54 |
| BobBallAway | but then maybe not - your tox run says that Requirement already up-to-date: jsonpointer>=1.0... which is what was pulling in the bad six for me... | 16:55 |
| BobBallAway | I dunno! | 16:55 |
| BobBallAway | maybe I'll look more tomorrow | 16:55 |
| *** matty_dubs|lunch is now known as matty_dubs | 16:55 | |
| mordred | kk | 16:56 |
| mordred | BobBallAway: well, one difference I can see that sholdn't matter | 16:56 |
| mordred | is that run_tests.sh does "pip install -U -r requirements.txt ; pip install -U -r test-requirements.txt" - tox does pip install -U -r requirements.txt -r test-requirements.txt | 16:56 |
| *** dizquierdo has left #openstack-infra | 16:57 | |
| *** wenlock has quit IRC | 16:58 | |
| *** nati_ueno has joined #openstack-infra | 16:59 | |
| *** afazekas_ has joined #openstack-infra | 16:59 | |
| *** afazekas_ has quit IRC | 16:59 | |
| BobBallAway | eurgh - that's a horrible difference! so tox satisfies both at the same time but run_tests will upgrade things asked for by test-requirements even if they were capped by requirements? | 17:00 |
| *** julim has joined #openstack-infra | 17:00 | |
| BobBallAway | confirmed - thanks mordred. I updated tools/venv_common.py and it picked a valid version of six | 17:03 |
| BobBallAway | I'll tidy up and submit a patch tomorrow I hope! | 17:03 |
| *** zehicle_at_dell has quit IRC | 17:05 | |
| *** DinaBelova has quit IRC | 17:05 | |
| *** fbo is now known as fbo_away | 17:06 | |
| *** SergeyLukjanov has quit IRC | 17:06 | |
| *** gyee has joined #openstack-infra | 17:09 | |
| mordred | BobBallAway: yay! glad we found that | 17:11 |
| *** _TheDodd_ has joined #openstack-infra | 17:13 | |
| jog0 | huh http://logs.openstack.org/40/46840/3/check/gate-ci-docs/b13e62c | 17:16 |
| fungi | jog0: https://review.openstack.org/#/c/46984/1 | 17:16 |
| jog0 | fungi: haha thanks | 17:17 |
| *** thedodd has quit IRC | 17:17 | |
| *** vipul is now known as vipul-away | 17:17 | |
| fungi | np | 17:17 |
| fungi | short version, sphinx released 1.2b2 today and it seems to be buggy | 17:17 |
| *** comstud_ is now known as comstud | 17:18 | |
| fungi | not an issue most places we consume it, because using new enough pip won't install that beta version without extra coercing | 17:18 |
| *** pcrews has joined #openstack-infra | 17:21 | |
| *** nicedice has joined #openstack-infra | 17:22 | |
| *** vipul-away is now known as vipul | 17:23 | |
| openstackgerrit | Monty Taylor proposed a change to openstack-dev/pbr: Comply with pep440 https://review.openstack.org/46996 | 17:23 |
| mordred | ttx: ^^ that's the change I think we need to make to our versioning | 17:24 |
| mordred | zul: ^^ I do not _believe_ that should affect you - but I would like your feedback | 17:24 |
| zul | shouldnt affect us | 17:25 |
| mordred | zul: ossum | 17:26 |
| zul | in theory | 17:26 |
| *** Ryan_Lane has joined #openstack-infra | 17:31 | |
| *** mrmartin has quit IRC | 17:34 | |
| *** markmcclain has quit IRC | 17:34 | |
| *** burt has joined #openstack-infra | 17:37 | |
| *** SergeyLukjanov has joined #openstack-infra | 17:38 | |
| *** zehicle_at_dell has joined #openstack-infra | 17:39 | |
| *** atiwari has joined #openstack-infra | 17:45 | |
| *** pmoosh has joined #openstack-infra | 17:51 | |
| mgagne1 | How are public/private keys for cert handled by openstack infra? Is the raw content put in Hiera or is the file manually copied to the server? | 17:53 |
| *** mgagne1 is now known as mgagne | 17:53 | |
| *** mrodden has joined #openstack-infra | 17:57 | |
| *** DinaBelova has joined #openstack-infra | 17:59 | |
| fungi | mgagne: we usually put both parts in hiera (public cert and corresponding private key) | 18:01 |
| mgagne | fungi: in the yaml file? | 18:01 |
| mgagne | fungi: And I guess hiera is deployed from a private repo =) | 18:02 |
| *** sarob has quit IRC | 18:03 | |
| *** dkranz has joined #openstack-infra | 18:03 | |
| *** sarob has joined #openstack-infra | 18:03 | |
| *** rnirmal has quit IRC | 18:04 | |
| fungi | it's not so much deployed as manually edited, but we do have a local git repo for that file which resides on our puppet master, just so we can easily version/annotate changes we make to it | 18:04 |
| davidhadas | mordred: fungi: It was solved by doing easy_install to the pbr egg | 18:05 |
| davidhadas | for some reason teh pbr egg was downloaded to the locla directory but never installed on teh system | 18:05 |
| davidhadas | This happened in upgrading swift | 18:05 |
| fungi | davidhadas: yes, under normal circumstances pbr is not a run-time dependency so it's only needed to build the release state of the project (make an sdist tarball, et cetera) | 18:06 |
| *** sarob has quit IRC | 18:08 | |
| *** nicedice has quit IRC | 18:09 | |
| *** pblaho has joined #openstack-infra | 18:10 | |
| *** zehicle_at_dell has quit IRC | 18:15 | |
| *** nicedice has joined #openstack-infra | 18:15 | |
| *** zehicle_at_dell has joined #openstack-infra | 18:16 | |
| *** rnirmal has joined #openstack-infra | 18:23 | |
| *** sarob has joined #openstack-infra | 18:25 | |
| *** markmc_ has joined #openstack-infra | 18:25 | |
| *** markmc_ is now known as markmc | 18:26 | |
| *** sarob has quit IRC | 18:26 | |
| *** sarob has joined #openstack-infra | 18:26 | |
| *** sarob has quit IRC | 18:31 | |
| *** senk has quit IRC | 18:31 | |
| *** jcoufal has quit IRC | 18:33 | |
| *** DinaBelova has quit IRC | 18:37 | |
| *** cody-somerville has joined #openstack-infra | 18:47 | |
| *** cody-somerville has joined #openstack-infra | 18:47 | |
| *** jcoufal has joined #openstack-infra | 18:52 | |
| *** sarob has joined #openstack-infra | 18:52 | |
| *** mrmartin has joined #openstack-infra | 18:53 | |
| zul | heh down to 151 errors in python3 for novaclient | 18:54 |
| *** jhesketh has joined #openstack-infra | 18:57 | |
| *** jerryz has joined #openstack-infra | 19:00 | |
| *** Ryan_Lane has quit IRC | 19:00 | |
| *** Ryan_Lane has joined #openstack-infra | 19:00 | |
| *** tstevenson has quit IRC | 19:02 | |
| *** mriedem has quit IRC | 19:04 | |
| *** mriedem has joined #openstack-infra | 19:04 | |
| *** yportnova__ has joined #openstack-infra | 19:05 | |
| jhesketh | jeblair: no infra meeting this morning? | 19:05 |
| *** yportnova__ has quit IRC | 19:11 | |
| *** Ryan_Lane has quit IRC | 19:12 | |
| *** Ryan_Lane has joined #openstack-infra | 19:12 | |
| *** hashar has joined #openstack-infra | 19:14 | |
| *** SergeyLukjanov has quit IRC | 19:15 | |
| *** jhesketh has quit IRC | 19:15 | |
| *** hashar has quit IRC | 19:16 | |
| *** tstevenson has joined #openstack-infra | 19:18 | |
| fungi | jhesketh__: most everyone's busy, but we can have a quick one if you want | 19:18 |
| fungi | i think clarkb mentioned wanting to break away from what he's at to have a quick one, but i haven't heard anything from him today on it | 19:18 |
| *** lnxnut has quit IRC | 19:19 | |
| *** SergeyLukjanov has joined #openstack-infra | 19:21 | |
| mrmartin | hi | 19:21 |
| mrmartin | fungi: what we need to get an approval for groups-dev patch? | 19:21 |
| *** ryanpetrello has quit IRC | 19:22 | |
| fungi | mrmartin: i'm hoping at least one more infra core dev will have a chance to look back over the last few updates for it before approving, but i don't want to hold this up and they're all mostly away from irc/reviews this week | 19:23 |
| fungi | mrmartin: i suppose i can go ahead and fire up the virtual machine for it, test applying that change and see what else breaks which we didn't notice when looking over it | 19:24 |
| *** ryanpetrello has joined #openstack-infra | 19:25 | |
| fungi | i've got a backlog of a couple other things people are trying to get my help on, and then i'll give it a shot | 19:25 |
| Alex_Gaynor | mordred: May I complain that for stuff that isn't openstack doing releases is too much work? I hvae to manually bump version numbers, tag some things, create my sdists and wheels, push, etc. I want `git tag -S && git review` as my only process :/ | 19:25 |
| mrmartin | fung: ok, great, just drop me an email / review message if anything is missing | 19:26 |
| fungi | Alex_Gaynor: clearly you're preaching to the choir, composer, arranger and conductor | 19:26 |
| *** mestery has joined #openstack-infra | 19:26 | |
| mrmartin | fungi: I have some idea to enhance the entire drupal build process, but we need to start with something minimal that works | 19:27 |
| Alex_Gaynor | fungi: I know :) Basically though this comment was in the form of a complaint, really it's praise for how streamlined the OS process is | 19:27 |
| *** pmoosh has quit IRC | 19:28 | |
| clarkb | ya sorry very busy | 19:28 |
| clarkb | I will do a review on groups dev | 19:28 |
| fungi | Alex_Gaynor: we have the most effective and thorough automation here of any project i've ever worked on, open or proprietary. i'm continually impressed (ci!) | 19:28 |
| Alex_Gaynor | fungi: It'd be pretty awesome if someone turned all the openstack stuff into an as-a-service thing. | 19:29 |
| Alex_Gaynor | openstack-process-as-a-service | 19:29 |
| fungi | stackforge sorta does that | 19:30 |
| fungi | but not on separate systems | 19:30 |
| *** gyee has quit IRC | 19:30 | |
| fungi | just coexists with openstack on the same systems with some logical separation | 19:30 |
| *** vipul is now known as vipul-away | 19:34 | |
| *** mrmartin has quit IRC | 19:38 | |
| clarkb | fungi groups dev lgtm. did not approve as my attention is elsewherw. learning a lot about tripleo | 19:39 |
| fungi | clarkb: keep learning. i'll approve and then see what breaks here shortly | 19:39 |
| Alex_Gaynor | clarkb, fungi: Can one of you verify that I got the steps and behaviors of an OpenStack release right: https://gist.github.com/alex/8881eb1662bedf03a3ba (second set of bullet points) | 19:40 |
| fungi | having a look | 19:40 |
| fungi | Alex_Gaynor: we'd like to get there, but that's not (yet) how it works | 19:41 |
| fungi | gerrit lacks the ability to review git tags as separate refs, i think | 19:41 |
| fungi | we push to gerrit instead of review, and there's a special acl which allows the project team leads and their deputies/proxies that specific action | 19:41 |
| fungi | lemme get you a link to our doc | 19:42 |
| fungi | Alex_Gaynor: actual current recommended workflow is https://wiki.openstack.org/wiki/GerritJenkinsGithub#Tagging_a_Release | 19:42 |
| Alex_Gaynor | fungi: Ah, I probably shouldn't lie to people (making a short blog post). So the actual description is `git push --tags gerrit` instead? | 19:42 |
| Alex_Gaynor | oh, /me reads | 19:42 |
| fungi | push the specific tag, actually. --tags is... dangerous | 19:42 |
| Alex_Gaynor | k | 19:43 |
| *** mestery has quit IRC | 19:43 | |
| Alex_Gaynor | fungi: ok, is https://gist.github.com/alex/8881eb1662bedf03a3ba accurate now? | 19:43 |
| fungi | i also didn't know tag had -S in addition to -s | 19:45 |
| mordred | Alex_Gaynor: you are welcome to move all of your projects to here | 19:45 |
| * fungi rtfms | 19:45 | |
| Alex_Gaynor | fungi: let's just assume I typo'd a thing | 19:45 |
| *** cody-somerville has quit IRC | 19:45 | |
| Alex_Gaynor | mordred: pretty sure you don't want my projects for using sqlalchemy with twisted | 19:45 |
| mordred | Alex_Gaynor: why not? we like you :) | 19:45 |
| fungi | Alex_Gaynor: yes, seems to be a typo. you want -s (lower case) or --sign | 19:45 |
| Alex_Gaynor | fungi: cool, nice catch | 19:46 |
| Alex_Gaynor | Ok, posting a thing | 19:46 |
| fungi | also pushing doesn't send the tag to gerrit for review. it just adds the tag to gerrit's copy of the repository and gerrit then generates events about its addition in the event stream so our automation knows to do automated release stuff | 19:47 |
| fungi | as mentioned earlier, we want to get to the point where we can somehow review the tags and approve them, but this is a stop-gap and gets us partway to what we need | 19:48 |
| *** hashar has joined #openstack-infra | 19:48 | |
| fungi | we could probably do it with a separate "tags" project where the signed tag ref content goes into a file which gets reviewed and then the automation applies that ref manually to the repository keeping the original signature intact, but that would be hacky | 19:51 |
| fungi | a "tags" meta-branch in each project might be a similar way to get around that limitation | 19:52 |
| fungi | and has the benefit that it doesn't divorce the review annotation from the project to which the tags are being applied (similar to the meta "config" branch gerrit uses for acls and whatnot) | 19:53 |
| fungi | but far better would be if gerrit just grew support for reviewing tags | 19:53 |
| *** cody-somerville has joined #openstack-infra | 19:56 | |
| *** vipul-away is now known as vipul | 20:03 | |
| *** beagles has quit IRC | 20:03 | |
| jog0 | clarkb: join #openstack-recheck-watch-test | 20:05 |
| mordred | Alex_Gaynor: so - you can actually start using pbr in your project to get tag-based versions without having all of our stuff | 20:05 |
| jog0 | to see recheck elastic search at work | 20:05 |
| Alex_Gaynor | mordred: that's probably a thing I shoudl do | 20:05 |
| mordred | Alex_Gaynor: you can skip the manual edit step at that point - and you won't forget the tag step :) | 20:06 |
| Alex_Gaynor | mordred: I assume there's a way to turn off the autodoc stuff? | 20:06 |
| mordred | Alex_Gaynor: you could then just do git tag -S blah ; python setup.py sdist bdist_wheel upload -S - which while a little bit more verbose is at least closer | 20:06 |
| mordred | Alex_Gaynor: yup | 20:06 |
| mordred | actually, it defaults to off, ou have to turn it on now | 20:07 |
| Alex_Gaynor | mordred: and how about sphinx versioning, how does that work? | 20:07 |
| mordred | Alex_Gaynor: pbr injects things around the build_sphinx command that set the version | 20:07 |
| * mordred hates needing to remember to edit things | 20:08 | |
| Alex_Gaynor | mordred: ah, so you now longer use make html, it's now done through python setup.py build_sphinx ? | 20:08 |
| mordred | yes. becauseI don't understand why I'd use a makefile in a python project when I also already have a python based build system | 20:08 |
| Alex_Gaynor | that's reasonable, I should investigate using this for things | 20:08 |
| mordred | but then, you know, I'm entirely insane | 20:08 |
| mordred | Alex_Gaynor: lemme make you a patch real quick for you to look at | 20:09 |
| *** markmcclain has joined #openstack-infra | 20:15 | |
| mordred | Alex_Gaynor: https://github.com/alex/alchimia/pull/6 | 20:17 |
| Alex_Gaynor | mordred: boarding an airplane now. Will review. | 20:18 |
| mordred | Alex_Gaynor: have a good flight | 20:18 |
| ttx | mordred: looking | 20:19 |
| ttx | mordred: so... it's a bit wrong because we end up having the b1..b2.. versions out for milestones, and -devX is supposed to be always < bY. | 20:23 |
| ttx | mordred: but then, using aX is not really any better | 20:23 |
| mordred | ttx: yeah - I think it would have the exact same sorting semantics as we have right now - except that we'd be using dev instead of a | 20:24 |
| ttx | mordred: so the only remaiing gripe would be disruption to downstream consumers... which can be solved by properly communicating tha whay and the why | 20:25 |
| ttx | what* | 20:25 |
| mordred | the dev, b, dev, b, dev cycle is a little off from pep440 - but I don't that that's avoidable | 20:25 |
| mordred | ttx: I asked zul about it | 20:25 |
| mordred | and he said he doesn't think it'll affect him | 20:25 |
| mordred | thing is- our -a* versions aren't really usable by downstream packagers, so they're mostly skipping using them directly | 20:25 |
| dansmith | where should we report issues with the new closes-bug hotness? | 20:25 |
| mordred | dansmith: right here! (or in a bug on openstack-ci | 20:26 |
| dansmith | mordred: okay, well, sanity-check me first: | 20:26 |
| ttx | mordred: agree that being pep440-compliant at least gives them the option of following that | 20:26 |
| dansmith | it seems like closes-bug doesn't set the bug as in-progress like the old method did, | 20:26 |
| *** dkliban has quit IRC | 20:26 | |
| dansmith | which means we have lots of new bugs that are actually in progress | 20:26 |
| mordred | ttx: so I think that people are either a) only packaging milestones and release or b) using their own version or c) skipping packages and deploying directly from git | 20:26 |
| ttx | mordred: right. Will review and +1 libreally | 20:26 |
| mordred | ttx: which I _think_ is what we want them to do | 20:26 |
| ttx | liberally | 20:26 |
| mordred | hrm | 20:27 |
| mordred | ttx: dansmith's comment above | 20:27 |
| mordred | dansmith: do you have an example? | 20:27 |
| dansmith | mordred: I'll have to go look for one, hang on a sec | 20:27 |
| ttx | mordred: additional benefit would be to return to calling our milestones "alphas" rather than "betas" | 20:28 |
| mordred | ttx: yah. | 20:28 |
| ttx | not sure we'd do that, but having the option is nice | 20:28 |
| ttx | mordred: don't you have to update the zuul pipeline conf in parallel to this change ? | 20:29 |
| ttx | (have it support both -dev and -a) | 20:29 |
| mordred | ttx: nope | 20:29 |
| mordred | ttx: we're never going to tag a release as -dev | 20:29 |
| ttx | oh, that's only tag. ok | 20:29 |
| ttx | looking at dansmith question | 20:29 |
| dansmith | mordred: hmm, well, starting from reviews, it looks like it's working, but I saw some bugs this morning during triage that looked not right | 20:29 |
| mordred | yeah - actually, I think our current regex has a bug - but one that we've never triggered, and this will turn in to not a bug anymore | 20:30 |
| dansmith | mordred: so I'll keep an eye out for others when approaching from the bug side | 20:30 |
| mordred | dansmith: I believe that there are circumstances where it does not change things | 20:30 |
| fungi | dansmith: there are a variety of ways that can happen. if you get some examples i can probably identify why | 20:30 |
| mordred | dansmith: fungi knows everything | 20:30 |
| dansmith | fungi: okay | 20:30 |
| * fungi knows little, figures out much as needed | 20:30 | |
| dansmith | mordred: that's all I know, that he knows everything, which I figure is good enough | 20:30 |
| ttx | dansmith: so i've been seeing circumstantial evidence of failure, but so far we always traced it back to some other problem | 20:31 |
| dansmith | ttx: okay, must be something like that then | 20:31 |
| ttx | dansmith: but... | 20:31 |
| ttx | dansmith: the explanation used to be... "those changes just went in during the transition period, that would explain it" | 20:31 |
| dansmith | haha | 20:31 |
| ttx | and we're now very far fro mthat transition period | 20:32 |
| fungi | dansmith: all the logic is encapsulated within http://git.openstack.org/cgit/openstack-infra/jeepyb/tree/jeepyb/cmd/update_bug.py if you want to take a stab at identifying why a particular example you encounter would have been missed | 20:32 |
| dansmith | well, I dunno when these bugs were filed, as it was all of a few hours ago when I was looking | 20:32 |
| ttx | so if you have examples, I'd definitely look into them | 20:32 |
| dansmith | okay | 20:32 |
| dansmith | it's hard to find examples, other than looking at *lots* of bugs :D | 20:32 |
| ttx | dansmith: I certainly saw a surge in manual closings recently, due to people using the old "Bug" format | 20:33 |
| dansmith | heh, okay | 20:33 |
| *** pblaho has quit IRC | 20:33 | |
| fungi | also there have been a lot of reports of gerrit not adding a comment with a link to the bug, but in all the ones i looked at i was able to track it back to people adding the bug header to their commit message after the initial patchset | 20:33 |
| *** svarnau_ has joined #openstack-infra | 20:34 | |
| fungi | current behavior is to only add a change proposed comment for the first patchset | 20:34 |
| fungi | but it should set it to in progress on any patchset, pretty sure | 20:34 |
| ttx | fungi: btw https://review.openstack.org/#/c/46499/ | 20:35 |
| *** hashar has left #openstack-infra | 20:35 | |
| dansmith | fungi: would it fail if the Closes-bug: wasn't in the same paragraph as the Change-id? | 20:35 |
| dansmith | fungi: because I think normal git-review workflow leaves them with a line separating them | 20:35 |
| *** svarnau has quit IRC | 20:35 | |
| fungi | dansmith: oh, possible that the git-review behavior for adding the change-id header and ensuring it's in the last paragraph is making initially-added headers into a penultimate paragraph | 20:36 |
| fungi | good thinking | 20:36 |
| dansmith | fungi: okay | 20:37 |
| fungi | i'll go digging in git-review in a bit to see if that needs fixing | 20:37 |
| *** dizquierdo has joined #openstack-infra | 20:37 | |
| fungi | though it'll be a little touchy, since git-review will need to figure out if the last paragraph is header lines and not prose | 20:37 |
| fungi | i guess a regexp checking for header-looking lines will suffice though. if all the lines in the last paragraph match ^[^ ]+: then it's probably safe to add the change-id header to that paragraph instead of starting another | 20:39 |
| *** Ryan_Lane has quit IRC | 20:39 | |
| *** Ryan_Lane has joined #openstack-infra | 20:41 | |
| mordred | ttx: also, I'm thinking we should only do the merge-tags-back-to-master on release tags, not on pre-release tags | 20:46 |
| mordred | ttx: I think it will help make the master generated versions closer to being monotonically increasing | 20:46 |
| ttx | mordred: you mean.. not on milestones ? | 20:46 |
| ttx | I thought the original idea was to get the milestone tag (b1..b2..b3) back in tree so that git would return something more sensible | 20:48 |
| clarkb | fungi: dansmith: I have seen the changeid add them in the same paragraph | 20:49 |
| clarkb | changeid above bug header | 20:49 |
| dansmith | clarkb: hmm, mine doesn't seem to do that | 20:49 |
| dansmith | I wonder if that's changed in a later git-review? | 20:49 |
| fungi | dansmith: entirely possible. i'd need to pickaxe around that part to see when it went in | 20:51 |
| mordred | ttx: but the milestone tags don't affect versions on pre-verioned projects | 20:51 |
| mordred | ttx: wait. hrm. lemme go back and think about that again | 20:51 |
| mordred | ttx: pretend I didn't say anything | 20:52 |
| ttx | mordred: I can do that | 20:52 |
| *** rockyg has joined #openstack-infra | 20:52 | |
| ttx | In my current state I can't really wrap my head around the original problem we were fixing anyway | 20:53 |
| ttx | I blame nola food | 20:53 |
| *** sdake_ has joined #openstack-infra | 20:53 | |
| *** sdake_ has quit IRC | 20:53 | |
| *** sdake_ has joined #openstack-infra | 20:53 | |
| fungi | ttx: is the food there too far removed from its french origins? | 20:53 |
| fungi | i mean, most dishes still start with a roux | 20:54 |
| zaro | fungi: hey, any further thoughts on how to sign artifacts for pypi and maven? | 20:55 |
| ttx | fungi: My current state may derive from me actually enjoying the food ride | 20:55 |
| *** SergeyLukjanov has quit IRC | 20:56 | |
| fungi | zaro: it basically needs a broader discussion on whether we want machine-generated signatures on those, or a mechanism to wait for submission of a human-generated signature before continuing the uploading process | 20:56 |
| *** zul has quit IRC | 20:57 | |
| fungi | the latter would stop after tarball generation, and then separate jobs would pick up triggered off a detached signature upload | 20:57 |
| fungi | essentially, we can add more meaningful human signatures into the process, but that also means adding human delays | 20:58 |
| fungi | and to a great extent it's a project decision on whether we want more trust or more automatic | 20:59 |
| zaro | fungi: ok. when do you think discussion can happen? next infra meeting? | 20:59 |
| *** mestery has joined #openstack-infra | 21:00 | |
| *** mestery has quit IRC | 21:00 | |
| fungi | zaro: i think that's a good place to kick off talking about what the workflow for those two possibilities would look like and how supportable they'd be, but we likely want to take it to the dev ml or the project and release meeting after that | 21:00 |
| *** mestery has joined #openstack-infra | 21:00 | |
| fungi | it's entirely possible we might decide we like both, and can allow projects to make use of them in different but related ways (automatically upload with a machine-generated signature and then add human-generated signatures from people who want to vet the released material) | 21:02 |
| *** rfolco has quit IRC | 21:02 | |
| fungi | but on the whole i think whatever we do should be consistent across the project, which means discussing it with the project | 21:03 |
| zaro | fungi: cool. is human-generated signature automatable? | 21:03 |
| jeblair | fungi: https://jenkins02.openstack.org/job/gate-nova-python26/3585/console | 21:03 |
| jeblair | fungi: that looks weird, and has happened a few times on that node | 21:04 |
| fungi | jeblair: having a look | 21:04 |
| fungi | zaro: it's possible to provide scripts to simplify the vetting and signing process, but the human-generated signature is ultimately a person making a decision based on the outcome of the tarball generation step | 21:05 |
| mordred | fungi: I can see benefit of the machine generated signatures - it at least lets you know that this one _did_ come from our system, which could help should pypi get p0wned | 21:05 |
| *** beagles has joined #openstack-infra | 21:05 | |
| fungi | mordred: i concur | 21:05 |
| fungi | i can see benefit to both, and they serve different purposes/mean different things | 21:05 |
| mordred | yah | 21:07 |
| fungi | jeblair: eww, sigpipe... dead socket | 21:07 |
| fungi | the real question is what's up with the slave. i'll take it offline in jenkins for some further prodding | 21:07 |
| *** arosen has joined #openstack-infra | 21:08 | |
| arosen | mordred: random question -- you ever run into this error: http://codepad.org/gCwHqLoP ? | 21:08 |
| *** dkliban has joined #openstack-infra | 21:08 | |
| *** markmcclain has quit IRC | 21:08 | |
| * mordred looks | 21:09 | |
| mordred | arosen: yes. this usually means that you have attempted to install into a venv with an old setuptools and the install has cause that setuptools to get upgraded | 21:10 |
| mordred | workarounds are to manually upgrade pip/setuptools _firsT_ | 21:10 |
| mordred | or also in many cases running "pip install -e ." instead of "python setup.py develop" works too | 21:10 |
| mordred | as most recent pip has a built-in workaround for the breakage due to setuptools upgrade | 21:10 |
| *** beagles has quit IRC | 21:10 | |
| *** beagles has joined #openstack-infra | 21:11 | |
| *** mestery has quit IRC | 21:12 | |
| arosen | mordred: thanks that did the trick | 21:13 |
| *** mrodden has quit IRC | 21:15 | |
| *** vipul is now known as vipul-away | 21:16 | |
| mordred | arosen: woot! | 21:16 |
| *** cody-somerville has quit IRC | 21:28 | |
| *** rockyg has quit IRC | 21:29 | |
| *** zul has joined #openstack-infra | 21:32 | |
| *** pcm_ has quit IRC | 21:34 | |
| *** mriedem has quit IRC | 21:36 | |
| *** mrodden has joined #openstack-infra | 21:36 | |
| *** beagles has quit IRC | 21:36 | |
| zul | ttx/mordred: the artifacts of our ci would use the new version number but we prefix the versions with a 1: anyways | 21:37 |
| *** thomasm has quit IRC | 21:39 | |
| *** zul has quit IRC | 21:40 | |
| *** zul has joined #openstack-infra | 21:40 | |
| *** DennyZhang has joined #openstack-infra | 21:42 | |
| *** changbl has quit IRC | 21:44 | |
| *** mrodden has quit IRC | 21:45 | |
| *** fbo_away is now known as fbo | 21:46 | |
| *** mrodden has joined #openstack-infra | 21:47 | |
| *** alexpilotti has quit IRC | 21:48 | |
| *** gyee has joined #openstack-infra | 21:52 | |
| Alex_Gaynor | mordred: So, two follow questions: a) if I download a tarball from GH, this means it'll have no version info right? b) does the docs thing work with readthedocs? | 21:56 |
| *** vipul-away is now known as vipul | 21:58 | |
| fungi | Alex_Gaynor: github tarballs will be broken as you describe, yes. you need to download one of our tarballs (which have been sdist'd) or sdist a checkout of the tag you want | 21:58 |
| Alex_Gaynor | fungi: ok, just making sure, thanks | 21:58 |
| *** dcramer__ has quit IRC | 21:59 | |
| openstackgerrit | Russell Bryant proposed a change to openstack-infra/devstack-gate: Add docker support https://review.openstack.org/47043 | 21:59 |
| fungi | Alex_Gaynor: and we have a job template which publishes to rtfd.org, so presumably... openstack-infra/gear uses the rtfd upload hook anyway | 22:00 |
| fungi | and also uses pbr | 22:00 |
| *** beagles has joined #openstack-infra | 22:01 | |
| * fungi is heading out for a brisk walk around the neighborhood, but will return in a jiffy | 22:02 | |
| *** boris-42 has quit IRC | 22:03 | |
| *** weshay has quit IRC | 22:04 | |
| *** amotoki_ has joined #openstack-infra | 22:06 | |
| *** dkliban has quit IRC | 22:08 | |
| *** matty_dubs is now known as matty_dubs|gone | 22:09 | |
| *** doctorjnupe has quit IRC | 22:09 | |
| *** DennyZhang has quit IRC | 22:10 | |
| *** zehicle_at_dell has quit IRC | 22:12 | |
| *** pentameter has quit IRC | 22:12 | |
| *** tstevenson has quit IRC | 22:12 | |
| *** _TheDodd_ has quit IRC | 22:16 | |
| *** cody-somerville has joined #openstack-infra | 22:16 | |
| *** prad has quit IRC | 22:24 | |
| *** thomasm has joined #openstack-infra | 22:24 | |
| *** dripton has quit IRC | 22:30 | |
| *** dripton has joined #openstack-infra | 22:31 | |
| *** dizquierdo has left #openstack-infra | 22:31 | |
| *** sdake_ has quit IRC | 22:36 | |
| mordred | Alex_Gaynor: what fungi says is correct | 22:37 |
| * mgagne opens pandora's box | 22:38 | |
| mgagne | can tags be reviewed in gerrit? | 22:38 |
| jeblair | mgagne: no but we really want them to be | 22:38 |
| mordred | mgagne: patches happily accepted | 22:39 |
| mgagne | =) | 22:39 |
| mordred | because clearly that is the last missing piece | 22:39 |
| Alex_Gaynor | mordred: err, no it's not, gerrit can already accept patches | 22:40 |
| *** rnirmal has quit IRC | 22:40 | |
| * mordred bows head | 22:41 | |
| *** afazekas has quit IRC | 22:41 | |
| * Alex_Gaynor chuckles at his own bad joke | 22:42 | |
| *** svarnau has joined #openstack-infra | 22:47 | |
| *** jpich has quit IRC | 22:48 | |
| *** hashar has joined #openstack-infra | 22:49 | |
| *** svarnau_ has quit IRC | 22:50 | |
| *** Ryan_Lane has quit IRC | 22:50 | |
| *** Ryan_Lane has joined #openstack-infra | 22:52 | |
| *** hashar has quit IRC | 22:54 | |
| *** fbo is now known as fbo_away | 22:54 | |
| *** hashar has joined #openstack-infra | 22:54 | |
| *** julim has quit IRC | 22:55 | |
| *** julim has joined #openstack-infra | 22:57 | |
| *** cody-somerville has quit IRC | 22:57 | |
| *** UtahDave has joined #openstack-infra | 22:57 | |
| *** jhesketh__ has quit IRC | 22:58 | |
| *** thomasm has quit IRC | 22:58 | |
| *** hashar has quit IRC | 22:59 | |
| *** jhesketh__ has joined #openstack-infra | 23:00 | |
| *** jhesketh has joined #openstack-infra | 23:01 | |
| *** hashar has joined #openstack-infra | 23:01 | |
| russellb | are there any jenkins slaves running ubuntu 13.04? or just 12.04? | 23:04 |
| *** julim has quit IRC | 23:04 | |
| russellb | i'm looking at doing CI on the docker compute driver, which requires a newer kernel | 23:04 |
| Alex_Gaynor | russellb: I believe they're all precise right now | 23:05 |
| *** jcoufal has quit IRC | 23:05 | |
| russellb | ok, that's what it looked like, figured i'd ask | 23:05 |
| *** mrodden has quit IRC | 23:05 | |
| Alex_Gaynor | russellb: well, there's also some old centos ones, but I seriously doubt that helps you :) | 23:06 |
| russellb | heh, right | 23:06 |
| russellb | and i think docker only works on ubuntu right now, anyway | 23:07 |
| clarkb | russellb: Alex_Gaynor: correct only 12.04 because short term 13.04 support (9months) is painful | 23:08 |
| russellb | yep makes sense | 23:08 |
| mordred | also, we have a weird puppet issue to work out | 23:08 |
| mordred | that said - jeblair and I had a good discussion about dib support for nodepool this week | 23:08 |
| Alex_Gaynor | dib? | 23:08 |
| mordred | so it's conceivable we could start to think about having a plan for having maybe an answer for something the russellb could think about | 23:09 |
| mordred | Alex_Gaynor: diskimage-builder | 23:09 |
| mordred | Alex_Gaynor: using the diskimage-builder tool as a way to describe, build and boot the base images for things like devstack-gate | 23:09 |
| russellb | heh | 23:10 |
| russellb | 12.04 is fine, and there's a newer kernel package available ... but having to get every node for docker testing updated and rebooted with it every time seems like a pain | 23:11 |
| russellb | and that doesn't really work if you want to reuse nodes | 23:11 |
| russellb | i may just punt this back to the docker folks to sort out :-p | 23:11 |
| russellb | you, do CI on your thing! | 23:12 |
| fungi | on the thing! | 23:13 |
| mordred | russellb: well, we usually don't like to reuse nodes that have run devstack | 23:13 |
| russellb | yeah! | 23:13 |
| russellb | ah ok, for some reason i thought you had started doing that | 23:13 |
| mordred | so rebooting actually wouldn't be terrible - especially if you just kexec the new kernel ... | 23:13 |
| mordred | which - I even have the script cantrip for | 23:14 |
| russellb | looks like it's as simple as installing linux-generic-lts-raring and then rebooting | 23:14 |
| russellb | (at least on this rackspace node i'm testing on) | 23:14 |
| mordred | russellb: last four lines here: https://review.openstack.org/#/c/46482/3/scripts/reboot-image.sh | 23:15 |
| russellb | fancy | 23:15 |
| mordred | russellb: so in that case, doing that as part of devstack or something might not be the world's _worst_ thing | 23:15 |
| * mordred will ask jeblair about it at dinner tonight | 23:16 | |
| pleia2 | fwiw, those new kernels for precise have been pretty good so I don't think it's a bad idea | 23:17 |
| *** pcrews has quit IRC | 23:19 | |
| *** pcrews has joined #openstack-infra | 23:19 | |
| mordred | pleia2: are those in cloud-archive? | 23:19 |
| *** hashar is now known as hasharBusy | 23:20 | |
| mordred | (should we just consider installing them on all the devstack nodes and being done with it?) | 23:20 |
| clarkb | sounds reasonable to me. we could even build a second set of images with that kernel already installed | 23:20 |
| russellb | sounds easier than doing it dynamically with kexec :) | 23:21 |
| clarkb | mordred no not cloud archive. they are for lts hardware support | 23:21 |
| pleia2 | mordred: no, even better, they're in the standard precise repos | 23:21 |
| mordred | oh! nice! | 23:21 |
| * mordred votes we just add them to d-g nodes - because it's the least amount of work of anything and it would make russellb happy | 23:21 | |
| russellb | \o/ | 23:21 |
| clarkb | mordred iscsid can't run in containers so sad :( | 23:22 |
| mordred | but, as usual, it's entirely possible I'm ENTIRELY wrong | 23:22 |
| russellb | for reference: http://docs.docker.io/en/latest/installation/kernel/ | 23:22 |
| mordred | clarkb: wait - what? | 23:22 |
| russellb | ^^ is why i'm asking about all this | 23:22 |
| uvirtbot | russellb: Error: "^" is not a valid command. | 23:22 |
| russellb | uvirtbot: oh hush | 23:22 |
| uvirtbot | russellb: Error: "oh" is not a valid command. | 23:22 |
| clarkb | mordred iscsi isnt namespaced in the kernel | 23:22 |
| *** mrodden has joined #openstack-infra | 23:22 | |
| clarkb | mordred so iscsid in a container derps | 23:22 |
| pleia2 | mordred: yeah :( https://bugs.launchpad.net/ubuntu/+source/lxc/+bug/1226855 | 23:22 |
| uvirtbot | Launchpad bug 1226855 in lxc "Cannot use open-iscsi inside LXC container" [Undecided,New] | 23:22 |
| mordred | wow. that's great | 23:22 |
| pleia2 | we're pinging some LXC+Ubuntu people to look at it | 23:23 |
| mordred | so, that pretty much drop-kicks the idea of using containers for tripleo testing? | 23:23 |
| pleia2 | it's a setback, we may be able to figure something out | 23:23 |
| mordred | kk | 23:23 |
| mordred | pleia2: you could always do some kernel hacking. I always find that a simple solution to such problems | 23:23 |
| * mordred ducks | 23:23 | |
| pleia2 | haha | 23:23 |
| mrodden | that seems like a large issue... | 23:24 |
| fungi | though we still might want to reopen bug 1082815 and resurrect https://review.openstack.org/22228 | 23:24 |
| uvirtbot | Launchpad bug 1082815 in openstack-ci "Add ubuntu-cloud-archive repo to ubuntu slaves" [High,Won't fix] https://launchpad.net/bugs/1082815 | 23:24 |
| *** michchap has joined #openstack-infra | 23:24 | |
| *** pcrews has quit IRC | 23:24 | |
| jeblair | https://review.openstack.org/#/c/45896/ | 23:24 |
| mordred | pleia2, clarkb: wouldn't that have general implications for lxc support in openstack across the board? | 23:25 |
| *** michchap has quit IRC | 23:25 | |
| jeblair | fungi: ^ | 23:25 |
| mordred | I mean, that shoudl impact using cinder volumes too right? | 23:25 |
| fungi | jeblair: yep, reviewing it now. missed that one so far | 23:25 |
| mordred | jeblair: ++ | 23:25 |
| *** michchap has joined #openstack-infra | 23:25 | |
| pleia2 | mordred: hm, I don't know about cinder | 23:25 |
| russellb | well, more nova's use of cinder | 23:26 |
| mrodden | cinder usually mounts things at the host level and then abstracts them into some pure block device format to the guests | 23:26 |
| mordred | yah. that's what I meant | 23:26 |
| russellb | docker doesn't support cinder volumes at all | 23:26 |
| mordred | ah. ok | 23:26 |
| russellb | and libvirt+LXC ... *looks* | 23:26 |
| mordred | well, I meant for the libvirt lxc | 23:26 |
| russellb | probably doesn't either though | 23:26 |
| pleia2 | right now this is just tripleo's mechanism for doing stuff is iscsi, and the host iscsi gets really unhappy when the lxc guest tries to use it | 23:27 |
| * mrodden hasn't used lxc in years | 23:27 | |
| pleia2 | unhappy as it "no, *dies*" | 23:27 |
| mordred | loverly | 23:27 |
| russellb | does not support it - https://wiki.openstack.org/wiki/HypervisorSupportMatrix#Hypervisor_feature_support_matrix | 23:27 |
| pleia2 | we figured out all the other issues with seed+lxc :) | 23:29 |
| *** cody-somerville has joined #openstack-infra | 23:30 | |
| *** cody-somerville has quit IRC | 23:30 | |
| *** cody-somerville has joined #openstack-infra | 23:30 | |
| mordred | pleia2: so the only remaining issues are that linux doesn't work that way | 23:30 |
| pleia2 | yeah, that's all | 23:30 |
| *** danger_fo is now known as danger_fo_away | 23:31 | |
| mordred | wow. I didn't realize hyperv was so far along | 23:33 |
| fungi | it has an ungodly amount of funding | 23:33 |
| mordred | it kinda looks like the third best driver basd on that matrix | 23:33 |
| *** dims has quit IRC | 23:36 | |
| *** senk has joined #openstack-infra | 23:36 | |
| *** senk has quit IRC | 23:36 | |
| *** kpavel has joined #openstack-infra | 23:44 | |
| jhesketh__ | Is it possible to restore an abandoned changeset that isn't yours? | 23:47 |
| *** pcrews has joined #openstack-infra | 23:49 | |
| *** UtahDave has quit IRC | 23:51 | |
| *** dims has joined #openstack-infra | 23:52 | |
| clarkb | jhesketh__: no, but do you need one reopened so that you can push to it? | 23:52 |
| clarkb | jhesketh__: I can do that for if you give me a change number | 23:52 |
| jhesketh__ | clarkb: https://review.openstack.org/#/c/40222/ | 23:53 |
| markmc | clarkb, look at this: https://review.openstack.org/#/c/47050/ | 23:53 |
| markmc | clarkb, a comment on a deleted file, but the comment isn't visible | 23:53 |
| jhesketh__ | it just expired because of Jenkins giving a bad review | 23:53 |
| jhesketh__ | and Richard doesn't work on openstack so he hasn't paid attention since submitting it | 23:53 |
| clarkb | jhesketh__: done | 23:53 |
| clarkb | markmc: looking | 23:54 |
| markmc | clarkb, I'll file a bug if you don't recognize it as a known issue | 23:54 |
| clarkb | markmc: weird, I would file a bug | 23:54 |
| markmc | clarkb, grand, thanks | 23:54 |
| markmc | clarkb, https://bugs.launchpad.net/openstack-ci/+bug/1226880 | 23:56 |
| uvirtbot | Launchpad bug 1226880 in openstack-ci "gerrit: comment on a deleted file not show in diff view" [Undecided,New] | 23:56 |
| jhesketh__ | thanks cla | 23:58 |
| jhesketh__ | *clarkb | 23:58 |
| *** sarob_ has joined #openstack-infra | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!