| *** jamesmcarthur has quit IRC | 00:00 | |
| *** hamzy_ has joined #openstack-infra | 00:01 | |
| *** hamzy_ is now known as hamzy | 00:02 | |
| *** Swami has quit IRC | 00:04 | |
| *** jamesmcarthur has joined #openstack-infra | 00:07 | |
| *** jamesmcarthur has quit IRC | 00:08 | |
| *** jamesmcarthur has joined #openstack-infra | 00:11 | |
| *** jamesmcarthur has quit IRC | 00:11 | |
| *** lathiat has quit IRC | 00:14 | |
| *** slagle has quit IRC | 00:15 | |
| *** slagle has joined #openstack-infra | 00:15 | |
| *** rcernin has quit IRC | 00:27 | |
| *** rcernin has joined #openstack-infra | 00:29 | |
| *** longkb has joined #openstack-infra | 00:36 | |
| *** ijw has joined #openstack-infra | 00:38 | |
| *** dave-mccowan has quit IRC | 01:16 | |
| *** mrsoul has joined #openstack-infra | 01:17 | |
| *** jamesmcarthur has joined #openstack-infra | 01:17 | |
| *** lathiat has joined #openstack-infra | 01:17 | |
| *** felipemonteiro has joined #openstack-infra | 01:20 | |
| *** jamesmcarthur has quit IRC | 01:22 | |
| *** diablo_rojo has quit IRC | 01:24 | |
| *** aluria has quit IRC | 01:29 | |
| *** felipemonteiro has quit IRC | 01:36 | |
| *** felipemonteiro has joined #openstack-infra | 01:36 | |
| *** edmondsw has joined #openstack-infra | 01:45 | |
| *** rickhlwong has joined #openstack-infra | 01:53 | |
| *** jamesmcarthur has joined #openstack-infra | 01:54 | |
| *** hongbin has joined #openstack-infra | 01:57 | |
| *** agopi has joined #openstack-infra | 01:57 | |
| rickhlwong | exit | 01:58 |
|---|---|---|
| rickhlwong | quit | 01:58 |
| *** rickhlwong has quit IRC | 01:58 | |
| *** gyee has quit IRC | 02:01 | |
| *** ramishra has joined #openstack-infra | 02:09 | |
| *** ijw has quit IRC | 02:14 | |
| *** felipemonteiro has quit IRC | 02:17 | |
| *** jamesmcarthur has quit IRC | 02:20 | |
| *** yamamoto has quit IRC | 02:23 | |
| *** yamamoto has joined #openstack-infra | 02:23 | |
| *** jamesmcarthur has joined #openstack-infra | 02:24 | |
| *** jamesmcarthur has quit IRC | 02:35 | |
| *** dosaboy has quit IRC | 02:35 | |
| *** jamesmcarthur has joined #openstack-infra | 02:36 | |
| *** vivsoni has quit IRC | 02:56 | |
| *** eernst has joined #openstack-infra | 02:56 | |
| *** rh-jelabarre has quit IRC | 02:57 | |
| *** annp has joined #openstack-infra | 03:01 | |
| *** lathiat has quit IRC | 03:08 | |
| *** felipemonteiro has joined #openstack-infra | 03:12 | |
| *** agopi has quit IRC | 03:21 | |
| *** agopi has joined #openstack-infra | 03:21 | |
| *** rlandy has quit IRC | 03:22 | |
| *** sung-il9 has joined #openstack-infra | 03:23 | |
| *** lathiat has joined #openstack-infra | 03:34 | |
| openstackgerrit | Dominique Martinet proposed openstack/gertty master: gitrepo DiffFile: convert tab to ยป + spaces https://review.openstack.org/608154 | 03:35 |
| openstackgerrit | Dominique Martinet proposed openstack/gertty master: typo - s/fileojb/fileobj/ https://review.openstack.org/608155 | 03:35 |
| *** kaiokmo has quit IRC | 03:41 | |
| *** kaiokmo has joined #openstack-infra | 03:41 | |
| *** hongbin has quit IRC | 03:42 | |
| *** udesale has joined #openstack-infra | 03:49 | |
| *** jamesmcarthur has quit IRC | 03:57 | |
| *** witek has quit IRC | 04:06 | |
| *** ykarel__ has joined #openstack-infra | 04:15 | |
| *** eernst has quit IRC | 04:15 | |
| *** eernst has joined #openstack-infra | 04:16 | |
| *** ykarel__ is now known as ykarel | 04:34 | |
| *** roman_g has quit IRC | 04:41 | |
| *** rcernin has quit IRC | 04:42 | |
| *** eernst has quit IRC | 04:45 | |
| *** rcernin has joined #openstack-infra | 04:46 | |
| *** kukacz has quit IRC | 04:52 | |
| *** flaper87 has quit IRC | 04:52 | |
| *** niedbalski has quit IRC | 04:52 | |
| *** jbadiapa has quit IRC | 04:52 | |
| *** openstackgerrit has quit IRC | 04:52 | |
| *** d0ugal has quit IRC | 04:52 | |
| *** dhill_ has quit IRC | 04:52 | |
| *** kukacz has joined #openstack-infra | 04:57 | |
| *** niedbalski has joined #openstack-infra | 04:57 | |
| *** flaper87 has joined #openstack-infra | 04:57 | |
| *** jbadiapa has joined #openstack-infra | 04:57 | |
| *** openstackgerrit has joined #openstack-infra | 04:57 | |
| *** d0ugal has joined #openstack-infra | 04:57 | |
| *** dhill_ has joined #openstack-infra | 04:57 | |
| *** Bhujay has joined #openstack-infra | 04:59 | |
| *** Bhujay has quit IRC | 05:00 | |
| *** Bhujay has joined #openstack-infra | 05:00 | |
| *** agopi has quit IRC | 05:03 | |
| *** felipemonteiro has quit IRC | 05:03 | |
| *** jamesmcarthur has joined #openstack-infra | 05:04 | |
| AJaeger | frickler, could you review https://review.openstack.org/601583 and https://review.openstack.org/608069 , please? | 05:05 |
| *** jamesmcarthur has quit IRC | 05:09 | |
| *** kjackal has joined #openstack-infra | 05:15 | |
| openstackgerrit | Merged openstack-infra/zuul-jobs master: support passing extra arguments to bdist_wheel in build-python-release https://review.openstack.org/607900 | 05:22 |
| *** agopi has joined #openstack-infra | 05:24 | |
| *** kjackal has quit IRC | 05:33 | |
| *** e0ne has joined #openstack-infra | 05:34 | |
| *** jtomasek has joined #openstack-infra | 05:45 | |
| *** e0ne has quit IRC | 05:50 | |
| openstackgerrit | Matthew Thode proposed openstack-infra/project-config master: enable sqlite in python https://review.openstack.org/608161 | 05:51 |
| prometheanfire | :( | 05:51 |
| *** e0ne has joined #openstack-infra | 05:53 | |
| *** Emine has quit IRC | 05:56 | |
| *** kjackal has joined #openstack-infra | 06:01 | |
| *** e0ne has quit IRC | 06:12 | |
| AJaeger | frickler: could you review prometheanfire's change above as well, please? ^ | 06:27 |
| AJaeger | thanks! | 06:27 |
| openstackgerrit | Merged openstack-infra/project-config master: remove job settings for trove repositories https://review.openstack.org/601583 | 06:34 |
| *** udesale has quit IRC | 06:39 | |
| *** markvoelker has joined #openstack-infra | 06:40 | |
| *** kopecmartin|scho is now known as kopecmartin|ruck | 06:41 | |
| openstackgerrit | Merged openstack-infra/project-config master: always build universal wheels https://review.openstack.org/607902 | 06:42 |
| *** markvoelker has quit IRC | 06:45 | |
| *** Bhujay has quit IRC | 06:46 | |
| *** quiquell|off is now known as quiquell|brb | 06:50 | |
| *** udesale has joined #openstack-infra | 06:53 | |
| *** pcaruana has joined #openstack-infra | 06:57 | |
| *** kjackal has quit IRC | 06:58 | |
| openstackgerrit | Merged openstack-infra/project-config master: enable sqlite in python https://review.openstack.org/608161 | 07:00 |
| *** ijw has joined #openstack-infra | 07:03 | |
| *** rcernin has quit IRC | 07:04 | |
| *** roman_g has joined #openstack-infra | 07:07 | |
| *** ijw has quit IRC | 07:07 | |
| *** witek has joined #openstack-infra | 07:08 | |
| prometheanfire | AJaeger: mind kicking off another gentoo image? | 07:08 |
| *** quiquell|brb is now known as quiquell | 07:08 | |
| *** rtjure has quit IRC | 07:09 | |
| AJaeger | prometheanfire: I don't have those permissions, only infra-root can help ^ | 07:10 |
| *** jpena|off is now known as jpena | 07:10 | |
| *** kjackal has joined #openstack-infra | 07:10 | |
| prometheanfire | ah, k | 07:11 |
| *** rtjure has joined #openstack-infra | 07:12 | |
| *** njohnston has quit IRC | 07:17 | |
| *** amoralej|off is now known as amoralej | 07:17 | |
| *** shardy has joined #openstack-infra | 07:28 | |
| *** olivierb_ has joined #openstack-infra | 07:36 | |
| *** dosaboy has joined #openstack-infra | 07:38 | |
| *** bauzas is now known as PapaOurs | 07:39 | |
| *** tosky has joined #openstack-infra | 07:45 | |
| *** ykarel has quit IRC | 07:46 | |
| *** ykarel has joined #openstack-infra | 07:46 | |
| *** aluria has joined #openstack-infra | 07:47 | |
| *** e0ne has joined #openstack-infra | 07:49 | |
| *** lpetrut has joined #openstack-infra | 07:54 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: Revert "Revert "web: rewrite interface in react"" https://review.openstack.org/607479 | 08:13 |
| *** priteau has joined #openstack-infra | 08:25 | |
| *** aojeagarcia has joined #openstack-infra | 08:26 | |
| quiquell | Hello any infra-root here that can prioritiza promotion blocker review https://review.openstack.org/#/c/608080/ at tripleo gate ? | 08:35 |
| *** derekh has joined #openstack-infra | 08:35 | |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/project-config master: remove job settings for neutron repositories https://review.openstack.org/597865 | 08:37 |
| quiquell | prometheanfire: Are you infra-root ? | 08:38 |
| *** panda|off is now known as panda | 08:40 | |
| *** markvoelker has joined #openstack-infra | 08:41 | |
| *** electrofelix has joined #openstack-infra | 08:42 | |
| *** ykarel_ has joined #openstack-infra | 08:44 | |
| *** ykarel has quit IRC | 08:47 | |
| *** ykarel_ is now known as ykarel|lunch | 08:51 | |
| *** gfidente has joined #openstack-infra | 09:01 | |
| openstackgerrit | Pierre Riteau proposed openstack/diskimage-builder master: Fail build due to missing kauditd only when SELinux is enabled https://review.openstack.org/608187 | 09:03 |
| openstackgerrit | Colleen Murphy proposed openstack-infra/puppet-pip master: [DNM] debugging https://review.openstack.org/607048 | 09:03 |
| *** janki has joined #openstack-infra | 09:05 | |
| *** chandankumar has joined #openstack-infra | 09:06 | |
| *** agopi has quit IRC | 09:14 | |
| *** markvoelker has quit IRC | 09:15 | |
| *** efried has quit IRC | 09:17 | |
| *** efried1 has joined #openstack-infra | 09:17 | |
| *** efried1 is now known as efried | 09:19 | |
| openstackgerrit | melissaml proposed openstack/diskimage-builder master: fix a typo https://review.openstack.org/608191 | 09:26 |
| *** Emine has joined #openstack-infra | 09:32 | |
| *** stephenfin is now known as finucannot | 09:35 | |
| *** chandankumar has quit IRC | 09:48 | |
| *** ykarel|lunch is now known as ykarel | 09:54 | |
| *** chandankumar has joined #openstack-infra | 09:59 | |
| *** yamamoto has quit IRC | 10:07 | |
| *** yamamoto has joined #openstack-infra | 10:08 | |
| *** yamamoto has quit IRC | 10:09 | |
| *** yamamoto has joined #openstack-infra | 10:10 | |
| *** markvoelker has joined #openstack-infra | 10:12 | |
| *** yamamoto has quit IRC | 10:13 | |
| *** yamamoto has joined #openstack-infra | 10:14 | |
| AJaeger | dhellmann: https://review.openstack.org/#/c/597576 is ready for review, isn't it? And please WIP https://review.openstack.org/597865 again | 10:14 |
| *** yamamoto has quit IRC | 10:16 | |
| *** yamamoto has joined #openstack-infra | 10:17 | |
| *** Emine has quit IRC | 10:17 | |
| *** udesale has quit IRC | 10:18 | |
| *** yamamoto has quit IRC | 10:21 | |
| *** chandankumar has quit IRC | 10:26 | |
| *** yamamoto has joined #openstack-infra | 10:27 | |
| *** yamamoto has quit IRC | 10:32 | |
| *** yamamoto has joined #openstack-infra | 10:33 | |
| *** sshnaidm is now known as sshnaidm|off | 10:35 | |
| *** yamamoto has quit IRC | 10:38 | |
| *** longkb has quit IRC | 10:41 | |
| *** longkb has joined #openstack-infra | 10:42 | |
| *** longkb has quit IRC | 10:43 | |
| *** jtomasek has quit IRC | 10:43 | |
| *** markvoelker has quit IRC | 10:44 | |
| *** kjackal has quit IRC | 10:44 | |
| *** ijw has joined #openstack-infra | 10:45 | |
| *** ijw has quit IRC | 10:50 | |
| *** jamesmcarthur has joined #openstack-infra | 10:51 | |
| *** annp has quit IRC | 10:53 | |
| *** kjackal has joined #openstack-infra | 10:56 | |
| *** jamesmcarthur has quit IRC | 10:58 | |
| *** jchhatbar has joined #openstack-infra | 10:59 | |
| *** janki has quit IRC | 11:00 | |
| *** jchhatbar has quit IRC | 11:01 | |
| *** jchhatbar has joined #openstack-infra | 11:02 | |
| *** jpena is now known as jpena|lunch | 11:04 | |
| *** jesusaur has quit IRC | 11:06 | |
| *** jchhatba_ has joined #openstack-infra | 11:06 | |
| *** jtomasek has joined #openstack-infra | 11:07 | |
| *** yamamoto has joined #openstack-infra | 11:08 | |
| *** jchhatbar has quit IRC | 11:09 | |
| *** dtantsur|afk is now known as dtantsur | 11:09 | |
| *** jesusaur has joined #openstack-infra | 11:14 | |
| *** rpittau_ has quit IRC | 11:17 | |
| *** yamamoto has quit IRC | 11:18 | |
| *** yamamoto has joined #openstack-infra | 11:19 | |
| *** quiquell is now known as quiquell|lunch | 11:20 | |
| *** yamamoto has quit IRC | 11:23 | |
| *** tpsilva has joined #openstack-infra | 11:23 | |
| *** yamamoto has joined #openstack-infra | 11:24 | |
| fungi | quiquell|lunch: i've promoted 608080,2 just now | 11:38 |
| *** markvoelker has joined #openstack-infra | 11:41 | |
| *** quiquell|lunch is now known as quiquell | 11:42 | |
| quiquell | fungi: thanks ! | 11:42 |
| mrhillsman | hey folks, any thoughts on this: | 11:45 |
| mrhillsman | Oct 5 11:44:56 openlab-zuul zuul-scheduler[6129]: 2018-10-05 11:44:56,309 WARNING kazoo.client: Transition to CONNECTING | 11:45 |
| mrhillsman | Oct 5 11:45:03 openlab-zuul zuul-scheduler[6129]: 2018-10-05 11:45:03,108 WARNING kazoo.client: Connection dropped: outstanding heartbeat ping not received | 11:45 |
| dhellmann | AJaeger : done | 11:45 |
| mrhillsman | `nc -v -w2 192.168.200.147 2181` returns success so i know the issue is not the port being inaccessible | 11:46 |
| dhellmann | fungi , lifeless : I think we don't need to care about the contents of the setup.cfg files. My reasoning is in http://lists.openstack.org/pipermail/openstack-dev/2018-October/135461.html | 11:46 |
| dhellmann | let me know if my logic is faulty, of course | 11:46 |
| *** jtomasek has quit IRC | 11:48 | |
| sekharvajjula | Hi guys. I am trying to setup a 3rdparty CI for running Ironic testcases on Nokia Hardware. Still having issues related to zuul configurations and the project dependencies. Planning to post my logs and config files in some mailing list, to get some support. Do you think openstack-infra@lists.openstack.org is right place? | 11:51 |
| *** e0ne has quit IRC | 11:54 | |
| *** rh-jelabarre has joined #openstack-infra | 11:56 | |
| fungi | dhellmann: i've also replied. depends on what you mean by "we" | 11:56 |
| fungi | dhellmann: if "we" means "openstack, don't care about anybody else" then sure | 11:57 |
| dhellmann | I'm not sure why updating the setup.cfg in oslo.config affects anyone else? | 11:57 |
| fungi | sekharvajjula: yes | 11:58 |
| *** agopi has joined #openstack-infra | 11:58 | |
| dhellmann | I'm not saying no one anywhere needs to update that setting. I'm saying the reason given for doing it in all of *our* files doesn't make sense to me. | 11:58 |
| fungi | dhellmann: the concern seemed to be that if pbr updates its setup.cfg file to respect the transition in latest wheel, then there will be an avalanche of changes proposed to other projects by people who don't know any better? | 11:59 |
| *** agopi has quit IRC | 12:00 | |
| dhellmann | fungi : there is already such an avalanche | 12:00 |
| *** jtomasek has joined #openstack-infra | 12:00 | |
| dhellmann | there are at least 3 people writing those patches | 12:00 |
| dhellmann | the idea that it is somehow "more correct" confuses me | 12:01 |
| *** jtomasek has quit IRC | 12:01 | |
| fungi | i just want to make it clear when we provide such advice that we're only talking about projects using openstack-project-specific release workflow not needing to worry about it | 12:01 |
| dhellmann | what benefit do you perceive in any project anywhere including that flag in setup.cfg? | 12:01 |
| *** jtomasek has joined #openstack-infra | 12:02 | |
| fungi | what benefit do you perceive in any project anywhere building universal wheels? | 12:02 |
| fungi | or ceasing to use deprecated section names? | 12:02 |
| fungi | (which was the actual question in the pbr change) | 12:02 |
| dhellmann | universal wheels published to PyPI are more useful to more consumers. I don't see any real point in using them locally. | 12:02 |
| fungi | i'll repeat: i just want to make it clear when we provide such advice that we're only talking about projects using openstack-project-specific release workflow not needing to worry about it | 12:03 |
| dhellmann | having the flag in that file is a convenience so that when running "python setup.py bdist_wheel" you don't have to add the extra argument. So projects that publish packages by hand may want it. | 12:03 |
| dhellmann | ok, I guess I assumed that if we're talking about this on the openstack mailing list that was implied | 12:03 |
| fungi | other non-openstack projects will still want a mechanism to build universal wheels, using standard commands recommended by python packaging guides | 12:04 |
| dhellmann | are you talking about non-openstack projects hosted on openstack.org? or globally? | 12:04 |
| fungi | more the former | 12:04 |
| dhellmann | do we have any of those who are not using the CI system to publish packages after a tag is applied? | 12:05 |
| dhellmann | I imagine we do | 12:05 |
| fungi | i'm not going to recommend openstack's special release build jobs to non-openstack projects hosted on out infrastructure | 12:05 |
| fungi | s/out/our/ | 12:05 |
| *** boden has joined #openstack-infra | 12:05 | |
| dhellmann | sure | 12:05 |
| fungi | there are also potentially non-openstack projects following code review discussions on pbr changes. it's quite popular | 12:06 |
| *** jpena|lunch is now known as jpena | 12:06 | |
| *** felipemonteiro has joined #openstack-infra | 12:07 | |
| fungi | and one of the initial takeaways which started that ml thread was that someone got the impression configuring your setup.cfg to build universal wheels was actively harmful and should be avoided | 12:07 |
| *** udesale has joined #openstack-infra | 12:12 | |
| fungi | (a ptl, in fact) | 12:13 |
| *** markvoelker has quit IRC | 12:15 | |
| sekharvajjula | fungi: Thanks. should I subscribe to the mailing list first in order to send mails? | 12:15 |
| fungi | sekharvajjula: i recommend doing so, yes, for two reasons. first, posts from non-subscribers will get stuck in the moderation queue until someone gets around to approving them; second, replies are likely to go to the list only and subscribing will still get you a copy without having to periodically check the list archives | 12:17 |
| mrhillsman | any idea why i see success status and logs but zuul is not reporting back to github | 12:17 |
| mrhillsman | and all of our nodepool nodes are stuck in-use | 12:18 |
| mrhillsman | payloads are successfully delivered | 12:18 |
| mrhillsman | jobs are queued up | 12:18 |
| fungi | mrhillsman: i'm not really familiar with github... you may have more luck asking in #zuul | 12:18 |
| mrhillsman | thx | 12:19 |
| sekharvajjula | fungi: Sure. I have sent a subscription request. I will await for its approval, before sending my query. | 12:19 |
| AJaeger | fungi, could you review next python3-first change, please? https://review.openstack.org/#/c/597576 openstack-ansible is ready... | 12:19 |
| fungi | sekharvajjula: subscription confirmation is automated. if you try to subscribe you should immediately receive a request confirmation reply with instructions to follow | 12:20 |
| TheJulia | fungi: I am aware of the arm64 notes, but I haven't ever gotten far enough to give them a spin and build a job on them. They essentially would be needed to prepare a ramdisk since we can't cross-os build it. Maybe it might make sense to try and move one of our IPA jobs to arm64 at some point. | 12:23 |
| *** rlandy has joined #openstack-infra | 12:25 | |
| *** amoralej is now known as amoralej|lunch | 12:27 | |
| *** agopi has joined #openstack-infra | 12:27 | |
| fungi | TheJulia: out of curiosity, is it that you really can't use a multiarch environment to cross-build those from amd64, or simply that it's a lot more complicated to set up than it's worth? | 12:27 |
| *** dtantsur is now known as dtantsur|brb | 12:27 | |
| fungi | i've done a fair amount of arm/armel/armhf/arm64 compiles on amd64 systems in service of slow socs | 12:28 |
| fungi | and i agree at least that it's a pain | 12:29 |
| *** e0ne has joined #openstack-infra | 12:29 | |
| sekharvajjula | fungi: Hmm. Did not get any mail still. waiting :) | 12:29 |
| fungi | sekharvajjula: did you subscribe using the web form at http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-infra or by sending a message to openstack-infra-request@lists.openstack.org with a subject line that says "subscribe"? | 12:32 |
| *** aidin has joined #openstack-infra | 12:32 | |
| sekharvajjula | fungi: 1st option (on the web page). | 12:34 |
| openstackgerrit | Merged openstack-infra/project-config master: remove job settings for OpenStackAnsible repositories https://review.openstack.org/597576 | 12:35 |
| *** dosaboy has quit IRC | 12:35 | |
| AJaeger | yeah! Only three repos to go - out of 65 - for python3-first ^ | 12:35 |
| *** dosaboy has joined #openstack-infra | 12:36 | |
| *** dosaboy has quit IRC | 12:36 | |
| fungi | sekharvajjula: looks like you tried twice? 12:03:53 and 12:26:38 utc | 12:40 |
| *** trown|outtypewww is now known as trown | 12:41 | |
| fungi | sekharvajjula: the confirmation messages were both delivered to your mta at least. 250 response from nokia-com.mail.protection.outlook.com | 12:41 |
| fungi | internal queue ids on your end are 2645699854801 and 2645699857133 respectively | 12:42 |
| fungi | sekharvajjula: have you checked your spam folder or similar? | 12:42 |
| sekharvajjula | fungi: I checked in spam, Junk. not found anywhere. :( | 12:45 |
| fungi | sekharvajjula: we've had problems with outlook.com/hotmail/office-365 not delivering messages from our listserv in the past (users of that service incorrectly reporting moderator notifications as spam), so if you have another e-mail address you can use you might have more luck | 12:47 |
| sekharvajjula | fungi: Tried from Gmail and it worked. Thank you! | 12:49 |
| fungi | glad you found a workaround. now i get to start keeping an eye out for microsoft blocking our mailing lists again | 12:49 |
| *** ansmith has joined #openstack-infra | 12:50 | |
| *** dosaboy has joined #openstack-infra | 12:56 | |
| *** dims_ has quit IRC | 12:58 | |
| sekharvajjula | fungi: I got a confirmation mail. So I have sent an email with log attachment to the mailing list. when does it show in the archives? | 12:58 |
| *** mriedem has joined #openstack-infra | 12:59 | |
| *** jcoufal has joined #openstack-infra | 13:01 | |
| * mordred waves to people | 13:05 | |
| fungi | sekharvajjula: if the attached log was large, it may be stuck in the moderation queue. i'll check in a few minutes | 13:08 |
| fungi | mordred: have you made it all the way back from texas to... texas now? | 13:08 |
| sekharvajjula | fungi: It has 21MB attachment. | 13:09 |
| fungi | yes, that's fairly large. i think anything over 10mb ends up needing a moderator to check it out first | 13:10 |
| mordred | fungi: yes! it was a long journey | 13:10 |
| fungi | beset with armadillos at every turn? | 13:10 |
| mordred | SO MANY DILLOS | 13:11 |
| *** ykarel has quit IRC | 13:12 | |
| *** efried is now known as fried_rice | 13:14 | |
| *** dave-mccowan has joined #openstack-infra | 13:15 | |
| *** Seb-Solon has joined #openstack-infra | 13:18 | |
| *** dave-mccowan has quit IRC | 13:21 | |
| *** kjackal has quit IRC | 13:24 | |
| openstackgerrit | Stephen Finucane proposed openstack-dev/pbr master: Special case long_description_content_type https://review.openstack.org/565177 | 13:24 |
| *** evrardjp has joined #openstack-infra | 13:25 | |
| openstackgerrit | Stephen Finucane proposed openstack-dev/pbr master: Use templates for cover and lower-constraints https://review.openstack.org/600943 | 13:26 |
| *** kjackal has joined #openstack-infra | 13:26 | |
| *** eernst has joined #openstack-infra | 13:27 | |
| *** kjackal_v2 has joined #openstack-infra | 13:28 | |
| *** ramishra has quit IRC | 13:28 | |
| TheJulia | fungi: The limitation is DIB has no understanding of it and when I took like five minutes to ponder it, it seemed very invasive. | 13:29 |
| TheJulia | fungi: part of that being because binaries are executed in the chroot which means the loader is going to go "nope nope nope" | 13:30 |
| *** kaiokmo has quit IRC | 13:30 | |
| *** panda is now known as panda|off | 13:31 | |
| fungi | sekharvajjula: i looked in the moderation queue and your post wasn't held or anything. what address did you send it to? openstack-infra@lists.openstack.org or something else? | 13:32 |
| sekharvajjula | fungi: openstack-infra-request@lists.openstack.org | 13:33 |
| sekharvajjula | Sorry. I will send it again to openstack-infra@lists.openstack.org | 13:33 |
| fungi | sekharvajjula: that's the problem. the request address is basically the listserv's api for sending commands (subscribe, unsubscribe, and a few others) | 13:33 |
| *** quiquell is now known as quiquell|off | 13:34 | |
| slaweq | mordred: hi, quick question, if I would like to tell devstack to clone neutron from specific patch, like: https://review.openstack.org/#/c/608259/1 how I should configure my local.conf file for that? do You know? | 13:35 |
| sekharvajjula | fungi: sent | 13:35 |
| AJaeger | slaweq: "depends-on: https://review.openstack.org/#/c/608259/" | 13:36 |
| AJaeger | slaweq: https://docs.openstack.org/infra/manual/developers.html#cross-repository-dependencies | 13:36 |
| slaweq | AJaeger: it will not work because I need to use it in grenade job as "old" version | 13:36 |
| slaweq | and depends-on not works like that there AFAIK | 13:37 |
| mordred | slaweq: it should actually | 13:37 |
| mordred | since that's a patch to stable/rocky, it'll get applied to the 'old' version in a grenade run triggered by a master patch | 13:37 |
| *** ykarel has joined #openstack-infra | 13:38 | |
| openstackgerrit | Merged openstack-dev/pbr master: Support wheel 0.32.0+ https://review.openstack.org/607894 | 13:38 |
| mordred | (zuul stacks patches onto branches for all of the branches involved in the dependency chain) | 13:38 |
| slaweq | so if I will do my "master patch" as depends-on this one from rocky it will deploy old version from this one? | 13:38 |
| *** amoralej|lunch is now known as amoralej | 13:39 | |
| *** ykarel has quit IRC | 13:39 | |
| *** ykarel_ has joined #openstack-infra | 13:39 | |
| mordred | slaweq: yes - at least it should | 13:40 |
| slaweq | mordred: AJaeger: thx, I will check that then :) | 13:41 |
| fungi | it will deploy any dependencies targeting stable/rocky on the stable/rocky branches of those projects which grenade will use to populate the "old" state and any master branch dependencies on the master branches of those projects which grenade will use to set up the "new" state | 13:41 |
| sekharvajjula | fungi: thanks for the support. I can now see my mail in the archive. | 13:43 |
| sekharvajjula | Can someone help me on this issue http://lists.openstack.org/pipermail/openstack-infra/2018-October/006203.html ? | 13:43 |
| prometheanfire | can someone kick another gentoo rebiuld? I think the last one happened just before https://review.openstack.org/608161 merged | 13:44 |
| fungi | sekharvajjula: it may make sense to continue discussing in the #openstack-third-party-ci channel too, since in theory there are a greater number of openstack third-party ci operators there | 13:45 |
| *** eernst has quit IRC | 13:45 | |
| mriedem | mordred: AJaeger: corvus: clarkb: is there any way to make zuulv3 required-projects branch-specific? | 13:46 |
| mriedem | i.e. only require a project on master for example | 13:47 |
| fungi | prometheanfire: most recent image completed building 5 hours 44 minutes ago so did indeed probably start building before 608161 merged | 13:47 |
| sekharvajjula | fungi: Yes I have this issue open there as well. But I am in Europe/Finland. I see that people in #openstack-third-party-ci channel are active in US timezones. It would be good to carry this issue in mailchain than in irc, where I can also attach logs easily! | 13:47 |
| prometheanfire | fungi: timezones :( | 13:47 |
| prometheanfire | I did a recheck, so should see | 13:47 |
| AJaeger | mriedem: what's the use case? | 13:48 |
| mriedem | AJaeger: devstack on master is going to require the openstack/placement repo | 13:48 |
| mriedem | and i'd rather not have to update 300 individual job definitions | 13:48 |
| *** dtantsur|brb is now known as dtantsur | 13:48 | |
| AJaeger | mriedem: devstack is branched, so you can update only master. | 13:49 |
| mriedem | legacy-devstack-dsvm-updown isn't, | 13:49 |
| AJaeger | mriedem: convert to Zuul v3 ;) | 13:49 |
| mriedem | that's still in openstack-zuul-jobs, | 13:49 |
| mriedem | there are probably going to be more | 13:49 |
| fungi | ahh, so for the legacy jobs specifically | 13:49 |
| mriedem | i've had to do this already for a few grenade jobs | 13:49 |
| mriedem | yes | 13:49 |
| mriedem | https://review.openstack.org/#/q/topic:cd/placement-solo+(status:open+OR+status:merged) | 13:50 |
| AJaeger | Just add it, a stable/rocky change would then checkout master of placement... | 13:50 |
| fungi | mriedem: the legacy devstack jobs don't automatically copy required-projects into the devstack projects list though, right? | 13:50 |
| AJaeger | fungi: they don't... | 13:50 |
| fungi | so zuul will simply be pushing a copy of the repo which devstack doesn't know to use anyway | 13:50 |
| AJaeger | mriedem: and we don't really touch legacy jobs anymore, these are old and should be thrown away... | 13:50 |
| mriedem | so i can just add openstack/placement to required-projects in legacy-dsvm-base? | 13:51 |
| fungi | i _think_ it will be effectively ignored by devstack on earlier branches | 13:51 |
| mriedem | and that will be ok for stable branch job runs too? | 13:51 |
| fungi | ni the legacy jobs that is | 13:51 |
| fungi | should be, but a depends-on will tell you | 13:51 |
| mriedem | ok | 13:52 |
| openstackgerrit | Matt Riedemann proposed openstack-infra/openstack-zuul-jobs master: Add openstack/placement to legacy-dsvm-base required-projects https://review.openstack.org/608266 | 13:54 |
| fungi | prometheanfire: i see gentoo-17-0-systemd-0000000896 building on nb02 for ~7 minutes now | 13:54 |
| openstackgerrit | Matt Riedemann proposed openstack-infra/devstack-gate master: Include openstack/placement in PROJECTS starting in Stein https://review.openstack.org/606853 | 13:55 |
| prometheanfire | fungi: that works too :D | 13:55 |
| fungi | prometheanfire: well, it's because i deleted the previous two | 13:56 |
| prometheanfire | ah | 13:56 |
| fungi | i suppose i should have clarified | 13:56 |
| *** kaiokmo has joined #openstack-infra | 13:56 | |
| prometheanfire | well, wonder if my rekick worked :P | 13:56 |
| prometheanfire | I think it might have | 13:56 |
| *** felipemonteiro has quit IRC | 13:57 | |
| *** EmilienM is now known as EvilienM | 13:57 | |
| *** eharney has joined #openstack-infra | 14:02 | |
| openstackgerrit | Jeremy Stanley proposed openstack-infra/system-config master: Add note about mounting one AFS volume in another https://review.openstack.org/608272 | 14:03 |
| *** sthussey has joined #openstack-infra | 14:06 | |
| *** ykarel_ is now known as ykarel | 14:09 | |
| *** dansmith is now known as SteelyDan | 14:20 | |
| *** ykarel is now known as ykarel|afk | 14:23 | |
| *** aidin has quit IRC | 14:24 | |
| *** kjackal has quit IRC | 14:25 | |
| *** olivierb_ has quit IRC | 14:27 | |
| *** auristor has quit IRC | 14:27 | |
| *** auristor has joined #openstack-infra | 14:28 | |
| *** panda|off has quit IRC | 14:36 | |
| *** panda has joined #openstack-infra | 14:37 | |
| *** udesale has quit IRC | 14:38 | |
| *** e0ne has quit IRC | 14:38 | |
| *** munimeha1 has joined #openstack-infra | 14:43 | |
| TheJulia | Hey infra folks, are we aware of any CI infrastructures that support our CI infrastructure that do not permit nested virt? | 14:46 |
| *** hwoarang has quit IRC | 14:47 | |
| fungi | TheJulia: if you mean cloud providers where we boot nodes not having nested virt support or, in some cases, having very broken behavior when you try to engage nested virt, definitely | 14:47 |
| fungi | i should clarify, nested virt acceleration. we of course support non-accelerated booting of virtual machines inside the test node virtual machines | 14:48 |
| * TheJulia wonders how we could detect or something | 14:48 | |
| fungi | TheJulia: https://docs.openstack.org/infra/manual/testing.html#known-differences-to-watch-out-for | 14:48 |
| fungi | "Nested virt is not available in all clouds. And in clouds where it is enabled we have observed a higher rate of crashed test VMs when using it. As a result we enforce qemu when running devstack and may further restrict the use of nested virt." | 14:48 |
| fungi | TheJulia: clouds which simply don't provide it are easy to detect from within the vm. clouds which break terribly when you try to use the advertised nested virt acceleration on the other hand are nontrivial to identify | 14:49 |
| fungi | also, obviously, we have no access to troubleshoot such breakage since we lack direct access to the hypervisor layer | 14:50 |
| fungi | so basically your test nodes just suddenly stop responding and you get no ability to debug the cause | 14:50 |
| TheJulia | Yeah, except I also don't want to extend timeouts to be 2.5-6x longer to not enable nested virt for a job. :( | 14:50 |
| * TheJulia sighs | 14:51 | |
| fungi | convince cloud providers to replace old hardware and upgrade the software they're using? | 14:51 |
| fungi | also we have a mix of kvm and xen providers, so nesting acceleration options differ among them | 14:51 |
| fungi | mix of intel and amd cpus and varying generations of those even within providers leads to inconsistency there as well | 14:53 |
| *** jamesmcarthur has joined #openstack-infra | 14:56 | |
| logan- | the last few cycles of nested virt breakage on xenial have been triggered by guest kernel updates in the nodepool images | 14:59 |
| *** hwoarang has joined #openstack-infra | 15:00 | |
| logan- | currently, with xenial running in both the guest and hv, nested virt seems to only work if the hv is running a 4.15 hwe kernel. | 15:01 |
| *** boden has quit IRC | 15:01 | |
| *** dkrol has joined #openstack-infra | 15:02 | |
| dkrol | Hello, I'm openstack trove developer and I have a problem with running tests on openstack infra. As a part of our testing job we are building vm image to be used during tests. During this job I'm getting an error - "Sorry, user stack is not allowed to execute '/bin/bash -c dd if=/opt/stack/trove/integration/scripts/files/requirements/ubuntu-requirements.txt of=/tmp/dib_build.HmlRR06T/hooks/requirements.txt' as stack on ubuntu | 15:04 |
| dkrol | Do you know what maybe the reason of this and what can I do about it ? | 15:04 |
| fungi | dkrol: looks like you need devstack troubleshooting help. i recommend the #openstack-qa channel since they're the devstack maintainers | 15:10 |
| dkrol | great, thank you | 15:11 |
| openstackgerrit | Ilya Shakhat proposed openstack-infra/project-config master: Move Stackalytics jobs to project repository https://review.openstack.org/608286 | 15:15 |
| *** munimeha1 has quit IRC | 15:22 | |
| *** Emine has joined #openstack-infra | 15:23 | |
| prometheanfire | cloudnull: /win 20 | 15:23 |
| prometheanfire | well, oops | 15:23 |
| *** gyee has joined #openstack-infra | 15:24 | |
| *** bnemec is now known as beekneemech | 15:27 | |
| *** munimeha1 has joined #openstack-infra | 15:27 | |
| clarkb | TheJulia: fungi: at the very least building relationships with cloud providers so that we can work together with them to debug issues. We've been successful more recently doing this with logan- and mnaser and dpawlik. | 15:27 |
| openstackgerrit | Ilya Shakhat proposed openstack-infra/project-config master: Move performa jobs to project repository https://review.openstack.org/608289 | 15:28 |
| clarkb | mriedem: fyi https://review.openstack.org/#/c/608038/ is an e-r query that would be good to get in | 15:29 |
| *** ykarel|afk is now known as ykarel|ptotillwe | 15:33 | |
| *** ykarel|ptotillwe is now known as ykarel|ptotwedne | 15:34 | |
| prometheanfire | nb01 and nb02 are not mirrored to eachother? | 15:34 |
| clarkb | prometheanfire: no, they are independent distributed builders | 15:34 |
| fungi | as is nb03 | 15:35 |
| *** pcaruana has quit IRC | 15:39 | |
| prometheanfire | ah | 15:39 |
| pabelanger | +1 open cloud ops! | 15:39 |
| *** boden has joined #openstack-infra | 15:44 | |
| *** e0ne has joined #openstack-infra | 15:50 | |
| *** e0ne has quit IRC | 15:55 | |
| *** diablo_rojo has joined #openstack-infra | 15:55 | |
| TheJulia | clarkb: Yeah, and I recognize its not great that we've enabled some jobs with nested virt because of timeouts and constraints, and not wanting to keep things tied up too long running tests. Kind of created my own nightmare in a sense. | 15:58 |
| openstackgerrit | Clark Boylan proposed openstack-infra/zuul-jobs master: Retry failed git pushses on workspace setup https://review.openstack.org/608303 | 15:58 |
| *** lpetrut has quit IRC | 15:59 | |
| openstackgerrit | Dean Troyer proposed openstack-infra/project-config master: Add publish jobs for StarlingX api-ref and releasenotes https://review.openstack.org/608304 | 15:59 |
| openstackgerrit | Merged openstack-infra/elastic-recheck master: Generic OOM killer query https://review.openstack.org/608038 | 15:59 |
| *** emerson has quit IRC | 16:00 | |
| openstackgerrit | Clark Boylan proposed openstack-infra/zuul-jobs master: Retry failed git pushses on workspace setup https://review.openstack.org/608303 | 16:00 |
| *** jchhatba_ has quit IRC | 16:00 | |
| *** emerson has joined #openstack-infra | 16:01 | |
| clarkb | TheJulia: ironic may be in a special place due to its tight coupling to "hardware", but I'v etried to push containers more to support the arbitrary compute in a VM use case. Unfortunately it hasn't caught on much | 16:01 |
| *** PapaOurs is now known as bauzas | 16:08 | |
| TheJulia | I guess it kind of is a special case in a sense, but still :( | 16:08 |
| *** armax has quit IRC | 16:09 | |
| clarkb | ya whether virtual or baremetal ironic wants to talk to a "machine" and containers are all the bits you get after the machine stuff is sorted | 16:09 |
| *** mriedem has quit IRC | 16:09 | |
| *** armax has joined #openstack-infra | 16:11 | |
| *** jpena is now known as jpena|off | 16:12 | |
| openstackgerrit | Clark Boylan proposed openstack-infra/zuul-jobs master: Retry failed git pushses on workspace setup https://review.openstack.org/608303 | 16:12 |
| *** ginopc has quit IRC | 16:17 | |
| *** ianychoi_ is now known as ianychoi | 16:19 | |
| *** derekh has quit IRC | 16:19 | |
| *** jtomasek has quit IRC | 16:21 | |
| johnsom | clarkb FYI, I finally got a good test run using a bionic nodepool instance with OVH hosts that have vmx enabled. They are crashing out the same way as the xenial instances are. So, we are stuck waiting for another kernel rev that fixes the last regression. | 16:26 |
| clarkb | johnsom: logan- has indicated that the xenial hardware enablement kernels fix things in limestone. I don't think that bionic has hwe kernels yet? | 16:27 |
| johnsom | Yeah, maybe not. I know it was working on the last round of kernels, but this round has a regression. sigh | 16:28 |
| logan- | yeah i've noticed bionic guests living on base xenial kernel hvs crash the same way as xenial guests | 16:28 |
| johnsom | FYI: http://logs.openstack.org/39/600539/18/check/octavia-v2-dsvm-scenario/6c294a7/controller/logs/libvirt/qemu/instance-0000000a_log.txt.gz | 16:28 |
| johnsom | I'm going to update my kernel bug with the new kernel version info. | 16:28 |
| logan- | xenial hwe magically allows regular xenial guests to work, no idea why. we are about 75% upgraded to hwe on the limestone nodepool cloud. | 16:28 |
| *** trown is now known as trown|lunch | 16:30 | |
| *** aojeagarcia has quit IRC | 16:32 | |
| TheJulia | oh joy :( | 16:33 |
| *** gfidente has quit IRC | 16:37 | |
| *** ansmith has quit IRC | 16:42 | |
| *** pcaruana has joined #openstack-infra | 16:42 | |
| johnsom | logan- Good to know. I might keep randomly doing re-checks to see if I win the limestone lottery with my test patch and see if they are booting there. | 16:44 |
| *** ansmith has joined #openstack-infra | 16:45 | |
| logan- | johnsom: i'll ping ya once I get the rest of the HVs cycled thru for updates to hwe too. hopefully it is a stable fix.. it seems like we get new brokenness every time there's a new kernel package | 16:46 |
| johnsom | Yeah, recently it has been alternating for sure. | 16:46 |
| *** dims has joined #openstack-infra | 16:46 | |
| *** pcaruana has quit IRC | 16:50 | |
| *** jtomasek has joined #openstack-infra | 16:52 | |
| *** jamesmcarthur has quit IRC | 17:02 | |
| *** amoralej is now known as amoralej|off | 17:05 | |
| *** dave-mccowan has joined #openstack-infra | 17:08 | |
| *** yamamoto has quit IRC | 17:08 | |
| *** yamamoto has joined #openstack-infra | 17:09 | |
| openstackgerrit | Jeremy Stanley proposed openstack-infra/puppet-gerritbot master: Use default IRC port https://review.openstack.org/608313 | 17:13 |
| *** yamamoto has quit IRC | 17:14 | |
| *** dave-mccowan has quit IRC | 17:14 | |
| openstackgerrit | Jeremy Stanley proposed openstack-infra/gerritbot master: Identify with SASL https://review.openstack.org/608314 | 17:14 |
| fungi | clarkb: mwhahaha: proposed fix for join race in gerritbot ^ | 17:14 |
| openstackgerrit | Merged openstack-infra/nodepool master: Run release-zuul-python on release https://review.openstack.org/607649 | 17:17 |
| *** dtantsur is now known as dtantsur|afk | 17:18 | |
| *** aidin has joined #openstack-infra | 17:25 | |
| openstackgerrit | Jeremy Stanley proposed openstack-infra/puppet-gerritbot master: Use port 6697 for IRC over SSL/TLS https://review.openstack.org/608313 | 17:25 |
| *** ansmith has quit IRC | 17:25 | |
| *** aidin has quit IRC | 17:29 | |
| *** florianf has quit IRC | 17:29 | |
| *** ansmith has joined #openstack-infra | 17:30 | |
| *** shardy has quit IRC | 17:31 | |
| openstackgerrit | Jeremy Stanley proposed openstack-infra/zuul-website master: Update events lists/banner after 2018 Ansiblefest https://review.openstack.org/608320 | 17:31 |
| *** trown|lunch is now known as trown | 17:32 | |
| *** ykarel|ptotwedne has quit IRC | 17:33 | |
| *** eharney has quit IRC | 17:34 | |
| *** jamesmcarthur has joined #openstack-infra | 17:35 | |
| *** jamesmcarthur has quit IRC | 17:39 | |
| *** felipemonteiro has joined #openstack-infra | 17:41 | |
| *** jamesmcarthur has joined #openstack-infra | 17:48 | |
| *** eharney has joined #openstack-infra | 17:49 | |
| *** felipemonteiro has quit IRC | 17:49 | |
| openstackgerrit | Jeremy Stanley proposed openstack-dev/specs-cookiecutter master: Clean up .gitignore references to personal tools https://review.openstack.org/608329 | 17:50 |
| AJaeger | infra-root, is nodepool somehow miscounting nodes? Looking at http://grafana.openstack.org/d/T6vSHcSik/zuul-status?panelId=20&fullscreen&orgId=1 , we always stay around 950 - but have a capacity of 1114. And there's 1000 node request waiting... | 17:51 |
| AJaeger | wow, 668 of our used nodes are CentOS 7... | 17:52 |
| clarkb | AJaeger: dmsimard pointed out at ansiblefest that that is a really good number to look at to see how much of the usage is tripleo | 17:52 |
| clarkb | since most of the centos usage is tripleo | 17:52 |
| clarkb | as for why the counting seems off Iam not sure | 17:53 |
| *** jamesmcarthur has quit IRC | 17:53 | |
| clarkb | hrm looking at ovh bhs1 we aren't using the full capacity there | 17:53 |
| pabelanger | AJaeger: looks like packethost could maybe be dropped in max-servers: http://grafana.openstack.org/dashboard/db/nodepool-packethost | 17:53 |
| clarkb | I think that implies a quota issue in that region | 17:53 |
| clarkb | probably a quota issue in packethost depressing usage too | 17:53 |
| pabelanger | and will reduce nodepool launch errors too | 17:53 |
| clarkb | my guess is leaked ports in both | 17:54 |
| AJaeger | pabelanger: that's 20 nodes for packethost - but where are the other 144? | 17:54 |
| clarkb | AJaeger: ovh bhs1 | 17:54 |
| clarkb | this is a side effect of nodepool checking quota I think | 17:54 |
| clarkb | we set max-servers and boot min(max-servers, whatever cloud quota says we can boot) | 17:54 |
| pabelanger | AJaeger: OVH is the rest: http://grafana.openstack.org/dashboard/db/nodepool-ovh | 17:54 |
| AJaeger | clarkb, pabelanger : indeed ovh bhs1 | 17:54 |
| AJaeger | thanks! | 17:55 |
| pabelanger | lots of errors there | 17:55 |
| clarkb | pabelanger: likely due to port quota | 17:55 |
| pabelanger | Yah | 17:55 |
| clarkb | I'm checking bhs1 one | 17:55 |
| clarkb | s/one/now/ | 17:55 |
| AJaeger | clarkb: yeah, centos = tripleo looks like a good approximination. | 17:55 |
| clarkb | ya 263 DOWN ports | 17:55 |
| clarkb | I will start a loop to delete those | 17:55 |
| AJaeger | thanks, clarkb | 17:56 |
| pabelanger | but yah, centos-7 is hungry today: http://grafana.openstack.org/d/rZtIH5Imz/nodepool?orgId=1&from=now-1h&to=now | 17:56 |
| AJaeger | pabelanger: for CentOS, I think the problem is that they had guide problems - abandoned everything in gate and restored. | 17:56 |
| AJaeger | And a restore pushes everything through check again | 17:56 |
| pabelanger | yah | 17:57 |
| AJaeger | mwhahaha ^ | 17:57 |
| pabelanger | we have zuul dequeue CLI now | 17:57 |
| AJaeger | (can't find other tripleo folks on IRC) | 17:57 |
| pabelanger | but, that would still pushin it via check | 17:57 |
| clarkb | I don't think dequeing is the solution though, it is going to be prioritizing stabilization fixes over features | 17:57 |
| AJaeger | might be worth discussing options with tripleo. promotion is also an option. | 17:58 |
| pabelanger | AJaeger: is tripleo doing a release this week? | 17:58 |
| clarkb | we promoted a change early today (thanks fungi) but I read the change and it was to fix periodic jobs | 17:58 |
| AJaeger | no idea ;( | 17:58 |
| clarkb | I don't think those changes should be promoted | 17:58 |
| mwhahaha | the rocky release screwed our upgrade jobs | 17:58 |
| mwhahaha | which hosed the gate | 17:58 |
| AJaeger | mwhahaha: your abandon and restore crashes our CI resources ;/ | 17:58 |
| AJaeger | mwhahaha: we need a better solution since restore means everything goes through check again... | 17:59 |
| *** yamamoto has joined #openstack-infra | 17:59 | |
| fungi | i'll admit i didn't read the change quiquell|off asked to have promoted, just knew they were struggling with a bunch of new issues and took his word for it that it was urgent. i'll look more closely in the future | 17:59 |
| *** josephrsandoval has joined #openstack-infra | 17:59 | |
| mwhahaha | AJaeger: so would have failing the gate and reseting every patch | 18:00 |
| clarkb | fungi: no worries, but something we should check in the future as promoting a change resets the entire gate queue (so should really be reserved for changes that give the existing changes in the gate a better chance at merging) | 18:00 |
| AJaeger | mwhahaha: right now 660 out of 950 nodes are CentOS-7... | 18:00 |
| AJaeger | mwhahaha: I agree - I'm just saying, we need to find a better approach together. | 18:00 |
| mwhahaha | yes i am working on reducing the load | 18:00 |
| mwhahaha | https://review.openstack.org/#/q/topic:reduce-consumption+(status:open+OR+status:merged) | 18:00 |
| mwhahaha | so everything takes time | 18:01 |
| clarkb | AJaeger: bhs1 availability is increasing as the ports are deleted | 18:01 |
| fungi | clarkb: yep, it basically didn't even cross my mind that someone would ask to have a change promoted for tripleo without understanding what it did | 18:01 |
| AJaeger | mwhahaha: I mean also for abandon/restore - and promote. I hope we can do this together in a better way... | 18:01 |
| AJaeger | clarkb: great, thanks | 18:01 |
| mwhahaha | sure | 18:01 |
| mwhahaha | what was promoted earlier? | 18:01 |
| AJaeger | mwhahaha: 608080,2 | 18:01 |
| pabelanger | fungi: maybe a discussion for smaller window size in gate pipeline can help minimize the number of changes that would reset on a promote | 18:01 |
| AJaeger | pabelanger: we limit it for tripleo ;) | 18:02 |
| clarkb | pabelanger: I'm not concerned about it if we promote changes that actually fix bugs in the gate | 18:02 |
| mwhahaha | AJaeger: yea that didn't cause the problem | 18:02 |
| AJaeger | (at least on request) | 18:02 |
| clarkb | pabelanger: the problem is that the changes being promoted don't always do that | 18:02 |
| mwhahaha | https://review.openstack.org/#/c/608056/ that broke the gate | 18:03 |
| mwhahaha | cause upgrade jobs suddenly weren't getting the latest version of code | 18:03 |
| mwhahaha | because version numbers :/ | 18:03 |
| mwhahaha | happy friday | 18:03 |
| AJaeger | ah, a release... | 18:03 |
| pabelanger | AJaeger: clarkb: yah, if concern is in gate, it is not stable, maybe a new window-floor change, reduces the any given number of nodes from nodepool is something to try. Not a longer solution, since all other projects also get bumped down | 18:04 |
| clarkb | pabelanger: right, I'm saying I'd much rather put effort on directly addressing the problems | 18:04 |
| clarkb | this is why I keep pushing elastic-recheck so that we can understand what the problems are and work from there to fix them | 18:05 |
| clarkb | and I've been picking infra related ones off of that list and trying to push fixes too | 18:05 |
| pabelanger | clarkb: Yup, agree. hopefully everybody is onboard with that | 18:05 |
| clarkb | (like the stackviz python3 issue and the git push retries) | 18:05 |
| pabelanger | ++ | 18:05 |
| AJaeger | clarkb: much appreciated - thanks | 18:05 |
| *** felipemonteiro has joined #openstack-infra | 18:06 | |
| AJaeger | anybody else to review clarkb's git push retries? https://review.openstack.org/#/c/608303/ | 18:06 |
| clarkb | AJaeger: we may want to test that one a bit before merging just because it will affect most jobs | 18:07 |
| AJaeger | clarkb: ovh bhs1 looks good again... | 18:07 |
| AJaeger | clarkb: yes, indeed... | 18:07 |
| AJaeger | Do we know what the fix for packethost is? | 18:09 |
| clarkb | AJaeger: not yet, I suspect it is also port leaks | 18:09 |
| clarkb | Ican check as soon as current bhs1 port delete scrpt completes | 18:10 |
| clarkb | basically both clouds exhibit the same issue in neutron | 18:10 |
| clarkb | and soon gra1 will too | 18:10 |
| AJaeger | argh ;( | 18:10 |
| *** electrofelix has quit IRC | 18:10 | |
| clarkb | this is quite literally a bug in openstack affecting our ability to consume openstack for testing | 18:10 |
| clarkb | unfortuantely I think it was fixed in newer openstack and so the answer from openstack is going to be "upgrade" | 18:11 |
| pabelanger | is it something we can handle in nodepool, like we do for clean-floating-ips? | 18:12 |
| clarkb | pabelanger: possibly, we'd have to sort out the lifecycle of states and metadata info for a port and whether or not it is safe to delete any DOWN port that doens't have an instance that maps to a nodepool instance | 18:13 |
| fungi | clarkb: are you sure it's leaks and not just a lag in them getting deleted? the response from ovh seemed to suggest it was a performance problem on their backend | 18:13 |
| clarkb | fungi: yes aiui when you delete a nova instance the nova service tells neutron to delete the port. If that request from nova to neutron times out then the port leaks | 18:13 |
| clarkb | fungi: the performance issue is in handling that request between nova and neutron | 18:13 |
| clarkb | if it fails/timesout the port leaks | 18:14 |
| clarkb | fixing the performance issue makes neutron more reliable for nova and you leak less | 18:14 |
| fungi | except when we turned off our use of the region, all the seemingly "leaked" ports freed up on their own a few minutes later | 18:14 |
| clarkb | fungi: I think those may not have leaked yet | 18:14 |
| clarkb | the outstanding requests to remove those ports were still in fligh? | 18:14 |
| fungi | then again, might also have been some poor sysadmin doing cleanup behind the scenes too | 18:15 |
| clarkb | when I deleted ports a previous time some of them were definitely days old | 18:15 |
| fungi | good to know | 18:15 |
| clarkb | unfrotauntely I didn't care to chekc the age of the ports in this list | 18:15 |
| fungi | so, yeah, maybe that bug was still present in newton (what they were upgrading to)? | 18:15 |
| clarkb | newton ish iirc | 18:16 |
| clarkb | | created_at | 2018-10-05T10:37:53Z is a randomly selected packethost port | 18:17 |
| *** jcoufal has quit IRC | 18:18 | |
| clarkb | that is DOWN | 18:18 |
| *** jcoufal has joined #openstack-infra | 18:18 | |
| clarkb | {"forbidden": {"message": "Quota exceeded for cores: Requested 8, but already used 3600 of 3600 cores", "code": 403}} is packethost issue | 18:19 |
| clarkb | it thinks we've booted ~450 instances | 18:20 |
| clarkb | quota isn't updating their I guess | 18:20 |
| openstackgerrit | Jeremy Stanley proposed openstack-infra/puppet-infra-cookiecutter master: Dissuade .gitignore references to personal tools https://review.openstack.org/608334 | 18:20 |
| *** imacdonn has quit IRC | 18:22 | |
| *** imacdonn has joined #openstack-infra | 18:22 | |
| clarkb | cleaning up the ports in packethost likely won't help like it did in bhs1 | 18:22 |
| clarkb | I've got to pop out fo ra few right now too, back after lunch things | 18:22 |
| fungi | yeah, if it's like last time, scaling waaaay back on max-servers in that region is likely the only solution until they resolve the performance issue there again | 18:23 |
| openstackgerrit | Dean Troyer proposed openstack-infra/project-config master: Add publish jobs for StarlingX api-ref and releasenotes https://review.openstack.org/608304 | 18:29 |
| *** Swami has joined #openstack-infra | 18:36 | |
| *** slaweq has quit IRC | 18:37 | |
| *** slaweq has joined #openstack-infra | 18:38 | |
| *** eharney has quit IRC | 18:40 | |
| *** felipemonteiro has quit IRC | 18:49 | |
| *** jamesmcarthur has joined #openstack-infra | 18:49 | |
| *** jmorgan1 has quit IRC | 18:51 | |
| *** jmorgan1 has joined #openstack-infra | 18:56 | |
| *** ansmith has quit IRC | 18:58 | |
| *** josephrsandoval has quit IRC | 19:02 | |
| *** zul has quit IRC | 19:02 | |
| *** zul has joined #openstack-infra | 19:03 | |
| *** graphene has joined #openstack-infra | 19:10 | |
| *** ansmith has joined #openstack-infra | 19:14 | |
| *** felipemonteiro has joined #openstack-infra | 19:14 | |
| *** diablo_rojo has quit IRC | 19:14 | |
| *** jesusaur has quit IRC | 19:16 | |
| *** jamesmcarthur has quit IRC | 19:17 | |
| *** jesusaur has joined #openstack-infra | 19:23 | |
| *** dave-mccowan has joined #openstack-infra | 19:23 | |
| *** diablo_rojo has joined #openstack-infra | 19:28 | |
| openstackgerrit | Merged openstack-infra/zuul-website master: Update events lists/banner after 2018 Ansiblefest https://review.openstack.org/608320 | 19:29 |
| *** felipemonteiro has quit IRC | 19:30 | |
| openstackgerrit | sebastian marcet proposed openstack-infra/openstackid-resources master: Fix on heuristic to get Summit Pages https://review.openstack.org/608341 | 19:30 |
| openstackgerrit | Clark Boylan proposed openstack-infra/zuul-jobs master: Retry failed git pushses on workspace setup https://review.openstack.org/608303 | 19:31 |
| openstackgerrit | Clark Boylan proposed openstack-infra/zuul-jobs master: Add test workspace setup role https://review.openstack.org/608342 | 19:31 |
| openstackgerrit | Clark Boylan proposed openstack-infra/project-config master: Test workspace setup role changes https://review.openstack.org/608343 | 19:33 |
| *** slagle has quit IRC | 19:35 | |
| *** e0ne has joined #openstack-infra | 19:38 | |
| *** jcoufal has quit IRC | 19:38 | |
| *** ansmith has quit IRC | 19:39 | |
| *** ansmith has joined #openstack-infra | 19:40 | |
| *** e0ne has quit IRC | 19:43 | |
| *** kjackal_v2 has quit IRC | 19:46 | |
| *** eharney has joined #openstack-infra | 19:46 | |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: WIP: docker-compose quickstart example https://review.openstack.org/608344 | 19:46 |
| *** Emine has quit IRC | 19:47 | |
| openstackgerrit | Michael Vollman proposed openstack-infra/project-config master: Add os_manila role https://review.openstack.org/608345 | 19:53 |
| *** e0ne has joined #openstack-infra | 19:55 | |
| *** e0ne has quit IRC | 19:56 | |
| openstackgerrit | sebastian marcet proposed openstack-infra/openstackid-resources master: Fix on heuristic to get Summit Pages https://review.openstack.org/608341 | 19:57 |
| *** agopi is now known as agopi|away | 19:59 | |
| *** kjackal has joined #openstack-infra | 19:59 | |
| *** ansmith has quit IRC | 20:03 | |
| openstackgerrit | Michael Vollman proposed openstack-infra/project-config master: Add os_manila role https://review.openstack.org/608345 | 20:16 |
| openstackgerrit | Merged openstack-infra/nodepool master: Fix race in test_launchNode_delete_error https://review.openstack.org/604678 | 20:18 |
| *** graphene has quit IRC | 20:23 | |
| *** graphene has joined #openstack-infra | 20:24 | |
| openstackgerrit | Michael Vollman proposed openstack-infra/project-config master: Add os_manila role https://review.openstack.org/608345 | 20:27 |
| *** e0ne has joined #openstack-infra | 20:43 | |
| openstackgerrit | Dean Troyer proposed openstack-infra/project-config master: Add publish jobs for StarlingX api-ref and releasenotes https://review.openstack.org/608304 | 20:47 |
| *** e0ne has quit IRC | 20:49 | |
| *** munimeha1 has quit IRC | 20:52 | |
| mtreinish | fungi: did you ever get a chance to check the proxy for the subunit2sql db? | 20:56 |
| fungi | mtreinish: thanks for the reminder! and no ;) | 20:59 |
| fungi | checking now | 20:59 |
| mtreinish | fungi: heh, no worries. Thanks for looking into it | 20:59 |
| fungi | what's the hostname again? | 20:59 |
| mtreinish | the proxy runs on logstash.o.o | 21:00 |
| *** kopecmartin|ruck is now known as kopecmartin|off | 21:00 | |
| mtreinish | I don't know the hostname for the trove instance where the db actually lives though | 21:00 |
| clarkb | I think the firewall on the proxy node (logstash.o.o) is what needs fixing | 21:01 |
| fungi | np | 21:01 |
| fungi | clarkb: yeah, that was my first hunch | 21:01 |
| fungi | just trying to remember which server ;) | 21:01 |
| fungi | we've got a listening service on 0.0.0.0:3306 but no firewall entry for that | 21:02 |
| mtreinish | fungi: heh, well at least simpleproxy is still up and running | 21:03 |
| fungi | https://git.openstack.org/cgit/openstack-infra/system-config/tree/playbooks/group_vars/logstash.yaml#n3 | 21:03 |
| fungi | looks like it _should_ be open? | 21:04 |
| fungi | oh, bad grep | 21:04 |
| fungi | ACCEPT tcp -- anywhere anywhere state NEW tcp dpt:mysql | 21:04 |
| fungi | okay, so maybe not the fw | 21:04 |
| clarkb | iptables -n ftw | 21:05 |
| fungi | telnet logstash.openstack.org mysql | 21:05 |
| fungi | i can issue a `help` to it | 21:05 |
| fungi | mtreinish: what symptom are you seeing? | 21:05 |
| mtreinish | fungi: it's not able to open a mysql session with the server: http://paste.openstack.org/show/731606/ | 21:07 |
| *** rtjure has quit IRC | 21:07 | |
| mtreinish | I'm able to open up telnet on logstash.o.o 3306 and issue help (it doesn't get a response though) | 21:08 |
| fungi | interesting. i definitely get help output returned | 21:09 |
| fungi | oh, but only sometimes | 21:09 |
| fungi | i got lucky the first try, it seems | 21:10 |
| fungi | i just got it to return some help output again | 21:10 |
| fungi | but that took a lot of tries | 21:10 |
| fungi | okay, so likely something screwy here | 21:10 |
| clarkb | fungi: probably test the backend from logstash.o.o and see if that connectivity is bad? | 21:13 |
| fungi | something in rackspace is hitting that port at the moment | 21:13 |
| clarkb | we point the health status app/server at it | 21:14 |
| clarkb | iir | 21:14 |
| fungi | how do we configure simpleproxy? | 21:14 |
| mtreinish | fungi: there was a puppet file in puppet-subunit2sql for simpleproxy iirc | 21:14 |
| mtreinish | one sec I'll dig it up | 21:14 |
| clarkb | fungi: https://git.openstack.org/cgit/openstack-infra/system-config/tree/modules/openstack_project/manifests/logstash.pp#n43 is the start of the config | 21:15 |
| mtreinish | heh, clarkb is faster | 21:15 |
| fungi | oic, it's all cli opts | 21:16 |
| fungi | yeah, i'm getting "no route to host" when i try hitting the backend | 21:17 |
| fungi | from logstash.o.o | 21:17 |
| clarkb | weird I'd point at the trove instance except you get help sometimes | 21:18 |
| fungi | incomplete arp for the backend routed gateway where that address lies | 21:19 |
| fungi | 10.208.160.1 dev eth1 INCOMPLETE | 21:19 |
| fungi | must be getting an arp response for it occasionally and then dropping out again | 21:20 |
| fungi | 10.208.160.1 dev eth1 FAILED | 21:20 |
| fungi | it's retrying quite a bit | 21:20 |
| *** trown is now known as trown|outtypewww | 21:21 | |
| fungi | oh, fun, there's some layered routing here | 21:22 |
| fungi | so the db server lies in 10.176.0.0/12 | 21:22 |
| fungi | which is reached as such: | 21:22 |
| fungi | 10.176.0.0/12 via 10.208.160.1 dev eth1 | 21:22 |
| *** bobh has joined #openstack-infra | 21:23 | |
| fungi | oh, nevermind, the other i was seeing is the local source route | 21:23 |
| fungi | 10.208.160.0/19 dev eth1 proto kernel scope link src 10.208.160.29 | 21:23 |
| fungi | anyway, 10.208.160.29 seems to be unable to get arp replies for 10.208.160.1 | 21:23 |
| fungi | *most* of the time | 21:24 |
| fungi | review.o.o can reach it fine | 21:25 |
| fungi | yeah, from review i can connect to the subunit2sql backend database and query it, no problem | 21:27 |
| fungi | logstash.o.o on the other had seems to have layer 2 connectivity problems to the backend gateway | 21:27 |
| fungi | yay clouds! | 21:27 |
| fungi | same backend gateway address for both of them (10.208.160.1) | 21:28 |
| *** kjackal has quit IRC | 21:29 | |
| lifeless | dhellmann: fungi: re: univeral setting in setup.cfg. Its better to set it in setup.cfg because for projects that *aren't* univeral wheel compatible, building a universal wheel by adding the CLI flag will create corrupt wheels. | 21:30 |
| lifeless | dhellmann: fungi: noone anywhere should ever use the CLI flag without understanding the details of what that particular project does, to know that its universal wheel compatible. | 21:31 |
| lifeless | dhellmann: fungi: *our* projects are (*now*) universal wheel compatible because we tracked down and made sure noone was doing conditional dependencies via any other method than the declaritive metadata one | 21:32 |
| lifeless | dhellmann: fungi: but in general thats not true. So its unreasonable to expect that folk outside of our community, building wheels of our projects, will know whether its safe or not. | 21:33 |
| lifeless | until they check. etc. | 21:33 |
| lifeless | if I can make an observation, the actual issue here seems to be that we're doing these 600 patches by hand. Surely an appropriately privileged bot could just commit them itself. | 21:33 |
| *** graphene has quit IRC | 21:35 | |
| *** graphene has joined #openstack-infra | 21:37 | |
| fungi | yeah, that was sort of where i was going with the earlier discussion of bypassing review (and perhaps even ci) for mechanically-generated changes | 21:38 |
| mtreinish | fungi: hah, well that's fun | 21:38 |
| fungi | mtreinish: yeah, i have a feeling a reboot of logstash.o.o _might_ fix it, but there's a good chance it'll require a trouble ticket with the provider instead | 21:39 |
| *** boden has quit IRC | 21:39 | |
| fungi | clarkb: how disruptive would rebooting logstash.o.o be? | 21:39 |
| clarkb | fungi: not very, we'll lose log indexing for jobs that complete while it is offline but that should be fine | 21:40 |
| clarkb | especially now that job demand has fallen way off | 21:40 |
| fungi | okay, i'll give that a shot next | 21:40 |
| fungi | unless you have other suggestions | 21:40 |
| clarkb | Only other idea would be to trigger a hypervisor move somehow | 21:43 |
| *** priteau has quit IRC | 21:46 | |
| fungi | not sure how to do that in rackspace | 21:47 |
| fungi | giving the reboot a shot | 21:47 |
| *** jamesmcarthur has joined #openstack-infra | 21:47 | |
| fungi | it's on its way back up now | 21:49 |
| fungi | i can ssh into it | 21:49 |
| fungi | 10.208.160.1 dev eth1 lladdr 00:00:0c:9f:f0:01 REACHABLE | 21:49 |
| fungi | i can hit the db from it | 21:50 |
| fungi | i can hit the db through the proxy again | 21:50 |
| fungi | mtreinish: give it another try? | 21:50 |
| clarkb | I had to start the jenkins log client to get gearman running again. I seem to remember writing a change to switch us over to boring old geard | 21:51 |
| clarkb | but I guess that never merged | 21:51 |
| fungi | #status log rebooted logstash.openstack.org to "fix" broken layer 2 connectivity to backend gateway | 21:51 |
| openstackstatus | fungi: finished logging | 21:51 |
| *** jamesmcarthur has quit IRC | 21:52 | |
| *** slagle has joined #openstack-infra | 22:00 | |
| mtreinish | fungi: yay, it works for me now | 22:00 |
| mtreinish | thanks | 22:00 |
| *** bobh has quit IRC | 22:04 | |
| clarkb | lifeless: fungi: in the past at least we've automated the proposal of changes so that testing can happen and to give project teams autonomy over whether or not things go into their project. Good example is the translations. Whether or not testing or even ability to refuse merging is desirable in this case I have no idea | 22:04 |
| clarkb | I don't really know what happened other than wheel updated and didn't do so in a backward compat manner | 22:04 |
| *** EvilienM is now known as EmilienM | 22:06 | |
| clarkb | lifeless: and for 4) in most cases zuul job config is testable pre merge. So you can totally push up ad hoc testing changes to have things tested in unique/new/odd/etc ways per patchset in gerrit | 22:09 |
| clarkb | lifeless: and for at least one cloud we are doing continuous deployment to it from Zuul then running zuul jobs on it :) | 22:16 |
| *** rlandy has quit IRC | 22:24 | |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: WIP: docker-compose quickstart example https://review.openstack.org/608344 | 22:25 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: WIP: docker-compose quickstart example https://review.openstack.org/608344 | 22:26 |
| lifeless | clarkb: thats awesome | 22:28 |
| lifeless | clarkb: so thats an infra-run cloud? | 22:29 |
| clarkb | lifeless: no it is a limestone networks run cloud by logan-, but logan- is willing to have us/zuul run the deploys post merge | 22:29 |
| lifeless | clarkb: (pkging) yes, I know we push up proposals programmatically; I rather suspect that at 600 contact points it would be better to deal with fallout from some % being inappropriate than manual review of them all. | 22:30 |
| lifeless | clarkb: wahoo - that is brilliant. | 22:30 |
| lifeless | clarkb: is infra the only customer of the cloud? | 22:30 |
| clarkb | lifeless: no I think logan- said their business developmenty stuff runs on that cloud too, but it isn't exposed to their customers | 22:30 |
| lifeless | clarkb: (4) - yeah, I know you can poke at the metadata and the code under test now, but its still very action at a distance. Having some contained but direct mechanism might be useful - I'm specifically thinking about vendor drivers here | 22:31 |
| lifeless | e.g. if I wanted to do adhoc exploration of an ESXi connected nova-compute, how would I? | 22:32 |
| clarkb | well you'd need vmware licenses first and those cost something like what my house does | 22:32 |
| clarkb | but if you figured that out, I'd store that info in zuul secrets and deploy them directly? | 22:32 |
| lifeless | clarkb: so no - what I'm propsing there is that there is an ESXi wired into deployment logic, and I both can get my nova-compute deployed on it, or get a limited shell of some sort on it, brokered by infra | 22:33 |
| clarkb | lifeless: yes if you, vmware, said we have an esxi pool over "here" you'd put the crdentials for it in zuul secrets and talk to that pool | 22:34 |
| lifeless | clarkb: ESXi is perhaps a bad example, because I work at VMware, and I'm not actually thinking VMware specific here. Uh, whats that proprietary container driver - windriver or something... etc | 22:34 |
| clarkb | then your jobs can interact with that remote set of resources as necessary. | 22:35 |
| lifeless | clarkb: and I'm *definitely* not wearing a VMware hat for this discussion | 22:35 |
| clarkb | sure | 22:35 |
| clarkb | I guess what I'm suggesting is that regardless of what it is vmware or cisco switching gear etc it has to exist somewhere (and be paid for) | 22:35 |
| *** eernst has joined #openstack-infra | 22:35 | |
| lifeless | yes, absolutely. | 22:36 |
| clarkb | but if you've done that zuul allows you to feed the secrets to authenticate into it. | 22:36 |
| lifeless | the delta vs what the 3rd party test rigs do today is being part of a persistent cloud, rather than being a deploy-a-cloud-via-devstack-or-whatever job endpoint | 22:36 |
| lifeless | persistent community-run cloud | 22:36 |
| clarkb | Unfortunatley I think we've found that our ability to run such a thing is basically nil | 22:37 |
| clarkb | it is incredibly difficult to get access to those resources in a community space | 22:38 |
| clarkb | and the one set we've had was taken from us without even a single email | 22:38 |
| lifeless | for the proprietary bits, and for the actual-hardware bits - yes, thats certainly a big thing; I don't think we can shift that needle easily. | 22:39 |
| lifeless | For the devstack jobs, - the vms that infra has access to don't have that problem; so the basic architecture I describe could be realised - and give benefits I think - and if/when vendors see the light they could join in | 22:40 |
| *** eernst has quit IRC | 22:42 | |
| clarkb | yes the proposal is more feasible if constrained to what is possible given current resource access | 22:43 |
| *** eernst has joined #openstack-infra | 22:43 | |
| fungi | clarkb: sorry, the wheel update is a bit of a red herring. the actual situation is that there are still a lot of openstack projects which are pure python but don't declare in setup.cfg that they should build universal wheels, which makes running python3 by default in some jobs install them from sdist. dhellmann wrote a set of changes (which i believe we merged) to have the openstack-specific wheel | 22:47 |
| fungi | builds (overrideably) pass a command-line option to build universal wheels by default so that projects don't need to update their configuration | 22:47 |
| clarkb | gotcha | 22:47 |
| fungi | mainly so as to avoid yet another round of mass patches | 22:48 |
| *** eernst has quit IRC | 22:48 | |
| *** eernst has joined #openstack-infra | 22:48 | |
| fungi | it came up on the pbr change to update testing for latest wheel release because wheel has now deprecated the [wheel] section in favor of [bdist_wheel] which has also been supported for a very long time. the pbr change updated pbr's setup.cfg to change that section name so as to silence the deprecation warning | 22:49 |
| fungi | this caused AJaeger to ask whether we should also consider porting that setup.cfg change to projects with an existing [wheel] section | 22:49 |
| fungi | so that was the (tenuous) connection | 22:51 |
| fungi | some discussions which arose from that led some people to misinterpret the situation as suggesting that configuring packages in setup.cfg to build universal wheels was actively harmful and to be avoided | 22:52 |
| *** eernst has quit IRC | 22:54 | |
| lifeless | Sounds like an antipattern that has never happened before. | 22:56 |
| *** eernst has joined #openstack-infra | 23:00 | |
| *** Swami has quit IRC | 23:04 | |
| *** eernst has quit IRC | 23:05 | |
| *** pbourke has quit IRC | 23:05 | |
| *** eernst has joined #openstack-infra | 23:05 | |
| *** pbourke has joined #openstack-infra | 23:05 | |
| fungi | well, the main concern raised is that we have people not particularly communicative within the community who are prone to push massive batches of identical changes to every project, without warning and uncoordinated with the rest of the cross-project efforts we have going on | 23:05 |
| lifeless | I know, been there, seen it. | 23:07 |
| lifeless | Google has a team specifically chartered to do such holistic cross-the-code-base maintenance. | 23:08 |
| fungi | it's been asserted that these individuals are tasked with pumping their employers' contribution statistics | 23:08 |
| lifeless | (On their code, not ours :P) | 23:08 |
| *** eernst has quit IRC | 23:12 | |
| *** ijw has joined #openstack-infra | 23:23 | |
| *** rfolco has quit IRC | 23:23 | |
| *** felipemonteiro has joined #openstack-infra | 23:24 | |
| *** gyee has quit IRC | 23:31 | |
| openstackgerrit | Merged openstack-dev/pbr master: Special case long_description_content_type https://review.openstack.org/565177 | 23:32 |
| *** bdodd_ has joined #openstack-infra | 23:34 | |
| *** roman_g has quit IRC | 23:35 | |
| *** bdodd has quit IRC | 23:35 | |
| *** sthussey has quit IRC | 23:36 | |
| *** ijw has quit IRC | 23:53 | |
| *** felipemonteiro has quit IRC | 23:56 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!