| fungi | mgagne: one showing in use for over 2 days, e8c4c1df-743e-44ea-b576-60d655b9f58e | 00:00 |
|---|---|---|
| fungi | something tells me it's lost | 00:01 |
| mgagne | ok, this one exists on my side | 00:01 |
| fungi | other than that, nothing | 00:01 |
| mgagne | cool | 00:01 |
| mgagne | thanks | 00:01 |
| clarkb | mordred: mostly thinking maybe the existing servers go in a new group for the extra ansible | 00:02 |
| clarkb | mordred: then not being in that is the basest of ansible | 00:02 |
| mordred | clarkb: yah - totes | 00:04 |
| mordred | clarkb: although this is all predicated on actually knowing what the basest of ansible is first | 00:04 |
| mordred | clarkb: which I thought I knew, but now I'm less sure | 00:04 |
| *** wolverineav has quit IRC | 00:05 | |
| corvus | clarkb: if you have a chance to +3 https://review.openstack.org/635358 today i'd appreciate it (so it can wend its way through and we can find the next error) | 00:07 |
| *** wolverineav has joined #openstack-infra | 00:07 | |
| *** markvoelker has joined #openstack-infra | 00:11 | |
| *** wolverineav has quit IRC | 00:12 | |
| fungi | mgagne: after some playing around with awk i've managed to ask nodepool to delete all 24 images it thinks it uploaded to inap-mtl01 | 00:15 |
| mgagne | fungi: ok, will nodepool reupload them? | 00:15 |
| fungi | and i see a bunch new in uploading status, yes | 00:15 |
| mordred | fungi: any task that can't be solved with awk is not worth solving | 00:22 |
| *** dmsimard has quit IRC | 00:25 | |
| corvus | cgit urls with line numbers (eg http://git.openstack.org/cgit/openstack/nova/tree/requirements.txt#n31) use url fragments, which aren't sent to the server, which means we can't redirect them. cgit uses "#n", gitea uses "#L". the only choices i can think of are to ignore the problem, so line-number links won't work anymore (but at least you'll get to the right file, and can probably find the line | 00:27 |
| corvus | by looking at the url). or coming up with some javascript or something that we stick in gitea that causes it to work with both #n and #L (possibly by mutating the url to #L). | 00:27 |
| mordred | corvus: seems like adding javascript to gitea to accept #n as well shouldnt' be SUPER terrible, right? | 00:28 |
| corvus | i guess it's worth a try? | 00:29 |
| corvus | maybe it's upstreamable, maybe not... | 00:30 |
| corvus | if not, if it works for a couple years before bitrotting, that's probably good enough... | 00:30 |
| mordred | corvus: well, even if it's not, we can squeeze it in in a theme | 00:30 |
| mordred | corvus: yeah | 00:30 |
| mordred | corvus: I'm looking at gitea source code ... | 00:30 |
| mordred | public/js/index.js has function selectRange - which has a line: changeHash('#L' + a + '-' + 'L' + b); | 00:31 |
| mordred | so - something something | 00:31 |
| corvus | mordred: if you feel like taking that on while i continue working through the wonderful world of regexes, i'd be most appreciative. | 00:33 |
| mordred | corvus: yeah - on it | 00:33 |
| mordred | corvus: cgit only supports single lines and not ranges, right? | 00:33 |
| corvus | mordred: yep | 00:33 |
| *** sdake has quit IRC | 00:37 | |
| *** jamesmcarthur has quit IRC | 00:40 | |
| *** sdake has joined #openstack-infra | 00:41 | |
| *** jamesmcarthur has joined #openstack-infra | 00:41 | |
| *** sdake has quit IRC | 00:41 | |
| *** hwoarang has quit IRC | 00:42 | |
| *** tosky has quit IRC | 00:43 | |
| *** markvoelker has quit IRC | 00:44 | |
| *** hwoarang has joined #openstack-infra | 00:44 | |
| *** jamesmcarthur has quit IRC | 00:46 | |
| clarkb | corvus: done. Is the client cert data non secret? | 00:47 |
| clarkb | I guess the key is necessary | 00:48 |
| corvus | clarkb: afaik | 00:48 |
| openstackgerrit | Monty Taylor proposed openstack-infra/system-config master: Rewrite #nXXX to #LXXX in incoming gitea urls https://review.openstack.org/635379 | 00:49 |
| mordred | corvus: ^^ there's stab #1 | 00:49 |
| *** jamesmcarthur has joined #openstack-infra | 00:50 | |
| *** dmsimard has joined #openstack-infra | 00:57 | |
| *** wolverineav has joined #openstack-infra | 00:57 | |
| *** hwoarang has quit IRC | 00:58 | |
| *** wolverineav has quit IRC | 01:01 | |
| *** hwoarang has joined #openstack-infra | 01:01 | |
| mordred | corvus: https://github.com/go-gitea/gitea/pull/5987 | 01:03 |
| *** sthussey has quit IRC | 01:09 | |
| *** roman_g has quit IRC | 01:11 | |
| *** sdake has joined #openstack-infra | 01:12 | |
| openstackgerrit | Roman Gorshunov proposed openstack-infra/openstack-zuul-jobs master: Replace deprecated docs job in docs-on-readthedocs template https://review.openstack.org/635381 | 01:19 |
| *** roman_g has joined #openstack-infra | 01:20 | |
| *** gyee has quit IRC | 01:22 | |
| openstackgerrit | Merged openstack-infra/system-config master: Configure .kube/config on bridge https://review.openstack.org/635358 | 01:22 |
| *** dklyle_ has joined #openstack-infra | 01:27 | |
| *** hwoarang has quit IRC | 01:29 | |
| *** david-lyle has quit IRC | 01:31 | |
| *** hwoarang has joined #openstack-infra | 01:31 | |
| *** yamamoto has quit IRC | 01:41 | |
| *** bobh has joined #openstack-infra | 01:42 | |
| *** bobh has quit IRC | 01:44 | |
| *** armax has quit IRC | 01:46 | |
| Shrews | infra-root: just checking in on the new builder... no issues as of yet. But I did notice that opensuse-423 has not been successfully built for quite a few days. Also, I am afk tomorrow, so just a heads up. | 01:47 |
| *** sdake has quit IRC | 01:48 | |
| *** sdake has joined #openstack-infra | 01:50 | |
| *** jamesmcarthur has quit IRC | 01:50 | |
| *** bobh has joined #openstack-infra | 01:52 | |
| *** jamesmcarthur has joined #openstack-infra | 01:53 | |
| *** hwoarang has quit IRC | 02:05 | |
| *** yamamoto has joined #openstack-infra | 02:05 | |
| *** hwoarang has joined #openstack-infra | 02:06 | |
| *** jamesmcarthur has quit IRC | 02:06 | |
| *** wolverineav has joined #openstack-infra | 02:07 | |
| *** bobh has quit IRC | 02:09 | |
| *** david-lyle has joined #openstack-infra | 02:12 | |
| *** jamesmcarthur_ has joined #openstack-infra | 02:13 | |
| *** wolverineav has quit IRC | 02:14 | |
| *** manjeets has quit IRC | 02:15 | |
| *** dklyle_ has quit IRC | 02:15 | |
| *** manjeets has joined #openstack-infra | 02:16 | |
| *** sdake has quit IRC | 02:19 | |
| *** bobh has joined #openstack-infra | 02:23 | |
| *** diablo_rojo has quit IRC | 02:27 | |
| *** wolverineav has joined #openstack-infra | 02:35 | |
| *** wolverineav has quit IRC | 02:40 | |
| *** ijw has quit IRC | 02:43 | |
| *** armax has joined #openstack-infra | 02:52 | |
| *** bobh has quit IRC | 02:53 | |
| *** jamesmcarthur_ has quit IRC | 02:54 | |
| fungi | noted, thanks! | 02:56 |
| *** psachin has joined #openstack-infra | 02:56 | |
| fungi | Shrews: seems like that could also be relevant to proposed openstack/governance change https://review.openstack.org/633460 | 02:57 |
| *** jamesmcarthur has joined #openstack-infra | 02:59 | |
| *** rcernin has joined #openstack-infra | 03:07 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: web: add /{tenant}/buildsets route https://review.openstack.org/630035 | 03:08 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: web: add buildsets page https://review.openstack.org/630041 | 03:08 |
| *** ekcs has quit IRC | 03:09 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: web: add /{tenant}/buildset/{uuid} route https://review.openstack.org/630078 | 03:10 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul-jobs master: add-build-sshkey: remove previously authorized build-sshkey https://review.openstack.org/632620 | 03:12 |
| *** jamesmcarthur has quit IRC | 03:14 | |
| *** jamesmcarthur has joined #openstack-infra | 03:15 | |
| *** adriant has quit IRC | 03:18 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool master: Implement zookeeper-auth https://review.openstack.org/619155 | 03:23 |
| *** wolverineav has joined #openstack-infra | 03:26 | |
| *** adriant has joined #openstack-infra | 03:27 | |
| *** hwoarang has quit IRC | 03:31 | |
| *** hwoarang has joined #openstack-infra | 03:32 | |
| *** sdake has joined #openstack-infra | 03:43 | |
| *** rlandy|bbl is now known as rlandy | 03:43 | |
| *** sdake has quit IRC | 03:44 | |
| *** jamesmcarthur has quit IRC | 03:53 | |
| *** jamesmcarthur has joined #openstack-infra | 03:53 | |
| *** jamesmcarthur has quit IRC | 03:56 | |
| *** jamesmcarthur has joined #openstack-infra | 03:58 | |
| *** jamesmcarthur has quit IRC | 03:58 | |
| *** jamesmcarthur has joined #openstack-infra | 03:58 | |
| *** jamesmcarthur has quit IRC | 04:02 | |
| *** rlandy has quit IRC | 04:06 | |
| *** sdake has joined #openstack-infra | 04:10 | |
| *** ykarel has joined #openstack-infra | 04:14 | |
| *** ramishra has joined #openstack-infra | 04:14 | |
| *** udesale has joined #openstack-infra | 04:25 | |
| *** rcernin has quit IRC | 04:35 | |
| *** ramishra has quit IRC | 04:45 | |
| *** jtomasek has joined #openstack-infra | 04:49 | |
| *** ykarel has quit IRC | 05:03 | |
| *** jtomasek has quit IRC | 05:07 | |
| *** armax has quit IRC | 05:09 | |
| *** ramishra has joined #openstack-infra | 05:14 | |
| *** hwoarang has quit IRC | 05:15 | |
| *** jtomasek has joined #openstack-infra | 05:15 | |
| *** hwoarang has joined #openstack-infra | 05:17 | |
| *** sdake has quit IRC | 05:18 | |
| *** ykarel has joined #openstack-infra | 05:19 | |
| *** jtomasek has quit IRC | 05:24 | |
| *** noama has joined #openstack-infra | 05:26 | |
| *** ramishra has quit IRC | 05:29 | |
| *** ramishra has joined #openstack-infra | 05:30 | |
| *** chandankumar has joined #openstack-infra | 05:45 | |
| *** chandankumar is now known as chkumar|ruck | 05:46 | |
| *** janki has joined #openstack-infra | 05:51 | |
| *** jamesmcarthur has joined #openstack-infra | 06:02 | |
| *** hwoarang has quit IRC | 06:04 | |
| *** adriant has quit IRC | 06:06 | |
| *** hwoarang has joined #openstack-infra | 06:06 | |
| *** jamesmcarthur has quit IRC | 06:06 | |
| *** markvoelker has joined #openstack-infra | 06:07 | |
| *** adriant has joined #openstack-infra | 06:07 | |
| *** e0ne has joined #openstack-infra | 06:10 | |
| *** markvoelker has quit IRC | 06:11 | |
| *** wolverineav has quit IRC | 06:21 | |
| *** wolverineav has joined #openstack-infra | 06:22 | |
| *** wolverineav has quit IRC | 06:30 | |
| *** quiquell|off is now known as quiquell|rover | 06:36 | |
| *** diablo_rojo has joined #openstack-infra | 06:37 | |
| *** chkumar|ruck has quit IRC | 06:48 | |
| *** chandankumar has joined #openstack-infra | 06:48 | |
| *** chandankumar is now known as chkumar|ruck | 06:49 | |
| *** quiquell|rover is now known as quique|rover|r-- | 06:53 | |
| *** kopecmartin|off is now known as kopecmartin | 07:01 | |
| *** pcaruana has joined #openstack-infra | 07:02 | |
| *** betherly has joined #openstack-infra | 07:07 | |
| *** markvoelker has joined #openstack-infra | 07:08 | |
| *** e0ne has quit IRC | 07:10 | |
| *** efried has quit IRC | 07:11 | |
| *** jtomasek has joined #openstack-infra | 07:12 | |
| *** efried has joined #openstack-infra | 07:12 | |
| *** hwoarang has quit IRC | 07:15 | |
| *** yamamoto has quit IRC | 07:15 | |
| *** wolverineav has joined #openstack-infra | 07:15 | |
| *** slaweq has joined #openstack-infra | 07:16 | |
| *** diablo_rojo has quit IRC | 07:16 | |
| *** yamamoto has joined #openstack-infra | 07:16 | |
| *** quique|rover|r-- is now known as quiquell|rover | 07:18 | |
| *** hwoarang has joined #openstack-infra | 07:18 | |
| *** e0ne has joined #openstack-infra | 07:20 | |
| *** slaweq has quit IRC | 07:21 | |
| *** pgaxatte has joined #openstack-infra | 07:21 | |
| *** hwoarang has quit IRC | 07:24 | |
| *** hwoarang has joined #openstack-infra | 07:25 | |
| *** janki is now known as janki- | 07:27 | |
| *** janki- is now known as janki | 07:27 | |
| *** e0ne has quit IRC | 07:31 | |
| *** gfidente has joined #openstack-infra | 07:33 | |
| *** quiquell|rover is now known as quiquell|rover|b | 07:33 | |
| *** quiquell|rover|b is now known as quique|rover|brb | 07:33 | |
| *** slaweq has joined #openstack-infra | 07:37 | |
| *** markvoelker has quit IRC | 07:41 | |
| *** slaweq has quit IRC | 07:42 | |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/project-config master: Fix translation jobs due to opendev/base-job change https://review.openstack.org/635443 | 07:44 |
| *** janki is now known as janki|lunch | 07:44 | |
| AJaeger | infra-root, the change above ^ follows https://review.openstack.org/635347 , could you review to unbreak all translation jobs, please? | 07:44 |
| *** slaweq has joined #openstack-infra | 07:47 | |
| *** wolverineav has quit IRC | 07:50 | |
| *** slaweq has quit IRC | 07:51 | |
| ianw | AJaeger: lgtm, was discussed a little before and i think the idea of just fixing like this was considered best | 07:51 |
| openstackgerrit | Merged openstack/diskimage-builder master: pip-and-virtualenv: handle centos image-based builds https://review.openstack.org/635365 | 07:54 |
| *** yboaron has quit IRC | 07:56 | |
| *** quique|rover|brb is now known as quiquell|rover | 07:58 | |
| *** hwoarang has quit IRC | 07:58 | |
| *** hwoarang has joined #openstack-infra | 07:59 | |
| *** Emine has joined #openstack-infra | 08:02 | |
| *** jamesmcarthur has joined #openstack-infra | 08:03 | |
| *** rpittau has joined #openstack-infra | 08:06 | |
| *** jamesmcarthur has quit IRC | 08:07 | |
| *** aojea has joined #openstack-infra | 08:09 | |
| AJaeger | thanks, ianw | 08:10 |
| *** tkajinam has quit IRC | 08:15 | |
| *** ginopc has joined #openstack-infra | 08:15 | |
| *** ykarel is now known as ykarel|lunch | 08:16 | |
| *** slaweq has joined #openstack-infra | 08:18 | |
| *** ccamacho has joined #openstack-infra | 08:20 | |
| *** rossella_s has joined #openstack-infra | 08:23 | |
| *** wolverineav has joined #openstack-infra | 08:25 | |
| *** wolverineav has quit IRC | 08:31 | |
| *** shardy has joined #openstack-infra | 08:32 | |
| *** slaweq has quit IRC | 08:36 | |
| *** markvoelker has joined #openstack-infra | 08:38 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: web: add /{tenant}/buildsets route https://review.openstack.org/630035 | 08:39 |
| *** jpena|off is now known as jpena | 08:45 | |
| *** tosky has joined #openstack-infra | 08:46 | |
| *** jpich has joined #openstack-infra | 08:48 | |
| *** zbr|ssbarnea has joined #openstack-infra | 08:49 | |
| *** xek__ has joined #openstack-infra | 08:49 | |
| *** chkumar|ruck has quit IRC | 08:53 | |
| *** slaweq has joined #openstack-infra | 08:53 | |
| *** slaweq has quit IRC | 08:58 | |
| *** yboaron has joined #openstack-infra | 09:00 | |
| openstackgerrit | Merged openstack-infra/project-config master: Fix translation jobs due to opendev/base-job change https://review.openstack.org/635443 | 09:03 |
| *** slaweq has joined #openstack-infra | 09:03 | |
| *** slaweq has quit IRC | 09:08 | |
| *** ykarel|lunch is now known as ykarel | 09:10 | |
| *** zbr has joined #openstack-infra | 09:11 | |
| *** markvoelker has quit IRC | 09:12 | |
| *** zbr|ssbarnea has quit IRC | 09:12 | |
| *** panda|off is now known as panda | 09:17 | |
| *** kjackal has joined #openstack-infra | 09:21 | |
| openstackgerrit | Quique Llorente proposed openstack-infra/zuul master: Escape jinja2 stuff from inventory https://review.openstack.org/633930 | 09:22 |
| openstackgerrit | Quique Llorente proposed openstack-infra/zuul master: Escape jinja2 stuff from inventory https://review.openstack.org/633930 | 09:22 |
| *** ccamacho has quit IRC | 09:23 | |
| *** wolverineav has joined #openstack-infra | 09:27 | |
| *** wolverineav has quit IRC | 09:31 | |
| *** derekh has joined #openstack-infra | 09:32 | |
| *** slaweq has joined #openstack-infra | 09:33 | |
| *** chandankumar has joined #openstack-infra | 09:34 | |
| *** chandankumar is now known as chkumar|ruck | 09:35 | |
| *** jtomasek has quit IRC | 09:36 | |
| *** jtomasek has joined #openstack-infra | 09:37 | |
| *** yamamoto has quit IRC | 09:38 | |
| *** zbr has quit IRC | 09:40 | |
| *** zbr|ssbarnea has joined #openstack-infra | 09:40 | |
| *** ociuhandu has joined #openstack-infra | 09:43 | |
| *** yamamoto has joined #openstack-infra | 09:47 | |
| *** jbadiapa has quit IRC | 09:47 | |
| *** jbadiapa has joined #openstack-infra | 09:48 | |
| *** priteau has joined #openstack-infra | 09:49 | |
| *** chandankumar has joined #openstack-infra | 09:50 | |
| *** chkumar|ruck has quit IRC | 09:51 | |
| *** oanson has joined #openstack-infra | 09:51 | |
| *** chandankumar is now known as chkumar|ruck | 09:51 | |
| *** ociuhandu has quit IRC | 09:56 | |
| *** ociuhandu has joined #openstack-infra | 09:58 | |
| *** markmcd has quit IRC | 10:00 | |
| *** slaweq has quit IRC | 10:00 | |
| *** e0ne has joined #openstack-infra | 10:03 | |
| *** zbr|ssbarnea has quit IRC | 10:04 | |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/project-config master: Fix releasenotes publishing after opendev/base-jobs change https://review.openstack.org/635470 | 10:06 |
| AJaeger | config-core, another failing job, please review ^ | 10:07 |
| *** zbr|ssbarnea has joined #openstack-infra | 10:07 | |
| *** kjackal has quit IRC | 10:08 | |
| *** markvoelker has joined #openstack-infra | 10:09 | |
| AJaeger | infra-root, there's a change for openstack-infra/tripleo-ci in zuul for 69 hours ;( Change is 634410, probably waiting for Fedora 28 node | 10:09 |
| *** slaweq has joined #openstack-infra | 10:17 | |
| *** slaweq has quit IRC | 10:22 | |
| openstackgerrit | Quique Llorente proposed openstack-infra/zuul master: Escape jinja2 stuff from inventory https://review.openstack.org/633930 | 10:23 |
| *** betherly has quit IRC | 10:25 | |
| *** betherly has joined #openstack-infra | 10:25 | |
| Shrews | AJaeger: i looked in the nodepool logs yesterday and fedora 28 nodes were being created. i suspect something weird with that job we'll have to dig into zuul logs to figure out | 10:29 |
| openstackgerrit | Monty Taylor proposed openstack-infra/system-config master: Add opendev kubernetes nodes to ansible inventory https://review.openstack.org/635472 | 10:30 |
| zbr|ssbarnea | AJaeger: indeed, I created one job using fedora-28 node about an hour ago and it did start. i suspect something else. | 10:31 |
| *** ociuhandu has quit IRC | 10:31 | |
| AJaeger | Shrews, zbr|ssbarnea, so what can be done with that change? Abort it? | 10:35 |
| zbr|ssbarnea | AJaeger: let me ask on #oooq, | 10:36 |
| zbr|ssbarnea | AJaeger: ok to abort. | 10:38 |
| frickler | infra-root: seems there was a weird zuul error when that job started http://paste.openstack.org/show/744662/ | 10:38 |
| *** betherly has quit IRC | 10:40 | |
| frickler | AJaeger: did you see the error on 635470? | 10:41 |
| *** markvoelker has quit IRC | 10:41 | |
| AJaeger | frickler: no - thanks for telling me | 10:42 |
| AJaeger | fixing | 10:42 |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/project-config master: Fix releasenotes publishing after opendev/base-jobs change https://review.openstack.org/635470 | 10:43 |
| AJaeger | frickler: ^ | 10:43 |
| AJaeger | thanks | 10:43 |
| roman_g | Hello team. If anyone has capacity, please, have a look at openstack-infra/openstack-zuul-jobs change https://review.openstack.org/#/c/635381/1. I would like to get an advice whether this change could be tested somehow, before being reviewed (because Zuul check coverage is very basic). It potentially could break many jobs (I can add release notes, if needed). Thank you. | 10:45 |
| roman_g | Change: Replace deprecated docs job in docs-on-readthedocs template | 10:45 |
| openstackgerrit | Fabien Boucher proposed openstack-infra/zuul master: [WIP] - URLTrigger based on the timer trigger https://review.openstack.org/635241 | 10:57 |
| *** fdegir has quit IRC | 10:57 | |
| *** fdegir has joined #openstack-infra | 10:58 | |
| *** ociuhandu has joined #openstack-infra | 11:01 | |
| *** ociuhandu has quit IRC | 11:05 | |
| *** slaweq has joined #openstack-infra | 11:13 | |
| *** ociuhandu has joined #openstack-infra | 11:14 | |
| *** ociuhandu has quit IRC | 11:14 | |
| *** ociuhandu has joined #openstack-infra | 11:15 | |
| *** sshnaidm|afk is now known as sshnaidm | 11:17 | |
| *** ociuhandu_ has joined #openstack-infra | 11:22 | |
| *** ociuhand_ has joined #openstack-infra | 11:23 | |
| *** ociuhandu has quit IRC | 11:25 | |
| *** ociuhandu has joined #openstack-infra | 11:26 | |
| *** ociuhandu_ has quit IRC | 11:26 | |
| *** chkumar|ruck has quit IRC | 11:29 | |
| *** wolverineav has joined #openstack-infra | 11:29 | |
| *** ociuhand_ has quit IRC | 11:29 | |
| *** ociuhandu_ has joined #openstack-infra | 11:30 | |
| *** chandankumar has joined #openstack-infra | 11:30 | |
| *** chandankumar is now known as chkumar|ruck | 11:30 | |
| *** ociuhand_ has joined #openstack-infra | 11:31 | |
| *** ociuhandu has quit IRC | 11:31 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/system-config master: Add opendev kubernetes nodes to ansible inventory https://review.openstack.org/635472 | 11:32 |
| openstackgerrit | Monty Taylor proposed openstack-infra/system-config master: Run base-repos as ubuntu on k8s nodes to bootstrap https://review.openstack.org/635481 | 11:32 |
| openstackgerrit | Monty Taylor proposed openstack-infra/system-config master: Update k8s-on-openstack for ssh known_hosts https://review.openstack.org/635482 | 11:32 |
| *** wolverineav has quit IRC | 11:33 | |
| *** udesale has quit IRC | 11:33 | |
| *** udesale has joined #openstack-infra | 11:34 | |
| *** ociuhandu_ has quit IRC | 11:34 | |
| *** yamamoto has quit IRC | 11:37 | |
| *** rfolco is now known as rfolco_doctor | 11:38 | |
| *** markvoelker has joined #openstack-infra | 11:38 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/zone-opendev.org master: Add kubernetes nodes to DNS https://review.openstack.org/635485 | 11:40 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zone-opendev.org master: Switch opendev.org to point to gitea https://review.openstack.org/635486 | 11:40 |
| *** yamamoto has joined #openstack-infra | 11:40 | |
| *** yamamoto has quit IRC | 11:45 | |
| *** yamamoto has joined #openstack-infra | 11:46 | |
| openstackgerrit | Quique Llorente proposed openstack-infra/zuul master: Escape jinja2 stuff from inventory https://review.openstack.org/633930 | 11:47 |
| fnordahl | Anyone familiar with the Depends-On: GitHub integration? We see intermittent issues with gate not picking up +2A +WF events even though external dependencies have merged and are not sure how to debug them. Example review: https://review.openstack.org/#/c/633464/ | 11:48 |
| *** ccamacho has joined #openstack-infra | 11:49 | |
| *** yamamoto has quit IRC | 11:50 | |
| *** slaweq has quit IRC | 11:54 | |
| frickler | fnordahl: this is what I found in the zuul log, not sure whether this is a bug or whether the setup is bad, but I assume that you'll need to remove the dependency for now http://paste.openstack.org/show/744664/ | 11:56 |
| *** e0ne has quit IRC | 11:56 | |
| frickler | infra-root: ^^ maybe someone knows more about this scenario? | 11:56 |
| *** e0ne has joined #openstack-infra | 11:57 | |
| mordred | frickler: yeah: https://review.openstack.org/#/c/613143/ corvus has a WIP patch to fully support this | 11:58 |
| *** electrofelix has joined #openstack-infra | 11:58 | |
| *** electrofelix has quit IRC | 12:04 | |
| *** bassemessam has joined #openstack-infra | 12:04 | |
| *** jamesmcarthur has joined #openstack-infra | 12:04 | |
| frickler | mordred: ah, thx. fnordahl ^^ | 12:05 |
| *** electrofelix has joined #openstack-infra | 12:05 | |
| frickler | infra-root: how does one set up a launchpad project so that it's bugs can be updated? I haven't found a reference to that in the project creation guide. seems the watcher project has some issues with it, see e.g. https://bugs.launchpad.net/watcher/+bug/1802217 | 12:06 |
| openstack | Launchpad bug 1802217 in watcher "'watcher audit show' can't display the value of 'start_time' and 'end_time'" [Undecided,New] - Assigned to licanwei (li-canwei2) | 12:06 |
| *** jamesmcarthur has quit IRC | 12:09 | |
| *** markvoelker has quit IRC | 12:11 | |
| *** priteau has quit IRC | 12:12 | |
| AJaeger | mordred: could you +2A 635470, please? | 12:12 |
| *** aojeagarcia has joined #openstack-infra | 12:15 | |
| mordred | AJaeger: done! | 12:17 |
| AJaeger | thanks! | 12:18 |
| *** aojea has quit IRC | 12:18 | |
| openstackgerrit | Merged openstack-infra/project-config master: Fix releasenotes publishing after opendev/base-jobs change https://review.openstack.org/635470 | 12:24 |
| *** yamamoto has joined #openstack-infra | 12:27 | |
| *** wolverineav has joined #openstack-infra | 12:29 | |
| *** iurygregory has quit IRC | 12:30 | |
| *** wolverineav has quit IRC | 12:34 | |
| *** markmcd has joined #openstack-infra | 12:39 | |
| openstackgerrit | Tobias Henkel proposed openstack-infra/zuul master: Perform per repo locking on the executor https://review.openstack.org/635495 | 12:40 |
| openstackgerrit | Tobias Henkel proposed openstack-infra/zuul master: Optionally parallelize update threads https://review.openstack.org/635496 | 12:40 |
| *** jpena is now known as jpena|lunch | 12:44 | |
| *** sdake has joined #openstack-infra | 12:45 | |
| *** sdake has quit IRC | 12:51 | |
| *** sshnaidm is now known as sshnaidm|afk | 12:52 | |
| *** sdake has joined #openstack-infra | 12:52 | |
| *** jtomasek has quit IRC | 12:55 | |
| *** jtomasek has joined #openstack-infra | 12:55 | |
| *** sigmavirus24 has joined #openstack-infra | 13:00 | |
| *** bobh has joined #openstack-infra | 13:01 | |
| * sigmavirus24 waves | 13:02 | |
| sigmavirus24 | I tripped over BonnyCI again and remembered that Zuul v3 is using github3.py. I just wanted to check in and make sure we wren't blocking anything for y'all | 13:02 |
| *** rh-jelabarre has joined #openstack-infra | 13:03 | |
| *** rlandy has joined #openstack-infra | 13:04 | |
| *** priteau has joined #openstack-infra | 13:06 | |
| *** markvoelker has joined #openstack-infra | 13:08 | |
| *** jamesmcarthur has joined #openstack-infra | 13:09 | |
| *** jamesmcarthur has quit IRC | 13:09 | |
| *** jamesmcarthur has joined #openstack-infra | 13:09 | |
| *** sshnaidm|afk is now known as sshnaidm | 13:11 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul master: WIP Run python with jemalloc in containers https://review.openstack.org/635504 | 13:14 |
| *** slaweq has joined #openstack-infra | 13:17 | |
| mordred | sigmavirus24: I think we're eagerly awaiting the landing of the tasks api stuff - but I don't think anything is actively blocking us | 13:19 |
| mordred | sigmavirus24: ooh - looks like you landed that | 13:20 |
| mordred | tobiash: was there anything you were blocked on from github3.py ^^ ? | 13:20 |
| mordred | (checks api, not tasks api) | 13:21 |
| tobiash | the checks api landed right? | 13:21 |
| *** mriedem has joined #openstack-infra | 13:21 | |
| tobiash | but not release yet? | 13:21 |
| *** bobh has quit IRC | 13:22 | |
| mordred | tobiash: looks like it's been released! | 13:22 |
| tobiash | oh, then we can move on with checks api in zuul | 13:22 |
| mordred | tobiash: https://pypi.org/project/github3.py/ 1.3.0 is out, and https://github.com/sigmavirus24/github3.py/blob/master/docs/source/release-notes/1.3.0.rst mentions it | 13:23 |
| tobiash | so if that doesn't break us, everything is fine ;) | 13:23 |
| * mordred hands sigmavirus24 a cake | 13:23 | |
| tobiash | oh, looks like we're already running 1.3.0 in production, so everything looks good :) | 13:25 |
| mordred | woot! | 13:27 |
| *** ramishra has quit IRC | 13:29 | |
| *** wolverineav has joined #openstack-infra | 13:30 | |
| fnordahl | great, thanks frickler mordred | 13:31 |
| *** yboaron_ has joined #openstack-infra | 13:32 | |
| *** ramishra has joined #openstack-infra | 13:33 | |
| *** yboaron has quit IRC | 13:35 | |
| *** wolverineav has quit IRC | 13:35 | |
| *** jamesmcarthur has quit IRC | 13:37 | |
| *** jamesmcarthur has joined #openstack-infra | 13:38 | |
| mnaser | infra-root: i had a hold on an os_mistral job that i no longer need :) | 13:39 |
| *** roman_g has quit IRC | 13:40 | |
| *** roman_g has joined #openstack-infra | 13:41 | |
| *** markvoelker has quit IRC | 13:41 | |
| *** jamesmcarthur has quit IRC | 13:43 | |
| *** lpetrut has joined #openstack-infra | 13:45 | |
| *** kgiusti has joined #openstack-infra | 13:47 | |
| *** sdake has quit IRC | 13:47 | |
| *** ykarel is now known as ykarel|away | 13:47 | |
| cmurphy | clarkb: I'm mostly not here till next week but can you send me the whole log of one puppet run on etherpad-dev and I can try to see what's up | 13:48 |
| *** ramishra has quit IRC | 13:48 | |
| *** jpena|lunch is now known as jpena | 13:48 | |
| frickler | mnaser: thanks for the update, deleted the node. there are also two old nodes held by you/fungi for magnum-kubernetes-conformance, are these obsolete, too? | 13:51 |
| frickler | there's also "mordred: debugging the dockers" three weeks old ... | 13:52 |
| *** lucas-afk is now known as lucasagomes | 13:53 | |
| mordred | frickler: I do not believe I need to debug the dockers any more if you are deleting things | 13:56 |
| *** janki|lunch has quit IRC | 13:57 | |
| *** rfolco_doctor is now known as rfolco | 13:58 | |
| openstackgerrit | Quique Llorente proposed openstack-infra/zuul master: Escape jinja2 stuff from inventory https://review.openstack.org/633930 | 13:59 |
| *** ykarel|away has quit IRC | 14:00 | |
| *** quiquell|rover is now known as quique|rover|eat | 14:06 | |
| *** jamesmcarthur has joined #openstack-infra | 14:08 | |
| *** yboaron_ has quit IRC | 14:08 | |
| *** yboaron_ has joined #openstack-infra | 14:13 | |
| *** jamesmcarthur has quit IRC | 14:14 | |
| *** jcoufal has joined #openstack-infra | 14:15 | |
| *** jtomasek has quit IRC | 14:18 | |
| *** bobh has joined #openstack-infra | 14:20 | |
| *** psachin has quit IRC | 14:26 | |
| *** irclogbot_1 has joined #openstack-infra | 14:30 | |
| *** wolverineav has joined #openstack-infra | 14:31 | |
| *** sdake has joined #openstack-infra | 14:34 | |
| *** wolverineav has quit IRC | 14:36 | |
| *** ykarel|away has joined #openstack-infra | 14:37 | |
| *** quique|rover|eat is now known as quiquell|rover | 14:38 | |
| *** markvoelker has joined #openstack-infra | 14:38 | |
| *** bobh has quit IRC | 14:45 | |
| *** jamesmcarthur has joined #openstack-infra | 14:46 | |
| *** jamesmcarthur has quit IRC | 14:52 | |
| *** d0ugal has quit IRC | 14:52 | |
| *** sdake has quit IRC | 14:54 | |
| *** jtomasek has joined #openstack-infra | 14:56 | |
| *** eharney has joined #openstack-infra | 14:57 | |
| *** david-lyle has quit IRC | 14:57 | |
| *** iurygregory has joined #openstack-infra | 15:04 | |
| clarkb | cmurphy: http://paste.openstack.org/show/744682/ that is the full log of a puppet4 run on etherpad-dev01 | 15:08 |
| clarkb | cmurphy: it is worth noting that I'm sort of assuming it is talking about the group in hiera.yaml since the item before it is complaining about hiera.yaml version 3 | 15:09 |
| clarkb | but it doesn't actually print the file or line so that is a guess. | 15:09 |
| *** sdake has joined #openstack-infra | 15:09 | |
| *** markvoelker has quit IRC | 15:11 | |
| *** mpeterson has quit IRC | 15:12 | |
| *** mpeterson has joined #openstack-infra | 15:12 | |
| *** dansmith has quit IRC | 15:13 | |
| *** sigmavirus24 has quit IRC | 15:13 | |
| *** dansmith has joined #openstack-infra | 15:14 | |
| clarkb | cmurphy: also if I run hiera command with -d flag I don't get that error message when not setting the group var on the command line | 15:14 |
| *** sigmavirus24 has joined #openstack-infra | 15:15 | |
| *** jamesmcarthur has joined #openstack-infra | 15:18 | |
| clarkb | https://tickets.puppetlabs.com/si/jira.issueviews:issue-html/PUP-8887/PUP-8887.html does seem to have some related info. Basically you need to set things as a fact, via ENC, or at top scope so setting it at node level is bad? | 15:20 |
| clarkb | what is extra weird is that it works | 15:20 |
| clarkb | or at least the subsequent puppet run doesn't complain about failed lookups | 15:21 |
| fungi | frickler: thanks for the reminder. i've tried to ask mnaser a few times if we can release those nodes but i know he's busy... i expect we can just go ahead and drop them at this point as it's been weeks already | 15:22 |
| mnaser | sorry we can release those as well, i apologize frickler and fungi :) | 15:22 |
| *** jamesmcarthur has quit IRC | 15:23 | |
| fungi | no worries! | 15:23 |
| clarkb | or https://tickets.puppetlabs.com/browse/PUP-7336 too | 15:23 |
| fungi | frickler: on your lp bug updating question, if the short repository name (the part after the /) matches the name of a project in lp and the openstack infra account is added to the bug supervisor group for that lp project, then it should just work | 15:24 |
| *** ccamacho has quit IRC | 15:24 | |
| fungi | frickler: if you have a more specific example i can take a peek | 15:24 |
| clarkb | cmurphy: fungi: ok I wonder if it is this lookup that fails https://git.openstack.org/cgit/openstack-infra/system-config/tree/manifests/site.pp#n8 since that is at top scope before we set a group scope | 15:25 |
| clarkb | cmurphy: fungi if that is the case I think we can replace that hiera lookup with a static definition of that data ( we don't lose much if we do that) then the warning should go away and we can move forward | 15:26 |
| clarkb | I'll get a patch up for that which should let us test it at least | 15:26 |
| fungi | makes sense | 15:26 |
| frickler | fungi: I think that account is missing here, what's the exact name for it? also, is that part documented somewhere and I just didn't search deep enough? https://launchpad.net/~watcher-drivers https://bugs.launchpad.net/watcher/ | 15:26 |
| guilhermesp | infra-root: I'm hitting a failure on a job that is quit tricky to reproduce in a local vm. Is it possible to get a hold on openstack-ansible-upgrade-aio_lxc-ubuntu-bionic over project openstack/openstack-ansible (review #627782 ) | 15:26 |
| *** slaweq has quit IRC | 15:27 | |
| frickler | fungi: you might also want to respond to Dantalion directly in #openstack-dev where this issue came up earlier today | 15:27 |
| *** jtomasek has quit IRC | 15:27 | |
| *** quiquell|rover is now known as quiquell|off | 15:27 | |
| fungi | frickler: huh... it looks like maybe we removed all mention of launchpad from https://docs.openstack.org/infra/manual/creators.html at some point? that does indeed make it hard to point people to documentation we have^H^Hd about it | 15:30 |
| *** wolverineav has joined #openstack-infra | 15:32 | |
| frickler | fungi: indeed https://review.openstack.org/561753 may have been a bit too eager | 15:34 |
| fungi | looks like it happened last april in https://review.openstack.org/561753 | 15:34 |
| fungi | yep | 15:34 |
| fungi | was just hunting that down myself | 15:34 |
| *** zbr has joined #openstack-infra | 15:35 | |
| *** yboaron_ has quit IRC | 15:35 | |
| *** yboaron_ has joined #openstack-infra | 15:36 | |
| openstackgerrit | Clark Boylan proposed openstack-infra/system-config master: Remove top level hiera lookups https://review.openstack.org/635543 | 15:36 |
| clarkb | cmurphy: fungi ^ something like that maybe | 15:36 |
| clarkb | guilhermesp: give me a moment and I'll have that set up for you | 15:37 |
| *** wolverineav has quit IRC | 15:37 | |
| *** zbr|ssbarnea has quit IRC | 15:37 | |
| *** slaweq has joined #openstack-infra | 15:39 | |
| guilhermesp | thanks clarkb | 15:39 |
| clarkb | guilhermesp: | openstack | git.openstack.org/openstack/openstack-ansible | openstack-ansible-upgrade-aio_lxc-ubuntu-bionic | refs/changes/82/627782/.* | 1 | guilhermesp debugging hard to reproduce failure | done | 15:39 |
| *** sthussey has joined #openstack-infra | 15:46 | |
| clarkb | guilhermesp: let us know when the next failure on that job happens and we should be able to get you access to it then | 15:47 |
| *** diablo_rojo has joined #openstack-infra | 16:01 | |
| openstackgerrit | Fabien Boucher proposed openstack-infra/zuul master: Propose the URL driver to implement an URL change Require Filter https://review.openstack.org/635554 | 16:03 |
| *** yboaron_ has quit IRC | 16:04 | |
| *** yboaron_ has joined #openstack-infra | 16:05 | |
| *** yboaron_ has quit IRC | 16:05 | |
| *** apetrich has quit IRC | 16:06 | |
| fungi | infra-root: we've been getting e-mails to the infra-root@o.o alias from cloudbees for a while about the fact that our gearman-plugin is incompatible with csrf and api permissions restrictions in supported jenkins releases. they're handling it as an embargoed vulnerability report in their jira even though it's pretty clearly not something which needs to be kept under embargo. today they updated saying | 16:08 |
| fungi | they're going to issue an advisory at the end of march if we don't "fix it" before then... should we ask them to not wait until then and just go ahead now? should we request they remove the plugin? any suggestions? | 16:08 |
| *** markvoelker has joined #openstack-infra | 16:08 | |
| fungi | i can try to reply by e-mail and see if that works (i'm not super interested in trying to set up a jira account to be able to correspond with them if that turns out to be required) | 16:09 |
| clarkb | it doesnt http though | 16:09 |
| clarkb | do they say ehat needs fixing or is the actual purpose of the thing (triggering jobs) the problem? | 16:09 |
| odyssey4me | Hi folks - if an ansible project on github, not in gerrit, wishes to test using zuul in a similar way to how ansible's openstack modules are tested - what's the process? Is there a document somewhere? | 16:09 |
| clarkb | fungi I have little interest in maintaining it though so if there are problems hsing it in modern jenkins we should say that and see if anyone shows up to maintain it? | 16:10 |
| guilhermesp | ok clarkb I just recheck the patch and when I get the failure I'll let you know | 16:10 |
| fungi | clarkb: "CSRF vulnerability and missing permission checks in Gearman Plugin allowed server-side request forgery, capturing credentials [...] Please see the documentation for advice how to address this issue. I'd be happy to review any changes beforehand, in a private repository I can create on request, if you're unsure how to address this issue for your plugin." | 16:11 |
| clarkb | fungi thats weird there isnt any http to csrf in that plugin iirc | 16:11 |
| fungi | clarkb: i have similar sentiments. i lean toward telling them it's basically abandonware and better if they just go ahead and warn users if they think it's unsafe | 16:11 |
| fungi | (or remove it from their plugins site completely) | 16:12 |
| clarkb | fungi we should be careful we dont mischaracterize whatever the issue is though. If the problem is any gearman client can submit work well that is by design | 16:13 |
| *** gyee has joined #openstack-infra | 16:13 | |
| *** rossella_s has quit IRC | 16:14 | |
| *** ekultails has joined #openstack-infra | 16:14 | |
| fungi | yeah, i honestly don't know enough about the design of the plugin, or of the jenkins plugin interface, to be able to accurately assess the risks they're trying to describe, nor do i have much interest in gaining that degree of knowledge about them as we're really not maintaining or recommending it these days | 16:14 |
| *** noama has quit IRC | 16:16 | |
| clarkb | odyssey4me: so far we've been trying to limit the scopenof that to third party testing of opensta k integrations in github. eg not the primary ci system and not gating | 16:18 |
| clarkb | odyssey4me: currently doing that for the openstack ansible module and zuul's github api lin github3 | 16:18 |
| fungi | that sounds like what he's requesting | 16:19 |
| fungi | unless i misread | 16:19 |
| *** jamesmcarthur has joined #openstack-infra | 16:20 | |
| odyssey4me | clarkb we're discussing moving the ansible-hardening repo function to ansible-lockdown, and OSA has test coverage in terms of distro which ansible-lockdown cannot have with their current test provider | 16:20 |
| odyssey4me | so we're looking to help out there | 16:20 |
| *** pcaruana has quit IRC | 16:21 | |
| odyssey4me | although the preference is to test everything in one infrastructure, of course | 16:21 |
| *** cyberpear has joined #openstack-infra | 16:23 | |
| zbr | out of curiosity, can bindep check for alternative package names? like checking for foo and bar, passing if one of them is istalled? | 16:23 |
| *** ccamacho has joined #openstack-infra | 16:24 | |
| *** jamesmcarthur has quit IRC | 16:24 | |
| *** jamesmcarthur has joined #openstack-infra | 16:24 | |
| clarkb | so we would be testing openstacks usage of a third party repo/function. That is similar to the github3 case. As for process you ha e to add thr repo to zuul's config (github3 a good example) and then manage the jobs on our side since we cant gate the remote | 16:25 |
| clarkb | zbr I dont think it does | 16:25 |
| *** armax has joined #openstack-infra | 16:25 | |
| clarkb | zbr but the package manager should? I guess if used in config mgmt that may result in attempting to install the same package over and over | 16:25 |
| odyssey4me | clarkb ok, the ansible-lockdown folks will send an email to the infra list describing what they're hoping to achieve and we can take it from there | 16:27 |
| *** eharney has quit IRC | 16:28 | |
| odyssey4me | fungi on a separate topic, I've tried several times to subscribe to openstack-discuss, and I'm getting nothing back from the list server... and no emails | 16:28 |
| odyssey4me | fungi my subscription is with my rackspace email address | 16:28 |
| *** slaweq has quit IRC | 16:29 | |
| *** jamesmcarthur has quit IRC | 16:30 | |
| fungi | odyssey4me: i'll take a look | 16:31 |
| fungi | odyssey4me: are you trying to subscribe via smtp or http? | 16:31 |
| odyssey4me | fungi http | 16:32 |
| odyssey4me | fungi I can login to 'manage' my subscription, if that helps. | 16:32 |
| *** udesale has quit IRC | 16:33 | |
| *** wolverineav has joined #openstack-infra | 16:33 | |
| *** agopi has quit IRC | 16:34 | |
| fungi | Jan 14 13:26:03 2019 (2583) openstack-discuss: pending Jesse Pretorius <... | 16:34 |
| fungi | i'll see if the cnofirmation messages are getting rejected by the mta for some reason | 16:34 |
| *** wolverineav has quit IRC | 16:37 | |
| fungi | hrm, we only keep 10-11 days of mta logs on the lists.o.o server | 16:38 |
| fungi | odyssey4me: when was the last time you tried to subscribe? | 16:38 |
| odyssey4me | fungi I think that was the last time - I can retry now. | 16:38 |
| *** chkumar|ruck has quit IRC | 16:38 | |
| fungi | yeah, if you have a sec to do so, i can take a look in the mta logs | 16:38 |
| *** jamesmcarthur has joined #openstack-infra | 16:38 | |
| *** chandankumar has joined #openstack-infra | 16:39 | |
| odyssey4me | fungi done | 16:39 |
| *** apetrich has joined #openstack-infra | 16:40 | |
| fungi | odyssey4me: 2019-02-07 16:39:32 1grmi2-0007Bq-AW => [REDACTED]@rackspace.co.uk R=mailman_verp_router T=mailman_verp_smtp H=cust65406-1-in.mailcontrol.com [208.87.233.190] X=TLS1.2:DHE_RSA_AES_256_CBC_SHA256:256 C="250 2.0.0 x17GdUl4039216 Message accepted for delivery" | 16:41 |
| fungi | redaction mine | 16:41 |
| odyssey4me | hmm, ok so the rax spam thing must be blocking it or something | 16:41 |
| odyssey4me | ok, thanks for confirming - I'll go digging there | 16:41 |
| *** markvoelker has quit IRC | 16:41 | |
| fungi | so cluster-g.mailcontrol.com (208.87.233.190) claims to have accepted the message and assigned an internal queue id of x17GdUl4039216 to it | 16:42 |
| *** pgaxatte has quit IRC | 16:42 | |
| fungi | if you manage to identify why they're flagging it as spam, please do let us know | 16:42 |
| clarkb | fwiw I'm mostly going to try and catch up on expenses and mail and rebooting for updates today. But do ping me if there are reviews I should be looking at | 16:42 |
| efried | Howdy infra, quick question: | 16:44 |
| efried | Patch has a dependency on a lib change. I put Depends-On in my patch, meaning it will build but can't merge. | 16:44 |
| efried | Then the dependency merges, but as yet hasn't been released, requirements bumped, etc. | 16:44 |
| efried | But my patch still has the Depends-On, and the dependency has now merged; so will my patch actually merge, thereby breaking the world? | 16:44 |
| odyssey4me | thanks fungi - appreciate your time... I've not seen any notifications saying that this is being quarantined, so there might be something funky going on... I might ask again for some diagnostic info next week when I'm back at the office | 16:44 |
| fungi | no worries, happy to help | 16:44 |
| clarkb | efried: ya its not a perfect blocker if you need a release for the depends on | 16:44 |
| clarkb | efried: in general though if your testing is set up right you should still fail testing until the release happens | 16:45 |
| efried | clarkb: Hm, how so? | 16:45 |
| fungi | efried: yeah, the real way to do that is have your project test both against unreleased source of that dependency and also against the latest released version (e.g., from pypi) | 16:45 |
| fungi | so perhaps two similar jobs | 16:46 |
| efried | "test against" - it's just using requirements.txt for like UT etc. | 16:46 |
| fungi | with that as the only distinction | 16:46 |
| clarkb | efried: right if you depend on it you would expect that your unittests would fail until tehy can consume it | 16:46 |
| clarkb | efried: and if consuming it requires a release that means failing until then | 16:46 |
| efried | okay, but tldr Depends-On isn't quiiite smart enough to be a total no-op once the dep merges. | 16:47 |
| clarkb | correct | 16:47 |
| efried | and I need to put a procedural hold on such patches, e.g. https://review.openstack.org/#/c/634866/ | 16:47 |
| efried | That's what I needed to know. Thanks clarkb, fungi! | 16:47 |
| clarkb | efried: procedural hold or tests fail until it works hold | 16:47 |
| openstackgerrit | Fabien Boucher proposed openstack-infra/zuul master: URLTrigger driver time based https://review.openstack.org/635567 | 16:48 |
| fungi | right, in a one-off case like that, would dropping the depends-on once you've verified it works prevent the change from passing its jobs until the dependency gets released? | 16:48 |
| efried | yeah, but it sounds like I can't count on tests failing because the dep will get pulled into the test envs even after it merges. | 16:48 |
| efried | fungi: Yes it would | 16:48 |
| fungi | so that may be the simplest route | 16:48 |
| efried | Either way, I need to hold it until that happens. | 16:48 |
| efried | so it doesn't accidentally merge. | 16:49 |
| ttx | mordred: is https://review.openstack.org/#/c/634975/ supposed to fix the weird fails in direct-linking in Zuul status website? Is there anything more to do to get zuul.opendev.org to get the fix, or will it just pick it up at the next release / update ? | 16:49 |
| clarkb | fungi: I think https://review.openstack.org/#/c/635543/ is ready for review to try and fix that puppet4 group var not defined warning | 16:49 |
| fungi | basically 1. develop depending change with depends-on to new feature in dependency, 2. once it's observed working frop depends-on from commit message, 3. wait for jobs in formerly-depending change to pass (indicating the dependency has been released with the necessary change) | 16:49 |
| fungi | efried: ^ | 16:49 |
| fungi | s/from/drop/ | 16:50 |
| efried | fungi: ack | 16:50 |
| clarkb | ttx: the js stuff should get automatically picked up | 16:50 |
| fungi | though frop needs to be a word | 16:50 |
| clarkb | ttx: python changes requires a daemon restart but that isn't python | 16:50 |
| ttx | clarkb: you run master? | 16:51 |
| clarkb | ttx: we do | 16:51 |
| clarkb | well we install master, then run whatever is current on the python side when daemons get restarted | 16:51 |
| fungi | it's continuously deployed but not restarted for every change which merges (because that would be really disruptive) | 16:51 |
| ttx | clarkb: ok.. then I think it did not fix the issue. let me recheck | 16:51 |
| clarkb | ttx: you might need a hard refresh? | 16:51 |
| ttx | Now you sound like Jimmy | 16:52 |
| fungi | yeah, since it's static javascript, it will likely get cached client-side | 16:52 |
| zbr | clarkb: re bindep example, i can tell you that I can do "yum install /usr/bin/docker" and it will success regardless the package name that provides docker (which can be different). So, far I dont think that this is possible with bindep. | 16:52 |
| ttx | fungi, clarkb: I still have random 404s despite hard-refreshing. Trying to get a clear reproducer | 16:53 |
| zbr | yum accepts executable names instead of package names for installation, quite handy when the exec could be provided by rpms with various names (based on how you configured your repos). | 16:53 |
| clarkb | zbr: that would work if the yum lookup that bindep performs to check if package is installed also honors that | 16:54 |
| clarkb | I have no idea if yum does that and it would be platform specific behavior in that case | 16:54 |
| ttx | yeah, i can confirm direct linking to https://zuul.opendev.org/t/openstack/builds on a clean browser just 404s | 16:54 |
| zbr | clarkb: yeah, i know it could be tricky. i could look into bindep myself and check if this is possible. i just wanted to explain or validate the use-case for the moment. | 16:55 |
| ttx | maybe we enable that feature in our deployment ? That "fix" just changed the default | 16:56 |
| clarkb | zbr: I wouldn't rely on that behavior from bindep generaly. It is possible it would work due to yum behavior on platforms that use yum | 16:56 |
| *** aojeagarcia has quit IRC | 16:57 | |
| fungi | ttx: confirmed, if i go to https://zuul.opendev.org/tenants and select the builds link for the openstack tenant, i get cnotent. if i then refresh that page i get a 404 | 17:00 |
| *** rpittau has quit IRC | 17:02 | |
| corvus | mordred, ttx, fungi, clarkb: by visiting "about:debugging" i see that zuul.openstack.org has a service worker registered in my browser, so the unregister() method seems not to have worked | 17:04 |
| corvus | also, fwiw, my understanding is that when mordred's change works, it will cause *more* 404s, but at that point, we'll be able to reason more easily about what the problem is with the rewrite rules and fix them. | 17:05 |
| corvus | i don't have a service worker for zuul.opendev.org, but i'm not sure if i even did | 17:06 |
| fungi | until this moment, i had no idea service workers existed | 17:07 |
| *** sdake has quit IRC | 17:07 | |
| *** e0ne has quit IRC | 17:08 | |
| fungi | i just now went through and unregistered all the service workers in my browser. i didn't have any for zuul.opendev.org either | 17:08 |
| fungi | they were all, unsurprisingly, for sites i almost never visit | 17:09 |
| *** sdake has joined #openstack-infra | 17:10 | |
| corvus | fatal: [insecure-ci-registry01.opendev.org]: UNREACHABLE! => {"changed": false, "msg": "SSH Error: data could not be sent to r | 17:10 |
| corvus | emote host \"2001:4800:7818:101:be76:4eff:fe04:67f5\". Make sure this host can be reached over ssh", "unreachable": true} | 17:10 |
| corvus | fungi, clarkb: ^ i ssh'd to that host as root on bridge... what am i missing? | 17:10 |
| openstackgerrit | Sorin Sbarnea proposed openstack-infra/bindep master: Allow mentioning tools paths in bindep https://review.openstack.org/635576 | 17:11 |
| fungi | corvus: strange indeed. i can also ssh from bridge to the bare ipv6 address there as root, no problem | 17:11 |
| clarkb | corvus: did you sudo (which will use your user's known_hosts) or did you run as root root? Other than that I can't think of anything off the top of my head | 17:11 |
| fungi | ohh... maybe there's a conflicting ssh host key cached by the root account? | 17:12 |
| corvus | clarkb: i did 'sudo su -' ssh | 17:12 |
| corvus | as 2 separate commands | 17:12 |
| fungi | yeah, that should be fine | 17:12 |
| fungi | and wfm as well with the bare address too | 17:12 |
| *** bobh has joined #openstack-infra | 17:13 | |
| ttx | I'm not sure I want to learn what a "service worker" is | 17:13 |
| ttx | I miss HTML 1.0 | 17:13 |
| tosky | ttx: I'd say 3.0, but you are not the only one | 17:14 |
| openstackgerrit | Sorin Sbarnea proposed openstack-infra/bindep master: Allow mentioning tools paths in bindep https://review.openstack.org/635576 | 17:16 |
| clarkb | fungi: to follow up on jenkins gearman plugin stuff I think we should go ahead and let them know it is largely unmaintained. We should communicate that and see if anyone wants to fix it. And also ask for more information on the issue so that if someone does show up we can point them to that (and also sanity check the issue isn't that any gearman client can trigger jobs as intended) | 17:18 |
| fungi | i miss xhtml/1.0 | 17:18 |
| fungi | it was nicely self-consistent | 17:18 |
| fungi | sure, i can try to update them via e-mail reply | 17:19 |
| corvus | clarkb: ++ | 17:19 |
| *** jtomasek has joined #openstack-infra | 17:19 | |
| fungi | also, i'm running out of bandwidth, but if any infra-root wants to pick up the effort on https://launchpad.net/bugs/1721803 that would help (or we can just let it fester i suppose) | 17:20 |
| openstack | Launchpad bug 1721803 in OpenStack Community Project "Not able to log in using google/launchpad account - confirmation email is not received" [Critical,Confirmed] - Assigned to Jeremy Stanley (fungi) | 17:20 |
| clarkb | the one place where it may tie into http/api things is the configuration stuff | 17:20 |
| clarkb | which hopefully is straightforward to fix if anyone wants to do that | 17:20 |
| corvus | fungi: i do not understand that bug report. it seems to be at least 3 different things. | 17:21 |
| corvus | fungi: are you aware of any current problems with review.o.o? | 17:22 |
| fungi | corvus: i'm not aware of any current problems with review.o.o. it's one of several dumping-ground bugs about mail delivery issues with ask.o.o | 17:23 |
| corvus | the only thing on that bug from this year seems to be about launchpad itself | 17:23 |
| corvus | oh, yeah, the reverse ip on the last thing says ask | 17:24 |
| clarkb | corvus: the last comment is from 9 hours ago and talks about spf records | 17:24 |
| clarkb | oh you mean re ask ya | 17:24 |
| corvus | i'm going to yield to the ask maintainers. | 17:24 |
| clarkb | seems the latest comment is suggesting that our ?all entry in the spf record isn't positive enough for some mail servers to accept mail from foo.openstack.org | 17:25 |
| clarkb | and that we should individual list all domains? | 17:25 |
| corvus | i disagree | 17:26 |
| corvus | but if the ask maintainers want to do that for ask.openstack.org, i have no objection. | 17:26 |
| clarkb | corvus: aren't we the ask maintainers? | 17:26 |
| corvus | i'm not | 17:27 |
| *** dklyle has joined #openstack-infra | 17:29 | |
| openstackgerrit | Tobias Henkel proposed openstack-infra/zuul master: Optionally parallelize update threads https://review.openstack.org/635496 | 17:29 |
| *** xek__ has quit IRC | 17:30 | |
| clarkb | rereading really quickly on spf records ?all is a neutral/none policy on everything else not explicitly specified. I don't know enough about how things are done in practice to know if mail servers are treating that as less than good | 17:30 |
| *** xek has joined #openstack-infra | 17:30 | |
| openstackgerrit | Tobias Henkel proposed openstack-infra/zuul master: Optionally parallelize update threads https://review.openstack.org/635496 | 17:30 |
| *** ginopc has quit IRC | 17:31 | |
| corvus | clarkb: if i were to debug this, i would suggest avoid making guesses and instead ask for an actual delivery failure. | 17:31 |
| openstackgerrit | Tobias Henkel proposed openstack-infra/zuul master: Parallelize update threads https://review.openstack.org/635496 | 17:31 |
| *** gfidente is now known as gfidente|afk | 17:36 | |
| *** jamesmcarthur has quit IRC | 17:36 | |
| fungi | yep, i found the pbl issue mentioned in the second-to-last comment by looking in the exim log on ask.o.o | 17:37 |
| *** jamesmcarthur has joined #openstack-infra | 17:37 | |
| *** jamesmcarthur has quit IRC | 17:37 | |
| *** jamesmcarthur has joined #openstack-infra | 17:38 | |
| fungi | i'm guessing if there are other similar sorts of rejections they'll be in the mta log as well. beyond that i have limited interest in speculating on what makes random rfc-violating mass freemail providers less likely or more likely to deliver a legitimate e-mail message | 17:38 |
| corvus | this stuff isn't magic, it's just computers. | 17:38 |
| *** trown is now known as trown|lunch | 17:38 | |
| *** markvoelker has joined #openstack-infra | 17:38 | |
| zbr | clarkb: to my surprise fixing the bindep issue was extremely easy: https://review.openstack.org/#/c/635576/ | 17:39 |
| *** wolverineav has joined #openstack-infra | 17:39 | |
| openstackgerrit | Mohammed Naser proposed openstack-infra/nodepool master: docker: don't daemonize when starting images https://review.openstack.org/635584 | 17:39 |
| clarkb | zbr: you may want to add a test case that uses a / in the name to avoid unintended future regressions | 17:41 |
| zbr | clarkb: you read my mind. | 17:41 |
| *** wolverineav has quit IRC | 17:42 | |
| *** wolverineav has joined #openstack-infra | 17:43 | |
| openstackgerrit | James E. Blair proposed openstack-infra/system-config master: Tell ansible to use py3 on insecure-ci-registry01 https://review.openstack.org/635587 | 17:45 |
| corvus | clarkb, fungi: apparently *that's* what that ssh error really means | 17:45 |
| clarkb | huh | 17:45 |
| corvus | (diagnosed by manually running ansible and throwing lots of "v"s at it) | 17:45 |
| *** priteau has quit IRC | 17:45 | |
| fungi | huh | 17:46 |
| fungi | wow | 17:46 |
| corvus | i wonder why we don't see that in tests? | 17:47 |
| clarkb | we have python2 installed on our test nodes is my guess | 17:48 |
| corvus | doh. | 17:48 |
| fungi | yep, that would be why | 17:48 |
| *** slaweq has joined #openstack-infra | 17:50 | |
| *** ykarel|away has quit IRC | 17:52 | |
| clarkb | re mail delivery failures we should be able to get the logs for that if we had the destination IP or the recipient could provide them from their end right? | 17:58 |
| corvus | clarkb: even an email address and approximate time would be a start | 17:58 |
| clarkb | cool | 17:58 |
| *** jpena is now known as jpena|off | 17:59 | |
| *** kopecmartin is now known as kopecmartin|off | 18:00 | |
| fungi | yeah. in the pbl case earlier this week i simply skimmed the exim log since the rejections were numerous and jumped out at me rather quickly | 18:00 |
| clarkb | I've posted a comment basically saying we need that additional info and synchronous debugging on irc may help so that when we have a followup question we aren't waiting hours/days for a response | 18:01 |
| fungi | thanks! | 18:01 |
| openstackgerrit | Sorin Sbarnea proposed openstack-infra/bindep master: Allow mentioning tools paths in bindep https://review.openstack.org/635576 | 18:01 |
| fungi | corvus: in related news, http://lists.openstack.org/pipermail/openstack-discuss/2019-February/002499.html is the first message i've seen to openstack-discuss with a dkim validation failure since our changes went into effect. going to dig into it in a moment and see if i can find the cause | 18:02 |
| corvus | fungi: thanks | 18:02 |
| *** derekh has quit IRC | 18:03 | |
| fungi | whatever is causing kfox111's messages to be invalid is apparently not related to deduplication in mailman, at least | 18:03 |
| corvus | mordred: i think we need to rethink how we run the k8s-on-openstack playbook; "TASK [Run kubernetes deploy playbook]" is all one line in the run_all log file -- and it failed. somewhere. | 18:05 |
| corvus | also, did we figure out how to install kubectl on bridge? | 18:06 |
| *** wolverineav has quit IRC | 18:06 | |
| *** jesusaur has quit IRC | 18:07 | |
| *** wolverineav has joined #openstack-infra | 18:07 | |
| *** wolverineav has quit IRC | 18:09 | |
| *** wolverineav has joined #openstack-infra | 18:09 | |
| clarkb | ianw: fungi as a heads up I've put the letsencrypt spec on the meeting agenda for tuesday. I think it would be great if we can resurrect that line of discussion so that we can reach a conclusion and start moving in some direction towards that. ianw has rebased the spec. Fungi if you find time (or maybe this can be part of the meeting) it would be good to write down or capture your thoughts on that as | 18:09 |
| clarkb | I know you had some alternative ideas before | 18:09 |
| *** markvoelker has quit IRC | 18:11 | |
| openstackgerrit | Sorin Sbarnea proposed openstack-infra/bindep master: Adds job to test bindep on Fedora 28 https://review.openstack.org/635590 | 18:11 |
| fungi | appreciated! | 18:11 |
| *** ekcs has joined #openstack-infra | 18:13 | |
| *** defionscode has joined #openstack-infra | 18:13 | |
| fungi | ianw and i were coming at it from separate directions, i think. his design was around how to get it into production as quickly as possible, my ideas were more about how to improve our security posture over the way we currently handle private keys while also reducing our technical debt in the long-run | 18:14 |
| *** jtomasek has quit IRC | 18:14 | |
| *** e0ne has joined #openstack-infra | 18:15 | |
| openstackgerrit | James E. Blair proposed openstack-infra/system-config master: Install kubectl on bridge https://review.openstack.org/635593 | 18:15 |
| corvus | fungi, clarkb, mordred: ^ so that's one idea. | 18:15 |
| clarkb | hrm weren't we uninstalling snap tooling? | 18:15 |
| * clarkb checks | 18:15 | |
| corvus | yes! | 18:16 |
| clarkb | ya we do remove snapd | 18:16 |
| corvus | so, er, if we go this way, we should probably remove that | 18:16 |
| clarkb | ya | 18:16 |
| corvus | i'm open to other options though if we want to continue to avoid snap | 18:16 |
| openstackgerrit | Sorin Sbarnea proposed openstack-infra/bindep master: Adds job to test bindep on Fedora https://review.openstack.org/635590 | 18:16 |
| openstackgerrit | James E. Blair proposed openstack-infra/system-config master: Install kubectl on bridge https://review.openstack.org/635593 | 18:17 |
| clarkb | ya their deb repo has no bionic packages, interesting | 18:18 |
| fungi | they likely assume you'll use whatever's packaged directly i bionic? | 18:18 |
| clarkb | https://packages.ubuntu.com/search?keywords=kubectl&searchon=all&suite=bionic§ion=all seems unpackaged? | 18:19 |
| clarkb | a random blogpost says the xenial packages work on bionic, but otherwise ya snap seems like the "correct" method | 18:20 |
| *** e0ne has quit IRC | 18:22 | |
| *** jpich has quit IRC | 18:25 | |
| *** wolverineav has quit IRC | 18:25 | |
| *** wolverineav has joined #openstack-infra | 18:26 | |
| *** wolverineav has quit IRC | 18:26 | |
| *** wolverineav has joined #openstack-infra | 18:26 | |
| *** eharney has joined #openstack-infra | 18:29 | |
| clarkb | corvus: I think it was mordred that wanted to remove snapd so should probably get mordred's input | 18:32 |
| clarkb | I'm not opposed to using snaps if that is where tools are being packaged. They are also self contained so potential future cleanup should be straightforward | 18:33 |
| corvus | i didn't want it on there if we weren't using it either. but i think it's one of those "it's useless until it's not" things. | 18:33 |
| clarkb | corvus: fungi have time for quick review on https://review.openstack.org/#/c/635543/ I think that will clear up pupet 4 things on etherpad-dev01 so that we can continue to make progress on the puppet upgrade | 18:36 |
| zbr | clarkb: corvus : https://review.openstack.org/#/c/635576/ -- bindep fix, now with regression testing. | 18:37 |
| *** e0ne has joined #openstack-infra | 18:40 | |
| *** e0ne has quit IRC | 18:41 | |
| *** e0ne has joined #openstack-infra | 18:42 | |
| *** ociuhandu has joined #openstack-infra | 18:43 | |
| *** wolverineav has quit IRC | 18:45 | |
| *** agopi has joined #openstack-infra | 18:45 | |
| corvus | mordred: another thing we need to figure out -- how to get the secret hostvars to those 3 playbooks. | 18:46 |
| *** ociuhand_ has quit IRC | 18:46 | |
| corvus | this is another case where it would make so much more sense for the k8s cluster to be in the inventory, then we'd attach them as hostvars | 18:47 |
| *** wolverineav has joined #openstack-infra | 18:47 | |
| *** ociuhandu has quit IRC | 18:48 | |
| *** slaweq has quit IRC | 18:49 | |
| *** jamesmcarthur has quit IRC | 18:51 | |
| *** wolverineav has quit IRC | 18:52 | |
| *** jesusaur has joined #openstack-infra | 18:52 | |
| *** wolverineav has joined #openstack-infra | 18:54 | |
| guilhermesp | clarkb: I already have a failure, would be possible to add my keys ? https://github.com/guilhermesteinmuller.keys | 18:54 |
| openstackgerrit | James E. Blair proposed openstack-infra/system-config master: Add gitea-cluster extra vars https://review.openstack.org/635599 | 18:55 |
| corvus | clarkb, fungi, mordred: ^ that's my idea for now | 18:56 |
| *** ociuhandu has joined #openstack-infra | 18:56 | |
| corvus | oh neat, the ansible snap module is unreleased | 18:57 |
| *** sdake has quit IRC | 18:58 | |
| openstackgerrit | James E. Blair proposed openstack-infra/system-config master: Install kubectl on bridge https://review.openstack.org/635593 | 18:59 |
| openstackgerrit | James E. Blair proposed openstack-infra/system-config master: Add gitea-cluster extra vars https://review.openstack.org/635599 | 18:59 |
| corvus | it arrives in 2.8 | 18:59 |
| fungi | corvus: so far i've confirmed that dkimpy says the version of that message recorded by exim has a valid signature and the version saved from my mta does not. after much diffing, it appears that the second and third item in the references list from that message are only separated by "," in the original but are separated by ", " when i receive my copy. removing the extra " " from what i received causes | 19:00 |
| fungi | the signature to verify | 19:00 |
| *** ociuhandu has quit IRC | 19:00 | |
| fungi | so my best guess is that mailman is helpfully correcting the format of the references list to add a space where there was not one originally | 19:01 |
| fungi | i'm going to go take a brief walk since we've been gifted with truly pleasant weather this week (70f and sunny with a light breeze), but will return shortly | 19:03 |
| *** ociuhandu has joined #openstack-infra | 19:03 | |
| clarkb | wow, we finally have witner here | 19:03 |
| guilhermesp | clarkb: and to be helpful, the job is openstack-ansible-upgrade-aio_lxc-ubuntu-bionic on #627782 | 19:04 |
| *** remi_ness has joined #openstack-infra | 19:04 | |
| clarkb | guilhermesp: ok give me a few. I've moved to my desktop and need to load my keys for that | 19:05 |
| guilhermesp | clarkb: np, thanks! | 19:06 |
| *** ociuhandu has quit IRC | 19:07 | |
| *** markvoelker has joined #openstack-infra | 19:08 | |
| corvus | fungi: interestingly, in the rfc, references is not described as comma-separated at all, only separated by foldable-whitespace. | 19:09 |
| corvus | so it's correcting from technically-wrong to different-technically-wrong | 19:10 |
| clarkb | guilhermesp: ssh root@23.253.218.253 | 19:11 |
| openstackgerrit | Merged openstack-infra/system-config master: Tell ansible to use py3 on insecure-ci-registry01 https://review.openstack.org/635587 | 19:12 |
| guilhermesp | nice, I'm in. Thank you clarkb ! | 19:13 |
| *** trown|lunch is now known as trown | 19:15 | |
| mordred | corvus: reading scrollback | 19:29 |
| mordred | corvus: wow. that's how one installs kubectl? | 19:29 |
| openstackgerrit | Merged openstack-infra/zuul master: Update git connection logging https://review.openstack.org/635204 | 19:29 |
| clarkb | mordred: or run xenial or use xenial packages on bionic | 19:31 |
| mordred | ah - yeah. k8s-on-openstack installs from packages, but is using the xenial packages because there are no bionic packages for k8s yet | 19:32 |
| aspiers | Where's the mechanism for auto-updating blueprint whiteboards when Gerrit reviews are submitted? | 19:32 |
| *** wolverineav has quit IRC | 19:34 | |
| *** wolverineav has joined #openstack-infra | 19:35 | |
| *** roman_g has quit IRC | 19:37 | |
| mordred | corvus: re: k8s-on-openstack - I'm kind of tempted to reorganize how we're doing that into consuming the roles and making our own playbook | 19:39 |
| mordred | corvus: since we also have other ansible we run on those nodes, and it's starting to feel a little rube goldberg | 19:40 |
| *** jesusaur has quit IRC | 19:40 | |
| *** lpetrut has quit IRC | 19:40 | |
| *** jesusaur has joined #openstack-infra | 19:41 | |
| *** markvoelker has quit IRC | 19:42 | |
| aspiers | Never mind, melwitt pointed me to it | 19:43 |
| mordred | aspiers: it's where melwitt pointed you to | 19:47 |
| *** sdake has joined #openstack-infra | 19:47 | |
| * mordred is so super helpful | 19:47 | |
| * aspiers falls over himself in gratitude :) | 19:47 | |
| *** sdake has quit IRC | 19:49 | |
| clarkb | mriedem: as a followup to the e-r things it looks happy again after I restarted services yesterday. However we still have a too new index from the future. I'm going to take al ook at that now | 19:49 |
| mriedem | clarkb: yeah i noticed this morning, probably 5am your time - wasn't sure what to make of the future date but was happy with the graphs populating again | 19:50 |
| *** sdake has joined #openstack-infra | 19:51 | |
| clarkb | looks like we have indexes from 2106 and november of this year | 19:51 |
| clarkb | they are relatively tiny compared to the expected indexes. I think I'll just delete them since the data is confusing | 19:51 |
| clarkb | but first let me see if we can identify how they got in there | 19:52 |
| openstackgerrit | Mohammed Naser proposed openstack-infra/zuul master: docker: add state folder https://review.openstack.org/635618 | 19:52 |
| *** sdake has quit IRC | 19:53 | |
| *** sdake has joined #openstack-infra | 19:54 | |
| openstackgerrit | Mohammed Naser proposed openstack-infra/zuul master: docker: start process in foreground https://review.openstack.org/635619 | 19:55 |
| clarkb | mriedem: http://logs.openstack.org/91/628691/17/check/neutron-fullstack/46d2a72/job-output.txt.gz#_2106-02-07_06_29_26_855112 it legit thought it was in the future | 19:57 |
| mriedem | ha wtf | 19:58 |
| clarkb | the good news is that implies it isn't a logstash bug | 19:58 |
| clarkb | the bad news is why | 19:58 |
| clarkb | perhaps a stray ntp server? | 19:58 |
| mriedem | so we've been watching "dark" on netflix | 19:58 |
| mriedem | and i think that might help explain this | 19:58 |
| mriedem | if we had some german friends around | 19:58 |
| clarkb | mriedem: I don't want to live in a world where that show explains things | 19:58 |
| clarkb | I might never go outside again | 19:59 |
| clarkb | in any case I think it is safe to delete these indexes with the confusing data | 19:59 |
| clarkb | if we see this happen more we can investigate potentially bad ntp servers and so on | 19:59 |
| mriedem | neutron-fullstack POST_FAILURE in 5m 41s | 19:59 |
| clarkb | I wonder if dns fails there due to dnssec | 19:59 |
| mriedem | so it posted logs | 19:59 |
| mriedem | but still failed the post | 19:59 |
| mriedem | http://logs.openstack.org/91/628691/17/check/neutron-fullstack/46d2a72/ara-report/result/f994f6c6-c827-48c6-b7f4-31feed3b8148/ | 20:00 |
| clarkb | it had problems with journal collection | 20:00 |
| mriedem | yeah | 20:00 |
| mriedem | /bin/sh: 1: cannot create /home/zuul/workspace/logs/journal.log: Directory nonexistent | 20:00 |
| *** wolverineav has quit IRC | 20:01 | |
| *** wolverineav has joined #openstack-infra | 20:01 | |
| clarkb | #status log Cleaned up Elasticsearch indexes from the future. One was from the year 2106 (job logs actually had those timestamps) and others from November 2019. Total data was a few megabytes. | 20:02 |
| openstackstatus | clarkb: finished logging | 20:02 |
| clarkb | mriedem: ^ that should clean up the remaining confusion in the e-r data | 20:02 |
| *** wolverineav has quit IRC | 20:06 | |
| openstackgerrit | Mohammed Naser proposed openstack-infra/zuul master: docker: start process in foreground https://review.openstack.org/635619 | 20:07 |
| openstackgerrit | Mohammed Naser proposed openstack-infra/zuul master: docker: add state folder https://review.openstack.org/635618 | 20:07 |
| *** remi_ness has quit IRC | 20:07 | |
| openstackgerrit | Adam Spiers proposed openstack-infra/project-config master: Announce changes to governance-{sigs,website} on IRC https://review.openstack.org/635625 | 20:08 |
| mriedem | clarkb: cool thanks | 20:08 |
| *** ijw has joined #openstack-infra | 20:09 | |
| aspiers | mordred: since you were so helpful last time, perhaps you can continue the winning streak ;-) by advising whether the os-traits entry in gerrit/projects.yaml should have "groups: - nova"? | 20:11 |
| aspiers | and if so, melwitt just suggested a list of other projects which should get similar treatment | 20:11 |
| mordred | aspiers: tell me what an os-traits is? | 20:11 |
| aspiers | mordred: https://docs.openstack.org/os-traits/latest/ | 20:12 |
| aspiers | and https://git.openstack.org/cgit/openstack/os-traits/tree/ | 20:12 |
| aspiers | it's the registry for built-in traits used by nova etc. | 20:12 |
| mordred | hrm | 20:13 |
| clarkb | the groups are mappings onto the bug trackers iirc. The behavior is slightly different between storyboard and launchpad but I wouldn't expect it to be part of the nova group as that would cause bugs to be auto associated with nova when they show up in os-traits? | 20:14 |
| aspiers | clarkb: that was kind of the angle which started this conversation in #openstack-nova - I noticed that submitting reviews to os-traits didn't update the corresponding nova blueprint | 20:15 |
| aspiers | melwitt may have a view on whether auto-association of bugs is one bridge too far though | 20:16 |
| clarkb | aspiers: the assumption there is that every bug/blueprint in os-traits is also a nova bug/blueprint I think | 20:16 |
| clarkb | whcih may be correct | 20:16 |
| aspiers | yeah ... or not, I don't know | 20:17 |
| aspiers | however, the same question applies for python-novaclient which is also not associated with the nova group | 20:17 |
| melwitt | yeah, I'm trying to think... whether os-traits bugs are tracked as nova bugs. probably, but it doesn't really come up because os-traits is basically just a library full of constants. bugs haven't really happened yet and might never | 20:17 |
| aspiers | and os-{vif,resource-classes} as melwitt already spotted | 20:18 |
| melwitt | I'm trying to understand if, in general, if it would hurt anything to have nova as a group for those projects. because IIUC, if a bug is opened under os-traits, it would still be updated when referenced in the commit, or are you saying it wouldn't if we added the nova group? | 20:18 |
| clarkb | melwitt: it would still be updated but I think nova might be added to the bug | 20:19 |
| melwitt | that is, would the nova group overwrite the fact that it's os-traits, or would it act as a list of both: os-traits, nova | 20:19 |
| melwitt | oh, I see. ok | 20:19 |
| melwitt | yeah, so I think that would be weird for os-vif, if someone opened a bug in os-vif and uploaded a patch, and then it adds nova to the bug | 20:20 |
| melwitt | it only works cleanly if the project is using the bug tracker under the nova project. which I think os-traits is, but I'd have to double check | 20:21 |
| aspiers | https://launchpad.net/~os-traits-bugs | 20:21 |
| melwitt | a-ha, thanks | 20:22 |
| aspiers | "Nova Bug Team" is listed as a member but I'm not sure if that's the same thing | 20:22 |
| fungi | if memory serves (i haven't double-checked), python-novaclient has its own lp project as well | 20:22 |
| aspiers | fungi: bingo! https://bugs.launchpad.net/python-novaclient | 20:22 |
| *** wolverineav has joined #openstack-infra | 20:22 | |
| melwitt | it was os-resource-classes that is sharing the nova bug tracker | 20:23 |
| fungi | basically that "groups" parameter in the yaml used to be called something like lp-project, and in the case of launchpad-using projects it's intended to provide a mapping of a gerrit project to a launchpad project | 20:23 |
| melwitt | yeah, novaclient does. I was only thinking from the update-blueprint angle. it would mess things up for the bugs angle | 20:23 |
| fungi | we renamed it to "groups" when we overloaded the setting to alternatively specify project group membership for storyboard-using projects | 20:24 |
| aspiers | yeah, if those two updating mechanisms are inseparable then that changes things somewhat | 20:24 |
| melwitt | aye | 20:24 |
| fungi | in hindsight, we probably should have just left the old parameter name alone and used a different one for sb project group configuration | 20:24 |
| fungi | but regardless, the point of the setting from the perspective of lp is to say "the name of this project is something different in launchpad" | 20:25 |
| *** jamesmcarthur has joined #openstack-infra | 20:25 | |
| fungi | and was then also (ab)used to handle situations where multiple gerrit projects shared a single lp project name | 20:25 |
| aspiers | OK, well it's a minor thing so maybe we don't need to change anything right now | 20:26 |
| aspiers | good to have put it under the microscope though :) | 20:26 |
| *** cyberpear has quit IRC | 20:26 | |
| aspiers | things can be added to the whiteboard manually anyway, right? | 20:26 |
| fungi | as far as i know, yes | 20:26 |
| aspiers | a very hurried glance at the code suggested that it wouldn't overwrite the whole field | 20:26 |
| fungi | that may require specific lp permissions from the project owner/drivers group | 20:27 |
| *** wolverineav has quit IRC | 20:27 | |
| melwitt | yeah, since the gerrit topic gets posted to the blueprint (as long as something in nova has been uploaded, usually the spec), one can find all related patches by gerrit topic | 20:27 |
| *** jamesmcarthur_ has joined #openstack-infra | 20:27 | |
| fungi | but i really don't know much about the lp bp implementation tbh | 20:27 |
| aspiers | fungi: in this case I submitted the bp so I was allowed to do it | 20:27 |
| aspiers | at least that's the reason I'm guessing | 20:27 |
| fungi | very well could be | 20:27 |
| aspiers | alright, well thanks for all the advice folks - I'm off to cook dinner! | 20:28 |
| fungi | any time! | 20:28 |
| *** ccamacho has quit IRC | 20:28 | |
| aspiers | o/ | 20:28 |
| *** sshnaidm is now known as sshnaidm|off | 20:29 | |
| *** jamesmcarthur has quit IRC | 20:30 | |
| *** sdake has quit IRC | 20:31 | |
| *** sdake has joined #openstack-infra | 20:32 | |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: Remove default user for fingergw https://review.openstack.org/635632 | 20:37 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: Remove default zookeeper hosts https://review.openstack.org/635633 | 20:37 |
| *** markvoelker has joined #openstack-infra | 20:39 | |
| *** ociuhandu has joined #openstack-infra | 20:42 | |
| *** priteau has joined #openstack-infra | 20:46 | |
| openstackgerrit | Mohammed Naser proposed openstack-infra/zuul master: doc: clarify sshkey option usage in github connection https://review.openstack.org/635640 | 20:47 |
| *** ociuhandu has quit IRC | 20:51 | |
| *** wolverineav has joined #openstack-infra | 20:53 | |
| *** priteau has quit IRC | 20:55 | |
| *** wolverineav has quit IRC | 20:58 | |
| *** mkarray has joined #openstack-infra | 21:02 | |
| fungi | corvus: okay, digging deeper on remaining dkim validation failures, we got two more from kfox111 and in every case the penultimate and final item in the references header are separated by a "," so those are at least a consistent cause. i also see failures from two different dell addresses which i'll look into next. also two messages from different gmail addresses fail to validate, that will be | 21:03 |
| fungi | interesting to spot since ~95% of the messages i see from gmail are successfully validating after we removed the deduplication pipeline | 21:03 |
| fungi | also one (perhaps unsurprising) side-effect of removing the deduplication pipeline entry is that if someone sends a message to both the old openstack-dev alias and the openstack-discuss address, the ml forwards both copies to subscribers | 21:04 |
| *** mkarray has quit IRC | 21:04 | |
| corvus | fungi: i looked *very* briefly at the mm source code and didn't see any smoking guns wrt references headers and so suspect that round-trips through the email module may be at play. but i'm not remotely confident in that assessment. | 21:06 |
| fungi | thanks for checking | 21:06 |
| *** mkarray has joined #openstack-infra | 21:07 | |
| *** wolverineav has joined #openstack-infra | 21:07 | |
| *** wolverineav has quit IRC | 21:09 | |
| *** markvoelker has quit IRC | 21:11 | |
| *** priteau has joined #openstack-infra | 21:12 | |
| fungi | corvus: good news on the two gmail posts and one of the dell posts... i dump several mailing lists into one inbox and got tunnel vision so didn't notice they were for other lists which we know break dkim signatures, so i just have one dell post to try and identify the cause on; though i'm going to have to resort to tools again because my eye isn't picking out the differences | 21:15 |
| *** e0ne has quit IRC | 21:19 | |
| *** kgiusti has left #openstack-infra | 21:20 | |
| *** xek has quit IRC | 21:22 | |
| *** wolverineav has joined #openstack-infra | 21:24 | |
| openstackgerrit | Tobias Henkel proposed openstack-infra/zuul master: Add foreground option https://review.openstack.org/635649 | 21:25 |
| openstackgerrit | Matthieu Huin proposed openstack-infra/zuul master: [WIP] web: add tenant and project scoped, JWT-protected actions https://review.openstack.org/576907 | 21:26 |
| *** sdake has quit IRC | 21:29 | |
| *** ekultails has quit IRC | 21:29 | |
| corvus | mordred: yeah, reworking that all into a new playbook sounds good | 21:30 |
| *** jamesmcarthur_ has quit IRC | 21:30 | |
| *** jamesmcarthur has joined #openstack-infra | 21:30 | |
| openstackgerrit | James E. Blair proposed openstack-infra/system-config master: Correct insecure-ci-registry01 hostvar filename https://review.openstack.org/635652 | 21:33 |
| *** jamesmcarthur has quit IRC | 21:33 | |
| corvus | fungi, clarkb: ^ brown paper patch. | 21:33 |
| openstackgerrit | Tobias Henkel proposed openstack-infra/nodepool master: docker: don't daemonize when starting images https://review.openstack.org/635584 | 21:33 |
| *** jamesmcarthur has joined #openstack-infra | 21:33 | |
| *** bassemessam has quit IRC | 21:36 | |
| clarkb | corvus: we are probably going to get really good at making that distinction in the future :) | 21:36 |
| openstackgerrit | Tobias Henkel proposed openstack-infra/nodepool master: docker: don't daemonize when starting images https://review.openstack.org/635584 | 21:36 |
| corvus | clarkb: at least one of us is not there yet :) | 21:37 |
| *** sdake has joined #openstack-infra | 21:37 | |
| fungi | corvus: wow! i'm increasingly disheartened by the moment over dkim inconsistencies... http://lists.openstack.org/pipermail/openstack-discuss/2019-February/002470.html fails validation after passing through mailman because the original (as recorded by exim) inserts a line break followed by a hard tab between the quoted name and address parts of the to header. the copy i received has that hard tab | 21:38 |
| fungi | rewritten to a single space instead. that alone is enough to invalidate the dkim signature, but that's clearly not the only difference breaking the signature because just repairing that in the later version doesn't get it to validate successfully | 21:38 |
| corvus | fungi: does the original validate? | 21:40 |
| fungi | yep | 21:40 |
| *** priteau has quit IRC | 21:40 | |
| fungi | the message also has two DKIM-Signature headers, one with its folded lines indented by only spaces and the other with its folded lines indented by a mix of hard tabs and hard tabs followed by spaces | 21:40 |
| fungi | so presumably passing through different mtas both of which are signing the message near the point of origin | 21:41 |
| corvus | i'm of half a mind to make exim run incoming messages through the python email module and if the result doesn't validate dkim, reject it. | 21:42 |
| *** jamesmcarthur has quit IRC | 21:42 | |
| fungi | that seems reasonable, though this message would have made it through | 21:43 |
| fungi | at least the copy exim saved validates | 21:43 |
| corvus | fungi: i mean, run it through whatever transformations mailman is doing | 21:43 |
| fungi | ahh, yeah that could work | 21:43 |
| *** gfidente|afk has quit IRC | 21:43 | |
| clarkb | do we expect users to beable to address that locally beyond "get a new email account"? | 21:44 |
| corvus | it at least puts the onus on the operators of questionable dkim-signing systems | 21:44 |
| fungi | i suspect mailman thinks normalizing whitespace in headers is a "good idea[tm]" (and it might be, but dkim signatures don't seem to treat all whitespace characters equally it seems) | 21:44 |
| clarkb | (I just worry that for many that might not be something they can fix on their end) | 21:44 |
| corvus | fungi: it's such a well-thought out system isn't it? | 21:44 |
| fungi | clarkb: it's something they can fix on their end by using a different address somewhere else | 21:44 |
| clarkb | fungi: right, but haven't we heard that some people are not allowed to do that? | 21:45 |
| fungi | corvus: i mean, i never really thought it was well-thought-out but this exercise is reminding me how loosely rfc 822 has been treated over the years | 21:45 |
| clarkb | it is unfortunate that dkim isn't more forgiving, though I suppose the whole point is to avoid ambiguity by signing the actual email and fuzzing whitespace could introduce ambiguity? | 21:46 |
| clarkb | ugh | 21:46 |
| fungi | clarkb: we've heard that some people aren't allowed to use personal e-mail addresses when corresponding with our mailing lists, but we can at least set a policy that we won't accept messages which we can't safely forward | 21:46 |
| *** jamesmcarthur has joined #openstack-infra | 21:47 | |
| *** priteau has joined #openstack-infra | 21:48 | |
| *** jcoufal has quit IRC | 21:48 | |
| *** priteau has quit IRC | 21:49 | |
| clarkb | fungi: right wouldn't that potentially mean we would stop accepting messages from those individuals (if their mail servers and mailman don't agree on formatting) then they wouldn't be able to use a new account elsewhere where format is saner? | 21:49 |
| corvus | it's worth noting that it's the originating system that sets the dkim policy, including what headers it signs, and formats the (presumably nonconforming) message, and, in my little strawman proposal, would be receiving the rejection. | 21:49 |
| corvus | clarkb: correcting the sending system is an option. | 21:49 |
| *** jamesmcarthur has quit IRC | 21:51 | |
| *** sdake has quit IRC | 21:51 | |
| clarkb | and mailman is formatting the the rfc right? | 21:51 |
| corvus | unclear | 21:51 |
| fungi | in this case it's quite likely that dell would need to convince cisco to alter the behavior of the software in their ironport appliances to take care of this (or switch vendors, or run their own systems), but it *is* still an option of a sort | 21:52 |
| corvus | clarkb: this is why i'm only of half a mind to do this. i'd want to have a lot more data first. | 21:53 |
| corvus | i'm just brainstorming. | 21:53 |
| sigmavirus24 | mordred: tobiash well if you need anything from github3.py, feel free to invite me here or pop into #github3.py | 21:53 |
| *** sigmavirus24 has left #openstack-infra | 21:53 | |
| *** jamesmcarthur has joined #openstack-infra | 21:54 | |
| clarkb | corvus: rgr | 21:55 |
| *** sdake has joined #openstack-infra | 21:55 | |
| corvus | clarkb, fungi: back to fun things -- http://logs.openstack.org/93/635593/3/check/system-config-run-base/7504cb2/ara-report/result/bbf8d5e5-15cc-4ec4-b0ca-c71efb66b774/ | 21:55 |
| corvus | let's assume that kubectl really was installed... is there some path thing we need to think about? | 21:56 |
| *** sdake has quit IRC | 21:56 | |
| corvus | it should be installed in /snap/bin/kubectl | 21:56 |
| corvus | which is in root's path on a random host i'm on | 21:56 |
| * clarkb checks a random host too | 21:57 | |
| corvus | and that ansible command that testinfra runs should be opening a new shell with the path, even if the profile was only recently updated by the reinstallation of snapd, right? | 21:57 |
| clarkb | I have /snap/bin in my path on random machine too | 21:57 |
| dtroyer | Zuul/yaml question: I am building a set of plugins for DevStack that have dependencies with each other so order matters, it works locally with enabled_service calls in local.conf. It appears that they are initialized in alpha sort order in the devstack Zuul jobs which is a different order from what I have in devstack_plugins in .zuul.yaml. Can this order be controlled outside of renaming things? | 21:58 |
| *** sdake has joined #openstack-infra | 21:58 | |
| fungi | corvus: clarkb: looking on bridge.o.o, /snap/bin is last in its path so we don't have to worry about snaps shadowing more legitimate system executables at least | 21:58 |
| *** weshay has quit IRC | 21:58 | |
| corvus | dtroyer: yes, 1 sec | 21:58 |
| clarkb | /etc/profile.d/apps-bin-path.sh is where that comes from on my ubuntu host | 21:59 |
| corvus | dtroyer: see "plugin_requires" on https://docs.openstack.org/devstack/latest/plugins.html | 21:59 |
| dtroyer | Zuul uses that? nice | 22:00 |
| clarkb | corvus: fungi is profile the bit thato nly runs on login? | 22:00 |
| corvus | dtroyer: the devstack roles use that, yes | 22:00 |
| dtroyer | cool, I didn't know anything had implemented it, thanks | 22:00 |
| clarkb | or is rc only on login? I always have to read manpages when it gets to that level of distinction, but maybe if it only runs at login the we need to "login" again for the path to be loaded after snapd is installed | 22:00 |
| corvus | dtroyer: yeah, i don't think anyones gotten around to doing that in devstack itself yet. | 22:00 |
| clarkb | ya profile appears to be the login loaded stuff | 22:01 |
| corvus | clarkb: shouldn't that ansible module invocation be a "login"? | 22:01 |
| clarkb | corvus: ya, but if we run ansible to install snapd then it won't be there when we login (potentially) | 22:01 |
| corvus | (even if it's using a controlpersist socket, that's just maintaining the tcp connection, not the ssh session) | 22:01 |
| clarkb | corvus: then if we are in the same login for testinfra we'll have the older path | 22:01 |
| corvus | well, i think that's what i'm saying -- isn't it a new login for each task? | 22:02 |
| clarkb | oh that I don't know | 22:02 |
| corvus | maybe i should hold this job | 22:02 |
| clarkb | corvus: if we don't want to think too hard on it you could use a fully qualified paht to the executable | 22:03 |
| openstackgerrit | sebastian marcet proposed openstack-infra/puppet-openstackid master: Removed nodejs and added nodejs and npm as main packages https://review.openstack.org/635666 | 22:03 |
| tobiash | corvus: zuul enables pipelining so there is not neccessarily a new login per task | 22:03 |
| corvus | clarkb: that would make me sad | 22:03 |
| corvus | tobiash: ooooh | 22:04 |
| tobiash | So you would need to reset the connection to enforce a new login | 22:04 |
| corvus | got it | 22:05 |
| clarkb | corvus: another option is to make the task sh --login -c kubectl | 22:05 |
| clarkb | and force a new login that way | 22:05 |
| *** markvoelker has joined #openstack-infra | 22:08 | |
| clarkb | fungi: bah good catch on ubuntu distro release names :P | 22:11 |
| *** jamesmcarthur has quit IRC | 22:12 | |
| * clarkb steps out for a bit, back to email and reviews after | 22:14 | |
| *** jamesmcarthur has joined #openstack-infra | 22:15 | |
| openstackgerrit | Mohammed Naser proposed openstack-infra/zuul master: doc: fix sqlalchemy database url docs path https://review.openstack.org/635670 | 22:17 |
| *** jamesmcarthur has quit IRC | 22:20 | |
| *** jamesmcarthur has joined #openstack-infra | 22:21 | |
| openstackgerrit | James E. Blair proposed openstack-infra/system-config master: Install kubectl on bridge https://review.openstack.org/635593 | 22:25 |
| *** jamesmcarthur has quit IRC | 22:25 | |
| *** jamesmcarthur has joined #openstack-infra | 22:27 | |
| *** sthussey has quit IRC | 22:29 | |
| fungi | clarkb: sure, just didn't want the potential newcomer to think we were *that* far behind the curve and run away screaming | 22:33 |
| clarkb | fungi: ya it was a good catch. | 22:34 |
| *** mkarray has quit IRC | 22:34 | |
| *** sdake has quit IRC | 22:34 | |
| fungi | they may well still run away screaming for other reasons, but at least it won't be that one ;) | 22:35 |
| corvus | i had been trying to find the right way to say that :) | 22:35 |
| clarkb | fungi: any idea what the message size limit is on the openstack-infra list? I'm guessing a 1MB message might get trapped for moderation (have a pdf attachment) | 22:38 |
| *** markvoelker has quit IRC | 22:42 | |
| *** agopi is now known as agopi|pto | 22:42 | |
| *** mriedem is now known as mriedem_afk | 22:43 | |
| fungi | looking now | 22:45 |
| fungi | Maximum length in kilobytes (KB) of a message body: 40 | 22:47 |
| fungi | so, er, yeah | 22:47 |
| clarkb | fungi: do attachments count as message body? | 22:47 |
| clarkb | its mostly the slightly less than 1MB attachment | 22:47 |
| clarkb | I guess I can stick that on tarballs.o.o instead | 22:47 |
| clarkb | maybe I'll just do that an avoid troule with mailman | 22:47 |
| fungi | i think everything following the message headers is considered the message body | 22:48 |
| clarkb | ok I'll just stick it on tarballs then | 22:48 |
| *** agopi|pto has quit IRC | 22:48 | |
| fungi | you can always just approve it through moderation and then let people who care you sent them 1mb exploded into however much mime makes that blame you personally ;) | 22:48 |
| clarkb | my mua knows how to not make that look terrible :P | 22:49 |
| clarkb | I'm guessing mutt has a sad though | 22:49 |
| fungi | meh, it just shows me that there's a binary attachment and then i can hit v to view the attachments list, highlight the part i want and hit s to save it to a local file | 22:50 |
| corvus | pretty sure all the muas support mime now :) | 22:50 |
| clarkb | ah | 22:51 |
| fungi | granted, that file is "local" to the shell server where i run my mua, so i then scp it to my workstation where i have a graphical pdf viewer (though i do also have a pdf2txt filter in mutt for mostly-text pdfs) | 22:51 |
| *** ociuhandu has joined #openstack-infra | 22:52 | |
| *** tosky has quit IRC | 22:52 | |
| clarkb | well now its a nice browseable link instaed. Mail sent too | 22:52 |
| fungi | corvus: i finished dissecting the dell post. verdict is that i have to undo the normlization of indent whitespace for the wrapped to header as well as both the wrapped dkim-signature headers | 22:52 |
| fungi | after that, it validates | 22:52 |
| *** hwoarang has quit IRC | 22:54 | |
| *** tkajinam has joined #openstack-infra | 22:55 | |
| *** ociuhandu has quit IRC | 22:56 | |
| fungi | interestingly, the dkim-signature header which is only wrapped with (varying amounts of) spaces has to be put back to its original wonky layout to work, not just the one with mixed spaces and tabs | 22:56 |
| fungi | i can't begin to speculate why that matters | 22:56 |
| fungi | aha, specifically it's the wrapping of the continuation for the h= list | 22:57 |
| fungi | indenting that by one or two spaces won't cut it, has to be three. i guess the hash covers that subfield and is embedding some of the whitespace | 22:58 |
| *** hwoarang has joined #openstack-infra | 22:59 | |
| *** Emine has quit IRC | 22:59 | |
| *** bobh has quit IRC | 22:59 | |
| *** panda is now known as panda|off | 22:59 | |
| *** jtomasek has joined #openstack-infra | 23:00 | |
| *** jamesmcarthur has quit IRC | 23:01 | |
| *** whoami-rajat has quit IRC | 23:03 | |
| *** wolverineav has quit IRC | 23:04 | |
| clarkb | three space tabs! the horror :) | 23:05 |
| clarkb | is that header included in the list of dkim signed headers? iirc they are all explicitly listed | 23:05 |
| *** rascasoft has quit IRC | 23:08 | |
| *** jamesmcarthur has joined #openstack-infra | 23:15 | |
| *** jtomasek has quit IRC | 23:16 | |
| *** takamatsu has quit IRC | 23:19 | |
| *** jamesmcarthur has quit IRC | 23:20 | |
| *** smarcet has joined #openstack-infra | 23:20 | |
| smarcet | fungi: please review https://review.openstack.org/#/c/635666/ when u get a chance , seems that problem is that nodejs puppet module is not installing npm and i dont see any option at the module for it, so i opt to install by hand nodejs and npm and declare those 2 on the main_packages list cheers | 23:20 |
| corvus | fungi, clarkb, mordred: i've got a first cut at rewrite rules for cgit at http://git-test.openstack.org/.htaccess along with a regression test script at http://git-test.openstack.org/test.py | 23:20 |
| corvus | you should be able to go to any git.o.o url, change the host to git-test.o.o and try it out | 23:21 |
| clarkb | corvus: the comments are quite helpful | 23:21 |
| *** smarcet has quit IRC | 23:21 | |
| clarkb | hrm smarcet is already gone | 23:22 |
| *** sdake has joined #openstack-infra | 23:23 | |
| clarkb | corvus: http://git-test.openstack.org/cgit/openstack/nova/tag/?h=17.0.7 that doesn't seem to be handled? | 23:25 |
| clarkb | but http://git-test.openstack.org/cgit/openstack/nova/commit/?h=17.0.7&id=a48c1123cd0dedf0c0a8978b76bacb1987766255 does work | 23:25 |
| corvus | clarkb: yep, looks like i missed the tag/ endpoint | 23:26 |
| clarkb | other than the recursive escaped /s rule most of these look "sane". i mean its regexes but they are all [L] so easy to evaluate individualy | 23:27 |
| clarkb | not a short list but I expect this is maintainable? | 23:27 |
| corvus | yeah, i tried to make it as comprehensible as possible | 23:29 |
| corvus | clarkb: hrm, i wonder what http://git.openstack.org/cgit/openstack/nova/tag/?h=17.0.9 should redirect to | 23:29 |
| *** wolverineav has joined #openstack-infra | 23:29 | |
| *** eharney has quit IRC | 23:29 | |
| clarkb | cgit renders the actual tag message | 23:30 |
| *** wolverineav has quit IRC | 23:30 | |
| clarkb | which maybe we can't get from that data there in gitea? | 23:30 |
| * clarkb browses gitea | 23:30 | |
| *** wolverineav has joined #openstack-infra | 23:30 | |
| clarkb | corvus: maybe http://38.108.68.64/openstack/nova/src/tag/17.0.9 | 23:31 |
| clarkb | its not quite 1:1 | 23:31 |
| corvus | clarkb: well, if you go to http://38.108.68.64/openstack/nova/src/tag/17.0.9 and click "..." after the thingy, you get the message | 23:31 |
| corvus | yeah | 23:31 |
| clarkb | I think that is probably the closest thing | 23:31 |
| corvus | http://38.108.68.64/openstack/nova/commit/9929a7082d78a19478f942431601728b32c357ee is nice, but requires the sha | 23:31 |
| clarkb | http://38.108.68.64/openstack/nova/commit/9929a7082d78a19478f942431601728b32c357ee is the real thing but we don't have that data | 23:31 |
| clarkb | yup | 23:31 |
| clarkb | it is discoverable from the earlier url though so probably good enough | 23:32 |
| corvus | so seems like src/ is the best thing | 23:32 |
| *** wolverineav has quit IRC | 23:32 | |
| clarkb | fungi do you know why smarcet thinks that npm isn't installed? openstackid-dev.openstack.org shows npm 6.1.0 at /usr/local/bin/npm | 23:33 |
| clarkb | am I looking at the wrong server? | 23:33 |
| clarkb | and the apt source for nodesource is installed | 23:34 |
| fungi | clarkb: it is actually part _of_ the header which contains the dkim signature, just a different subfield of that header | 23:34 |
| clarkb | fungi: huh | 23:34 |
| fungi | i think some data other than the body and select headers (such as the list of headers to cover in the signature) must also be fed into the signature but i haven't read the dkim spec | 23:35 |
| *** jamesmcarthur has joined #openstack-infra | 23:36 | |
| corvus | fungi: yes, the list of headers is signed | 23:36 |
| corvus | (so you can't bypass it) | 23:36 |
| fungi | figures... just interesting that they don't strip all whitespace before calculating | 23:37 |
| *** markvoelker has joined #openstack-infra | 23:39 | |
| fungi | basically, if the h= subfield contains any whitespace or is continued onto a subsequent line, you apparently can't alter the newline/spaces/tabs which appear in it | 23:39 |
| corvus | fungi: the dkim "spec" supports both: https://tools.ietf.org/html/rfc6376#section-3.4.1 | 23:39 |
| corvus | the one that permits no whitespace changes is default | 23:39 |
| * fungi sighs | 23:40 | |
| *** wolverineav has joined #openstack-infra | 23:40 | |
| *** jamesmcarthur has quit IRC | 23:40 | |
| fungi | c=simple/simple; | 23:41 |
| fungi | yep, that's it i guess | 23:41 |
| fungi | so maybe we can suggest to dell that they figure out how to change that from simple to relaxed? | 23:41 |
| corvus | fungi: is gmail using relaxed? | 23:42 |
| fungi | c=relaxed/relaxed; | 23:42 |
| fungi | yep | 23:42 |
| fungi | taken from a gmail.com message which just arrived on the ml | 23:42 |
| fungi | pnnl.gov is also using c=relaxed/relaxed; fwiw, so that wouldn't solve the problem we're seeing with the references header on their messages | 23:43 |
| *** wolverineav has quit IRC | 23:45 | |
| *** wolverineav has joined #openstack-infra | 23:45 | |
| corvus | fungi: yeah, that was ',' to ', ', which doesn't match any of the canonicalization rules. but ',' isn't rfc compliant to start with (', ' may not be either; that's unclear to me) | 23:46 |
| corvus | page 8 says how to set it to relaxed: https://www.cisco.com/c/en/us/td/docs/security/esa/esa11-0/user_guide_fs/b_ESA_Admin_Guide_11_0/b_ESA_Admin_Guide_chapter_010101.pdf | 23:48 |
| fungi | right, it looks like it's taking the existing references from the message he's replying to, and concatenating them with the message-id of the message being replied to, with a "," between them rather than a space | 23:49 |
| fungi | the mta seems to be ms exchange, but the mua might be what's actually writing the noncompliant references header | 23:49 |
| corvus | clarkb: tag support added to regexes | 23:50 |
| fungi | the election officials and the tc have bits of software which fetch individual file contents from git.o.o via the cgit webui, so going to try and find time to verify we have them covered with these rewrite rules | 23:51 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: web: add buildsets page https://review.openstack.org/630041 | 23:52 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: web: add buildset page https://review.openstack.org/630079 | 23:53 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: web: add /{tenant}/buildset/{uuid} route https://review.openstack.org/630078 | 23:53 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: web: add buildset page https://review.openstack.org/630079 | 23:53 |
| clarkb | corvus: seems to work now | 23:53 |
| *** harlowja has quit IRC | 23:54 | |
| *** wolverineav has quit IRC | 23:58 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!