| *** tetsuro has joined #openstack-infra | 00:06 | |
| openstackgerrit | Merged opendev/system-config master: Add initial Ansible for nodepool hosts https://review.opendev.org/710888 | 00:10 |
|---|---|---|
| *** jamesmcarthur has joined #openstack-infra | 00:11 | |
| openstackgerrit | Merged opendev/system-config master: Add install zookeeper role; use for nodepool-builder testing https://review.opendev.org/710908 | 00:12 |
| openstackgerrit | Ian Wienand proposed opendev/system-config master: nodepool-builder: deploy from container https://review.opendev.org/710891 | 00:14 |
| openstackgerrit | Ian Wienand proposed opendev/system-config master: nodepool-builder: Add webserver https://review.opendev.org/711574 | 00:14 |
| openstackgerrit | Ian Wienand proposed opendev/system-config master: Add nb01.opendev.org https://review.opendev.org/693119 | 00:14 |
| *** Goneri has quit IRC | 00:14 | |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: Implement zookeeper-auth https://review.opendev.org/619156 | 00:15 |
| clarkb | ianw: https://review.opendev.org/693119's dependency needs a rebase too. 693119 got a -1 because its dpeends on can't be merged | 00:15 |
| *** matt_kosut has joined #openstack-infra | 00:15 | |
| clarkb | (its gonna be the serial on the zone file) | 00:15 |
| openstackgerrit | Ian Wienand proposed opendev/zone-opendev.org master: Add nb01.opendev.org https://review.opendev.org/693118 | 00:17 |
| *** tetsuro has quit IRC | 00:18 | |
| *** tetsuro has joined #openstack-infra | 00:18 | |
| ianw | clarkb: ^thanks, should get it | 00:18 |
| *** jamesmcarthur has quit IRC | 00:19 | |
| *** matt_kosut has quit IRC | 00:20 | |
| *** jamesmcarthur has joined #openstack-infra | 00:21 | |
| *** tetsuro has quit IRC | 00:21 | |
| *** tetsuro has joined #openstack-infra | 00:22 | |
| *** dklyle has quit IRC | 00:22 | |
| *** jamesmcarthur has quit IRC | 00:26 | |
| clarkb | cool I think I've +2'd the whole stack at this point and we can make changes to mounts and config copies as followons | 00:30 |
| clarkb | (assuming we decide those changes make sense) | 00:31 |
| ianw | thanks! | 00:32 |
| *** diablo_rojo has quit IRC | 00:42 | |
| *** jamesmcarthur has joined #openstack-infra | 00:44 | |
| *** yamamoto has joined #openstack-infra | 00:54 | |
| openstackgerrit | Ian Wienand proposed openstack/diskimage-builder master: Add ensure-venv element, install glean with it https://review.opendev.org/707513 | 00:57 |
| ianw | clarkb: ^ all the boot tests passed, but i still added the mkdir -p you mentioned | 00:58 |
| ianw | if you're ok with that, i think we release dib with it soon. couple of other changes to merge too | 00:58 |
| *** jamesmcarthur has quit IRC | 01:08 | |
| *** jamesmcarthur has joined #openstack-infra | 01:12 | |
| openstackgerrit | Merged openstack/diskimage-builder master: Uncap hacking https://review.opendev.org/709332 | 01:16 |
| openstackgerrit | Merged opendev/glean master: Switch to Fedora 30 jobs https://review.opendev.org/711970 | 01:17 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: Implement zookeeper-auth https://review.opendev.org/619156 | 01:18 |
| *** jamesmcarthur has quit IRC | 01:24 | |
| *** jamesmcarthur has joined #openstack-infra | 01:24 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/project-config master: Rename x/devstack-plugin-nfs to openstack/devstack-plugin-nfs https://review.opendev.org/711834 | 01:24 |
| *** ccamacho has joined #openstack-infra | 01:25 | |
| gmann | fungi: ^^ updated. | 01:26 |
| fungi | lookin' | 01:28 |
| *** hongbin has joined #openstack-infra | 01:31 | |
| *** ricolin_ has joined #openstack-infra | 01:39 | |
| *** rosmaita has left #openstack-infra | 01:40 | |
| *** rkukura has quit IRC | 01:40 | |
| *** armax has quit IRC | 01:43 | |
| *** rlandy|bbl is now known as rlandy | 01:53 | |
| *** yamamoto has quit IRC | 02:01 | |
| *** happyhemant has quit IRC | 02:05 | |
| *** jamesmcarthur has quit IRC | 02:06 | |
| *** jamesmcarthur has joined #openstack-infra | 02:07 | |
| *** rkukura has joined #openstack-infra | 02:07 | |
| *** yamamoto has joined #openstack-infra | 02:07 | |
| *** nicolasbock has joined #openstack-infra | 02:26 | |
| openstackgerrit | Merged openstack/diskimage-builder master: Remove Fedora 29 job https://review.opendev.org/711972 | 02:50 |
| *** yamamoto has quit IRC | 02:52 | |
| *** yamamoto has joined #openstack-infra | 02:54 | |
| *** rkukura has quit IRC | 02:56 | |
| *** rkukura has joined #openstack-infra | 03:01 | |
| *** nicolasbock has quit IRC | 03:06 | |
| *** jamesmcarthur has quit IRC | 03:21 | |
| *** rlandy has quit IRC | 03:36 | |
| *** hongbin has quit IRC | 03:47 | |
| *** lbragstad_ has joined #openstack-infra | 03:54 | |
| *** roman_g has quit IRC | 04:02 | |
| *** rcernin has quit IRC | 04:04 | |
| *** rcernin has joined #openstack-infra | 04:04 | |
| *** matt_kosut has joined #openstack-infra | 04:15 | |
| *** matt_kosut has quit IRC | 04:20 | |
| *** ianychoi has quit IRC | 04:39 | |
| *** ianychoi has joined #openstack-infra | 04:40 | |
| *** dave-mccowan has quit IRC | 04:40 | |
| *** yamamoto has quit IRC | 04:57 | |
| openstackgerrit | Merged openstack/diskimage-builder master: Add ensure-venv element, install glean with it https://review.opendev.org/707513 | 05:08 |
| *** yamamoto has joined #openstack-infra | 05:27 | |
| *** lbragstad_ has quit IRC | 05:34 | |
| *** evrardjp has quit IRC | 05:35 | |
| *** evrardjp has joined #openstack-infra | 05:35 | |
| *** ociuhandu has joined #openstack-infra | 06:07 | |
| *** ijw_ has quit IRC | 06:08 | |
| *** ociuhandu has quit IRC | 06:12 | |
| openstackgerrit | Ian Wienand proposed openstack/project-config master: Move fedora builds to nb01.opendev.org https://review.opendev.org/693120 | 06:12 |
| *** ricolin_ has quit IRC | 06:23 | |
| *** ricolin has joined #openstack-infra | 06:23 | |
| *** eandersson has quit IRC | 06:23 | |
| *** eandersson has joined #openstack-infra | 06:24 | |
| *** ijw has joined #openstack-infra | 06:33 | |
| *** lmiccini has joined #openstack-infra | 06:34 | |
| *** ijw has quit IRC | 06:38 | |
| *** udesale has joined #openstack-infra | 06:41 | |
| *** AJaeger has joined #openstack-infra | 07:03 | |
| *** xinranwang has joined #openstack-infra | 07:08 | |
| AJaeger | gmann, fungi, clarkb, https://review.opendev.org/#/c/711834 (project rename) needs still a gerrit downtime, correct? clarkb, want to bring this up in the team meeting today? | 07:13 |
| AJaeger | config-core, before we can disable fedora-29, we need to update fedora-latest, please review https://review.opendev.org/711969 | 07:14 |
| *** dpawlik has joined #openstack-infra | 07:21 | |
| *** ociuhandu has joined #openstack-infra | 07:28 | |
| *** tetsuro has quit IRC | 07:29 | |
| *** Lucas_Gray has joined #openstack-infra | 07:30 | |
| *** yamamoto has quit IRC | 07:33 | |
| *** ociuhandu has quit IRC | 07:35 | |
| *** ccamacho has quit IRC | 07:35 | |
| *** ccamacho has joined #openstack-infra | 07:35 | |
| *** yamamoto has joined #openstack-infra | 07:35 | |
| *** pgaxatte has joined #openstack-infra | 07:38 | |
| *** tetsuro has joined #openstack-infra | 07:43 | |
| *** Lucas_Gray has quit IRC | 07:45 | |
| *** Lucas_Gray has joined #openstack-infra | 07:46 | |
| *** AJaeger has quit IRC | 07:49 | |
| *** Wryhder has joined #openstack-infra | 07:51 | |
| *** Lucas_Gray has quit IRC | 07:51 | |
| *** Wryhder is now known as Lucas_Gray | 07:52 | |
| *** slaweq has joined #openstack-infra | 07:56 | |
| *** xek_ has joined #openstack-infra | 07:57 | |
| *** AJaeger has joined #openstack-infra | 08:00 | |
| *** yboaron has joined #openstack-infra | 08:02 | |
| *** ociuhandu has joined #openstack-infra | 08:05 | |
| *** tkajinam has quit IRC | 08:08 | |
| *** ijw has joined #openstack-infra | 08:08 | |
| *** yboaron has quit IRC | 08:10 | |
| *** tesseract has joined #openstack-infra | 08:11 | |
| *** ijw has quit IRC | 08:13 | |
| *** ociuhandu has quit IRC | 08:18 | |
| *** tosky has joined #openstack-infra | 08:18 | |
| *** matt_kosut has joined #openstack-infra | 08:21 | |
| *** jcapitao has joined #openstack-infra | 08:22 | |
| *** pkopec has joined #openstack-infra | 08:28 | |
| *** amoralej|off is now known as amoralej | 08:29 | |
| *** ijw has joined #openstack-infra | 08:39 | |
| *** rpittau|afk is now known as rpittau | 08:43 | |
| *** ijw has quit IRC | 08:44 | |
| *** priteau has joined #openstack-infra | 08:44 | |
| *** ralonsoh has joined #openstack-infra | 08:50 | |
| *** jpena|off is now known as jpena | 08:50 | |
| *** Wryhder has joined #openstack-infra | 08:50 | |
| *** Lucas_Gray has quit IRC | 08:51 | |
| *** Wryhder is now known as Lucas_Gray | 08:51 | |
| *** ociuhandu has joined #openstack-infra | 08:53 | |
| *** yamamoto has quit IRC | 09:02 | |
| *** yamamoto has joined #openstack-infra | 09:06 | |
| *** ociuhandu has quit IRC | 09:09 | |
| *** ijw has joined #openstack-infra | 09:10 | |
| *** ijw has quit IRC | 09:15 | |
| *** priteau has quit IRC | 09:28 | |
| *** tetsuro has quit IRC | 09:31 | |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: Evaluate CODEOWNERS settings during canMerge check https://review.opendev.org/644557 | 09:41 |
| *** ijw has joined #openstack-infra | 09:42 | |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: Add optional support for circular dependencies https://review.opendev.org/685354 | 09:45 |
| *** ijw has quit IRC | 09:46 | |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: Optionally allow zoned executors to process unzoned jobs https://review.opendev.org/673840 | 09:49 |
| openstackgerrit | Andreas Jaeger proposed openstack/project-config master: Remove Fedora 29 from nodepool https://review.opendev.org/712032 | 09:51 |
| AJaeger | ianw: are you ok with the change above to remove Fedora 29 (once dependencies are merged)? ^ | 09:51 |
| *** derekh has joined #openstack-infra | 09:58 | |
| *** udesale has quit IRC | 10:01 | |
| *** gfidente|afk is now known as gfidente | 10:01 | |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: Use implied branch matcher for implied branches https://review.opendev.org/640272 | 10:01 |
| ianw | AJaeger: yes ... working to get f31 up soon with the builder changes | 10:01 |
| *** sshnaidm|afk is now known as sshnaidm | 10:01 | |
| rpittau | hey good morning! Is there a way to get a bionic image with kernel version 5.x for devstack in ci ? | 10:02 |
| *** dtantsur|afk is now known as dtantsur | 10:06 | |
| ianw | rpittau: umm, like a hwe kernel? not really i'm afraid. iirc reboots weren't handled too well when i last tried it | 10:10 |
| *** jcapitao has quit IRC | 10:13 | |
| *** yamamoto has quit IRC | 10:13 | |
| *** jcapitao has joined #openstack-infra | 10:15 | |
| *** ociuhandu has joined #openstack-infra | 10:17 | |
| openstackgerrit | Benjamin Schanzel proposed zuul/nodepool master: Kubernetes/OpenShift Provider: Don't Require Bash in Container Images https://review.opendev.org/712034 | 10:17 |
| rpittau | ianw: thanks, yeah didn't want to deal with restarts.... I see that bionic is shipping with 5.x by default but I guess that's not valid for the cloud image | 10:17 |
| *** xinranwang has quit IRC | 10:18 | |
| AJaeger | ianw: Yeah, I've seen those f31 changes. | 10:24 |
| zbr | ianw: low hanging https://review.opendev.org/#/c/708704/ thanks. | 10:26 |
| zbr | https://review.opendev.org/#/c/703053/ is also useful, we already use this for 2+ months on two other instances (rdo and ansible) | 10:28 |
| openstackgerrit | Sorin Sbarnea proposed zuul/zuul-jobs master: DNM: test triggers https://review.opendev.org/712037 | 10:33 |
| openstackgerrit | Sorin Sbarnea proposed zuul/zuul-jobs master: DNM: test triggers https://review.opendev.org/712037 | 10:33 |
| zbr | AJaeger: see my last comment on https://review.opendev.org/#/c/708642/16/zuul-tests.d/python-jobs.yaml@18 | 10:36 |
| AJaeger | zbr: let's discuss with corvus, it looks to me like you found a bug... | 10:38 |
| zbr | i doubt is a bug because some files have a huge number of jobs defined in them, and user may not want to trigger 100 jobs when they fix a typo in one description. | 10:39 |
| zbr | but i have no problem waiting | 10:39 |
| zbr | ideally zuul could be smart to figure-out that job definition dictionary was updated, but i have no idea if this was ever implemented. | 10:41 |
| AJaeger | zbr, we have https://zuul-ci.org/docs/zuul/reference/job_def.html#attr-job.match-on-config-updates , so it should run | 10:41 |
| AJaeger | zbr, I agree, it does not work - and a change of description might not be enough. Let's change some more - let me do that quickly... | 10:41 |
| zbr | sure, feel free to re-use my ^ DNM change. | 10:41 |
| openstackgerrit | Andreas Jaeger proposed zuul/zuul-jobs master: DNM: test triggers https://review.opendev.org/712037 | 10:42 |
| AJaeger | zbr:now they run... | 10:42 |
| openstackgerrit | Andreas Jaeger proposed zuul/zuul-jobs master: DNM: test triggers https://review.opendev.org/712037 | 10:43 |
| AJaeger | zbr, so all is fine ;) | 10:43 |
| AJaeger | zbr, the test for change in job does not seem to take description into account, which is fine for me | 10:43 |
| zbr | i guess the am_i_changed() logic is quite smart, ignoring minor changes | 10:44 |
| *** ijw has joined #openstack-infra | 10:44 | |
| AJaeger | zbr, so http://zuul.opendev.org/t/zuul/status/change/712037,4/ LGTM - to you as well? | 10:44 |
| AJaeger | zbr, yes, seems the logic only looks for real change - and a description change does not count. | 10:45 |
| zbr | yep, looks well | 10:45 |
| AJaeger | zbr, so I leave my -1 on the other change ;) | 10:45 |
| *** ijw has quit IRC | 10:49 | |
| openstackgerrit | Sorin Sbarnea proposed zuul/zuul-jobs master: Improve ensure-tox role https://review.opendev.org/708642 | 10:51 |
| AJaeger | zbr, please update commit message as well | 10:54 |
| *** andrewbonney has joined #openstack-infra | 10:55 | |
| *** yamamoto has joined #openstack-infra | 10:57 | |
| zbr | no need, replied. | 11:01 |
| *** Lucas_Gray has quit IRC | 11:04 | |
| openstackgerrit | Merged zuul/zuul-jobs master: Tests bindep role on all-platforms https://review.opendev.org/708704 | 11:06 |
| *** yamamoto has quit IRC | 11:07 | |
| *** Lucas_Gray has joined #openstack-infra | 11:08 | |
| *** jcapitao is now known as jcapitao_lunch | 11:11 | |
| *** tkajinam has joined #openstack-infra | 11:15 | |
| *** ianychoi has quit IRC | 11:16 | |
| *** rpittau is now known as rpittau|bbl | 11:16 | |
| *** ianychoi has joined #openstack-infra | 11:16 | |
| *** jamesmcarthur has joined #openstack-infra | 11:19 | |
| openstackgerrit | Slawek Kaplonski proposed openstack/project-config master: Remove neutron-tempest-dvr job from Neutron's dashboard https://review.opendev.org/712048 | 11:22 |
| *** yamamoto has joined #openstack-infra | 11:24 | |
| *** jamesmcarthur has quit IRC | 11:24 | |
| *** udesale has joined #openstack-infra | 11:25 | |
| *** Lucas_Gray has quit IRC | 11:26 | |
| *** Lucas_Gray has joined #openstack-infra | 11:28 | |
| *** udesale_ has joined #openstack-infra | 11:29 | |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: Optionally allow zoned executors to process unzoned jobs https://review.opendev.org/673840 | 11:30 |
| *** udesale has quit IRC | 11:32 | |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: Add spec for enhanced regional executor distribution https://review.opendev.org/663413 | 11:34 |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: Move fingergw config to fingergw https://review.opendev.org/664949 | 11:36 |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: WIP: Route streams to different zones via finger gateway https://review.opendev.org/664965 | 11:36 |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: Support ssl encrypted fingergw https://review.opendev.org/664950 | 11:36 |
| *** yamamoto has quit IRC | 11:37 | |
| *** rosmaita has joined #openstack-infra | 11:38 | |
| openstackgerrit | Balazs Gibizer proposed opendev/irc-meetings master: Reschedule Nova Team meeting https://review.opendev.org/712052 | 11:38 |
| *** udesale_ has quit IRC | 11:38 | |
| *** ociuhandu has quit IRC | 11:40 | |
| *** auristor has quit IRC | 11:40 | |
| *** ociuhandu has joined #openstack-infra | 11:40 | |
| *** udesale has joined #openstack-infra | 11:41 | |
| *** ociuhandu has quit IRC | 11:46 | |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: WIP: Route streams to different zones via finger gateway https://review.opendev.org/664965 | 11:47 |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: Support ssl encrypted fingergw https://review.opendev.org/664950 | 11:47 |
| *** udesale has quit IRC | 11:47 | |
| *** Goneri has joined #openstack-infra | 11:49 | |
| *** nicolasbock has joined #openstack-infra | 11:54 | |
| *** dchen has quit IRC | 11:59 | |
| *** rh-jelabarre has joined #openstack-infra | 11:59 | |
| AJaeger | zbr ,want to abandon https://review.opendev.org/712037 now? | 12:09 |
| zbr | done | 12:10 |
| AJaeger | thanks | 12:11 |
| *** Lucas_Gray has quit IRC | 12:12 | |
| *** Goneri has quit IRC | 12:12 | |
| *** auristor has joined #openstack-infra | 12:13 | |
| *** nicolasbock has quit IRC | 12:13 | |
| *** nicolasbock has joined #openstack-infra | 12:15 | |
| openstackgerrit | Sorin Sbarnea proposed zuul/zuul-jobs master: bindep: Add missing virtualenv and fixed repo install https://review.opendev.org/693637 | 12:17 |
| *** Lucas_Gray has joined #openstack-infra | 12:19 | |
| *** jamesmcarthur has joined #openstack-infra | 12:21 | |
| openstackgerrit | Sorin Sbarnea proposed zuul/zuul-jobs master: WIP: Make ensure-tox pass cross-platform https://review.opendev.org/707439 | 12:24 |
| *** jpena is now known as jpena|lunch | 12:25 | |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: Implement zookeeper-auth https://review.opendev.org/619156 | 12:25 |
| openstackgerrit | Sorin Sbarnea proposed zuul/zuul-jobs master: ensure-tox: use failed_when https://review.opendev.org/712062 | 12:28 |
| *** ociuhandu has joined #openstack-infra | 12:29 | |
| *** rh-jelabarre has quit IRC | 12:36 | |
| *** jamesmcarthur has quit IRC | 12:36 | |
| *** happyhemant has joined #openstack-infra | 12:40 | |
| *** ociuhandu has quit IRC | 12:43 | |
| *** Goneri has joined #openstack-infra | 12:43 | |
| *** ociuhandu has joined #openstack-infra | 12:44 | |
| *** ociuhandu has quit IRC | 12:49 | |
| *** ianychoi has quit IRC | 12:49 | |
| *** ijw has joined #openstack-infra | 12:49 | |
| *** ianychoi has joined #openstack-infra | 12:50 | |
| *** gfidente has quit IRC | 12:54 | |
| *** rpittau|bbl is now known as rpittau | 12:54 | |
| *** piotrowskim has joined #openstack-infra | 12:54 | |
| *** ijw has quit IRC | 12:54 | |
| *** jcapitao_lunch is now known as jcapitao | 12:57 | |
| *** eharney has quit IRC | 12:59 | |
| *** yamamoto has joined #openstack-infra | 12:59 | |
| *** xek_ is now known as xek | 13:01 | |
| *** lbragstad_ has joined #openstack-infra | 13:01 | |
| *** jpena|lunch is now known as jpena | 13:04 | |
| *** Lucas_Gray has quit IRC | 13:07 | |
| *** matt_kosut has quit IRC | 13:10 | |
| *** gfidente has joined #openstack-infra | 13:10 | |
| *** yamamoto has quit IRC | 13:11 | |
| *** lbragstad_ is now known as lbragstad | 13:12 | |
| *** lbragstad has quit IRC | 13:13 | |
| *** Lucas_Gray has joined #openstack-infra | 13:13 | |
| *** lbragstad has joined #openstack-infra | 13:16 | |
| *** zxiiro has joined #openstack-infra | 13:18 | |
| *** yamamoto has joined #openstack-infra | 13:19 | |
| *** yamamoto has quit IRC | 13:19 | |
| *** ijw has joined #openstack-infra | 13:21 | |
| *** Lucas_Gray has quit IRC | 13:21 | |
| *** amoralej is now known as amoralej|lunch | 13:22 | |
| *** yamamoto has joined #openstack-infra | 13:23 | |
| *** Lucas_Gray has joined #openstack-infra | 13:25 | |
| *** ijw has quit IRC | 13:25 | |
| *** yamamoto has quit IRC | 13:30 | |
| *** rkukura has quit IRC | 13:30 | |
| *** rkukura has joined #openstack-infra | 13:37 | |
| *** Wryhder has joined #openstack-infra | 13:37 | |
| *** Lucas_Gray has quit IRC | 13:38 | |
| *** Wryhder is now known as Lucas_Gray | 13:38 | |
| fungi | ianw: what problem did you run into with reboots, besides needing to restart the log streamer? | 13:38 |
| *** eharney has joined #openstack-infra | 13:43 | |
| *** ociuhandu has joined #openstack-infra | 13:44 | |
| smcginnis | Howdy folks. We have a release job pending on getting a publish-to-pypi job added for it. Can someone other than Andreas take a look at https://review.opendev.org/#/c/711612/ | 13:50 |
| *** yamamoto has joined #openstack-infra | 13:50 | |
| *** dklyle has joined #openstack-infra | 13:51 | |
| openstackgerrit | Lance Bragstad proposed openstack/project-config master: Add a new project and repository for tripleo-ipa https://review.opendev.org/711114 | 13:54 |
| openstackgerrit | Lance Bragstad proposed openstack/project-config master: Add jobs for tripleo-ipa project https://review.opendev.org/711115 | 13:54 |
| *** Liang__ has quit IRC | 14:00 | |
| *** dave-mccowan has joined #openstack-infra | 14:02 | |
| *** rh-jelabarre has joined #openstack-infra | 14:06 | |
| *** matt_kosut has joined #openstack-infra | 14:07 | |
| *** ianychoi has quit IRC | 14:10 | |
| *** ianychoi has joined #openstack-infra | 14:11 | |
| *** matt_kosut has quit IRC | 14:12 | |
| *** Lucas_Gray has quit IRC | 14:12 | |
| openstackgerrit | Cédric Jeanneret (Tengu) proposed openstack/project-config master: Add new Validation Framework projects https://review.opendev.org/711910 | 14:17 |
| *** yamamoto has quit IRC | 14:18 | |
| *** Lucas_Gray has joined #openstack-infra | 14:18 | |
| *** ijw has joined #openstack-infra | 14:23 | |
| openstackgerrit | Lance Bragstad proposed openstack/project-config master: Add a new project and repository for tripleo-ipa https://review.opendev.org/711114 | 14:25 |
| openstackgerrit | Lance Bragstad proposed openstack/project-config master: Add jobs for tripleo-ipa project https://review.opendev.org/711115 | 14:25 |
| *** ijw has quit IRC | 14:28 | |
| clarkb | rpittau: ianw fungi ya you can reboot in jobs, the reason we dont have the hwe kernel is they dont give it by default (and we install the default kernel). I think only desktop installs swotch by default? | 14:28 |
| *** amoralej|lunch is now known as amoralej | 14:28 | |
| clarkb | I know I had to opt in on my server install in order to get some driver fix for hardware | 14:28 |
| rpittau | clarkb: interesting, any help on how we can handle that ? | 14:30 |
| *** nicolasbock has quit IRC | 14:30 | |
| clarkb | rpittau: as fungi mentions you need to restart the zuul console streamer (there is a role you can use to do that) | 14:30 |
| clarkb | otherwise should be as normal, install kernel and reboot | 14:31 |
| rpittau | clarkb: thanks, didn't see fungi comment, I'll search for the role :) | 14:31 |
| openstackgerrit | Merged openstack/project-config master: Add publish-to-pypi template for tenks https://review.opendev.org/711612 | 14:32 |
| *** jamesmcarthur has joined #openstack-infra | 14:39 | |
| *** matt_kosut has joined #openstack-infra | 14:39 | |
| *** iurygregory has quit IRC | 14:39 | |
| *** iurygregory has joined #openstack-infra | 14:43 | |
| *** jamesmcarthur has quit IRC | 14:46 | |
| *** ijw has joined #openstack-infra | 14:46 | |
| *** armax has joined #openstack-infra | 14:51 | |
| *** jamesmcarthur has joined #openstack-infra | 14:54 | |
| *** jamesmcarthur has quit IRC | 14:56 | |
| *** jamesmcarthur has joined #openstack-infra | 14:56 | |
| *** yamamoto has joined #openstack-infra | 14:57 | |
| *** yamamoto has quit IRC | 15:04 | |
| *** Lucas_Gray has quit IRC | 15:10 | |
| *** mattw4 has joined #openstack-infra | 15:14 | |
| *** Lucas_Gray has joined #openstack-infra | 15:14 | |
| *** ociuhandu has quit IRC | 15:17 | |
| *** tkajinam has quit IRC | 15:17 | |
| *** ociuhandu has joined #openstack-infra | 15:17 | |
| openstackgerrit | Clark Boylan proposed opendev/base-jobs master: Cleanup workaround for importlib-resources bad version https://review.opendev.org/712107 | 15:22 |
| *** ociuhandu has quit IRC | 15:23 | |
| clarkb | infra-root ^ fyi I think all the images that are building right now have caught up after lceaning up builders yesterday and we can land that change. I also linked to a base-test run you can confirm with in the change | 15:23 |
| clarkb | rpittau: https://opendev.org/zuul/zuul-jobs/src/branch/master/roles/start-zuul-console/README.rst is the role. Not sure if you've found it yet | 15:26 |
| rpittau | clarkb: thanks, no! :) | 15:27 |
| *** armax has quit IRC | 15:33 | |
| *** pgaxatte has quit IRC | 15:37 | |
| *** KeithMnemonic has joined #openstack-infra | 15:38 | |
| *** gyee has joined #openstack-infra | 15:43 | |
| *** matt_kosut has quit IRC | 15:45 | |
| *** ociuhandu has joined #openstack-infra | 15:48 | |
| *** pkopec has quit IRC | 16:02 | |
| *** lbragstad has quit IRC | 16:07 | |
| openstackgerrit | Merged opendev/base-jobs master: Switch nodeset fedora-latest to fedora 30 https://review.opendev.org/711969 | 16:08 |
| *** lbragstad has joined #openstack-infra | 16:08 | |
| *** yamamoto has joined #openstack-infra | 16:10 | |
| clarkb | AJaeger: correct on the gerrit downtime for project renames | 16:11 |
| clarkb | AJaeger: maybe we should do openstack/infra-manual -> opendev/infra-manual now that there is another rename queued up | 16:11 |
| *** ianychoi has quit IRC | 16:11 | |
| *** ianychoi has joined #openstack-infra | 16:12 | |
| *** yamamoto has quit IRC | 16:15 | |
| Tengu | AJaeger: heya! do I need to do anything for https://review.opendev.org/#/c/711910/ to get merged? I think everything should be OK (the needed-by is there, just running a recheck on it) | 16:15 |
| clarkb | Tengu: what sort of renaming might happen later? renaming of the git repos? | 16:19 |
| *** michael-mc-aleer has joined #openstack-infra | 16:23 | |
| *** michael-mc-aleer has quit IRC | 16:23 | |
| *** michael-mc-aleer has joined #openstack-infra | 16:24 | |
| *** jcapitao has quit IRC | 16:25 | |
| Tengu | clarkb: heya! nope, not the repositories - more the "tripleo-validations" part that will be dropped later, so we might want to rename the... err... "group"? to validation-framework instead. | 16:26 |
| michael-mc-aleer | Hi Infrastructure team, I'm running into some issues stacking Devstack from master, Horizon has a dependency pyScss==1.3.4 that tries to import Fearure from setup tools but it was removed in setup tools 46.0.0, any advice on where I can go to raise it to get fixed? Thanks! | 16:26 |
| *** aarents has quit IRC | 16:27 | |
| *** aarents has joined #openstack-infra | 16:27 | |
| michael-mc-aleer | github.com/Kronuz/pyScss/issues/385 | 16:27 |
| michael-mc-aleer | Stacking process broken as a result | 16:28 |
| Tengu | clarkb: but maybe the "tox" error here might be due to the grouping I'm trying to do in the governance chance... https://zuul.opendev.org/t/openstack/build/1c34c13f61154a0d9ee7fd8b4892963f | 16:28 |
| clarkb | Tengu: I'm not sure, just wanted to make sure we didn't merge the repo creation early if the names were still up for debate | 16:29 |
| Tengu | clarkb: ah, nah, the repositories are fixed for good | 16:29 |
| Tengu | packages will get that name and all | 16:29 |
| *** bdodd has quit IRC | 16:29 | |
| clarkb | michael-mc-aleer: you can disable horizon and bypass the issue entirely. Otherwise you'll probably need pyScss to update | 16:29 |
| clarkb | michael-mc-aleer: Setuptools isn't a library we can easily control the version of so addressing bugs that come up with interactions with it is usually the best option | 16:30 |
| clarkb | (bceause users local version of setuptools is arbitrary) | 16:30 |
| Tengu | thanks clarkb :) | 16:30 |
| michael-mc-aleer | Sure, I can make that work for our dev team but our quality engineers may be impacted because they use horizon | 16:30 |
| michael-mc-aleer | Thanks for the heads up, that pyScss repo looks pretty stagnant, last commit was 2018 and there is a tonne of issues left unanswered | 16:31 |
| clarkb | michael-mc-aleer: possible that horizon may need to consider a different css compiler then | 16:31 |
| michael-mc-aleer | Yeah that's what my thoughts are too, I'll give their channel a msg to see if they are aware | 16:32 |
| michael-mc-aleer | Thanks again! | 16:32 |
| *** bdodd has joined #openstack-infra | 16:33 | |
| *** trident has quit IRC | 16:34 | |
| *** jcapitao has joined #openstack-infra | 16:35 | |
| *** matt_kosut has joined #openstack-infra | 16:36 | |
| *** trident has joined #openstack-infra | 16:36 | |
| *** yamamoto has joined #openstack-infra | 16:47 | |
| openstackgerrit | Clark Boylan proposed opendev/system-config master: Add periodic job to run goaccess reporting https://review.opendev.org/712121 | 16:47 |
| *** ianychoi has quit IRC | 16:47 | |
| clarkb | infra-root ^ thats a first pass at using go access for website stats. corvus I've chosen zuul-ci.org as the first site to run this | 16:48 |
| *** ianychoi has joined #openstack-infra | 16:48 | |
| *** pkopec has joined #openstack-infra | 16:49 | |
| clarkb | the job is set up to not publicly publsih the reports initially. That will give us the opportunity to review them first (to ensure the version on bionic does what we want and that things otherwise are as expected) | 16:50 |
| fungi | clarkb: michael-mc-aleer: i think there was a recent ml thread (last month?) on openstack-discuss about options for replacing it. i'll see if i can find that in the archive | 16:51 |
| *** yamamoto has quit IRC | 16:52 | |
| openstackgerrit | James E. Blair proposed zuul/zuul master: Implement zookeeper-auth https://review.opendev.org/619156 | 16:54 |
| *** armax has joined #openstack-infra | 16:55 | |
| lbragstad | AJaeger i think we finally reached consensus on https://review.opendev.org/#/c/711114/ - curious if you'd like to take another look when you get a minute? | 16:55 |
| clarkb | oh apparently I can use copy: with a content attribute instead of shell and a heredoc | 16:56 |
| Tengu | clarkb: yep, that's pretty nice :) | 16:56 |
| Tengu | ansible is full of tricks :] | 16:56 |
| openstackgerrit | David Shrewsbury proposed zuul/nodepool master: Implement zookeeper-auth https://review.opendev.org/619155 | 16:57 |
| openstackgerrit | Clark Boylan proposed opendev/system-config master: Add periodic job to run goaccess reporting https://review.opendev.org/712121 | 16:58 |
| clarkb | now with better ansibling | 16:58 |
| fungi | clarkb: michael-mc-aleer: nevermind, i was misremembering the discussion about finding a replacement for django-babel | 16:58 |
| clarkb | frickler: ^ sorry, I decided to get rid of the hacky heredoc once I realized the copy module could do what I wanted. Thank you for the review | 16:59 |
| michael-mc-aleer | Downgrading setuptools to N-1 release 45.3.0 works if anyone else runs into a similar problem, I was stacking in a clean environment suit was pulling the latest setuptools during setup | 17:04 |
| michael-mc-aleer | *it was pulling | 17:05 |
| *** tesseract has quit IRC | 17:07 | |
| *** matt_kosut has quit IRC | 17:09 | |
| clarkb | friendly reminder: For many of us in north america that observe dst the clocks shifted a coupel days ago. This means that our meeting is an hour later than it was last week if going off your local timezone clock | 17:09 |
| clarkb | Our meeting is in just under 2 hours | 17:09 |
| clarkb | unless you are fungi and only use UTC clocks | 17:10 |
| *** rpittau is now known as rpittau|afk | 17:10 | |
| *** ociuhandu has quit IRC | 17:10 | |
| fungi | yup | 17:11 |
| fungi | though i had a conference call earlier today which follows usa/canada dst so had to remember that it occurred an hour earlier relative to utc | 17:12 |
| * fungi lobbies his congresscritters again to abolish dst | 17:13 | |
| *** ccamacho has quit IRC | 17:15 | |
| *** jamesmcarthur has quit IRC | 17:20 | |
| *** ianychoi has quit IRC | 17:26 | |
| *** mattw4 has quit IRC | 17:26 | |
| *** mattw4 has joined #openstack-infra | 17:27 | |
| *** jamesmcarthur has joined #openstack-infra | 17:27 | |
| *** ianychoi has joined #openstack-infra | 17:27 | |
| fungi | apparently it currently costs us$800k/mo to operate pypi? https://groups.google.com/forum/#!topic/pypa-dev/Pdnoi8UeFZ8 | 17:27 |
| *** sshnaidm is now known as sshnaidm|afk | 17:31 | |
| corvus | fungi: fascinating! we have experience with at least 2 of their suggestions.... | 17:33 |
| corvus | fungi: i wonder if the reason we abandoned one of them (bandersnatch -- because of tensorflow data packages) has anything to do with the 800k/month | 17:33 |
| fungi | corvus: yeah, i've sent a reply detailing our prior and current approaches/experience there (per the reference implementation section) along with links to the relevant configs | 17:34 |
| corvus | fungi: is that moderated? | 17:35 |
| fungi | no idea how long it takes replies to threads on pypa-dev to show up, may be moderated (google's mta accepted my reply near-instantly) | 17:35 |
| fungi | so doesn't seem they're greylisting me | 17:35 |
| corvus | fungi: ack, thanks for doing that, i think we can be helpful | 17:35 |
| *** evrardjp has quit IRC | 17:35 | |
| *** evrardjp has joined #openstack-infra | 17:35 | |
| fungi | i hope so. if my message doesn't make it through to the list i'll pull it out of my fcc and stick it in an etherpad or somewhere for someone else to try to send | 17:36 |
| clarkb | corvus: judging based on freedesktop's rough cost break down it does seem like storage and transfer is not cheap | 17:36 |
| clarkb | corvus: and having multi gigabytes of tensorflow related packages pushed everyday would make a big dent I Think | 17:36 |
| *** dtantsur is now known as dtantsur|afk | 17:36 | |
| fungi | yeah, moc said they're currently paying amazon something like us$2m/yr and almost all of it is data transfer on storage of research data | 17:37 |
| *** matt_kosut has joined #openstack-infra | 17:37 | |
| clarkb | it does make me happy to know that our attempts at being friendly neighbors are appreciated | 17:37 |
| fungi | or rather fetching the research data they've stored (amazon storage is priced so that it's cheap to put things there and expensive to read them back) | 17:38 |
| corvus | clarkb: yeah, i mean, if they're pushing that TF data up there, i bet someone is downloading it too | 17:38 |
| clarkb | corvus: ya exactly | 17:38 |
| clarkb | fungi: thats how they get you! :) | 17:39 |
| fungi | indeed | 17:39 |
| clarkb | also ++ for providing feedback on how bandersnatch is basically unusable at this point unless your goal is to make a complete mirror | 17:39 |
| fungi | well, it grew a feature to filter specific packages around the time we stopped using it, but 1. that would have required a full rebuild of the mirror to filter out earlier files we'd already copied and 2. it would still need keeping on top of for any future packages following the same pattern | 17:40 |
| fungi | i mentioned those in my reply | 17:40 |
| fungi | which has still not appeared | 17:41 |
| clarkb | arg my change hit the single ' in comment in ansible playbook issue | 17:41 |
| clarkb | fungi: exactly | 17:41 |
| fungi | 2020-03-10 17:24:22 1jBic8-0003nD-SU => pypa-dev@googlegroups.com R=dnslookup T=remote_smtp H=gmr-smtp-in.l.google.com [2607:f8b0:400d:c0f::e] X=TLS1.2:ECDHE_RSA_CHACHA20_POLY1305:256 CV=yes DN="C=US,ST=California,L=Mountain View,O=Google LLC,CN=mx.google.com" C="250 2.0.0 OK 1583861062 q1si355258pff.4 - gsmtp" | 17:41 |
| fungi | c'mon google, it's been over 15 minutes now | 17:42 |
| *** michael-mc-aleer has quit IRC | 17:42 | |
| openstackgerrit | Clark Boylan proposed opendev/system-config master: Add periodic job to run goaccess reporting https://review.opendev.org/712121 | 17:42 |
| clarkb | you know when I was taught ot write a compiler the first thing we were told to do was to strip all the comments out and completely ignore them | 17:42 |
| clarkb | because you shouldn't apply your syntax rules to comment content | 17:42 |
| *** jpena is now known as jpena|off | 17:43 | |
| fungi | unfortunately it's the yaml parser which knows what should be a comment, and i guess ansible's substitution engine takes precedence? | 17:43 |
| *** jamesmcarthur has quit IRC | 17:43 | |
| *** jcapitao is now known as jcapitao_off | 17:44 | |
| clarkb | fungi: I'm not sure what layer is hitting it. The linter catches it, but I'm pretty sure real ansible fails on it too (so yay linter does catch issues) | 17:44 |
| fungi | or is it a shell comment? that's even harder to catch since the comment identification happens at runtime not during deserialization | 17:44 |
| clarkb | fungi: in this case it was actually a shell comment in a block string. But it happens in yaml comments too | 17:45 |
| clarkb | (so ya it must be evaluating the entire string before yaml parses any of it) | 17:45 |
| clarkb | the docker suggestions in that pep draft really only shift the problem to the docker registries | 17:47 |
| mordred | fungi: if only both vexxhost and ovh's object storage weren't half the cost | 17:47 |
| clarkb | which wouldn't be so bad if docker registries weren't so painful to cache | 17:48 |
| clarkb | but they are | 17:48 |
| clarkb | (so bad that lennart even wrote an entire new toolchain to address it as he does) | 17:48 |
| fungi | i did mention caching dockerhub and quay (and it's happening in the same file i linked to for our pypi caching example), but i didn't explicitly mention the struggles with caching them since they were less the topic of the pep | 17:49 |
| mordred | clarkb: to be fair - he also writes entire new toolchains to replace good software - so I don't know that that adds more credibility to the point that it's hard to cache | 17:49 |
| clarkb | mordred: ya | 17:49 |
| *** lmiccini has quit IRC | 17:49 | |
| clarkb | fungi: ya I think thats fine. I'm mostly just a little concerned the fix is to make it dockerhubs problem | 17:50 |
| clarkb | which will only make dockerhub even less reliable | 17:50 |
| fungi | right, i touched on the need to cache basically everything you can (within reason, so long as it doesn't undermine the integrity of the build itself of course) | 17:51 |
| clarkb | ++ | 17:51 |
| fungi | and that the internet is terribad | 17:51 |
| fungi | albeit with more elegant phrasing | 17:51 |
| *** jamesmcarthur has joined #openstack-infra | 17:56 | |
| clarkb | fungi: I've found documentation that says you must have a google account to post to a google group. | 17:58 |
| clarkb | https://support.google.com/groups/answer/1067205?hl=en | 17:58 |
| clarkb | that may explain the non post? | 17:59 |
| *** pkopec has quit IRC | 17:59 | |
| *** derekh has quit IRC | 18:00 | |
| openstackgerrit | Merged openstack/os-testr master: [ussuri][goal] Drop python 2.7 support and testing https://review.opendev.org/705522 | 18:01 |
| fungi | clarkb: i wonder if that's irrespective of whether it comes from a subscribed e-mail address? | 18:01 |
| clarkb | fungi: hrm it lists join a group as being somethign that requires a googel account too. Not sure if that is the same as subscribing though | 18:03 |
| *** Lucas_Gray has quit IRC | 18:03 | |
| fungi | right, i've been treating it as a public mailing list to which my e-mail address is subscribed. pretty sure i've posted to it in the past | 18:03 |
| fungi | via smtp | 18:04 |
| *** ijw_ has joined #openstack-infra | 18:05 | |
| fungi | i suspect it's just moderated | 18:06 |
| zbr | isn | 18:06 |
| zbr | isn't easier to reply to https://discuss.python.org/t/draft-pep-pypi-cost-solutions-ci-mirrors-containers-and-caching-to-scale/3681 ? | 18:07 |
| *** ijw has quit IRC | 18:07 | |
| fungi | no for two reasons, first i'm not subscribed to discuss.python.org and second the pep says feedback should go to the pypa-dev google group forum/ml | 18:07 |
| *** trident has quit IRC | 18:08 | |
| *** amoralej is now known as amoralej|off | 18:08 | |
| fungi | but if it doesn't eventually show up i can try signing into my infrequently-used google account and doing unspeakable $things | 18:08 |
| *** ianychoi has quit IRC | 18:10 | |
| zbr | please do, i am going to make some popcorn | 18:10 |
| AJaeger | clarkb: let me add opendev/infra-manual to the queue... | 18:11 |
| *** matt_kosut has quit IRC | 18:15 | |
| zbr | by looking at the stats of the google ML, it appears to be far less popular than the discuss "packaging" category. i kinda prefer discuss because is not isolated from python, which also have a PEPs section and more important: i can subscribe to the topics I care about, not as in the ML which is all or nothing. | 18:15 |
| *** armax has quit IRC | 18:15 | |
| *** ianychoi has joined #openstack-infra | 18:17 | |
| *** trident has joined #openstack-infra | 18:18 | |
| openstackgerrit | Andreas Jaeger proposed openstack/project-config master: Move openstack/infra-manuals to opendev https://review.opendev.org/712149 | 18:22 |
| AJaeger | clarkb: ^ | 18:22 |
| openstackgerrit | Andreas Jaeger proposed openstack/project-config master: Move openstack/infra-manuals to opendev https://review.opendev.org/712149 | 18:24 |
| clarkb | AJaeger: thanks. I'm eating early lunch then will catch up on meeting prep | 18:27 |
| *** diablo_rojo has joined #openstack-infra | 18:28 | |
| AJaeger | clarkb: enjoy! | 18:30 |
| *** factor has joined #openstack-infra | 18:36 | |
| openstackgerrit | Merged openstack/project-config master: Add new Validation Framework projects https://review.opendev.org/711910 | 18:38 |
| rm_work | mtreinish: hey do you have "core" on https://github.com/testing-cabal/subunit ? | 18:39 |
| rm_work | mtreinish: it's non-py3-able right now because the required fixes haven't been released yet... IE https://github.com/testing-cabal/subunit/commit/24a2465c92368784800562e105c920db0c997870 | 18:39 |
| fungi | rm_work: lifeless may also be worth trying to reach on that | 18:40 |
| rm_work | hmm looks like you're not in the maintainers list, just saw your name come up on an issue there and figured you might know the right people at least -- i've pinged a numbber of them previously | 18:40 |
| rm_work | yeah I think last time lifeless told me to ping jelmer? and I tracked him down in some channel (he has PMs disabled and is hidden) and he said "sure sometime" but hasn't done it | 18:41 |
| rm_work | also looking for a release of https://github.com/testing-cabal/testtools | 18:42 |
| rm_work | they release like once every two years and I'm dieing T_T | 18:42 |
| *** gfidente is now known as gfidente|Dafk | 18:49 | |
| *** gfidente|Dafk is now known as gfidente|afk | 18:49 | |
| fungi | clarkb: corvus: zbr: i think i figured it out. google groups was tossing my replies because they were pgp+mime signed | 18:52 |
| mordred | fungi: and they don't like signatures? | 18:52 |
| fungi | well, at least setting the message to not be signed got it to go through to the list | 18:52 |
| mordred | sigh | 18:52 |
| mtreinish | rm_work: I think I have commit and merge access on both subunit and testtools. But, I haven't been able to keep up with the maint on either package | 19:00 |
| rm_work | :( | 19:00 |
| mtreinish | before I push a release I need to check that everthing is working on master, and everytime I go to check things tend to be kind of stale | 19:01 |
| mtreinish | on the subunit side before we push a release I'd like to see stephenfin's PR: https://github.com/testing-cabal/subunit/pull/40 reach a conclusion | 19:01 |
| *** ijw has joined #openstack-infra | 19:01 | |
| mtreinish | because that's the top bug people are hitting with stestr/subunit on py3 | 19:02 |
| *** ijw_ has quit IRC | 19:05 | |
| mtreinish | rm_work: I will try to find time to review things this week (but I have a pretty busy schedule, I've got a release crunch on a work project) | 19:05 |
| rm_work | mtreinish: i hear you :D | 19:06 |
| rm_work | and yeah that makes sense | 19:06 |
| rm_work | no point doing a py3 release when not all the stuff is solved | 19:06 |
| *** ralonsoh has quit IRC | 19:07 | |
| *** eharney has quit IRC | 19:08 | |
| *** igordc has joined #openstack-infra | 19:09 | |
| rm_work | ah speaking of stestr: http://paste.openstack.org/show/790522/ <-- `stestr run` leads to `run: error: the following arguments are required: command` | 19:10 |
| rm_work | anyone seen this before? | 19:10 |
| rm_work | it's like it's getting a completely different set of args for "run" than it normally should | 19:11 |
| mtreinish | rm_work: I have vague memory of that, there was a bug in glance something or an oslo lib that was messing with argparse | 19:16 |
| rm_work | hmm | 19:16 |
| mtreinish | (or I remember getting pinged about something like that not too long ago) | 19:16 |
| mtreinish | bnemec: ^^^ do you remember the details? (I think you pinged me about it) | 19:17 |
| bnemec | I know there was a patch. I'm looking for it now. | 19:17 |
| *** matt_kosut has joined #openstack-infra | 19:18 | |
| bnemec | mtreinish: rm_work: I think this was the fix in glance: https://review.opendev.org/#/c/701580/ | 19:18 |
| mtreinish | bnemec: thanks, that's what I was sort-of remembering :) | 19:20 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: spec: add a zuul-runner cli https://review.opendev.org/681277 | 19:22 |
| *** yamamoto has joined #openstack-infra | 19:23 | |
| *** mugsie has quit IRC | 19:23 | |
| rm_work | oh, so pretty recent... hmm | 19:25 |
| *** matt_kosut has quit IRC | 19:25 | |
| *** mugsie has joined #openstack-infra | 19:26 | |
| *** yamamoto has quit IRC | 19:28 | |
| *** jamesmcarthur has quit IRC | 19:30 | |
| *** openstackgerrit has quit IRC | 19:32 | |
| *** ahosam has joined #openstack-infra | 19:35 | |
| *** armax has joined #openstack-infra | 19:36 | |
| *** lbragstad_ has joined #openstack-infra | 19:55 | |
| *** eharney has joined #openstack-infra | 19:56 | |
| *** lbragstad has quit IRC | 19:58 | |
| AJaeger | config-core, we can remove fedora-29 nodes now, please review https://review.opendev.org/#/c/712032/ | 20:01 |
| *** slaweq has quit IRC | 20:03 | |
| Shrews | AJaeger: I do not recommend removing from the builder nodepool.yaml at the same time you remove from the launcher nodepool.yaml unless you've confirmed that there are no active fedora-29 nodes in any provider | 20:05 |
| clarkb | Shrews: whats the concern there? | 20:06 |
| clarkb | the builder manages the images and as long as we don't try to boot new nodes that shouldn't cause any problems | 20:06 |
| * AJaeger sets a WIP quickly | 20:06 | |
| clarkb | and since we aren't removing any providers themselves it should know how to clean up nodes if told to directly | 20:06 |
| Shrews | if there are live fedora29 instances when the builder tries to delete an uploaded image that is in use, it will error | 20:07 |
| clarkb | Shrews: only in vexxhost and it will retry until the instance goes away | 20:07 |
| clarkb | it isn't fatal thankfully | 20:07 |
| AJaeger | Shrews: there should not be a job running, we rmoeved them 12 hours ago | 20:07 |
| AJaeger | (more than 12 hours even) | 20:08 |
| Shrews | there is a min-ready of 1 though | 20:08 |
| AJaeger | so, first set that to 0, then merge? | 20:08 |
| clarkb | Shrews: ya that just means we'll retry on deleting the image from vexxhost sjc1 until all volumes using the image there delete | 20:08 |
| clarkb | AJaeger: that won't acutllay delete the instances either. I'm checking now if there are any fedora-29 nodes and will manaully issue deletes if so | 20:08 |
| *** jcapitao_off has quit IRC | 20:09 | |
| Shrews | clarkb: i suppose it will retry, true. I guess it's ok then | 20:09 |
| clarkb | | 0003587456 | vexxhost-sjc1 | fedora-29-vexxhost | 62e01c16-019d-46b8-b92b-b3e409161e97 | 38.108.68.35 | | ready | 368:07:00:35 | unlocked | | 20:09 |
| AJaeger | clarkb: I see , so I'll remove the WIP | 20:09 |
| clarkb | something tells me that that server isn't going to delete on its own | 20:09 |
| clarkb | AJaeger: ya I think you are fine to merge it | 20:09 |
| AJaeger | thanks, clarkb and Shrews | 20:09 |
| *** openstackgerrit has joined #openstack-infra | 20:12 | |
| openstackgerrit | Merged openstack/project-config master: Remove Fedora 29 from nodepool https://review.opendev.org/712032 | 20:12 |
| AJaeger | and gone ^ | 20:12 |
| AJaeger | #status log Fedora 29 images are retired | 20:12 |
| AJaeger | did we loose the status bot again? | 20:13 |
| clarkb | seems like only gerritbot and meetbot are here | 20:13 |
| ianw | corvus: repsonded in change, but there's quite a few things mapped into the container that are (correctly, i would say) owned by the local "nodepool" user | 20:15 |
| *** jamesmcarthur has joined #openstack-infra | 20:17 | |
| *** ianychoi has quit IRC | 20:17 | |
| ianw | although there is probably a case for things like dib_tmp being some sort of ephemerally attached volume storage in the container | 20:17 |
| ianw | however, i'd like to KISS for the initial deployment | 20:18 |
| *** ianychoi has joined #openstack-infra | 20:18 | |
| clarkb | AJaeger: it last logged entries in its log file on march 5, I have restarted it, but it may take a minute to join this channel | 20:19 |
| *** openstackstatus has joined #openstack-infra | 20:19 | |
| *** ChanServ sets mode: +v openstackstatus | 20:19 | |
| clarkb | there it is | 20:19 |
| clarkb | AJaeger: if you want to rerun your status now | 20:19 |
| AJaeger | #status log Fedora 29 images are retired | 20:19 |
| openstackstatus | AJaeger: finished logging | 20:19 |
| AJaeger | thanks, clarkb ! | 20:19 |
| clarkb | the process itself was still running, I guess it got netsplit and stopped having messages to process and thus no more logging | 20:21 |
| corvus | ianw: i think you've convinced me on 3/4 of those; does what i wrote make sense? (i want to make sure i'm following the situation and requirements correctly) | 20:25 |
| corvus | mordred: can you take a look at the conversation between ianw and i on https://review.opendev.org/710891 -- it relates to container user ids, and i wonder if you have any left-field ideas | 20:26 |
| mordred | corvus: reading | 20:26 |
| *** dave-mccowan has quit IRC | 20:27 | |
| *** rh-jelabarre has quit IRC | 20:28 | |
| mordred | corvus: I left a comment which may be useful or may be completely off base | 20:34 |
| ianw | mordred: so we export the build logs via http so people can see their build status | 20:37 |
| ianw | while that could also be containerised ... yeah i'm back to KISS :) | 20:37 |
| corvus | mordred: i think the main thing is that we need to set the uid of the logs so that nodepool can write to it when it's mounted | 20:38 |
| corvus | mordred: i agree, we probably don't have to create a user for it, we could just set the uid to 1001 and be done -- but we do need to set it to 1001 if we're going to mount it in the host | 20:38 |
| corvus | (and i think we should mount it in the host so we don't have to exec into the container to debug) | 20:39 |
| ianw | corvus/mordred: i'm putting some tasks @ https://storyboard.openstack.org/#!/story/2007407 about what i think we can look at progressively | 20:39 |
| corvus | ianw: iiuc, your objection to making dib_tmp a volume right now is that it would be backed by the wrong filesystem on the host? | 20:40 |
| *** mattw4 has quit IRC | 20:40 | |
| ianw | yes, and that i feel like it's something we can try out once we have a POC | 20:42 |
| ianw | i'm just proposing we build fedora on this for now, as we learn how it will work | 20:42 |
| openstackgerrit | Clark Boylan proposed zuul/nodepool master: Install zypper on the nodepool-builder image https://review.opendev.org/712177 | 20:43 |
| corvus | ianw: fwiw, i agree with your kiss, i'm trying to see if we can make it simpler. :) | 20:43 |
| clarkb | mordred: if you have a moment https://review.opendev.org/#/c/712121/3 is a quick rereview after I fixed the unmatch ' in ansible yaml bug | 20:44 |
| clarkb | I can trigger that directly instead of waiting for the periodic run once it lands too | 20:44 |
| corvus | ianw: change lgtm, i think mordreds +0 comment about tag may be a -1; you should take a look at that | 20:44 |
| ianw | thanks, will do soon | 20:45 |
| mordred | yeah - I think otherwise we'd have to have the : in the tag name if we used it | 20:45 |
| mordred | (it's academic right now - there are no tags) | 20:45 |
| corvus | but ideally code review helps us avoid landing future bugs :) | 20:48 |
| clarkb | hrm actually thinking about https://review.opendev.org/#/c/712121/ a bit more do I need to worry about that zcat somehow ending up in the console log? | 20:51 |
| clarkb | fungi: ^ you probably know if that is a risk with bash | 20:51 |
| corvus | clarkb: zcat where? | 20:52 |
| clarkb | corvus: https://review.opendev.org/#/c/712121/3/playbooks/periodic/goaccess.yaml line 58 | 20:52 |
| *** rcernin has quit IRC | 20:53 | |
| clarkb | if goaccess were to crash early maybe? though in that case I think you get an error and not output to stdout? | 20:53 |
| corvus | clarkb: ah, you're asking if that command can somehow go wrong and, in that case, something be output to the zuul console log for the job | 20:54 |
| *** Lucas_Gray has joined #openstack-infra | 20:54 | |
| clarkb | yes | 20:54 |
| mordred | yeah - the only way I can see the zcat going to stdout would be if goaccess decided to crash with an error message and included the contents of stdin in that error message | 20:55 |
| corvus | clarkb: i can't think of one, but i'm in the same boat as you, and would love fungi's opinion on that. but if we want to be paranoid, we could put that in a subshell that redirects everything to a file or /dev/null -- so that if something goes wrong, there's something one level up to catch it. | 20:55 |
| corvus | mordred: yeah, which seems crazy | 20:56 |
| clarkb | mordred: ya I've done some testing of the goaccess side and its pretty quiet | 20:56 |
| fungi | i don't think the shell is at risk, but it's possible goaccess itself might echo loglines in error messages? | 20:56 |
| clarkb | fungi: I havne't encountered that yet but will double check their man page for clues | 20:56 |
| mordred | like - I've never seen bash start sending a stream to stdout when it was redirected to stdin of a process after that process failed | 20:56 |
| mordred | and I think if it did it would be a huge bug :) | 20:57 |
| clarkb | I can actually feed it junk pretty trivially and see if it echos those lines | 20:57 |
| clarkb | goaccess that is | 20:57 |
| fungi | i do still think it might be cleaner to cat/zcat all the files in a subshell (similar to corvus's suggestion) and then only process - | 20:58 |
| fungi | but that's really just from a readability perspective | 20:59 |
| fungi | if you were worried about it, you could concatenate them all into a tempfile and then clean up the tempfile after processing | 20:59 |
| fungi | i assume goaccess doesn't care that the lines it processes are traversed in chronological order | 21:01 |
| clarkb | fungi: ya it uses the time stamps in the logs themselves | 21:01 |
| clarkb | doing some testing locally feeding it bad input it gives you a very limited error (likely for this reason) | 21:01 |
| corvus | given that, i think the task as written is probably good? | 21:01 |
| clarkb | basically gives you the single token that doesn't match and then the parser function that tried to match it | 21:01 |
| clarkb | corvus: ya I think its probably safe given this testing locally | 21:02 |
| clarkb | fungi: what does that subshell look like $(cat log log.1 ; zcat log.*.gz) | goaccess - ? | 21:03 |
| fungi | clarkb: almost, drop the $ | 21:04 |
| fungi | and put spaces around the ( and ) | 21:04 |
| fungi | (i think they need spaces anyway) | 21:04 |
| clarkb | k, can do that in a followon (which we need anyway to publish results once happy with them) | 21:04 |
| fungi | yeah, spaces are optional. you can also log{,.1} | 21:06 |
| fungi | which may be more readable depending on if there's a lengthy leading path | 21:06 |
| *** xek has quit IRC | 21:07 | |
| *** lbragstad_ has quit IRC | 21:11 | |
| rm_work | hmmm are folks dealing with this already? https://github.com/pallets/markupsafe/issues/57 | 21:16 |
| rm_work | may hit stable branches, like glance and designate at least on rocky | 21:16 |
| openstackgerrit | Merged opendev/system-config master: Add periodic job to run goaccess reporting https://review.opendev.org/712121 | 21:16 |
| rm_work | i wonder if we're seeing it downstream due to some other bad pinning though... | 21:17 |
| rm_work | oh nm https://review.opendev.org/#/c/711930/ is happening already | 21:18 |
| clarkb | rm_work: horizon hit it with pyscss | 21:18 |
| rm_work | did setuptools actually un-deprecate? | 21:18 |
| rm_work | i assumed they wouldn't do that and it'd be up to us to upgrade other libs | 21:19 |
| *** factor has quit IRC | 21:19 | |
| clarkb | not sure I understand the question. un-deprecate? | 21:19 |
| rm_work | i mean, the issue is they deprecated (and removed) some code | 21:19 |
| rm_work | this is skipping (!=) release 46.0.0 | 21:20 |
| clarkb | rm_work: ya that won't work | 21:20 |
| rm_work | but when they release 46.0.1 we'll be in the same boat | 21:20 |
| clarkb | you either need to cap or fix the libs | 21:20 |
| rm_work | right | 21:20 |
| rm_work | so what's the point of this patch | 21:20 |
| rm_work | buying is an undetermined (could be MINUTES?) amount of time? lol | 21:20 |
| rm_work | *buying us | 21:20 |
| clarkb | rm_work: I don't think people understand the underlying problem if that is the proposed fix | 21:20 |
| clarkb | ianw: do you hvae an example zuul enqueue command for periodic jobs? (or is it enqueue-ref?) | 21:21 |
| clarkb | zuul enqueue-ref --tenant openstack --pipeline periodic --project opendev/system-config --ref refs/heads/master maybe? | 21:25 |
| clarkb | the vars in an example inventory file don't exactly line up with enqueue or enqueue-ref | 21:26 |
| *** jamesmcarthur has quit IRC | 21:28 | |
| *** jamesmcarthur has joined #openstack-infra | 21:28 | |
| rm_work | smcginnis: ^^ or see comment on https://review.opendev.org/#/c/711930/ | 21:30 |
| rm_work | prometheanfire: ^^ | 21:30 |
| ianw | clarkb: umm, did you check my history? | 21:30 |
| clarkb | ianw: I hadn't but will do so now | 21:31 |
| smcginnis | rm_work: It's a stopgap while the actual fix is done. We don't want to cap setuptools forever. | 21:31 |
| smcginnis | rm_work: Is there a replacement for psScss | 21:31 |
| rm_work | smcginnis: ok, so we just hope they don't release in the next ... day? | 21:31 |
| smcginnis | *pyScss | 21:31 |
| clarkb | ianw: ok its there thanks | 21:31 |
| rm_work | smcginnis: this is also a problem for markupsafe | 21:31 |
| rm_work | https://github.com/pallets/markupsafe/issues/57 | 21:31 |
| ianw | clarkb: heh, yeah that's the only way i remember :) | 21:31 |
| rm_work | which has a fix in 1.1.1 but most stable branches cap at 1.1.0 | 21:31 |
| rm_work | sorry, 1.0 | 21:32 |
| clarkb | ianw: though I don't see system-config in the list as I expected, but can adapt from the others easily enough | 21:32 |
| smcginnis | I see MarkupSafe in upper-constraints, but not in global-requirements. | 21:32 |
| rm_work | for example, designate and glance in their rocky branches | 21:32 |
| rm_work | yeah it's probably an indirect-dep only | 21:33 |
| *** ahosam has quit IRC | 21:33 | |
| rm_work | https://github.com/openstack/requirements/blob/stable/rocky/upper-constraints.txt#L109 | 21:33 |
| rm_work | stein+ is fine | 21:34 |
| *** ahosam has joined #openstack-infra | 21:34 | |
| clarkb | oh neat we don't start a console log streamer on the remote in those jobs so we don't actualyl stream that into the console log | 21:34 |
| clarkb | that might be advantageous here? in any case I think it failed because goaccess hasn't been installed yet | 21:34 |
| rm_work | I don't know that we can bump a requirement in rocky, may have to pin setuptools in rocky? | 21:35 |
| clarkb | I'll re enqueue once ansible installs goaccess | 21:35 |
| *** ociuhandu has joined #openstack-infra | 21:37 | |
| clarkb | rm_work: you can't reliably pin setuptools, it tends to already be there when you start doing things and when it downgrades itself that applies to the next process invocation not the current one | 21:40 |
| fungi | which means you can't easily control it in tox invocations either because tox wants to install everything in one shot | 21:42 |
| rm_work | eugh yeah | 21:43 |
| rm_work | so ... can we bump upper-constraints for things in stable branches? :/ | 21:43 |
| *** jamesmcarthur has quit IRC | 21:44 | |
| rm_work | we will need markupsafe==1.1.1 and probably pyscss greater than whatever is there (assuming they fix the issue?) | 21:44 |
| *** Lucas_Gray has quit IRC | 21:45 | |
| *** jamesmcarthur has joined #openstack-infra | 21:45 | |
| prometheanfire | hi | 21:48 |
| rm_work | :) | 21:48 |
| prometheanfire | rm_work: pyscss is the problem for horizon at least, and it may be a dead project | 21:50 |
| prometheanfire | fixing the actual issue is preferred but may not be possible, so stop using deps that are bad, mmmmk? | 21:51 |
| prometheanfire | stop hitting yourself :P | 21:51 |
| rm_work | lol | 21:52 |
| rm_work | yeah so how do we resolve this permanently? we can all agree this fix that is merging is not going to last, right? :D | 21:52 |
| rm_work | maybe not even more than a day or so | 21:52 |
| *** jamesmcarthur has quit IRC | 21:53 | |
| rm_work | for markupsafe (which hits a few projects in their stable/rocky branches) it is already fixed, so we'd just need to bump the upper-constraints for it in rocky requirements | 21:55 |
| *** matt_kosut has joined #openstack-infra | 21:55 | |
| *** matt_kosut has quit IRC | 21:56 | |
| prometheanfire | in python nothing is permament | 21:56 |
| rm_work | i'm not sure about the fix for pyscss | 21:56 |
| prometheanfire | markupsafe bump is fine | 21:56 |
| prometheanfire | not sure about pyscss | 21:56 |
| rm_work | we might need to submit PRs and rouse the authors from their slumber | 21:56 |
| prometheanfire | the skip patch just merged, so we are good now til we aren't again | 21:57 |
| prometheanfire | then again, we don't track setuptools in constraints, this may be fixed there, if only we didn't have the chicken/egg problem | 21:57 |
| *** verdurin has quit IRC | 21:58 | |
| *** sreejithp has joined #openstack-infra | 21:59 | |
| rm_work | do i just need to change upper-constraints or is there more? | 21:59 |
| prometheanfire | no, it's not in upper-constraint, setuptools chicken/egg etc | 22:00 |
| prometheanfire | it'd be nice if we could include it imo | 22:00 |
| clarkb | mordred won't be happy with this but the pyproject stuff would address it I think | 22:00 |
| clarkb | (because you can set the libs you need to do an install and their versions | 22:01 |
| *** jcapitao_off has joined #openstack-infra | 22:01 | |
| *** sreejithp has quit IRC | 22:02 | |
| prometheanfire | ya | 22:02 |
| prometheanfire | I've suggested that before | 22:03 |
| rm_work | prometheanfire: i mean this: https://review.opendev.org/#/c/712187/ | 22:03 |
| fungi | well, a big part of the solution for this is and has been the deprecation of setuptools and its eventual removal, yes | 22:04 |
| fungi | at least the "solution" as endorsed by the folks actively discussing in distutils-sig and pypa-dev and the like | 22:05 |
| *** jamesmcarthur has joined #openstack-infra | 22:05 | |
| fungi | and maintaining a lot of the related toolchain pieces | 22:05 |
| clarkb | are they actually replacing setuptools? the new tooling allows you to but I hvaen't really see any movement towarsd that | 22:05 |
| fungi | i forget the name of the new thing which is basically "splat files into a directory" | 22:06 |
| prometheanfire | rm_work: thanks | 22:07 |
| fungi | i think it gets a mention in pep 518 | 22:07 |
| rm_work | prometheanfire: actually, I *assume* this is an issue in upstream gates... but i've only actually seen it downstream when attempting to run tests given that it's really hard to ensure the right version of setuptools everywhere... | 22:08 |
| rm_work | i am THINKING it'll probably show up in the gate too but i haven't proven it. it'll definitely show up for random people doing builds (as it did for me) | 22:08 |
| clarkb | rm_work: it could be delay in image build cycle. Depending on when that setuptools release happened it could take 24 hours for the gate to update | 22:09 |
| rm_work | yeah, k | 22:09 |
| *** jamesmcarthur has quit IRC | 22:10 | |
| fungi | clarkb: nevermind, it was mentioned in discussions of pep 517: flit | 22:14 |
| fungi | https://pypi.org/project/flit/ | 22:14 |
| clarkb | thanks | 22:15 |
| fungi | i don't think the plan is for flit to be *the* setuptools replacement, but rather for peps 517/518 to make it possible to have a variety of setuptools replacements emerge filling different niches while still all interoperating at the package installation level | 22:15 |
| fungi | flit just happened to be the first viable (albeit trivial) example | 22:16 |
| fungi | really it's separating build-time operations from install-time operations which is the goal, and making sure install-time operations can be handled by pip directly | 22:17 |
| *** evrardjp has quit IRC | 22:17 | |
| *** matt_kosut has joined #openstack-infra | 22:17 | |
| openstackgerrit | Ian Wienand proposed opendev/system-config master: nodepool-builder: deploy from container https://review.opendev.org/710891 | 22:17 |
| openstackgerrit | Ian Wienand proposed opendev/system-config master: nodepool-builder: Add webserver https://review.opendev.org/711574 | 22:17 |
| openstackgerrit | Ian Wienand proposed opendev/system-config master: Add nb01.opendev.org https://review.opendev.org/693119 | 22:17 |
| clarkb | fungi: ya but you still need tools that can make sdists and wheels | 22:17 |
| clarkb | (so that pip can then consume that from pypi) | 22:17 |
| fungi | right | 22:18 |
| *** evrardjp has joined #openstack-infra | 22:18 | |
| fungi | idea being setuptools is one of many (and hopefully increasingly abandoned in favor of simpler alternatives not encumbered by its legacy of api backward-compatibility obligations) | 22:19 |
| *** ianychoi has quit IRC | 22:19 | |
| *** ianychoi has joined #openstack-infra | 22:20 | |
| *** matt_kosut has quit IRC | 22:21 | |
| clarkb | goaccess has been installed so I've reenqueued the periodic job again to see how it goes | 22:26 |
| clarkb | it seems to have worked. I'm trying to view the report now | 22:27 |
| ianw | cool, that was quick | 22:28 |
| *** weifan has joined #openstack-infra | 22:29 | |
| *** weifan has quit IRC | 22:29 | |
| clarkb | the gui view filtered as expected. I'm reading through the json export now to see if it did as well | 22:31 |
| *** evrardjp has quit IRC | 22:31 | |
| clarkb | static.opendev.org:/home/zuul/zuul-ci.org_goaccess_report.html if others want to see it | 22:31 |
| *** evrardjp has joined #openstack-infra | 22:33 | |
| *** dciabrin_ has joined #openstack-infra | 22:33 | |
| clarkb | the json looks clean too | 22:33 |
| ianw | lgtm | 22:34 |
| ianw | i wonder if you can filter out .css files ... it doesn't seem that useful to know they're the most requested | 22:35 |
| ianw | you really want to know what bits of content are accessed | 22:35 |
| clarkb | ianw: ya thats why they break it out into two different graphs | 22:35 |
| clarkb | requests and static requests | 22:35 |
| clarkb | I think you can largely ignore "static requests" | 22:35 |
| *** dciabrin__ has quit IRC | 22:35 | |
| ianw | ohh, yeah ok i see now | 22:36 |
| *** dpawlik has quit IRC | 22:40 | |
| *** ianychoi has quit IRC | 22:41 | |
| *** ociuhandu has quit IRC | 22:41 | |
| *** ianychoi has joined #openstack-infra | 22:42 | |
| mordred | clarkb: I'd been thinking that once 517/518 were a real thing we could teach pbr to not use setuptools and instead just make sdists and wheels directly, since we need WAY less flexibility anyway | 22:48 |
| mordred | but - you know - ENOTIME | 22:49 |
| *** jcapitao_off has quit IRC | 22:49 | |
| mordred | clarkb, prometheanfire: my biggest pushback against pyproject at _this_ point is that it's less featureful for a few use cases we actively use | 22:49 |
| mordred | the model is not a bad model - but it only has 2 sets of pacakges, prod and dev, so we'd lose having docs requirements split out | 22:51 |
| openstackgerrit | Clark Boylan proposed opendev/system-config master: Simplify goaccess command https://review.opendev.org/712201 | 22:51 |
| openstackgerrit | Clark Boylan proposed opendev/system-config master: Pull goaccess report into zuul logs https://review.opendev.org/712202 | 22:51 |
| clarkb | mordred: ^ I've sort of taken your work on making log stuff templated for zuul jobs and untemplated it because static.o.o isn't a test node with all those assumptions built in. Can you review that and make sure its sane that way? | 22:51 |
| openstackgerrit | Ian Wienand proposed openstack/project-config master: nodepool-elements: Use venv for utilities https://review.opendev.org/707750 | 22:51 |
| mordred | also - last I checked, you can't create the lock file while referencing a shared constraints file - so while the lock construct solves things on a per-project basis it doesn't really let us express the same thing | 22:51 |
| clarkb | mordred: ya we'd probably have to produce our own lockfiles | 22:52 |
| clarkb | (not sure if that is a good idea or not) | 22:52 |
| mordred | well - we could for sure | 22:53 |
| *** tkajinam has joined #openstack-infra | 22:53 | |
| mordred | along with the de-setuptools-ify pbr mentioned earlier | 22:53 |
| mordred | but I think we need for some of that to still solidify a bit more | 22:53 |
| clarkb | but also I think you can use pyproject.toml without a lockfile | 22:53 |
| mordred | clarkb: I'm assuming that report gets replaced each time right? | 22:53 |
| clarkb | and still feed it the -c constraints when piping | 22:53 |
| clarkb | mordred: yes | 22:53 |
| mordred | clarkb: you can - but the only real benefit to pyproject.toml is the lockfile, or? | 22:54 |
| clarkb | mordred: you can specify you need older setuptools as an example related to the current situation without a lockfile | 22:54 |
| clarkb | mordred: then you can furhter lock that specificaton to one version if you want | 22:54 |
| mordred | I don't think that really gets us anything | 22:55 |
| mordred | I mean - at least not quickly | 22:55 |
| mordred | and if we're gonna go into setuptools-less 518 land, I think we can do better than pyproject.toml for our use cases | 22:55 |
| clarkb | ya I don't actually know what the answer is here | 22:56 |
| mordred | clarkb: the logs thing looks fine assuming that's making a whole new report | 22:56 |
| clarkb | mordred: ya its not appending it replaces | 22:57 |
| mordred | cool. then looks great | 22:57 |
| clarkb | (I tested that locally with goaccess, ot will overwrite the output) | 22:57 |
| mordred | probably shoudl be an artifact | 22:57 |
| mordred | just for, you know, completeness - but that could be a followup | 22:57 |
| *** igordc has quit IRC | 22:58 | |
| clarkb | ya I didn't want to figure out the zuul_return right away | 22:58 |
| clarkb | (thats mostly me being lazy and wanting something published though) | 22:58 |
| mordred | clarkb: both are 2x+2 - I +A'd the second one but left off the +A on the first | 22:58 |
| mordred | so - fire away whenever | 22:58 |
| clarkb | ok. ianw ^ I've got about an hour left in my day which should be enough time to monitor that but if not any chance you are able to trigger the periodic job again and take a look at it? | 22:58 |
| ianw | clarkb: sure | 22:59 |
| mordred | clarkb: I think before we go anywhere near 518 or pyproject we need to step back and assess our goals an our actual current contraints | 22:59 |
| clarkb | thanks! | 22:59 |
| mordred | because 518/pyproject is essentially a COMPLETE redesign for us due to how complex our current thing is - and I think if we're going to do that it's worth stepping back and whiteboarding for a while | 22:59 |
| *** bdodd has quit IRC | 23:00 | |
| *** lbragstad_ has joined #openstack-infra | 23:00 | |
| ianw | infra-root: if i could get final eyes on https://review.opendev.org/#/c/711574 (export logs via web for new containered nodepool-builder) and https://review.opendev.org/#/c/693119/6 (add host to ansible) i can fiddle with that today and make sure it's coming up | 23:00 |
| mordred | it's possible pyproject might be the path forward - or there might be a door number 3 | 23:00 |
| clarkb | mordred: I've heard good things about poetry (which gloms itself into the plugin system that you describe in pyproject.toml) | 23:01 |
| ianw | infra-root: and i'd also like to get in https://review.opendev.org/707750 which will switch our images to venv tools, and build a new image type deployed in rax without any of the pip/virtuatlenv stuff ... then we can look at a job to restore it via ansible | 23:01 |
| mordred | clarkb: yeah -I looked at it a while ago - it wasn't super happy making - but I can look at it again | 23:01 |
| ianw | (dib has been released with changes and should deploy to builders soon) | 23:01 |
| fungi | mordred: the recommendation i saw recently was to use extras to specify documentation build and test requirements (apparently extras named "docs" and "test" are special in some way) | 23:04 |
| clarkb | mordred: I think it does address the global lockfile issue fwiw | 23:04 |
| clarkb | mordred: because you can install all the things into an env with it then lock that set rather than a specific package | 23:04 |
| clarkb | ianw: that is failing linting | 23:05 |
| ianw | yeah looking | 23:05 |
| clarkb | https://zuul.opendev.org/t/openstack/build/bb696536b6ae47988f65434657b2b9ef/log/job-output.txt#503 | 23:05 |
| clarkb | ianw: ^ | 23:05 |
| mordred | clarkb: I don't think it does - but I don't have the actual brainspace to dig in fully and express why right now | 23:06 |
| ianw | oh, that's because the new dib release uncapped hacking | 23:06 |
| *** Tengu has quit IRC | 23:11 | |
| openstackgerrit | Ian Wienand proposed openstack/project-config master: nodepool-elements: Use venv for utilities https://review.opendev.org/707750 | 23:12 |
| openstackgerrit | Ian Wienand proposed openstack/project-config master: Fix new dib-lint errors https://review.opendev.org/712204 | 23:12 |
| ianw | clarkb: ^ that should fix the project-config gate | 23:12 |
| *** Tengu has joined #openstack-infra | 23:12 | |
| *** zxiiro has quit IRC | 23:15 | |
| *** dchen has joined #openstack-infra | 23:19 | |
| *** dchen has quit IRC | 23:19 | |
| *** dchen has joined #openstack-infra | 23:21 | |
| *** lbragstad_ has quit IRC | 23:25 | |
| openstackgerrit | Merged opendev/system-config master: nodepool-builder: deploy from container https://review.opendev.org/710891 | 23:25 |
| *** rcernin has joined #openstack-infra | 23:26 | |
| *** armax has quit IRC | 23:32 | |
| mordred | clarkb: https://review.opendev.org/#/c/712177/ has 3x+2 | 23:32 |
| clarkb | mordred: we actually build that image as part of testing so it should be safe to approve it right? | 23:33 |
| clarkb | should I hit +A? | 23:33 |
| openstackgerrit | Merged openstack/project-config master: Fix new dib-lint errors https://review.opendev.org/712204 | 23:34 |
| openstackgerrit | Merged openstack/project-config master: nodepool-elements: Use venv for utilities https://review.opendev.org/707750 | 23:34 |
| mordred | clarkb: yeah | 23:35 |
| clarkb | done | 23:35 |
| mordred | woot | 23:35 |
| mordred | between that and ianw's work - I'm pretty excited there | 23:36 |
| ianw | mordred: if you could just look over the webserver bits at https://review.opendev.org/#/c/711574 the rest is just mechanical to add the host | 23:37 |
| *** lbragstad_ has joined #openstack-infra | 23:38 | |
| ianw | grabbing a tea then hopefully non-pip-virtualenv images will be building | 23:43 |
| *** ahosam has quit IRC | 23:46 | |
| clarkb | mm tea | 23:51 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!