| *** marios is now known as marios|ruck | 05:20 | |
| opendevreview | Merged zuul/zuul-jobs master: Fix install podman error on Ubuntu aarch64 Bionic https://review.opendev.org/c/zuul/zuul-jobs/+/803413 | 05:31 |
|---|---|---|
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 06:13 | |
| *** jpena|off is now known as jpena | 07:36 | |
| *** jcapitao is now known as jcapitao_lunch | 10:34 | |
| opendevreview | Matthieu Huin proposed zuul/zuul master: web UI: user login with OpenID Connect https://review.opendev.org/c/zuul/zuul/+/734082 | 10:39 |
| opendevreview | Matthieu Huin proposed zuul/zuul master: Add authentication-realm attribute to tenants https://review.opendev.org/c/zuul/zuul/+/735586 | 10:39 |
| opendevreview | Matthieu Huin proposed zuul/zuul master: web UI: allow a privileged user to dequeue a change https://review.opendev.org/c/zuul/zuul/+/734850 | 10:39 |
| tobias-urdin | is it possible to in a zuul run execute a ansible task that reboots the node, waits for the node to come back up, then continue executing? | 11:00 |
| avass[m] | tobias-urdin: yeah there's a `reboot` module in Ansible | 11:01 |
| avass[m] | tobias-urdin: https://docs.ansible.com/ansible/latest/collections/ansible/builtin/reboot_module.html | 11:02 |
| tobias-urdin | avass[m]: yeah, but i'm more wondering on how zuul will behave since i assume i might need to start for example the zuul_console log streamer on the node again etc. | 11:07 |
| avass[m] | I thinking it's enough to just rerun the start-zuul-console role again after the reboot (or whatever the role is called) | 11:08 |
| *** jpena is now known as jpena|lunch | 11:26 | |
| opendevreview | Sorin Sbârnea proposed zuul/zuul-jobs master: Include podman installation with molecule https://review.opendev.org/c/zuul/zuul-jobs/+/803471 | 11:39 |
| fungi | tobias-urdin: avass[m]: https://zuul-ci.org/docs/zuul-jobs/general-roles.html#role-start-zuul-console | 11:40 |
| fungi | tobias-urdin: here's an example including it in another role along with a rebooot: https://opendev.org/zuul/zuul-jobs/src/branch/master/roles/enable-fips/tasks/main.yaml | 11:44 |
| fungi | (you can skim to the end of that file) | 11:44 |
| tobias-urdin | thank you both :) | 11:58 |
| opendevreview | Sorin Sbârnea proposed zuul/zuul-jobs master: Include podman installation with molecule https://review.opendev.org/c/zuul/zuul-jobs/+/803471 | 12:17 |
| *** jcapitao_lunch is now known as jcapitao | 12:20 | |
| opendevreview | Sorin Sbârnea proposed zuul/zuul-jobs master: Include podman installation with molecule https://review.opendev.org/c/zuul/zuul-jobs/+/803471 | 12:24 |
| *** jpena|lunch is now known as jpena | 12:31 | |
| opendevreview | James E. Blair proposed zuul/zuul master: Fix race in component registry https://review.opendev.org/c/zuul/zuul/+/803760 | 14:26 |
| corvus | zuul-maint: ^ that's our #1 test race (turns out it's a real race); that should save us much grief | 14:26 |
| *** rpittau is now known as rpittau|afk | 14:43 | |
| *** jpena is now known as jpena|off | 15:34 | |
| opendevreview | Matthieu Huin proposed zuul/zuul master: web UI: allow a privileged user to re-enqueue a change https://review.opendev.org/c/zuul/zuul/+/736772 | 15:43 |
| *** marios|ruck is now known as marios|out | 15:59 | |
| opendevreview | Tobias Henkel proposed zuul/zuul master: DNM: Try out pyston https://review.opendev.org/c/zuul/zuul/+/803781 | 16:05 |
| clarkb | tobiash[m]: re ^ it will be interesting to see how much of the unittest slowness is io vs python | 16:17 |
| opendevreview | Tobias Henkel proposed zuul/zuul master: DNM: Try pyston in the docker images https://review.opendev.org/c/zuul/zuul/+/803782 | 16:18 |
| tobiash[m] | yeah | 16:18 |
| opendevreview | Tobias Henkel proposed zuul/zuul master: DNM: Try pyston in the docker images https://review.opendev.org/c/zuul/zuul/+/803782 | 16:19 |
| clarkb | corvus: ok, the refactor merger and executor api chagne is next up on my list. I feel like the brain isn't quite awake enough for that yet, but I intend on getting to it soon | 16:21 |
| opendevreview | Tobias Henkel proposed zuul/zuul master: DNM: Try out pyston https://review.opendev.org/c/zuul/zuul/+/803781 | 16:22 |
| opendevreview | James E. Blair proposed zuul/zuul master: DNM: review companion for merger/executor api refactor https://review.opendev.org/c/zuul/zuul/+/803784 | 16:28 |
| corvus | clarkb: thanks -- ^ in case it helps, that shows the main diff you can't see in the refactor change because of the file moves. | 16:29 |
| clarkb | oh thanks that looks helpful | 16:31 |
| opendevreview | Matthieu Huin proposed zuul/zuul master: Web UI: allow a privileged user to request autohold https://review.opendev.org/c/zuul/zuul/+/768115 | 16:32 |
| opendevreview | Tobias Henkel proposed zuul/zuul master: DNM: Try pyston in the docker images https://review.opendev.org/c/zuul/zuul/+/803782 | 16:36 |
| clarkb | corvus: the consumption side of the api all seems to have had minimal updates which is a good sign | 17:00 |
| corvus | yeah, if you strip out the s/merge// and s/build// lines it's a lot smaller :) | 17:01 |
| corvus | the biggest structural usage change is probably in the zone handling | 17:01 |
| nirmoy | corvus: is there any other way to find out why zuul isn't starting hourly pipeline trigger without -d option in scheduler ? I can queue a change in the hourly pipeline queue using "kubectl exec -n zuul scheduler -it -- zuul enqueue --tenant <tenant_name >--project <project_name> --pipeline periodic-hourly --change <change,patchset>" | 17:02 |
| clarkb | corvus: ya the merger api basically goes away and relies on the parent which is also a good sign | 17:04 |
| corvus | nirmoy: not that i can think offhand; that's where i would start. | 17:04 |
| corvus | nirmoy: changes don't get enqueued into periodic pipelines, only heads | 17:06 |
| corvus | (ie, branches) | 17:06 |
| fungi | nirmoy: it's easier with code review triggered builds because you can ask zuul to provide a debug report in a change comment explaining why it did or didn't do certain things, but for timer triggers there's no obvious place to dump that information so it's basically just the global debug logging | 17:06 |
| corvus | nirmoy: so that enqueue command is not equivalent to what zuul would do. | 17:06 |
| opendevreview | Matthieu Huin proposed zuul/zuul master: Web UI: add Autoholds, Autohold page https://review.opendev.org/c/zuul/zuul/+/768199 | 17:08 |
| corvus | nirmoy: enqueue-ref with --ref and --oldrev and --newrev would be the equivalent command | 17:12 |
| nirmoy | corvus: thanks let me try that. I just realized that I am actually working on a git repo without a master branch. I will also try to change project.default-branch | 17:13 |
| nirmoy | fungi: true, global debug seems like last resort if nothing works. | 17:14 |
| fungi | yeah, unfortunately if you have a very busy zuul install it can be overwhelmingly verbose | 17:15 |
| fungi | and if you're not the service administrator you won't have access to those logs either | 17:15 |
| nirmoy | fungi: same, I already messed up something while debugging this yesterday and had to restart some services. I am trying to be more careful now :) | 17:22 |
| opendevreview | Merged zuul/zuul master: Fix race in component registry https://review.opendev.org/c/zuul/zuul/+/803760 | 17:28 |
| nirmoy | corvus: enqueue-ref worked and queued the job that I assigned for that project in hourly pipeline | 17:34 |
| opendevreview | Matthieu Huin proposed zuul/zuul master: REST API, Web UI: add pipelines' manager, triggers data in status https://review.opendev.org/c/zuul/zuul/+/736968 | 17:34 |
| nirmoy | the hourly pipeline trigger looks like https://pastebin.com/aMWjLDQB | 17:40 |
| opendevreview | Matthieu Huin proposed zuul/zuul master: REST API, Web UI: add pipelines' manager, triggers data in status https://review.opendev.org/c/zuul/zuul/+/736968 | 17:45 |
| clarkb | corvus: I have +2'd https://review.opendev.org/c/zuul/zuul/+/803529 but did leave some comments/questions/nits if you want to take a look | 17:54 |
| corvus | clarkb: thanks! replied | 18:32 |
| clarkb | oh thats weird new gerrit allows you to reply within comment threads now | 18:36 |
| clarkb | I was trying to ack my comments to mark them resolfed and instaed of going at the end of the comment thread like old gerrit it went between yours and mine. Anyway all your comments make sense to me. I left the one about the override comment in there in case anyone felt strongly about it though | 18:37 |
| opendevreview | Clark Boylan proposed zuul/zuul master: DNM: Try out pyston https://review.opendev.org/c/zuul/zuul/+/803781 | 18:47 |
| clarkb | tobiash[m]: ^ pyston needs wheels rebuilt for it so we need things like libpq-dev for postgres bindings and rust for cryptography builds | 18:47 |
| corvus | clarkb: does that mean we would need to use pyston in the builder image for docker builds? | 18:50 |
| fungi | we might ought to list those as build requirements anyway | 18:50 |
| corvus | i wasn't sure if it did its magic at runtime, or also needed to be used for .pyc? | 18:51 |
| Clark[m] | Looks like we would yes | 18:53 |
| tobiash[m] | corvus: regarding the merge result, when we have the pipeline state in zk I think it might make sense to put the repo state then directly to the buildset instead of doing multiple copies via the event queue to get it there | 19:09 |
| tobiash[m] | (let the merger directly put it there) | 19:10 |
| opendevreview | Tobias Henkel proposed zuul/zuul master: Align Executor API test helper methods with Merger API https://review.opendev.org/c/zuul/zuul/+/802300 | 19:15 |
| corvus | tobiash: wfm -- we were already planning (discussed in a comment on the executor api changes) to have the executor read it from that single location; it also makes sense for the merger to write it only once. but i think that it would only be one extra read/write, which isn't terrible. so maybe we could make that as a standalone followup change. | 19:19 |
| tobiash[m] | sure, a standalone change makes perfect sense and we're not there yet anyway | 19:20 |
| tobiash[m] | Clark: thanks for fixing :) | 19:21 |
| tristanC | tobiash[m]: when you have a moment, could you please have a look at this multiple github app support fix: https://review.opendev.org/c/zuul/zuul/+/803652 | 19:26 |
| tobiash[m] | tristanC: q on that ^ | 19:33 |
| tristanC | tobiash[m]: thanks, i replied | 19:38 |
| tobiash[m] | corvus: is it on purpose that the keystorage is under /keystorage and not under /zuul/keystorage in zk? | 19:39 |
| corvus | tobiash: tbh, that came as a bit of a surprise to me. i am wondering if we should move it, or if we should keep it there to make operations like "delete all of the ephemeral state" easier. my guess is that may be why swest (i think?) put it there. either that or it was an accident. :) | 19:42 |
| corvus | (to elaborate: if we keep it separate then "rm -fr /zuul" is a relatively safe operation) | 19:43 |
| corvus | (if we combine them, then "delete all the ephemeral state" has to be implemented as: delete everything under /zuul except /zuul/keystore) | 19:44 |
| corvus | definitely doable, not as convenient | 19:44 |
| swest | corvus tobiash : I think the keystore was the first thing I implemented and we didn't have the /zuul namespace yet. I guess that counts as accident | 19:47 |
| tobiash[m] | corvus: is 771463 ready to go or do you like to have more reviews on that? | 20:08 |
| corvus | tobiash: i think it's ready to go. and if we merge it, i'm totally going to do an open restart today :) | 20:10 |
| corvus | s/open/opendev/ | 20:12 |
| *** dviroel is now known as dviroel|out | 20:23 | |
| opendevreview | Merged zuul/zuul master: Add MergerApi https://review.opendev.org/c/zuul/zuul/+/787686 | 20:27 |
| opendevreview | Merged zuul/zuul master: Remove deprecated DataWatch extra args https://review.opendev.org/c/zuul/zuul/+/802689 | 20:41 |
| opendevreview | Merged zuul/zuul-jobs master: Remove success-url https://review.opendev.org/c/zuul/zuul-jobs/+/802974 | 20:53 |
| opendevreview | Merged zuul/zuul master: Store runtime related system attributes together https://review.opendev.org/c/zuul/zuul/+/801749 | 20:59 |
| opendevreview | Merged zuul/zuul master: Cache system config in Zookeeper https://review.opendev.org/c/zuul/zuul/+/801859 | 21:07 |
| opendevreview | Merged zuul/zuul master: Support cross scheduler config loading https://review.opendev.org/c/zuul/zuul/+/771463 | 21:24 |
| opendevreview | Merged zuul/zuul master: Display branch of queue in status page https://review.opendev.org/c/zuul/zuul/+/777613 | 22:14 |
| opendevreview | Merged zuul/zuul master: Fix support for multiple github connection https://review.opendev.org/c/zuul/zuul/+/803652 | 22:17 |
| opendevreview | Merged zuul/zuul master: Fix test_gerrit.TestPolling.test_config_update https://review.opendev.org/c/zuul/zuul/+/773023 | 22:17 |
| pabelanger[m] | Hmm, just noticed this today from zuul-executor | 22:27 |
| pabelanger[m] | 2021-08-06 22:20:11,561 DEBUG zuul.AnsibleJob.output: [e: 6bedebe0-f6ff-11eb-9fd6-86beb1283741] [build: 1fbd92453a5d482c8018ed6e4bc59d63] Ansible output: b'fatal: [vyos-1.1.8]: FAILED! => {"msg": "Unable to automatically determine host network os. Please manually configure ansible_network_os value for this host"}' | 22:27 |
| pabelanger[m] | that's trying to get the frozen facts before the job runs | 22:28 |
| pabelanger[m] | actually, could just be me | 22:38 |
| opendevreview | James E. Blair proposed zuul/zuul master: Execute merge jobs via ZooKeeper https://review.opendev.org/c/zuul/zuul/+/802299 | 22:40 |
| opendevreview | James E. Blair proposed zuul/zuul master: Refactor Merger/Executor API https://review.opendev.org/c/zuul/zuul/+/803529 | 22:41 |
| opendevreview | James E. Blair proposed zuul/zuul master: Streamline unlocking in merger and builder run loops https://review.opendev.org/c/zuul/zuul/+/803657 | 22:41 |
| opendevreview | James E. Blair proposed zuul/zuul master: Align Executor API test helper methods with Merger API https://review.opendev.org/c/zuul/zuul/+/802300 | 22:41 |
| corvus | the merger and config stacks collided, and so those changes failed gating (yay gating!) but the collision was trivial, so i've gone ahead and fixed it and will just re-approve those | 22:41 |
| corvus | (the name of the cat job changed from "merger:cat" to "cat") | 22:42 |
| corvus | (and the config changes added a new test that looked for 'merger:cat') | 22:42 |
| opendevreview | Merged zuul/zuul master: [doc] Clarify how regexp matching works https://review.opendev.org/c/zuul/zuul/+/794987 | 23:20 |
| opendevreview | James E. Blair proposed zuul/zuul master: Update regex note https://review.opendev.org/c/zuul/zuul/+/803811 | 23:40 |
| corvus | fungi, tobiash: fungi had some good suggestions on that ^ | 23:40 |
| opendevreview | James E. Blair proposed zuul/zuul master: Update regex note https://review.opendev.org/c/zuul/zuul/+/803811 | 23:41 |
| corvus | now with missing letters added | 23:41 |
| opendevreview | James E. Blair proposed zuul/zuul master: Update regex note https://review.opendev.org/c/zuul/zuul/+/803811 | 23:44 |
| opendevreview | Merged zuul/zuul master: Execute merge jobs via ZooKeeper https://review.opendev.org/c/zuul/zuul/+/802299 | 23:45 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!