| -@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/zuul] 831959: Clarify anonymous nodeset docs https://review.opendev.org/c/zuul/zuul/+/831959 | 00:37 | |
| -@gerrit:opendev.org- Ian Wienand proposed: [zuul/zuul] 832486: web: Ensure API calls reject unkonwn methods https://review.opendev.org/c/zuul/zuul/+/832486 | 01:30 | |
| -@gerrit:opendev.org- Ian Wienand proposed: [zuul/zuul] 832486: web: Ensure API calls reject unkonwn methods https://review.opendev.org/c/zuul/zuul/+/832486 | 02:08 | |
| -@gerrit:opendev.org- yatin proposed: [zuul/zuul-jobs] 832497: [DNM] Skip ovs install https://review.opendev.org/c/zuul/zuul-jobs/+/832497 | 07:04 | |
| -@gerrit:opendev.org- yatin proposed: [zuul/zuul-jobs] 832497: [DNM] Skip ovs install https://review.opendev.org/c/zuul/zuul-jobs/+/832497 | 07:29 | |
| -@gerrit:opendev.org- Ian Wienand proposed: | 08:45 | |
| - [zuul/zuul] 832486: web: Ensure API calls reject unkonwn methods https://review.opendev.org/c/zuul/zuul/+/832486 | ||
| - [zuul/zuul] 832505: zuul/web : Make /project end-point reject PUT https://review.opendev.org/c/zuul/zuul/+/832505 | ||
| -@gerrit:opendev.org- Ian Wienand proposed: | 10:29 | |
| - [zuul/zuul] 832486: web: Ensure API calls reject unkonwn methods https://review.opendev.org/c/zuul/zuul/+/832486 | ||
| - [zuul/zuul] 832505: zuul/web : Make /project end-point reject PUT https://review.opendev.org/c/zuul/zuul/+/832505 | ||
| -@gerrit:opendev.org- Dong Zhang marked as active: [zuul/zuul] 831925: Inject bundle id to inventroy zuul.items https://review.opendev.org/c/zuul/zuul/+/831925 | 10:29 | |
| -@gerrit:opendev.org- yatin proposed: [zuul/zuul-jobs] 832497: [multi-node-bridge] Allow to skip openvswitch installation https://review.opendev.org/c/zuul/zuul-jobs/+/832497 | 10:59 | |
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] 832517: Include original path of renamed file for a PR https://review.opendev.org/c/zuul/zuul/+/832517 | 11:05 | |
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] 832518: Fix Github debugging script https://review.opendev.org/c/zuul/zuul/+/832518 | 11:13 | |
| -@gerrit:opendev.org- Matthieu Huin https://matrix.to/#/@mhuin:matrix.org proposed: [zuul/zuul] 810955: GUI: Add tenant dropdown to top menu https://review.opendev.org/c/zuul/zuul/+/810955 | 14:42 | |
| @jpew:matrix.org | Is there a variable that can be used to post a URL to the buildset when Zuul posts comments on gerrit? | 15:13 |
|---|---|---|
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] 832655: Support non-top-level dirs as extra config path https://review.opendev.org/c/zuul/zuul/+/832655 | 15:20 | |
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] 832655: Support non-top-level dirs as extra config path https://review.opendev.org/c/zuul/zuul/+/832655 | 15:23 | |
| @avass:vassast.org | > <@jpew:matrix.org> Is there a variable that can be used to post a URL to the buildset when Zuul posts comments on gerrit? | 16:52 |
| Depending on what you need to achieve there are warnings and file comments you can use to post comments in Gerrit: | ||
| https://zuul-ci.org/docs/zuul/latest/job-content.html#leaving-file-comments | ||
| Or artifact links but they show up in zuuls dashboard for a build: | ||
| https://zuul-ci.org/docs/zuul/latest/job-content.html#returning-artifact-urls | ||
| @jpew:matrix.org | Albin Vass: Not quite.... I'd like it to post a comment when it starts gating like: "Build Started: http://zuul.example.com/t/Tenant/buildsets/aaabbbccc" | 16:54 |
| @avass:vassast.org | jpew: | 16:55 |
| Something like this? | ||
| https://zuul-ci.org/docs/zuul/latest/config/pipeline.html#attr-pipeline.start-message | ||
| @jpew:matrix.org | Ah, I'll try status_url | 16:55 |
| @jpew:matrix.org | Looks that status_url is just the web server root, not the path to the buildset (which is what I would like) | 16:57 |
| @avass:vassast.org | jpew: i think you can filter using the change number in the url somehow (that's what we do), or link directly to the buildset | 17:11 |
| @tobias.henkel:matrix.org | zuul-maint: this fixes a wrong file list when renaming files in github: https://review.opendev.org/c/zuul/zuul/+/832517 | 17:28 |
| @tobias.henkel:matrix.org | corvus: regarding https://review.opendev.org/c/zuul/zuul/+/832361, in case submitWholeTopic is enabled on gerrit it will merge all changes of the topic at once as soon as zuul merges the first in the cycle. However zuul will instruct gerrit to merge the other as well. Does zuul gracefully handle already-merged changes in the gate? | 17:37 |
| @tobias.henkel:matrix.org | * corvus: regarding https://review.opendev.org/c/zuul/zuul/+/832361, in case submitWholeTopic is enabled on gerrit it will merge all changes of the topic at once as soon as zuul merges the first in the cycle. However zuul will instruct gerrit to merge the others as well. Does zuul gracefully handle already-merged changes in the gate? | 17:37 |
| @jim:acmegating.com | tobiash: yes, i tested that. i changed the error to an info level for that since it's now a potentially expected situation. | 17:38 |
| @jim:acmegating.com | tobiash, swest suggestion on 832517 | 17:38 |
| @tobias.henkel:matrix.org | oh, I overlooked that log level change | 17:38 |
| @jim:acmegating.com | i should have highlighted that in the commit msg, sorry | 17:39 |
| @tobias.henkel:matrix.org | then lgtm | 17:39 |
| @tobias.henkel:matrix.org | corvus: for when you have time, a re-review of 772695 (perform per tenant locking in getStatus) would be great, I've updated it a while ago to address a comment by Clark | 18:03 |
| @apevec:matrix.org | tristanC: CVE-2022-0235 (something in nodejs ) was filed for Fedora Zuul package https://bugzilla.redhat.com/show_bug.cgi?id=2061831 - but is Zuul webui really affected? | 18:41 |
| @tristanc_:matrix.org | apevec: I don't think zuul should be affected, node is only used to build the web interface, and it should not use the affected node-fetch. | 18:44 |
| @apevec:matrix.org | sigh, overzealous Fedora Security folks, please close as notabug if you can :) | 18:46 |
| @y2kenny:matrix.org | I am seeing Zuul not able to rebase certain patch in a series when automatic rebase is available on Gerrit. Is there something related to rebase strategy that can be configured in Zuul? | 18:56 |
| @tobias.henkel:matrix.org | Kenny Ho: the automatic rebase strategy is sub-optimal for use with zuul, better would be the default merge strategy in gerrit | 19:12 |
| @tobias.henkel:matrix.org | Kenny Ho: I guess you're using automatic rebase to enforce a linear history? | 19:13 |
| @y2kenny:matrix.org | tobiash: yes, that's correct. | 19:14 |
| @y2kenny:matrix.org | I am using rebase-if-necessary | 19:14 |
| @y2kenny:matrix.org | tobiash: can you elaborate more on why the rebase strategy is sub-optimal? | 19:22 |
| @tobias.henkel:matrix.org | Well, it doesn't work great together with zuul which works best with the merge strategy. You could mimik loosely the gerrit rebase strategy if you set zuul to use cherry pick for this project. But beware that there might be corner cases where zuul and gerrit behavior mismatch | 19:25 |
| @y2kenny:matrix.org | um... ok... I thought zuul's speculative building would try to line things up properly but I don't know the details under the hood of zuul | 19:28 |
| @tobias.henkel:matrix.org | zuul tries to replicate what gerrit/github/gitlab is doing under the hood which is by default merge | 19:30 |
| @tobias.henkel:matrix.org | you can adapt that per project: https://zuul-ci.org/docs/zuul/latest/config/project.html?highlight=strategy#attr-project.merge-mode | 19:31 |
| @y2kenny:matrix.org | ah ok | 19:31 |
| @y2kenny:matrix.org | merge-mode... I was searching for rebase that's why I didn't spot this one | 19:31 |
| @tobias.henkel:matrix.org | cherry-pick is the strategy that matches most closely the rebase-if-necessary mode of gerrit | 19:32 |
| @y2kenny:matrix.org | I think I can play with this | 19:32 |
| @tobias.henkel:matrix.org | but it's not exactly the same | 19:32 |
| @tobias.henkel:matrix.org | but you can try if it works well enough for you | 19:32 |
| @y2kenny:matrix.org | understood. Thanks for the tip | 19:33 |
| @tobias.henkel:matrix.org | however I was successfully talking my users into doing merges so far ;) | 19:33 |
| @avass:vassast.org | tobiash: i have not :( | 19:34 |
| @tobias.henkel:matrix.org | Albin Vass: so you're using the cherry-pick mode then? | 19:35 |
| @avass:vassast.org | tobiash: yeah, but with zuul 2 for now. But if cherry-pick has have some issues in v5 it could be easier to change that. | 19:50 |
| @avass:vassast.org | Not sure why v2 doesn't seem to rebuild the queue when Gerrit creates a new patch-set with cherry pick however | 19:52 |
| @tobias.henkel:matrix.org | it's really interesting how big the lock in effect can be on any ci system in large organizations | 19:58 |
| @tobias.henkel:matrix.org | * it's really interesting to see how big the lock in effect can be on any ci system in large organizations | 19:59 |
| @avass:vassast.org | Yeah.. it's taking time to align v5 with the same workflow to make sure the migration is smooth | 20:09 |
| @sean-k-mooney:matrix.org | o/ zuul used too have a doc for deploying zuul from scratch without usign the docker compose file | 21:04 |
| @sean-k-mooney:matrix.org | has that been remvoved | 21:04 |
| @jim:acmegating.com | sean-k-mooney: yes, it was unmaintained and wrong | 21:06 |
| @sean-k-mooney:matrix.org | im trying to figure out what is the quickets way to get a semi production install curerntly | 21:06 |
| @jim:acmegating.com | docker-compose | 21:06 |
| @sean-k-mooney:matrix.org | that wont restart after a reboot automatically | 21:07 |
| @clarkb:matrix.org | it can | 21:07 |
| @sean-k-mooney:matrix.org | well i i actully install docker i gues it might | 21:07 |
| @jim:acmegating.com | are you saying you don't want it to? | 21:07 |
| @jim:acmegating.com | if you do want it to, just set 'restart: always' | 21:07 |
| @sean-k-mooney:matrix.org | no i want it to but i was tryign to do this on centos 9 stream | 21:07 |
| @sean-k-mooney:matrix.org | so i have podman pretty simply | 21:07 |
| @sean-k-mooney:matrix.org | but not sure about docker | 21:07 |
| @sean-k-mooney:matrix.org | im not sure if podman compose has restart always support | 21:08 |
| @sean-k-mooney:matrix.org | i mean i can make it work by modifying the systemd service files i guess | 21:09 |
| @sean-k-mooney:matrix.org | so contaienr install is not the recommend way to install | 21:09 |
| @jim:acmegating.com | i think i just recommended it :) | 21:09 |
| @sean-k-mooney:matrix.org | are they on quay or any regeistry other then docker.io | 21:09 |
| @sean-k-mooney:matrix.org | im hitting the rate limit which is why i decided not to try the compose file | 21:10 |
| @jim:acmegating.com | just docker.io | 21:10 |
| @sean-k-mooney:matrix.org | ok so i would have to build them locally or pull them at home and push to a local regestry on the vm | 21:10 |
| @sean-k-mooney:matrix.org | whihc was one of the things i was trying to avoid | 21:11 |
| @sean-k-mooney:matrix.org | ok ill locally build them thanks | 21:12 |
| @sean-k-mooney:matrix.org | or not... may ill have time to do that in a few day but if you cand download from docker.io because fo the rate limit it makes install zuul kind of awkward now | 21:29 |
| @clarkb:matrix.org | is the rate limit an issue beuase you're trying to deploy it on many machines at once and have a shared IP for them all? Or just a general problem with other activity hitting docker and triggering the limit? | 21:33 |
| @sean-k-mooney:matrix.org | im using a vm in redhat internal cloud | 21:33 |
| @clarkb:matrix.org | I ask because OpenDev doesn't actually seem to trip it frequently in CI or production. | 21:33 |
| @clarkb:matrix.org | It does happen occasionally but very infrequently | 21:33 |
| @sean-k-mooney:matrix.org | so i assume that is a nated ip shared by lots of redhat | 21:33 |
| @sean-k-mooney:matrix.org | if i run this at home it proably wont be an issue | 21:34 |
| @clarkb:matrix.org | ya so probably shared nat | 21:34 |
| @sean-k-mooney:matrix.org | i mean i have instlled zuul et al many time in the past so its not exactly hard to get working but im kind of surprised this is not the only recomend way to do it | 21:36 |
| @sean-k-mooney:matrix.org | give when the example compose file was first intoduce it was just for a quick tutorial and not production | 21:36 |
| @sean-k-mooney:matrix.org | i was considering trying te zuul operator but look liek there isnt realy good supprot for centos 9 with k8s | 21:37 |
| @sean-k-mooney:matrix.org | anyway i should have finished work a few hours ago so im going to call it an evening o/ | 21:39 |
| @y2kenny:matrix.org | can multiple merger share the same merger.git_dir / /var/lib/zuul/merger-git at the same time? | 23:23 |
| @jim:acmegating.com | > <@y2kenny:matrix.org> can multiple merger share the same merger.git_dir / /var/lib/zuul/merger-git at the same time? | 23:25 |
| nope | ||
| @y2kenny:matrix.org | ok, thanks for confirming. | 23:25 |
| @jim:acmegating.com | it writes to the working tree when performing merges | 23:26 |
| @y2kenny:matrix.org | I was wondering if I can have multiple merger container in the same pod... but looks like I need to do statefulset | 23:26 |
| @jim:acmegating.com | (it can do a lot of stuff in memory, but merges isn't one of them) | 23:26 |
| @sean-k-mooney:matrix.org | you can use stateful sets in k8s so that each pod get its own copy of the data | 23:27 |
| @y2kenny:matrix.org | yup. That's what I have been using for executor | 23:28 |
| @sean-k-mooney:matrix.org | so you can have mulitple merger pods but not on pod with multiple mergers | 23:28 |
| @sean-k-mooney:matrix.org | i used zuul as my example application to learn the basics of k8s a few years ago and that was one of the things i remember biting me | 23:29 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!