| -@gerrit:opendev.org- Zuul merged on behalf of Clark Boylan: [zuul/zuul] 945418: Move zuul-nox-py311-multi-scheduler to the experimental pipeline https://review.opendev.org/c/zuul/zuul/+/945418 | 01:09 | |
| -@gerrit:opendev.org- Zuul merged on behalf of Simon Westphahl: [zuul/zuul] 944328: Avoid reconfig for quick create/delete of branches https://review.opendev.org/c/zuul/zuul/+/944328 | 10:15 | |
| -@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: | 10:22 | |
| - [zuul/zuul-jobs] 944812: Add upload-image-swift role https://review.opendev.org/c/zuul/zuul-jobs/+/944812 | ||
| - [zuul/zuul-jobs] 944813: Add upload-image-s3 role https://review.opendev.org/c/zuul/zuul-jobs/+/944813 | ||
| -@gerrit:opendev.org- Zuul merged on behalf of Dong Zhang: [zuul/zuul] 943677: Support custom claim and print meta info in zuul-admin create-auth-token https://review.opendev.org/c/zuul/zuul/+/943677 | 10:59 | |
| -@gerrit:opendev.org- Fabien Boucher proposed: [zuul/zuul-jobs] 943586: zuul_log_path: allow override for emit-job-header and upload-logs https://review.opendev.org/c/zuul/zuul-jobs/+/943586 | 11:06 | |
| -@gerrit:opendev.org- Fabien Boucher proposed: [zuul/zuul-jobs] 943586: zuul_log_path: allow override for emit-job-header and upload-logs https://review.opendev.org/c/zuul/zuul-jobs/+/943586 | 11:10 | |
| -@gerrit:opendev.org- Fabien Boucher proposed: [zuul/zuul-jobs] 943586: zuul_log_path: allow override for emit-job-header and upload-logs https://review.opendev.org/c/zuul/zuul-jobs/+/943586 | 11:13 | |
| @morucci:matrix.org | Hi, https://review.opendev.org/c/zuul/zuul-jobs/+/927582 (Update the set-zuul-log-path-fact scheme to prevent huge url) has been merged. Our users want to bring back jobs name into the scheme so then I've rebased https://review.opendev.org/c/zuul/zuul-jobs/+/943586, I understood that we could land that patch as a followup. | 11:43 |
|---|---|---|
| @tristanc_:matrix.org | It's exactly like the spacebar heating xkcd:) https://xkcd.com/1172/ | 12:58 |
| @jim:acmegating.com | fbo: i don't understand why users should care, or even see the url at all. we're moving all log interaction into the web ui, and that has a really big label with the job name on it. i don't think there's anything technically wrong with the change you wrote (it adds flexibility to some roles). but i do worry that you're doing it because your users expect some facility that won't be there in the future. | 14:08 |
| -@gerrit:opendev.org- Clark Boylan proposed: [zuul/zuul] 945521: Add python3.13 unittest job https://review.opendev.org/c/zuul/zuul/+/945521 | 14:45 | |
| @morucci:matrix.org | corvus: yes thanks, that's true that is a users request, so I'll try to understand why this matters for them. Sharing the zuul-web url to the build is the right way to share build's artifacts for sure. | 15:07 |
| @fungicide:matrix.org | in fact, i find it super annoying when our users ask for help with a build and then direct link to the raw log files, because it means i have to dig around to find the build id and reconstruct the build detail url for the dashboard manually | 15:14 |
| -@gerrit:opendev.org- Clark Boylan proposed: [zuul/zuul] 945521: Add python3.13 unittest job https://review.opendev.org/c/zuul/zuul/+/945521 | 16:04 | |
| @mnaser:matrix.org | Before I go off the deep end and rewrite some roles to be "wrapped with ansible" since they are not compatible with Zuul's Ansible anymore, is there a way I can override the role tag that gets checked out? | 16:12 |
| @mnaser:matrix.org | I have this: | 16:13 |
| ``` | ||
| roles: | ||
| - name: artis3n.tailscale | ||
| zuul: github.com/artis3n/ansible-role-tailscale | ||
| ``` | ||
| But now the role is not compatible with the version of Ansible in Zuul (and I should have known better by simply running Ansible in Ansible, but I'm running this role against the node itself. | ||
| I tried this but no bueno: | ||
| ``` | ||
| required-projects: | ||
| - name: github.com/artis3n/ansible-role-tailscale | ||
| override-checkout: v4.6.0 | ||
| ``` | ||
| @clarkb:matrix.org | I was going to suggest override checkout | 16:13 |
| @mnaser:matrix.org | Oh interesting, I thought it didnt work, but it seems it didnt work for other reaosns | 16:14 |
| @mnaser:matrix.org | I re-enqueue'd it via the UI (it was a deploy pipeline job) and got an error from our Vault .. invalid token / permission denied -- wonder if somehow the secrets didnt get added when I did the enqueue in that way | 16:15 |
| @mnaser:matrix.org | actually override-checkout did the trick, sorry for the noise, it seems to be failing elsewhere | 16:16 |
| -@gerrit:opendev.org- Clark Boylan proposed: [zuul/zuul] 945521: Add python3.13 unittest job https://review.opendev.org/c/zuul/zuul/+/945521 | 16:30 | |
| -@gerrit:opendev.org- Zuul merged on behalf of Clark Boylan: [zuul/zuul] 945274: Add a process limit governor to the executor https://review.opendev.org/c/zuul/zuul/+/945274 | 16:39 | |
| @morucci:matrix.org | Yes this is the thing as fungi mentioned. We have folks that help others to figure out CI jobs issues and others used to share the log url. So they can educate but usually that we still be a problem for them. But well the inventory file contains the job name so they can find quite easily the job name, the change, ... from that file | 16:39 |
| @clarkb:matrix.org | fwiw in opendev when people link to changes and say something failed or link to the 20k line raw file I tend to push back and ask for the zuul build page | 16:41 |
| @clarkb:matrix.org | its just far easier to work with when trying to help people debug something. It removes ambiguity about what specific failure they are concerned aboutwhen they can link directly to the thing in question | 16:42 |
| @jim:acmegating.com | agree, and if someone does give you the raw log link, like https://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_e6b/zuul/e6b85a20ea53417c9038b8008f4029ba/ you can now extract the build uuid from that url, which may make working back to the build page slightly easier than before | 16:43 |
| @clarkb:matrix.org | the zuul python3.13 change seems to be hitting stuff related to the problems at the end of https://github.com/google/re2/issues/516 | 17:18 |
| @clarkb:matrix.org | seems like maybe google-re2 needs newer abseil and pybind11 than is available on ubuntu noble (but 24.10 works?) | 17:18 |
| @clarkb:matrix.org | I'm going to see if I can find an obvious older version to pin to that might work | 17:18 |
| @clarkb:matrix.org | oh wow the libabsl version on noble is from 2022 | 17:21 |
| -@gerrit:opendev.org- Clark Boylan proposed: [zuul/zuul] 945521: Add python3.13 unittest job https://review.opendev.org/c/zuul/zuul/+/945521 | 17:23 | |
| @fungicide:matrix.org | looks like they slurp it in from debian | 17:23 |
| @fungicide:matrix.org | trixie is slated to release with 20240722 | 17:24 |
| @clarkb:matrix.org | and this is only a problem because google-re2 doesn't have python3.13 wheels on pypi yet | 17:24 |
| @clarkb:matrix.org | the current requirement is `google-re2>=1.1` so I switched it to `google-re2==1.1` to see if we can build the oldest version we supposedly work with | 17:31 |
| @clarkb:matrix.org | seems like google-re2 1.1 is too new to be built using noble system libraries | 18:04 |
| @clarkb:matrix.org | so we may be stuck on 3.13 for now | 18:04 |
| @fungicide:matrix.org | unless we want to start adding debian-trixie images in opendev | 18:06 |
| @fungicide:matrix.org | it's far enough into the freeze that it ought to be fairly stable/low-churn now | 18:06 |
| @clarkb:matrix.org | ya or build wheels out of band using bazel to fetch all the deps in its build env | 18:07 |
| @jim:acmegating.com | could add trixie to zuul-providers and we can continue to experiment in a lightweight way | 18:19 |
| @fungicide:matrix.org | hopefully dib doesn't need any new work to support it | 18:26 |
| @jim:acmegating.com | would find out before merging the change :) | 18:26 |
| @fungicide:matrix.org | yeah, i mean that's the part with the highest chance to fail the change in testing, in my estimation | 18:30 |
| @fungicide:matrix.org | usrmerge and the shift to systemd-networkd are the main suspects i have for trixie that might choke the current dib element | 18:31 |
| @fungicide:matrix.org | but probably it will be fine | 18:31 |
| @clarkb:matrix.org | is /etc/network/interfaces no longer supported? | 18:45 |
| @fungicide:matrix.org | it's supported, but no longer the default approach | 18:46 |
| @clarkb:matrix.org | I think as long as that works then glean will work | 18:46 |
| @fungicide:matrix.org | and ifupdown may need to be expressly installed in order to use it | 18:46 |
| @fungicide:matrix.org | as for merged-/usr, the only reason i bring it up is if the dib element has some baked-in assumptions about the rootfs layout | 18:47 |
| @fungicide:matrix.org | though it's been so long, i can't remember if that actually took effect in bullseye | 18:48 |
| -@gerrit:opendev.org- Zuul merged on behalf of Clark Boylan: [zuul/zuul] 945413: Make MySQL test fixture more robust https://review.opendev.org/c/zuul/zuul/+/945413 | 18:55 | |
| -@gerrit:opendev.org- Zuul merged on behalf of Joseph Kostreva: [zuul/zuul] 923183: Add commentable_files variable to GerritChange https://review.opendev.org/c/zuul/zuul/+/923183 | 21:23 | |
| -@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/zuul] 920212: Use latest patchset with Gerrit cherry-pick https://review.opendev.org/c/zuul/zuul/+/920212 | 21:52 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!