| @jim:acmegating.com | it reported failure | 00:00 |
|---|---|---|
| @clarkb:matrix.org | The check said it is looking for zuul to comment. Is it specifically looking for a successful report? | 00:05 |
| @clarkb:matrix.org | Maybe Ansible and python 3.12 is the problem | 00:05 |
| @clarkb:matrix.org | If so we can probably set the zuul executor to run in verbose mode in that change | 00:07 |
| @jim:acmegating.com | Clark: be sure to set "-d" too so `-d --verbose` | 00:13 |
| -@gerrit:opendev.org- Clark Boylan proposed: [zuul/zuul] 905136: DNM first pass checking python3.12 with Zuul https://review.opendev.org/c/zuul/zuul/+/905136 | 00:27 | |
| @clarkb:matrix.org | something like that I think | 00:27 |
| @clarkb:matrix.org | hrm --verbose isn't a valid flag. Do we have to set verbose after starting the service? | 01:03 |
| @clarkb:matrix.org | I can just toggle the flag in the code | 01:05 |
| -@gerrit:opendev.org- Clark Boylan proposed: [zuul/zuul] 905136: DNM first pass checking python3.12 with Zuul https://review.opendev.org/c/zuul/zuul/+/905136 | 01:05 | |
| @bridgefan:matrix.org | Not sure if anyone here knows the answer but I'm having trouble getting submit as a group working in Gerrit. | 14:30 |
| @bridgefan:matrix.org | I have been attempting to use Gerrit's topic field to cause a group of patches to be submtted at once (to keep git history easier to read) | 14:30 |
| @bridgefan:matrix.org | using something like: git push -u origin HEAD:refs/for/[branch]%topic=tester1 | 14:31 |
| @bridgefan:matrix.org | It seems to partially work in that the patches don't submit one by one like they normally do | 14:32 |
| @bridgefan:matrix.org | Patch 1, 2, and 3 all succeed their testing. However right before they should merge I get "This change depends on a change that failed to merge." on patches 2 and 3 | 14:32 |
| @bridgefan:matrix.org | Also Patch 1 spams comments (Build succeeded. noop - Success) while it waits for the other patches to finish their tests | 14:33 |
| @bridgefan:matrix.org | Any idea if there is a setting I need to change in Zuul to support this ability of submitting gerrit topics together? | 14:34 |
| @jim:acmegating.com | bridgefan: you just need submitWholeTopic set in gerrit and then allow-circular-dependencies set in zuul: https://zuul-ci.org/docs/zuul/latest/releasenotes.html#relnotes-5-2-0 | 14:38 |
| @bridgefan:matrix.org | Thanks corvus. I'll look into that | 14:40 |
| -@gerrit:opendev.org- Zuul merged on behalf of Benjamin Schanzel: [zuul/zuul] 905077: Gerrit driver: fix for topics containing white space https://review.opendev.org/c/zuul/zuul/+/905077 | 16:00 | |
| @clarkb:matrix.org | https://zuul.opendev.org/t/zuul/build/8550c11138de419e93665da44d0184b2/log/container_logs/executor.log#580-581 confirms it is ansible that has issues with python3.12 | 16:42 |
| @clarkb:matrix.org | in general though it seems like zuul proper is working. I suspect we can make this work with some conditional import stuff but I'm not quite sure what that module has been replaced with if anything | 16:42 |
| -@gerrit:opendev.org- Clark Boylan proposed: [zuul/zuul] 905136: DNM first pass checking python3.12 with Zuul https://review.opendev.org/c/zuul/zuul/+/905136 | 17:06 | |
| @clarkb:matrix.org | So this is actually related to how zuul overloads certain action plugins in ansible and not an ansible proper problem (at least not yet). I've done a conversion based on a python discourse thread but this gets into low level python import details and this may not be the correct conversion | 17:07 |
| @jim:acmegating.com | thanks, i can take a look in a little bit | 17:11 |
| -@gerrit:opendev.org- Clark Boylan proposed: [zuul/zuul] 905136: DNM first pass checking python3.12 with Zuul https://review.opendev.org/c/zuul/zuul/+/905136 | 17:22 | |
| @jim:acmegating.com | Clark: that looks like a reasonable incantation. :) | 17:35 |
| -@gerrit:opendev.org- Clark Boylan proposed: [zuul/zuul] 905136: DNM first pass checking python3.12 with Zuul https://review.opendev.org/c/zuul/zuul/+/905136 | 18:47 | |
| @fungicide:matrix.org | Clark: were the py311 and quickstart failures on that unrelated? | 18:49 |
| @clarkb:matrix.org | fungi: depends on the patchset. Earlier patchsets only the quickstart failures were related. Since I've started modifying zuul code to make quickstart function (attempting to anyway) I'ev broken other jobs | 18:51 |
| @clarkb:matrix.org | that latest patchset passes quickstart and stream jobs. I think that implies the code is working. Once unittests confirm I can split the change up as the importlib stuff should be safe as far back as python3.6 | 20:09 |
| -@gerrit:opendev.org- Clark Boylan proposed: | 21:10 | |
| - [zuul/zuul] 905136: DNM first pass checking python3.12 with Zuul https://review.opendev.org/c/zuul/zuul/+/905136 | ||
| - [zuul/zuul] 905269: Replace python imp module with importlib https://review.opendev.org/c/zuul/zuul/+/905269 | ||
| @clarkb:matrix.org | That reorganizes things to separate any python3.12 transition from updates to the code to make zuul more future proof | 21:10 |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul] 905275: Optimize db prune query https://review.opendev.org/c/zuul/zuul/+/905275 | 23:41 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!