| -@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/zuul] 902316: Fix repo state restore for zuul role tag override https://review.opendev.org/c/zuul/zuul/+/902316 | 07:20 | |
| -@gerrit:opendev.org- Benjamin Schanzel proposed: [zuul/zuul] 902419: Don't let admin action dropdown move off screen https://review.opendev.org/c/zuul/zuul/+/902419 | 14:27 | |
| -@gerrit:opendev.org- Benjamin Schanzel proposed: [zuul/zuul] 902426: Add more dark mode friedly sprites https://review.opendev.org/c/zuul/zuul/+/902426 | 15:30 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/nodepool] 902428: Fix gpu parameter type in openshiftpods https://review.opendev.org/c/zuul/nodepool/+/902428 | 16:05 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/nodepool] 902428: Fix gpu parameter type in openshiftpods https://review.opendev.org/c/zuul/nodepool/+/902428 | 16:06 | |
| -@gerrit:opendev.org- Ahmon Dancy proposed: [zuul/zuul] 902433: Reduce file descriptor leaks during tests https://review.opendev.org/c/zuul/zuul/+/902433 | 17:02 | |
| -@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: | 17:21 | |
| - [zuul/zuul] 899213: Add tests for admin api token usage with access-rules https://review.opendev.org/c/zuul/zuul/+/899213 | ||
| - [zuul/zuul-client] 899292: Use server supplied error strings https://review.opendev.org/c/zuul/zuul-client/+/899292 | ||
| -@gerrit:opendev.org- Ahmon Dancy proposed: [zuul/zuul] 902433: Reduce file descriptor leaks during tests https://review.opendev.org/c/zuul/zuul/+/902433 | 18:15 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/nodepool] 902437: Redact k8s connection info in node list https://review.opendev.org/c/zuul/nodepool/+/902437 | 18:30 | |
| -@gerrit:opendev.org- Ahmon Dancy proposed: [zuul/zuul] 902433: Reduce file descriptor leaks during tests https://review.opendev.org/c/zuul/zuul/+/902433 | 19:11 | |
| @melwitt:matrix.org | hi, I have a .zuul.yaml where what I want is to set some vars for all jobs in my project, which I can do with project.vars. the problem I'm running into is that one of the ancestor jobs (outside my project) of my jobs set the same variable, so it overrides my project.vars setting. is there any way to set the variable as a default in my project such that it can only be overridden if I override it in a job in my project? so far it's looking like I will need to set the var for each job definition in my project. if anyone has any ideas, I would appreciate it | 19:58 |
|---|---|---|
| @fungicide:matrix.org | could the parent job be abstracted to a version without that variable set, and then your project use that? | 20:00 |
| @fungicide:matrix.org | kinda ugly, admittedly | 20:01 |
| @melwitt:matrix.org | I was thinking no bc the job that sets it is pretty far up/early in the ancestry chain and various jobs in the project inherit from it indirectly, if that makes sense | 20:02 |
| -@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/zuul] 899291: Tidy some auth exceptions https://review.opendev.org/c/zuul/zuul/+/899291 | 20:04 | |
| @fungicide:matrix.org | https://zuul-ci.org/docs/zuul/latest/job-content.html#user-jobs-variable-inheritance tries to cover this, but doesn't make it clear what direction that list is in | 20:04 |
| @fungicide:matrix.org | do site-wide variables override everything else? or do parent job results? | 20:05 |
| @jim:acmegating.com | `Meaning that a site-wide variable with the same name as any other will override its value,` | 20:05 |
| @jim:acmegating.com | that paragraph is an attempt to make the direction of inheritance clear. i am sorry i failed | 20:05 |
| @fungicide:matrix.org | no, i think i failed at reading! | 20:05 |
| @melwitt:matrix.org | yeah, it says "Meaning that a site-wide variable with the same name as any other will override its value, and similarly, secrets override job variables of the same name which override data returned from parent jobs" which I took to mean that jobs override projects, no matter where that job is located, project-wise | 20:06 |
| @fungicide:matrix.org | so #1 takes precedence over everything after it | 20:06 |
| @melwitt:matrix.org | right | 20:06 |
| @fungicide:matrix.org | and job variables take precedence over project vars, in this case | 20:06 |
| @jim:acmegating.com | and yes, jobs always override projects | 20:06 |
| @melwitt:matrix.org | I need something like a "local" default, is probably the best way to describe it | 20:07 |
| @fungicide:matrix.org | or (nonexistent) `project.extra-vars` | 20:08 |
| @melwitt:matrix.org | yeah exactly. I actually looked for that after reading through the docs for awhile 😂 | 20:09 |
| @melwitt:matrix.org | so close! | 20:09 |
| @fungicide:matrix.org | trying to think if there's a way to abuse a secret that way, but probably a bad idea even if it does end up working | 20:09 |
| @melwitt:matrix.org | I began to think about that too and agreed probably a bad idea heh | 20:10 |
| @fungicide:matrix.org | for other, similar sorts of challenges, we've done entire parallel series of jobs descending from wherever we had to differentiate an ancestor's definition | 20:11 |
| @fungicide:matrix.org | but there's the obvious risk that they diverge unintentionally (always my concern with that sort of duplication) | 20:12 |
| @melwitt:matrix.org | exactly my concern in this case too | 20:12 |
| @fungicide:matrix.org | i'm tapped out, but maybe someone else in here has a better idea | 20:13 |
| @melwitt:matrix.org | also want it to be less confusing. bc if I copy paste it everywhere I'm guessing it will be more confusing | 20:13 |
| @melwitt:matrix.org | thanks though, it helps to make sure I'm not missing a better way | 20:14 |
| @jim:acmegating.com | two things you might be able to build on: yaml anchors instead of copying; or the multiple inheritance trick | 20:14 |
| @melwitt:matrix.org | oh yeah, anchors is a good point | 20:15 |
| @melwitt:matrix.org | I also wondered about multiple inheritance but didn't look further into it | 20:15 |
| @melwitt:matrix.org | one of those "what if that's worse and/or more obscure" | 20:16 |
| @jim:acmegating.com | as in any language, it only sometimes makes things easier | 20:16 |
| @jim:acmegating.com | exactly | 20:16 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!