| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: | 00:57 | |
| - [zuul/zuul] 966638: Make ZuulTreeCache.waitForSync more robust https://review.opendev.org/c/zuul/zuul/+/966638 | ||
| - [zuul/zuul] 966741: Launcher; use waitForSync as a fallback for older ZK https://review.opendev.org/c/zuul/zuul/+/966741 | ||
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: | 16:03 | |
| - [zuul/zuul] 966501: Ensure requests referred to in assignments are up to date https://review.opendev.org/c/zuul/zuul/+/966501 | ||
| - [zuul/zuul] 966637: Add min_request_version to assignUnassignedNode method https://review.opendev.org/c/zuul/zuul/+/966637 | ||
| - [zuul/zuul] 966638: Make ZuulTreeCache.waitForSync more robust https://review.opendev.org/c/zuul/zuul/+/966638 | ||
| - [zuul/zuul] 966741: Launcher; use waitForSync as a fallback for older ZK https://review.opendev.org/c/zuul/zuul/+/966741 | ||
| @jangutter:matrix.org | couple of zuul-jobs reviews have +2, but no +W yet: | 16:29 |
|---|---|---|
| https://review.opendev.org/c/zuul/zuul-jobs/+/962794 Update ensure-helm role to add more functionality | ||
| https://review.opendev.org/c/zuul/zuul-jobs/+/872237 promote-docker-image: some notes on manual replication | ||
| (that last one would probably have saved me an hour of cruising down the rabbit hole yesterday.) | ||
| -@gerrit:opendev.org- Zuul merged on behalf of Ian Wienand: [zuul/zuul-jobs] 872237: promote-docker-image: some notes on manual replication https://review.opendev.org/c/zuul/zuul-jobs/+/872237 | 17:22 | |
| @clarkb:matrix.org | jangutter: I left a question about ansible variable expansion behavior in the ensure-helm role update change | 17:36 |
| @jangutter:matrix.org | Clark: perhaps I should move those vars from `defaults` to `vars`? I have a bad habit of just stashing everything in defaults. | 18:02 |
| @jangutter:matrix.org | I think they function the same way, it's just precedence that's different. | 18:03 |
| @jangutter:matrix.org | (I think play vars override role defaults, but not role vars) | 18:05 |
| @clarkb:matrix.org | jangutter: ya I guess in my head defaults are sort of preexisting and then tasks run so its weird to me that you can set defaults based on task outputs. But ansible variable handling is often misunderstood and it wouldn't be the first time I've gotten it wrong | 18:06 |
| @clarkb:matrix.org | I think what is probably happening if the change works is that the value isn't evaluated until used | 18:06 |
| @clarkb:matrix.org | so as long as we define the parent value that we derive from before using the child derivative value its probably fine? | 18:07 |
| @jangutter:matrix.org | Aha, I see your point on that too - using that "default" before the task defines the dependent var is not safe indeed yeah. (lazy evaluation works as intended...) So it does make it a bit more difficult to maintain the script because the information is in another place. | 18:09 |
| @jangutter:matrix.org | I'll amend it and move it into a block, being a bit more verbose is preventing future accidents. | 18:10 |
| @clarkb:matrix.org | cool thanks | 18:10 |
| -@gerrit:opendev.org- Jan Gutter proposed: [zuul/zuul-jobs] 962794: Update ensure-helm role to add more functionality https://review.opendev.org/c/zuul/zuul-jobs/+/962794 | 18:21 | |
| @clarkb:matrix.org | corvus: https://review.opendev.org/c/zuul/zuul/+/966741 failed due to problems fetching the zookeeper version string. I've rechecked it to gather more data on whether or not this is a common/persistent issue. But we may need to consider if there should be a fallback for not getting the version (like using the compatible behavior perhaps?) | 19:08 |
| @jim:acmegating.com | yeah, there's a fallback, but that's not an expected failure mode... i'll look into that | 19:12 |
| @jim:acmegating.com | Clark: i'm unable to repro that locally; i set an autohold | 19:18 |
| @clarkb:matrix.org | corvus: the zookeeper log seems to indicate connections timed out. I wonder if this is some sort of system overload problem | 19:20 |
| @clarkb:matrix.org | https://zuul.opendev.org/t/zuul/build/d2e915451f2a455e8e039b2a865db661/log/container_logs/zk.log#135-137 | 19:20 |
| @jim:acmegating.com | i can't exclude that hypothesis, but i also think it could be that the connection got dropped due to the crash | 19:21 |
| @jim:acmegating.com | the nodepool-launcher is performing zk operations at 16:53:45, before the scheduler crashed, so this doesn't make a lot of sense yet. | 19:26 |
| @clarkb:matrix.org | good point that zookeeper's perspective could be unexpected process shutdown | 19:26 |
| @clarkb:matrix.org | The FOSDEM Testing and Continuous Delivery devroom is soliciting proposals for presentations: https://lists.fosdem.org/pipermail/fosdem/2025q4/003641.html. You have until 23:59 CET, 01 December 2025 to get those proposals in. I think there is a good amount of overlap with the neat things people are doing with Zuul and their list of ideas. I think a number of Zuulians are based in Europe too. | 20:52 |
| -@gerrit:opendev.org- Piotr Parczewski proposed: [zuul/zuul-jobs] 966977: Drop Python 2 support https://review.opendev.org/c/zuul/zuul-jobs/+/966977 | 21:12 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul] 966741: Launcher; use waitForSync as a fallback for older ZK https://review.opendev.org/c/zuul/zuul/+/966741 | 21:27 | |
| @jim:acmegating.com | Clark: explanation is in the two new commit message pgraphs | 21:28 |
| @clarkb:matrix.org | oh neat you make a node then delete it which triggers a watcher to check the attributes on the event | 21:38 |
Generated by irclog2html.py 4.0.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!