| -@gerrit:opendev.org- Ian Wienand proposed: | 01:36 | |
| - [zuul/zuul] 854284: zuul-tox-remote: capture callback errors in remote zuul jobs https://review.opendev.org/c/zuul/zuul/+/854284 | ||
| - [zuul/zuul] 854285: zuul_stream: handle failed_when tasks https://review.opendev.org/c/zuul/zuul/+/854285 | ||
| - [zuul/zuul] 853603: zuul_stream: handle non-string msg value https://review.opendev.org/c/zuul/zuul/+/853603 | ||
| -@gerrit:opendev.org- Ian Wienand proposed: | 03:34 | |
| - [zuul/zuul] 854285: zuul_stream: handle failed_when tasks https://review.opendev.org/c/zuul/zuul/+/854285 | ||
| - [zuul/zuul] 853603: zuul_stream: handle non-string msg value https://review.opendev.org/c/zuul/zuul/+/853603 | ||
| -@gerrit:opendev.org- Ian Wienand proposed: [zuul/nodepool] 853914: nodepool-functional-openshift: update nodpeool launcher to Fedora 36 https://review.opendev.org/c/zuul/nodepool/+/853914 | 04:28 | |
| -@gerrit:opendev.org- Zuul merged on behalf of Ian Wienand: [zuul/zuul] 853369: zuul-stream: validate on per node basis https://review.opendev.org/c/zuul/zuul/+/853369 | 05:04 | |
| -@gerrit:opendev.org- Ian Wienand proposed: [zuul/zuul-jobs] 854335: ansible-lint: pin to < 6.5 https://review.opendev.org/c/zuul/zuul-jobs/+/854335 | 05:17 | |
| -@gerrit:opendev.org- Zuul merged on behalf of Simon Westphahl: [zuul/zuul] 853144: Set remote URL after config was updated https://review.opendev.org/c/zuul/zuul/+/853144 | 07:12 | |
| -@gerrit:opendev.org- Marvin Becker proposed wip: [zuul/nodepool] 853993: [WIP] Add hold command to disable nodes https://review.opendev.org/c/zuul/nodepool/+/853993 | 08:50 | |
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/nodepool] 854417: Consider all node types when adjusting label quota https://review.opendev.org/c/zuul/nodepool/+/854417 | 13:40 | |
| @tristanc_:matrix.org | ianw: have you seen https://crc.dev/blog/posts/2022-08-22-okd-on-crc/ ? I wonder if that would work better in zuul-jobs | 14:33 |
|---|---|---|
| @pearcetyler:matrix.org | Any idea what would cause such variance in the `prepare-workspace` job? | 15:39 |
| 8/16/2022: VERY slow | ||
| `2022-08-16 20:36:48.176581 | TASK [prepare-workspace : Synchronize src repos to workspace directory.] | ||
| 2022-08-16 20:58:51.122806 | ubuntu-focal...` | ||
| 8/17/2022: Back to normal | ||
| `2022-08-17 15:07:05.359812 | TASK [prepare-workspace : Synchronize src repos to workspace directory.] | ||
| 2022-08-17 15:07:57.015787 | ubuntu-focal...` | ||
| 8/24/2022: VERY slow again | ||
| `2022-08-24 14:35:44.031803 | TASK [prepare-workspace : Synchronize src repos to workspace directory.] | ||
| 2022-08-24 14:54:55.579349 | ubuntu-focal | Output suppressed because no_log was given` | ||
| All jobs are experiencing similar durations for `prepare-workspace` but it seems very odd that there's such a huge gap here. Prior to 8/16 I had not seen this issue | ||
| @jim:acmegating.com | Tyler Pearce: because that involves moving data from the executor's disk to the remote nodes here are some things to consider: cpu or io load on the executor, network bandwidth to the remote nodes, and then of course the whole range of possible contention points on the remote nodes depending on what they are (bare metal, cloud vms, containers) | 15:41 |
| @pearcetyler:matrix.org | Thanks for the insight corvus Cloudwatch metrics don't look out of the ordinary compared to the last few days, and nothing has changed with the configuration for the node containers. I'll keep an eye on it 🤔 | 15:45 |
| @jim:acmegating.com | Tyler Pearce: (also, if either side is subject to noisy neighbors, that could affect performance in ways that are difficult to measure directly) | 15:47 |
| @pearcetyler:matrix.org | The environment should be pretty isolated. I have a dedicated EC2 instance that only runs Zuul. The heavy lifting from the jobs is all done by CircleCI (not on the Zuul machine) so load should be relatively consistent | 15:48 |
| @avass:vassast.org | Tyler Pearce: could also be affected by other jobs running on the same executor | 16:59 |
| @fungicide:matrix.org | also if remote nodes are reused, then the executor only needs to push new references which can be much faster on large repositories | 17:28 |
| @fungicide:matrix.org | oh, wait, this is the rsync version of that task not the git version, so maybe less efficient in that regard | 17:29 |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/nodepool] 854438: Add missing cloud/region/az/host_id info to nodes https://review.opendev.org/c/zuul/nodepool/+/854438 | 17:33 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul] 854458: WIP: Add support for configuring and testing tracing https://review.opendev.org/c/zuul/zuul/+/854458 | 22:36 | |
| @iwienand:matrix.org | > <@tristanc_:matrix.org> ianw: have you seen https://crc.dev/blog/posts/2022-08-22-okd-on-crc/ ? I wonder if that would work better in zuul-jobs | 23:37 |
| Isn't OpenShift a packaging of OKD? So would OKD be smaller? The initial problem is the 9GiB required to even start the vm; and i guess the secondary problem is if that vm can run effectively on our hosts that don't have nested KVM | ||
| @iwienand:matrix.org | https://microshift.io/ seems like it's in the right direction; https://github.com/code-ready/crc/issues/3309#issuecomment-1223599562 mentions that and I guess implies that they'd like crc to be able to deploy it. | 23:41 |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/nodepool] 854438: Add missing cloud/region/az/host_id info to nodes https://review.opendev.org/c/zuul/nodepool/+/854438 | 23:46 | |
| @iwienand:matrix.org | I feel like it's not particularly worth spending a week building a very bespoke zuul-jobs deployer that then deploys something that is likely to bitrot almost as fast as it's committed ... especially just for these two jobs. But I'm open to suggestions :) | 23:55 |
| @iwienand:matrix.org | updating the nodepool opensift job to f36 doesn't seem to have worked anyway | 23:58 |
| @iwienand:matrix.org | https://zuul.opendev.org/t/zuul/build/40f4609a03134446a005852a415f9e47/console#2/1/0/launcher | 23:59 |
| @iwienand:matrix.org | seems like ```oc login`` has segfaulted | 23:59 |
| @iwienand:matrix.org | * seems like ```oc login``` has segfaulted | 23:59 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!