| -@gerrit:opendev.org- Tony Breeds proposed: [zuul/zuul] 908962: [docs] Update autohold-delete and autohold-info examples https://review.opendev.org/c/zuul/zuul/+/908962 | 01:11 | |
| -@gerrit:opendev.org- Bernhard Berg proposed wip on behalf of Lukas Kranz: [zuul/zuul-jobs] 887917: prepare-workspace-git: Add ability to define synced pojects https://review.opendev.org/c/zuul/zuul-jobs/+/887917 | 09:21 | |
| -@gerrit:opendev.org- Jeremy Stanley https://matrix.to/#/@fungicide:matrix.org proposed: [zuul/zuul] 908855: Refresh Keycloak OpenID Howto for latest Keycloak https://review.opendev.org/c/zuul/zuul/+/908855 | 17:06 | |
| -@gerrit:opendev.org- Clark Boylan proposed: [zuul/zuul-jobs] 909029: Override DOCKER_MIN_API_VERSION for skopeo when installing docker https://review.opendev.org/c/zuul/zuul-jobs/+/909029 | 17:26 | |
| @clarkb:matrix.org | corvus: ^ something like that maybe? I don't expect it to cause trouble for docker itself as this is only the minimum version that can be negotiated. The docker tools should negotiate higher versions | 17:26 |
|---|---|---|
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul-jobs] 908671: Remove command.warn usage https://review.opendev.org/c/zuul/zuul-jobs/+/908671 | 17:31 | |
| @jim:acmegating.com | Clark: lgtm; i rebased that on it ^ | 17:31 |
| @clarkb:matrix.org | the microk8s jobs seems to be unhappy. I wonder if that means microk8s is still broken and we need to pin the version there | 17:32 |
| -@gerrit:opendev.org- Clark Boylan proposed: [zuul/zuul-jobs] 909029: Override DOCKER_MIN_API_VERSION for skopeo when installing docker https://review.opendev.org/c/zuul/zuul-jobs/+/909029 | 17:53 | |
| @clarkb:matrix.org | I'm going to need to hold a node to see why it appears to not be running with the correct env var set | 17:54 |
| @clarkb:matrix.org | I'll rebase the other change once I've got this happy ish | 17:54 |
| @clarkb:matrix.org | ya looks like microk8s hasn't made any udpates and the issue for debian is still open. I'll have to pin microk8s in this change to get it passing. I'll work on that after the current set of jobs complete | 18:10 |
| @clarkb:matrix.org | corvus: is openshift 3.11 no longer hosting all of its images: https://zuul.opendev.org/t/zuul/build/8a85b33d70b94b0cb42b75c43ffbbf11/log/docker/origin.txt#297 ? | 18:20 |
| -@gerrit:opendev.org- Clark Boylan proposed: [zuul/zuul-jobs] 909029: Override DOCKER_MIN_API_VERSION for skopeo when installing docker https://review.opendev.org/c/zuul/zuul-jobs/+/909029 | 18:21 | |
| -@gerrit:opendev.org- Clark Boylan proposed on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/zuul-jobs] 908671: Remove command.warn usage https://review.opendev.org/c/zuul/zuul-jobs/+/908671 | 18:21 | |
| @clarkb:matrix.org | I've rebased as I expect things to pass now except for openshift. But I half suspect we may need to disable openshift testing which can happen in a change outside this stack | 18:22 |
| @jim:acmegating.com | Clark: yes i think that's the case and we removed testing from nodepool because of it in https://review.opendev.org/899308 so i think that's the right course here | 18:34 |
| @clarkb:matrix.org | ok I'll work on that change too then | 18:34 |
| @clarkb:matrix.org | corvus: to do that I have to edit the same file with the other registry tests in it. Zuul will run all of the tests in that file right? Effectively meaning I need to squash these updates together? | 18:36 |
| @clarkb:matrix.org | actually zuul should know when a specific test changes and only run that subset. I'll try stacking it more normally | 18:39 |
| -@gerrit:opendev.org- Clark Boylan proposed: | 18:40 | |
| - [zuul/zuul-jobs] 909029: Override DOCKER_MIN_API_VERSION for skopeo when installing docker https://review.opendev.org/c/zuul/zuul-jobs/+/909029 | ||
| - [zuul/zuul-jobs] 909045: Remove openshift + zuul registry testing https://review.opendev.org/c/zuul/zuul-jobs/+/909045 | ||
| -@gerrit:opendev.org- Clark Boylan proposed on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/zuul-jobs] 908671: Remove command.warn usage https://review.opendev.org/c/zuul/zuul-jobs/+/908671 | 18:40 | |
| -@gerrit:opendev.org- Clark Boylan proposed: | 18:43 | |
| - [zuul/zuul-jobs] 909045: Remove openshift + zuul registry testing https://review.opendev.org/c/zuul/zuul-jobs/+/909045 | ||
| - [zuul/zuul-jobs] 909029: Override DOCKER_MIN_API_VERSION for skopeo when installing docker https://review.opendev.org/c/zuul/zuul-jobs/+/909029 | ||
| -@gerrit:opendev.org- Clark Boylan proposed on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/zuul-jobs] 908671: Remove command.warn usage https://review.opendev.org/c/zuul/zuul-jobs/+/908671 | 18:43 | |
| @clarkb:matrix.org | That almost mostly worked. I had to ninja it a bit. I hope that stack is mergable | 18:43 |
| @clarkb:matrix.org | corvus: when I was working to set up a held node I think I realized that the zk db clearing for the circular dep upgrade also cleared out the autohold entries. Should we make note of that in the upgrade notes for zuul 10.0.0? | 18:55 |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul] 909048: Enable TCP keepalives in zuul_stream https://review.opendev.org/c/zuul/zuul/+/909048 | 19:15 | |
| @jim:acmegating.com | Clark: i don't feel like that release note is strictly necessary (hopefully it's clear to operators that's part of the system state that's deleted when they run the "delete-state" command) but i don't object. | 19:18 |
| @clarkb:matrix.org | ya I can go either way on it. I think it has surprised people in the past, but it also makes sense when youconsider what happens to the databse when you clear it | 19:19 |
| @jim:acmegating.com | Clark: i think the confusion you highlighted in opendev is a good point; that it deletes the zuul side but not the nodepool side. i look forward to there not being two sides. :) | 19:20 |
| @jim:acmegating.com | Clark: thx for the container changes; lgtm! | 19:22 |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul] 909053: Increase timeouts in TestComponentRegistry (#2) https://review.opendev.org/c/zuul/zuul/+/909053 | 21:00 | |
| @jim:acmegating.com | Clark: fungi ^ i think we should merge that now on the same basis we did https://review.opendev.org/908834 | 21:01 |
| @fungicide:matrix.org | thanks! | 21:12 |
| @jim:acmegating.com | Two things about the the command.warn change: | 21:12 |
| * centos and bionic base roles jobs both failed with `stestr: 'init' is not a stestr command. See 'stestr --help'.` does that sound familiar to anyone? i'm holding nodes to inspect the error further. | ||
| * we need to remove the command.warn settings because they are invalid in ansible 9, but doing so causes ansible-lint errors. oy. | ||
| @jim:acmegating.com | (i think those are the only things left on that punch list) | 21:13 |
| @fungicide:matrix.org | that stestr error doesn't look familiar. have a link to the failing build? | 21:14 |
| @clarkb:matrix.org | Looking at tox files for projects I think you can just run `stester run` without an init stepfirst | 21:14 |
| @jim:acmegating.com | fungi: https://zuul.opendev.org/t/zuul/build/82c76e50ac3847b0bcaeeef46f878472 but skip to " | 21:14 |
| @clarkb:matrix.org | I'm not sure why it would work on some platforms not but others. maybe we can just drop the init line though | 21:15 |
| @jim:acmegating.com | * fungi: https://zuul.opendev.org/t/zuul/build/82c76e50ac3847b0bcaeeef46f878472/console#1/10/13/ubuntu-bionic | 21:15 |
| @jim:acmegating.com | well it also fails stestr run | 21:15 |
| @jim:acmegating.com | like it's a wrong stestr executable or something | 21:15 |
| @fungicide:matrix.org | huh, okay so the task literally runs `stestr init` | 21:16 |
| @fungicide:matrix.org | maybe it's installing a very old stestr due to python version and pip resolver shenanigans | 21:16 |
| @jim:acmegating.com | (and be careful; that task is actually expected to fail, but not like that; the actual job failure is the lack of a subunit file because that failed) | 21:16 |
| @jim:acmegating.com | i wondered that too, but scrolling up a few tasks it looks like it installs the same version as in working jobs | 21:17 |
| @jim:acmegating.com | here's centos8 from the same buildset https://zuul.opendev.org/t/zuul/build/d0788a8a69cc41eca8eae97f937149c1/console | 21:17 |
| @fungicide:matrix.org | which would be the commonality with centos 8 and ubuntu bionic: old python interpreters | 21:17 |
| @jim:acmegating.com | centos 7 is the failing one but yeah | 21:18 |
| @fungicide:matrix.org | oh, 7, so even older yes | 21:18 |
| @jim:acmegating.com | stestr-4.1.0 in all cases i think | 21:18 |
| @fungicide:matrix.org | oh, so i guess those platforms do still meet the python>=3.6 requirement. 3.2.1 was the last stestr to support python 3.5 fwiw | 21:19 |
| @clarkb:matrix.org | I wonder if entrypoints are involved | 21:24 |
| @clarkb:matrix.org | maybe pkg_resources can't load an entrypoint because of python incompatibility | 21:24 |
| @fungicide:matrix.org | that's where i was headed too. like something's breaking entrypoint inspection | 21:25 |
| @fungicide:matrix.org | importlib-metadata-4.8.3 and stevedore-3.5.2 (maybe also pbr-6.0.0) could be involved in the entrypoint equation, also cliff-3.10.1 from a command-line syntax parsing perspective | 21:31 |
| @fungicide:matrix.org | those were the versions the previous task said were installed along with stestr anyway | 21:32 |
| -@gerrit:opendev.org- Zuul merged on behalf of Jeremy Stanley https://matrix.to/#/@fungicide:matrix.org: | 21:36 | |
| - [zuul/zuul] 908869: Test with Keycloak 23 https://review.opendev.org/c/zuul/zuul/+/908869 | ||
| - [zuul/zuul] 908855: Refresh Keycloak OpenID Howto for latest Keycloak https://review.opendev.org/c/zuul/zuul/+/908855 | ||
| @fungicide:matrix.org | looks like the last successful run of zuul-jobs-test-base-roles-ubuntu-bionic was over 4 months ago, beginning of october | 21:36 |
| @fungicide:matrix.org | same for zuul-jobs-test-base-roles-centos-7 | 21:37 |
| @clarkb:matrix.org | which is after stestr 4.1.0 released. So ya maybe in the dep tree rather than stestr itself | 21:38 |
| @fungicide:matrix.org | also worth calling out, the last setuptools before the bumped the requires_python to >=3.7 was 59.6.0 | 21:43 |
| @fungicide:matrix.org | * also worth calling out, the last setuptools before they bumped the requires\_python to >=3.7 was 59.6.0 | 21:43 |
| @jim:acmegating.com | okay i think i have the lint fix prepared. so just the stestr mystery remains | 21:45 |
| -@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/zuul] 909053: Increase timeouts in TestComponentRegistry (#2) https://review.opendev.org/c/zuul/zuul/+/909053 | 21:50 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul-jobs] 908671: Remove command.warn usage https://review.opendev.org/c/zuul/zuul-jobs/+/908671 | 21:52 | |
| @jim:acmegating.com | that adds the lint fixes and should trigger the autoholds so we can poke at the stestr thing | 21:52 |
| @fungicide:matrix.org | yeah, without access to (a copy of) the venv it used, i'm at a loss to speculate usefully | 21:54 |
| @jim:acmegating.com | fungi: Clark https://pypi.org/project/voluptuous/#history it's voluptuous 0.14.0 | 22:45 |
| @jim:acmegating.com | needs to be 0.13.1 | 22:45 |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul-jobs] 909060: fix old stestr test https://review.opendev.org/c/zuul/zuul-jobs/+/909060 | 22:51 | |
| @fungicide:matrix.org | wow! | 22:51 |
| @fungicide:matrix.org | i would never have guessed it was in any way related | 22:51 |
| @jim:acmegating.com | yeah, i don't know the mechanics of that but i bet they're fascinating! | 22:52 |
| @fungicide:matrix.org | it's also showing as yanked, so i wonder how we're still installing it | 22:53 |
| @clarkb:matrix.org | too old pip to handle yanking? | 22:54 |
| @fungicide:matrix.org | yes | 22:54 |
| @fungicide:matrix.org | i bet that's it | 22:54 |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul-jobs] 908671: Remove command.warn usage https://review.opendev.org/c/zuul/zuul-jobs/+/908671 | 23:19 | |
| @jim:acmegating.com | fingers crossed! ^ | 23:20 |
| @fungicide:matrix.org | for confirmation, pip 19.2 added support for detecting yanked versions: https://pip.pypa.io/en/stable/news/#v19-2 | 23:47 |
| @fungicide:matrix.org | which works on python 2.7 and 3.5+: https://pypi.org/project/pip/19.2/ | 23:47 |
| @fungicide:matrix.org | but maybe we used older pip anyway? | 23:48 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!