| opendevreview | Pranali Deore proposed openstack/tempest master: Add tests for new add location api of glance https://review.opendev.org/c/openstack/tempest/+/891026 | 06:23 |
|---|---|---|
| dansmith | Are others seeing this failure? https://zuul.opendev.org/t/openstack/build/c4563b7d3246465389367013d1bef7ad | 13:43 |
| dansmith | happened to me a few times yesterday and surely looks unrelated to my oslo change | 13:43 |
| dansmith | opensearch says there were 64 hits suddenly yesterday | 13:43 |
| dansmith | none since yesterday though so I guess I'll try a recheck | 13:45 |
| dansmith | (another one) | 13:45 |
| dansmith | nope, failed again | 13:55 |
| dansmith | hmm, the patch below it seems fine | 14:41 |
| dansmith | frickler: gmann: I could use some help.. I guess I'm doing something in my oslo patches that is breaking the swift install, but I have no idea what that would be | 14:42 |
| dansmith | this one works fine (recheck is working currently): https://review.opendev.org/c/openstack/oslo.utils/+/924369/1 | 14:42 |
| dansmith | but the next one fails to install swift, I guess because swift has some dependency on oslo.utils, but I don't see what in that patch could be breaking it | 14:43 |
| frickler | dansmith: at first sight, I have no idea. did you try to reproduce this locally? I can hold a node for further debugging otherwise | 15:07 |
| dansmith | I'm working on trying to do it now, yeah | 15:07 |
| frickler | other than that, my only idea is to break your patch into pieces in order to try to better locate the issue | 15:08 |
| dansmith | it's literally modifying a new file that was only adding a few patches ago | 15:09 |
| frickler | I'm setting up a hold now anyway, don't have a jammy devstack handy currently | 15:09 |
| frickler | I don't see the issue on 24.04 with a simple install test | 15:09 |
| dansmith | there were also 64 hits yesterday in opensearch but a bunch were retries, I guess I should see if they were on any other patches | 15:50 |
| dansmith | frickler: here's an ironic example, so obviously unrelated to the oslo change I think: https://zuul.opendev.org/t/openstack/build/00e29c08ae5d43b191438655a80980a3 | 15:52 |
| frickler | dansmith: o.k., so something completely different likely | 15:54 |
| dansmith | hopefully for me :) | 15:55 |
| frickler | dansmith: https://github.com/pypa/setuptools/issues/4480, seems setuptools just yanked a release | 15:56 |
| dansmith | fun | 15:56 |
| frickler | also 71.0.0 came out yesterday, maybe another, but related bug | 15:57 |
| frickler | clarkb: fungi: ^^ just in case more weird failures pop up | 15:57 |
| frickler | see also https://pypi.org/project/setuptools/71.0.2/#history | 15:58 |
| fungi | usually we don't see issues from new setuptools until it gets vendored in a new release of virtualenv | 16:01 |
| frickler | devstack does " pip_install -U pip setuptools" for the venv. the failing ironic job pulled 71.0.0 in | 16:03 |
| fungi | aha | 16:03 |
| fungi | yeah i guess that'd do it | 16:03 |
| fungi | looking at the issues list, seems like setuptools vendoring a copy of importlib.metadata is problematic now | 16:05 |
| fungi | also importing pkg_resources looks to be problematic now | 16:07 |
| frickler | dansmith: fyi the recheck passed the installation stage with setuptools 71.0.2 | 16:24 |
| dansmith | oh sure, because you have the special "if user == frickler: succeed()" condition I'm _sure_ exists | 16:25 |
| dansmith | (thanks :)) | 16:25 |
| clarkb | fungi: problematic in what sense? Like the official stance is to not do it anymore? That seems like it will be painful especially if importlib isn't working | 16:34 |
| fungi | clarkb: looks like probably a regression: https://github.com/pypa/setuptools/issues/4487 | 17:09 |
| clarkb | phew | 17:10 |
| *** elodilles is now known as elodilles_ooo | 17:10 | |
| opendevreview | Ghanshyam proposed openstack/tempest master: DNM: testing oslo policy RBAC default change https://review.opendev.org/c/openstack/tempest/+/924435 | 19:29 |
| opendevreview | Merged openstack/devstack-plugin-ceph master: Fix ingress deamon https://review.opendev.org/c/openstack/devstack-plugin-ceph/+/921389 | 20:50 |
| opendevreview | melanie witt proposed openstack/devstack master: nova: Add [keystone] config for unified limits https://review.opendev.org/c/openstack/devstack/+/924453 | 23:38 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!