| ianw | i'd like to sort out the python 3.6 things before a release | 00:16 |
|---|---|---|
| ianw | https://bugs.launchpad.net/tripleo/+bug/1977888 | 00:17 |
| opendevreview | Ian Wienand proposed openstack/diskimage-builder master: [wip] test f36 builds https://review.opendev.org/c/openstack/diskimage-builder/+/844912 | 01:35 |
| opendevreview | Ian Wienand proposed openstack/diskimage-builder master: [wip] test f36 builds https://review.opendev.org/c/openstack/diskimage-builder/+/844912 | 02:01 |
| opendevreview | Ian Wienand proposed openstack/diskimage-builder master: [wip] test f36 builds https://review.opendev.org/c/openstack/diskimage-builder/+/844912 | 04:39 |
| opendevreview | Ian Wienand proposed openstack/diskimage-builder master: CentOS 9-stream : work around selinux permissions issue https://review.opendev.org/c/openstack/diskimage-builder/+/845189 | 04:39 |
| frickler | kolla with bifrost/kayobe is in a similar situation. IMO we can readd the py3.6+7 classifiers, even if we don't run test jobs | 05:09 |
| opendevreview | Dr. Jens Harbott proposed openstack/diskimage-builder master: Revert "Remove py 3.6 support and update jobs" https://review.opendev.org/c/openstack/diskimage-builder/+/845217 | 05:16 |
| opendevreview | Dr. Jens Harbott proposed openstack/diskimage-builder master: Update jobs to use the latest openstack template https://review.opendev.org/c/openstack/diskimage-builder/+/845218 | 05:16 |
| frickler | ^^ I'm thinking a full revert and then redoing the jobs part is cleaner, let me know if you want to merge these into a partial revert instead | 05:17 |
| opendevreview | Ian Wienand proposed openstack/diskimage-builder master: CentOS 9-stream : work around selinux permissions issue https://review.opendev.org/c/openstack/diskimage-builder/+/845189 | 06:34 |
| opendevreview | Ian Wienand proposed openstack/diskimage-builder master: [wip] test f36 builds https://review.opendev.org/c/openstack/diskimage-builder/+/844912 | 06:34 |
| ianw | frickler: why don't we just add a python3.6 job too? | 06:35 |
| ianw | i feel like it's just a matter of time till we put in some non-3.6 syntax and then have to emergency release a fix when we don't see it in the gate | 06:35 |
| frickler | ianw: istr py36 was failing, but I can check that | 06:41 |
| ianw | right, cause the requirements were pulling in a newer networkx | 06:42 |
| ianw | but i do wonder why we run these with openstack requirements | 06:42 |
| ianw | seems like it's just always been the openstack-python jobs ... I235a0584d2b0156406b2d07848d484a9148d064e is where they were moved in-repo | 06:44 |
| frickler | maybe we can simply keep running py36 with yoga reqs? should be easy to add an override-checkout | 06:47 |
| frickler | with the py36 failure I need to squash my revert anyway. | 06:49 |
| frickler | maybe finish that partial revert first to unblock ooo and kolla? and have readding the py36 job in a second patch then? | 06:50 |
| ianw | i've made a comment in 845218. i just wonder if we're being a bit, for the want of a better term, masochistic by trying to limit ourselves around the requirements that nobody really cares about? | 06:52 |
| ianw | i'm not sure if we've actually discussed it before, or just been moving between openstack-* tox jobs because that's how it's always been | 06:52 |
| opendevreview | Dr. Jens Harbott proposed openstack/diskimage-builder master: Revert "Remove py 3.6 support and update jobs" https://review.opendev.org/c/openstack/diskimage-builder/+/845217 | 06:53 |
| frickler | ianw: I'm not sure how bifrost installs dib, but yes, using the openstack templates was likely mostly done because it was easy | 06:54 |
| frickler | maybe dtantsur or TheJulia can say more about bifrost | 06:56 |
| ianw | i don't want to be needlessly obstructionist about it, but it feels like we are much more likely to use string formatting or something that breaks python 3.6 than introduce an incompatible requirement | 06:58 |
| ianw | i don't think any of us want to bother with a branching model, and trying to decide what goes into backports etc. | 06:59 |
| frickler | yes, I agree with that | 07:00 |
| frickler | also I just saw that we won't be able to merge anything currently as long as cs9 builds are broken | 07:00 |
| frickler | unless we want to make them non-voting for now? | 07:01 |
| opendevreview | Dr. Jens Harbott proposed openstack/diskimage-builder master: Test using plain python jobs instead of openstack https://review.opendev.org/c/openstack/diskimage-builder/+/845218 | 07:06 |
| frickler | ^^ guess I'll iterate on that a bit | 07:07 |
| opendevreview | Dr. Jens Harbott proposed openstack/diskimage-builder master: Test using plain python jobs instead of openstack https://review.opendev.org/c/openstack/diskimage-builder/+/845218 | 07:33 |
| opendevreview | Dr. Jens Harbott proposed openstack/diskimage-builder master: Test using plain python jobs instead of openstack https://review.opendev.org/c/openstack/diskimage-builder/+/845218 | 07:42 |
| frickler | ianw: seems there was a similar issue some cycles ago already, maybe just cargo-cult that? https://review.opendev.org/c/openstack/diskimage-builder/+/845218/4/tox.ini | 07:43 |
| ianw | frickler: yeah, i mean it's an option i guess. but if we can just use the regular jobs to make it clear what's going on i think that might be best? | 08:13 |
| ianw | if we can confirm with bifrost i think that might be enough convincing. certainly nodepool doesn't care | 08:15 |
| opendevreview | Dr. Jens Harbott proposed openstack/diskimage-builder master: Test using plain python jobs instead of openstack https://review.opendev.org/c/openstack/diskimage-builder/+/845218 | 08:29 |
| frickler | ianw: ack, ^^ is testing everything without constraints now | 08:29 |
| opendevreview | Ronelle Landy proposed openstack/diskimage-builder master: DNM: Add seinux files - work around selinux https://review.opendev.org/c/openstack/diskimage-builder/+/845288 | 15:20 |
| opendevreview | dasm proposed openstack/diskimage-builder master: DNM: Add seinux files - work around selinux https://review.opendev.org/c/openstack/diskimage-builder/+/845288 | 16:25 |
| opendevreview | dasm proposed openstack/diskimage-builder master: CentOS 9-stream : work around selinux permissions issue https://review.opendev.org/c/openstack/diskimage-builder/+/845189 | 17:01 |
| opendevreview | dasm proposed openstack/diskimage-builder master: CentOS 9-stream : work around selinux permissions issue https://review.opendev.org/c/openstack/diskimage-builder/+/845189 | 17:39 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!