| opendevreview | OpenStack Release Bot proposed openstack/devstack-plugin-ceph stable/2023.2: Update .gitreview for stable/2023.2 https://review.opendev.org/c/openstack/devstack-plugin-ceph/+/895085 | 01:34 |
|---|---|---|
| opendevreview | OpenStack Release Bot proposed openstack/devstack-plugin-ceph stable/2023.2: Update TOX_CONSTRAINTS_FILE for stable/2023.2 https://review.opendev.org/c/openstack/devstack-plugin-ceph/+/895086 | 01:35 |
| opendevreview | Merged openstack/devstack-plugin-ceph stable/2023.2: Update .gitreview for stable/2023.2 https://review.opendev.org/c/openstack/devstack-plugin-ceph/+/895085 | 02:35 |
| opendevreview | Merged openstack/devstack-plugin-ceph stable/2023.2: Update TOX_CONSTRAINTS_FILE for stable/2023.2 https://review.opendev.org/c/openstack/devstack-plugin-ceph/+/895086 | 02:41 |
| *** osmanlicilegi is now known as Guest0 | 04:33 | |
| opendevreview | Jakub Skunda proposed openstack/tempest master: Remove test duplication between tempest and n-t-p QuotasTest https://review.opendev.org/c/openstack/tempest/+/892677 | 08:58 |
| sean-k-mooney | gmann: i have a question/objection i woudl like to raice for the 2024.1 testing runtime | 10:36 |
| sean-k-mooney | why are there refence to ubuntu 20.04/python 3.8 in that docs | 10:36 |
| sean-k-mooney | https://github.com/openstack/governance/blob/master/reference/runtimes/2024.1.rst | 10:36 |
| sean-k-mooney | 20.04 is not expected to be supported in caracal | 10:37 |
| sean-k-mooney | its last realeas was antilope and it has no supprot in bobcat | 10:37 |
| sean-k-mooney | i assume this is a bug so i was going to submit a patch to fix it | 10:38 |
| sean-k-mooney | do you know if it was included intentionally because i really dont think its approciate to incldue it | 10:38 |
| sean-k-mooney | i filed https://review.opendev.org/c/openstack/governance/+/895160 to correct what i think is an error but i am wonderign why this was not cuaght in review | 11:00 |
| sean-k-mooney | frickler: ^ | 11:02 |
| frickler | sean-k-mooney: that's essentially a tc topic, not qa. one could argue the same about python 3.9 possibly | 11:05 |
| sean-k-mooney | well python 3.9 was a testign runtime for centos 9 last cycle | 11:08 |
| sean-k-mooney | baiscally i want to remvoe the py3.8 job in caracal | 11:08 |
| sean-k-mooney | so im not partically happy to see it reinotduced | 11:08 |
| sean-k-mooney | when it was not requried to be tested at all in bobcat | 11:09 |
| frickler | yeah, I understand that, it's just nothing the QA team can do anything about | 11:11 |
| sean-k-mooney | frickler: you are on the tc right? would you mind bring this up in the next tc meeting | 11:11 |
| sean-k-mooney | well i was pinging gmann because they wrote the patch and then motivation was to remove the py38 jobs | 11:12 |
| sean-k-mooney | and ubuntu 20.04 jobs | 11:12 |
| frickler | sean-k-mooney: I'm just a candidate so far. I can still do that, but I assume your patch proposal will be on the agenda anyway | 11:12 |
| frickler | and yes, gmann is wearing multiple hats ;) | 11:13 |
| sean-k-mooney | am well dont go to any trouble i jsut dont normally attend tc calls | 11:14 |
| sean-k-mooney | but ya it will proably come up organically | 11:14 |
| dtantsur | A nasty problem in grenade: https://a86af8e24720e7d1aa6e-c7277a3e95ece459b8b2ecfff9ffad89.ssl.cf5.rackcdn.com/894015/1/check/ironic-inspector-grenade/1b5ab34/controller/logs/grenade.sh_log.txt | 11:56 |
| dtantsur | TL;DR the Nova's resource code expects Cirros version from new (2023.2) devstack (0.6.2), while the image is uploaded using the code from old devstack (2023.1, with cirros 0.5.2) | 11:56 |
| dtantsur | I guess we could override the version in our jobs... | 11:56 |
| sean-k-mooney | personly i alwasy have wanted to rename the image to just cirros | 12:02 |
| sean-k-mooney | so that it would not change when we change the version | 12:02 |
| sean-k-mooney | but you could jsut upload both images in the grenade job | 12:02 |
| sean-k-mooney | devstack can take multipel images in its config | 12:03 |
| sean-k-mooney | so i woudl just endure both are in the list as part fo the iniall install | 12:03 |
| sean-k-mooney | i think you would just need to add IMAGE_URLS+="https://download.cirros-cloud.net/0.6.2/cirros-0.6.2-${CIRROS_ARCH}-disk.img" | 12:07 |
| sean-k-mooney | in the local.conf | 12:07 |
| opendevreview | Rodolfo Alonso proposed openstack/tempest master: Mark "test_live_migration_with_trunk" as unstable https://review.opendev.org/c/openstack/tempest/+/895167 | 13:04 |
| *** tosky_ is now known as tosky | 14:06 | |
| opendevreview | Luigi Toscano proposed openstack/tempest master: DNM test rbd retype https://review.opendev.org/c/openstack/tempest/+/890360 | 14:26 |
| opendevreview | Goutham Pacha Ravi proposed openstack/tempest master: Revert "Revert "Mark "test_live_migration_with_trunk" as unstable"" https://review.opendev.org/c/openstack/tempest/+/895028 | 15:13 |
| gmann | sean-k-mooney: np! to ping me anywhere :) I replied on the gerrit. py3.8 testing is intentionally kept as we do not want to bump the python min version unless it is EOL or cannot be tested | 16:21 |
| gmann | sean-k-mooney: for ubuntu 20.04 ref, that is just a info on from where we can test python3.8 so it is not we do test/support ubuntu 20.04 in 2024.1 cycle | 16:21 |
| sean-k-mooney | ok cauuse i kind of want ot bump the min python version | 16:22 |
| sean-k-mooney | i guess that coudl be a ptg topic. i care leass baout the 3.8 to 3.9 change | 16:22 |
| gmann | yeah, we faced the issue in 2023.2 for that and explicitly made clear about it in PTI also https://review.opendev.org/c/openstack/governance/+/882154 | 16:22 |
| gmann | sean-k-mooney: we dropped that and added back after finding the issue. ^^ | 16:23 |
| sean-k-mooney | so can we drop all py38 testing on master/caracal? | 16:23 |
| gmann | we do not want to until py3.8 is EOL or there is no way to test it | 16:24 |
| gmann | it is minimum testing expectation just unit test job will make sure code are compatible so no big cost to keep them | 16:24 |
| sean-k-mooney | ... | 16:24 |
| sean-k-mooney | im not sure that is a good thing but ok that woudl be drop it in 2024.2 | 16:25 |
| gmann | that is how we can give more compatibility to old and new distro | 16:25 |
| sean-k-mooney | its eol 2024-10 | 16:25 |
| sean-k-mooney | the problem is we dont expect to support ubuntu 20.04 with master | 16:25 |
| sean-k-mooney | its not the most recent lts | 16:25 |
| gmann | distro version and pythin version are kept separate and pythin version is soemthing we want to suppor as mush as possible | 16:26 |
| sean-k-mooney | thats prblmeantic based on the fact that we are also trying to supprot 3.11 adn soone 3.12 | 16:26 |
| gmann | which is ok, if we can support as much as possible so that operator with old distro and new openstack can also work at least does not block them for python version compatibility | 16:27 |
| sean-k-mooney | well we are not oging to be abel to test it with tempest anymore | 16:29 |
| sean-k-mooney | unless we use a pyenv of something to install an older python with newer libvirt/qemu | 16:29 |
| sean-k-mooney | i guess we can hold off dropting the tox jobs until 2024.2 | 16:30 |
| sean-k-mooney | it was one of the first things i planed to do once master repoended | 16:30 |
| gmann | yeah, tox job at least make sure code is compatible | 16:30 |
| sean-k-mooney | im still not sold that that is a good idea by the way | 16:31 |
| gmann | and at least setup.cfg does not block installation on that which happended in this cycle | 16:31 |
| sean-k-mooney | but ok | 16:31 |
| sean-k-mooney | specific i dotn think we shoudl be encuriging peopel to dpeloy new openstack on old distors | 16:31 |
| sean-k-mooney | to mee this feels a bit liek the extended maintaince discusion | 16:32 |
| sean-k-mooney | we should not block it but i dont think we shoudl really imply its supported | 16:32 |
| sean-k-mooney | that is a wider comunity discussion but i dont want to get into a situraiton where our thrid party libs start droping py38 suport | 16:33 |
| sean-k-mooney | and we cant bump uc because of that | 16:34 |
| sean-k-mooney | i assume that is why the upstream eol timeframe was selected | 16:35 |
| gmann | we will see the situation and if something deps or so make us not tested on py3.8 or so then we can always think or that | 16:35 |
| sean-k-mooney | the other reason i wnated to dro py 38 this cycle by the way | 16:35 |
| sean-k-mooney | is i wanted it to stilil be supprot upstream for the lenght of the stable branch | 16:35 |
| sean-k-mooney | so i woudl have prefered if we required a python to have 18 months of upstream support till its eol to be in the tested runtimes | 16:36 |
| sean-k-mooney | distros will supprot it longer but its still fagile in my view | 16:37 |
| gmann | it will be very less chance that if any backport will cause issue on py3.8 if those are tested on py3.9 or 3.10 in master in future | 16:41 |
| gmann | and it is not difficult to keep python code compatible if we find such cases. | 16:42 |
| gmann | py2.7 drop was the same case for all stable branches kept py2.7 but not master | 16:42 |
| sean-k-mooney | ya i guess the disconnect is iw wanted the supproted pythons to be supproted for the full supproted lifecylce of the branch | 16:42 |
| sean-k-mooney | not just when its released | 16:42 |
| sean-k-mooney | which is why i didnt what to support py38 in bobcat of caracal | 16:43 |
| gmann | kopecmartin: I am ok with this but we need to make sure we keep tempest tests as much as possible and cleanup duplicate tests on plugin side https://review.opendev.org/c/openstack/tempest/+/892677?tab=comments | 16:43 |
| -opendevstatus- NOTICE: The lists.airshipit.org and lists.katacontainers.io sites will be offline briefly for migration to a new server | 16:48 | |
| opendevreview | Merged openstack/tempest master: Mark "test_live_migration_with_trunk" as unstable https://review.opendev.org/c/openstack/tempest/+/895167 | 19:32 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!