| *** tkajinam is now known as Guest1349 | 01:24 | |
| tkajinam | it's quite fun that people do not pay effort to fixing problems but insist that we should pay more effort so that they are not annoyed :-D | 01:46 |
|---|---|---|
| frickler | tkajinam: fun or not so fun, yes. thanks for all the effort you are putting into this | 07:05 |
| frickler | regarding fixing problems, the feature/nats-driver branch for oslo.messaging needs to either get its zuul config repaired or get removed, is it still needed or some old relict? https://zuul.opendev.org/t/openstack/config-errors?project=openstack%2Foslo.messaging&skip=0 | 07:07 |
| tkajinam | frickler, thank you, too. | 07:08 |
| tkajinam | frickler, let me look into it... I found the same problem in telemetry dsvm jobs and we have to fix it as well. | 07:08 |
| tkajinam | tobias-urdin, I wonder if you have any idea about the features/nats-driver branch ? I remember there were some discussions about NATS driver but I didn't follow all details really | 07:09 |
| frickler | tkajinam: yes, this is indeed the same issue that you found. so likely can be fixed by a simple patch and we can defer discussing branch deletion then | 07:14 |
| tkajinam | yeah I hope so | 07:34 |
| tkajinam | ugh. dib-utils is still required by os-refresh-config... | 07:59 |
| tkajinam | we can probably import the tiny script instead of restoring the repo. it's also used in sahara image elements but idk how we may update these | 07:59 |
| tkajinam | https://review.opendev.org/q/topic:%22dib-utils-removal%22 | 08:10 |
| mnasiadka | Hi there - I'm not at all knowledgable in oslo.db - trying to remove autocommit in Magnum in https://review.opendev.org/c/openstack/magnum/+/910512 - I'm not entirely sure the approach is correct, so would like a second pair of eyes ;-) | 08:22 |
| *** elodilles_pto is now known as elodilles | 08:29 | |
| frickler | probably stephenfin's eyes would work best there ^^ | 08:58 |
| tobias-urdin | tkajinam: that's a very old branch, the first implementation that was years before my effort | 09:22 |
| tobias-urdin | that can probably be retired | 09:22 |
| tkajinam | yeah the last update was made in 2019 and it still tires to run py27 job. | 09:26 |
| tkajinam | mnasiadka, yeah stephenfin is the best person to consult... my understanding is that you have to ensure all update queries are executed in sessions so the change seems reasonable though I don't know if that is enough | 09:38 |
| mnasiadka | so I wait for stephenfin ;-) | 09:55 |
| jamespage | JayF: https://review.opendev.org/c/openstack/oslo.service/+/910502 - based on some rework I did in the magnumclient for the same challenge (but client rather than server) | 10:37 |
| jamespage | I think that resolves 3.12 compat | 10:37 |
| *** tosky_ is now known as tosky | 11:48 | |
| opendevreview | Matthew Watkins proposed openstack/stevedore master: Add missing/undeclared setuptools dependency https://review.opendev.org/c/openstack/stevedore/+/910626 | 13:47 |
| *** ralonsoh_ is now known as ralonsoh | 13:59 | |
| frickler | tkajinam: so according to this, there was never any work done in that branch. if as release liaison you confirm, I would check what needs to be done to delete it https://review.opendev.org/q/project:openstack/oslo.messaging+branch:feature/nats-driver | 14:05 |
| tkajinam | frickler, yeah I confirm that this branch has only changes for branch cutoff from 10.4.0 is just useless | 14:14 |
| tkajinam | Update .gitreview for feature/nats-driver | 14:15 |
| tkajinam | Update TOX/UPPER_CONSTRAINTS_FILE for feature/nats-driver | 14:15 |
| tkajinam | these from release bot are the only diff from 10.4.0 it seems | 14:15 |
| tkajinam | that's a very good point | 14:15 |
| damani | hi | 14:20 |
| damani | tkajinam, do you think it's relevant https://bugs.launchpad.net/pbr/+bug/1494524 | 14:20 |
| tkajinam | damani, maybe, but I think it's eventually done when we get rid of wsgi_script entry point usage to support newer setuptools | 14:25 |
| tkajinam | once that work is done then we no longer use that code to generate wsgi script by pbr | 14:25 |
| damani | tkajinam, so we can can probably close it with a comment | 15:44 |
| stephenfin | mnasiadka: That looks too simple, but if it's passing CI and you have decent coverage then that might be all that's needed? The passing job is a good sign, though it could simply be that you don't have DB coverage | 20:27 |
| stephenfin | mnasiadka: tbh, my recommendation would be to deploy a DevStack, apply that change, bump SQLAlchemy and run your integration testsuite. You could probably do that in CI but idk how much harder that will be | 20:28 |
| *** jph6 is now known as jph | 22:24 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!