| opendevreview | OpenStack Proposal Bot proposed openstack/glance master: Imported Translations from Zanata https://review.opendev.org/c/openstack/glance/+/871930 | 03:34 |
|---|---|---|
| *** bhagyashris|ruck is now known as bhagyashris | 07:47 | |
| opendevreview | Guillaume Espanel proposed openstack/glance master: Limit CaptureRegion sizes in format_inspector for VMDK and VHDX https://review.opendev.org/c/openstack/glance/+/871831 | 11:24 |
| croelandt | so who could help with the Grenade issues on the CI? | 12:12 |
| croelandt | I'm not sure what that job does, where it's defined, why we lack rustc on the machines | 12:12 |
| croelandt | I wonder who understand the whole thing | 12:12 |
| jokke_ | croelandt: IIRC Grenade tests are the upgrade test, so it install current -1 and tries to upgrade it to current and IIRC it runs tempest against them | 12:16 |
| jokke_ | croelandt: what is it breaking on | 12:16 |
| croelandt | the CVE :) | 12:16 |
| croelandt | https://review.opendev.org/q/Idf561f6306cebf756c787d8eefdc452ce44bd5e0 | 12:16 |
| croelandt | I think all 3 failing backports are grenade-related | 12:17 |
| whoami-rajat | croelandt, for EM branches, here is our solution https://review.opendev.org/c/openstack/cinder/+/871790 | 12:19 |
| croelandt | hm | 12:20 |
| croelandt | jokke_: would you be fine moving the grenade jobs to experimental for train, ussuri and wallaby ? | 12:21 |
| *** tosky_ is now known as tosky | 12:55 | |
| jokke_ | croelandt: I don't think so. We rather need to fix the backports or just not apply them that far back those branches are EM anyways | 13:48 |
| rosmaita | croelandt: in cinder, we are hitting tempest problems (can't build bcrypt because Rust is missing) in ussuri and older ... are you seeing that too? | 14:15 |
| croelandt | rosmaita: yeah | 14:41 |
| croelandt | we have those because of Rust as well | 14:41 |
| croelandt | or the lack thereof | 14:41 |
| croelandt | rosmaita: I honestly don't really understand how the machines are provisioned | 14:44 |
| rosmaita | i was afraid you would say that | 14:45 |
| rosmaita | what pisses me off is that i pushed a patch to the cinder stable branches the week before the CVE was announced, and everything was fine | 14:45 |
| rosmaita | so suddenly a week later, all the stuff is broken | 14:46 |
| croelandt | yeah and btw when people tell you we got reproducible envs | 14:46 |
| croelandt | it's utter bs | 14:46 |
| croelandt | like I could not reproduce a pep8 issue because of dependencies | 14:46 |
| croelandt | and had to do it "blindly" instead | 14:46 |
| croelandt | Nix would get us actual reproducible builds | 14:47 |
| croelandt | jokke_: I feel like the CI should run at least once a day so we can pinpoint that shit more accurately | 14:47 |
| rosmaita | well, we could add a periodic stable tempest job | 14:47 |
| rosmaita | the default periodics are only unit tests and pep8 | 14:47 |
| rosmaita | maybe functional, but i don't think so | 14:48 |
| croelandt | rosmaita: how do I see the results of those? | 14:48 |
| croelandt | We had issues because of Python 3.10 *and* tox 4 at the same time | 14:48 |
| croelandt | and no one freaking noticed before it broke actual patches | 14:48 |
| rosmaita | 2 ways: there's a mailing list you can subscribe to, and there's a zuul query | 14:48 |
| rosmaita | gimme a sec and i will find it | 14:48 |
| croelandt | damn it how on Earth did I miss that | 14:49 |
| rosmaita | https://zuul.openstack.org/builds?project=openstack%2Fglance&pipeline=periodic-stable&skip=0 | 14:49 |
| rosmaita | the mailing list is https://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-stable-maint | 14:50 |
| rosmaita | i only look for cinder jobs in that email | 14:51 |
| rosmaita | but you don't want to worry until the same branch has failed for a few days in a row | 14:51 |
| rosmaita | shit is breaking all the time | 14:51 |
| croelandt | rosmaita: what makes me want to cry is that I'll retire at 65 | 14:52 |
| croelandt | and that's still going to be the same shit | 14:52 |
| rosmaita | man, in the USA i can't retire until at least 67, and i bet that gets changed soon | 14:52 |
| rosmaita | all though with all the covid deaths, maybe there is more social security money available | 14:53 |
| croelandt | rosmaita: join the protest on the 31st :) | 14:53 |
| rosmaita | croelandt: yeah, you won't be able to get your pension reliable because the software that cuts the checks will be hosed | 14:53 |
| croelandt | damn, that's probably right | 14:54 |
| croelandt | well it's wrong, but you're right | 14:54 |
| croelandt | Are we going to be releasing tarballs for the -em branches to which we're pushing backports for that CVE? | 14:54 |
| rosmaita | i don't think so, i believe that they are only generated by a release, and we don't release from -em branches | 14:56 |
| rosmaita | the patches we are pushing to those branches are a courtesy for people still running old releases, we're not even required to do it | 14:58 |
| croelandt | oh ok | 14:59 |
| croelandt | yeah so that grenade & rustc stuff is blocking 3 -em branches :/ | 14:59 |
| dansmith | croelandt: gmann is working on that IIUC | 15:03 |
| croelandt | dansmith: do you mean on *both* issues? | 15:03 |
| dansmith | croelandt: the grenade tempest dep problem | 15:04 |
| croelandt | ok | 15:04 |
| rosmaita | dansmith: that is good news! hooray for gmann! | 15:07 |
| gmann | yeah, fixes are up, I need to get them reviewed and merged - https://review.opendev.org/q/topic:wallaby-pin-tempest https://review.opendev.org/q/topic:bug%252F2003993 | 18:04 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!