| kata-irc-bot | <christophe> When you say "check", do we agree that it's just checking the format of the commit, and it's not about the CI, right? | 10:29 |
|---|---|---|
| kata-irc-bot | <christophe> Because a revert of something that is not the latest commit effectively creates a new combination of code. | 10:30 |
| kata-irc-bot | <fidencio> It's only checking the commit short log, Christophe. | 10:31 |
| kata-irc-bot | <fidencio> Whether the commit short log follows the format specified (as in "subsystem: short explanation") | 10:31 |
| kata-irc-bot | <gmaglion> Feel free to test it and report any error, I'll let you know once merged. | 10:49 |
| kata-irc-bot | <fidencio> @gmaglion, perfect! I will test it locally here and I think @simon.kaegi needs a day or two to test it on his environment. | 10:51 |
| kata-irc-bot | <fidencio> Thanks! | 10:51 |
| kata-irc-bot | <gmaglion> If you will move to the rust version I highly recommend you to use the new options, because the old ones (like -o) are deprecated. | 10:51 |
| kata-irc-bot | <gmaglion> No problem, I wanted to send it yesterday but I was distracted by something else | 10:52 |
| kata-irc-bot | <fidencio> Yeah, I think by the moment we switch to the rust agent officially we should advise using the new options (and do that ourselves on the options we hard-code) | 10:55 |
| kata-irc-bot | <gmaglion> about that, I want to send a patch to add 'announce_submounts' option, but I'm not sure how to check for the virtiofsd version, I meant what's the policy inKata check for the version string or try to execute with that option and fallback if failed | 11:24 |
| kata-irc-bot | <fidencio> > about that, I want to send a patch to add 'announce_submounts' option, but I'm not sure how to check for the virtiofsd version, I meant what's the policy inKata check for the version string or try to execute with that option and fallback if failed It depends on what we build with QEMU (as it's now). If this is something we can wait for the rust version to land so it becomes the default, better, because then we have more control to | 11:37 |
| kata-irc-bot | pin a specific version of virtiofsd (independently of the QEMU version) | 11:37 |
| kata-irc-bot | <fidencio> Ah! https://github.com/kata-containers/kata-containers/issues/3507#issuecomment-1027857675 :slightly_smiling_face: | 11:45 |
| kata-irc-bot | <eric.ernst> @ssheribe @fidencio @jose.carlos.venegas.m - I think my PR should help w the 2.3 kata deploy issues - anything more I can show re: testing to get acks/drop dnm? | 15:41 |
| kata-irc-bot | <eric.ernst> https://github.com/kata-containers/kata-containers/pull/3587 <- if anyone else is interested and can TAL! Thanks y'all! | 15:41 |
| kata-irc-bot | <fidencio> @eric.ernst, I'd love to get it tested, from your fork, on a branch different than `main`. | 15:42 |
| kata-irc-bot | <fidencio> Would be possible for you to test it and paste the results / links to the test? | 15:42 |
| kata-irc-bot | <eric.ernst> I see. Yeah. I'll create a stable and recreate. No | 15:48 |
| kata-irc-bot | <eric.ernst> Problem | 15:48 |
| kata-irc-bot | <fidencio> Sorry for being a pain, just asking explicitly becuase actions based on comments seem to have issues when dealing with different branches than main. | 15:49 |
| kata-irc-bot | <eric.ernst> No worries. Good to be thorough. If the target branch is non-main it behaves the same. Updated the PR w links | 16:09 |
| kata-irc-bot | <fidencio> Perfect, thanks! | 16:14 |
| kata-irc-bot | <fidencio> @eric.ernst, if you could already raise the backport ... :slightly_smiling_face: | 16:51 |
| kata-irc-bot | <eric.ernst> :slightly_smiling_face: done | 16:58 |
| kata-irc-bot | <ssheribe> @fidencio @eric.ernst backport is required? IIUC test-kata-deploy is trigger only by a comment so it always uses main's workflow no? | 17:14 |
| kata-irc-bot | <eric.ernst> the yaml is from the base, but the scripts/tooling that it exercises is from the PR. That’s the key part that is broken today and that this is fixing. | 17:14 |
| kata-irc-bot | <eric.ernst> ie, without my PR on stable, you don’t test any changes to the kata build scripts, or kata-deploy action, or kata-deploy install scripts. | 17:15 |
| kata-irc-bot | <ssheribe> O | 17:21 |
| kata-irc-bot | <eric.ernst> that make sense to you @ssheribe ? | 17:25 |
| kata-irc-bot | <ssheribe> Eric, sorry, I think i miss something | 17:26 |
| kata-irc-bot | <ssheribe> Your pr touches only the yaml no? | 17:26 |
| kata-irc-bot | <ssheribe> (I don't mind having it on stable anyhow..) | 17:26 |
| kata-irc-bot | <eric.ernst> It doesn’t use main’s yaml. | 18:00 |
| kata-irc-bot | <wmoschet> @eric.ernst just left one comment | 18:00 |
| kata-irc-bot | <eric.ernst> It uses the base yaml | 18:00 |
| kata-irc-bot | <eric.ernst> for the target branch. | 18:00 |
| kata-irc-bot | <eric.ernst> good suggestion @wmoschet | 18:02 |
| kata-irc-bot | <eric.ernst> ah. @ssheribe - I believe you’re correct. That for sure is a limitation. huh. | 18:14 |
| kata-irc-bot | <eric.ernst> i suppose more the reason to keep workflows as simple / non changing as possible, and do most of the lifting within the repo itself (ie, workflow yaml calls into our scripst). | 18:14 |
| kata-irc-bot | <wmoschet> Besides that having most of the logic on the repo helps when you need to run locally | 18:15 |
| kata-irc-bot | <eric.ernst> pushed update based on suggestion, and tested that touching versions.yaml results in the action triggering. | 18:24 |
| kata-irc-bot | <eric.ernst> see: https://github.com/egernst/kata-containers/pull/15 | 18:24 |
| kata-irc-bot | <wmoschet> great. I reviewed the full script and I don't have any other suggestion | 18:24 |
| kata-irc-bot | <eric.ernst> Looking for spicy takes on govet fieldAlignment in our static checks :point_down: :thread: | 18:38 |
| kata-irc-bot | <eric.ernst> PTAL: https://github.com/kata-containers/kata-containers/issues/3604 | 18:38 |
| kata-irc-bot | <fidencio> You're correct, Snir! | 19:08 |
| kata-irc-bot | <fidencio> And that was the main reason of the first issue we started with. | 19:09 |
| kata-irc-bot | <fidencio> Too tired to connect the dots, thanks for catching that up, Snir! :slightly_smiling_face: | 19:09 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!