| [06:44:34] | * adinfinitum[m] has joined #farmos |
| [06:51:17] | * farmBOT has joined #farmos |
| [08:38:18] | <symbioquine[m]> | <mstenta[m]> "on a scale of 1-5, how hard do..." <- Hmmm, pretty much every time I've had to mess with Gatsby dependencies it's been an exercise in frustration... |
| [08:54:05] | <mstenta[m]> | wotnak.... did you just... do it? 😲 |
| [08:58:54] | <wotnak[m]> | yes, I think so, might not be 1:1 when it comes to styles because of the required material ui updates, but generally it is updated and seems to work https://github.com/farmOS/farmOS.org/pull/79 |
| [08:59:46] | <mstenta[m]> | amazing |
| [09:00:13] | <mstenta[m]> | thank you!!! i'll do some side-by-side comparisons |
| [09:21:41] | <mstenta[m]> | ... after i release farmOS 2.1.0... working on that now! :-) |
| [10:36:51] | <mstenta[m]> | wotnak: the deploy preview of your PR LGTM! |
| [10:42:18] | <mstenta[m]> | farmOS 2.1.0 has been released! 🎉 https://github.com/farmOS/farmOS/releases/2.1.0 |
| [11:32:05] | <mstenta[m]> | symbioquine paul121 the mariadb test failures are suuuper annoying. i'm giving this a try to see if it helps: https://github.com/mstenta/farmOS/commit/e0995cbd4488ec9e15af8c67c54261d... |
| [11:32:38] | <mstenta[m]> | it might mean slower tests overall, but if I no longer have to keep clicking "re-run failed tests" i'll be happier :-) |
| [11:33:02] | <symbioquine[m]> | Makes sense |
| [11:33:42] | <symbioquine[m]> | "Slow and automatic" is probably better than "fast but flaky" |
| [11:34:35] | <symbioquine[m]> | It will probably take a while before we know if it really worked though if the race condition still exists but is just less likely... |
| [11:34:55] | <mstenta[m]> | yea... might try rerunning this job a few times |
| [11:35:12] | <symbioquine[m]> | In that case, it might be best to revert the change (if/when we discover that) and look for a root cause fix. |
| [11:35:34] | <symbioquine[m]> | Could be days/weeks before we know :) |
| [11:35:56] | <mstenta[m]> | > "Slow and automatic" is probably better than "fast but flaky" |
| [11:35:56] | <mstenta[m]> | and if the psql tests pass we can be pretty confident in the other two... those will still go normal speed, so in practice that can be a good indicator during development... and for merges/releases we just wait a little longer to be 100% sure |
| [11:36:15] | <mstenta[m]> | haha maybe... although it's been pretty consistently failing recently |
| [11:53:13] | <mstenta[m]> | passed on the first run! re-running... 🤞 |
| [11:59:02] | <paul121[m]> | yea especially if sqlite is already taking so long 🤷 |
| [11:59:14] | <mstenta[m]> | yea exactly - sqlite takes 15 mins already |
| [11:59:43] | <mstenta[m]> | i'm monitoring 4 more runs of the tests... will link to the results in the PR |
| [12:16:27] | <mstenta[m]> | dangit - some still failed |
| [12:18:41] | <mstenta[m]> | maybe we need to restructure so that paratest is only used for postgresql |
| [12:18:56] | <mstenta[m]> | (assuming that's the root of the issue) |
| [14:03:25] | * explore has quit (Quit: Connection closed for inactivity) |