Christian Nicolai

> Hmm, it looks at least more stable than before. I hoped/thought so as well but after running the numbers of `number(cancelled builds) divide number(all builds)` (excluding PRs) we can...

Not closing yet since we have to see if the 2nd fix in https://github.com/camunda/camunda/pull/21798 is effective. Note2self: CI health query used to see percentage of problems for BigQuery with a...

I looked at the CI statistics again with focus on the `optimize-backend-unit-tests` job on `main` and in the merge queue and have to say that I don't see the situation...

While we look for the root cause I think we should as a mitigation set the timeout to 15 minutes: better to have a slow build then a randomly failing...

> However, I'm not too sure that my testing is 100% accurate. I'm running the CI test when most developers are offline and there's plenty of resources to be used...

Things I see at a glance that can be removed: * all files below `.ci` folder since Preview Environments are not yet supported in this monorepo, will be added back...

Thanks for the PR and the detailed instructions, those are very helpful now that I start looking at it in more detail now and collect observations! > The changes to...

> If that's the case, maybe before reviewing this with the method I am suggesting, the update with the latest main should also be done I think you are correct...

I still see changes to `timeout-minutes` in `ci.yml` that are unnecessary and need removed: * `jobs.detect-changes.timeout-minutes`: from 10 to 20 while the job currently takes under 1min * `jobs.actionlint.timeout-minutes`: from...

There are several new subfolders created in `.github` after the merge, like `.github/configuration` and `.github/podSpecs` and `.github/scripts` which are all Optimize-specific. We need to make this more transparent (e.g. by...