-
-
Notifications
You must be signed in to change notification settings - Fork 25
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
postBuildStep's should probably be scheduled regardless of build skips #295
Comments
I think these postBuildSteps should be just used for custom binary cache upload logic rather than continuous delivery and I rather not want to make building individual packages more expensive by adding more steps since it's already quite slow (We see this on flakes building thousands of packages). This would be than only available when we merged: #68 |
If you want to make Hercules Ci happen faster, you can probably help us implementing it. |
Yeah, I can see how abusing the "build" metaphor isn't something you would want to do. That said, I can't wrap my head around how hercules-ci effects works, and how it's supposed to fit into buildbot-nix. Is there a design doc I can look at? |
We will probably write documentation/examples and use it our self once it is in. Otherwise I can just refer to hercules-ci own documentation. |
I get that - but how will hci effects integrate into buildbot? Will a worker act as the equivalent of the hercules-ci agent, or will that still require the agent to be in place? Probably should take that discussion to #68 (: |
I already implemented the runner interface of hercules-ci effects, so it will run in a buildbot worker as in a bubblewrap container. |
buildbot-nix/buildbot_effects/__init__.py Line 157 in 3ac54ba
|
I have an "effect" (not a hercules-ci-effect! no idea how to use those tbh) via a
postBuildStep
that I would like to run only on the default branch of my repo. If I make a pull request that affects that attribute, it properly gets run on the PR branch and decides it has nothing to do; but if I merge the PR, the build is skipped and the postBuildStep isn't executed - it isn't even scheduled.This is what a build on a PR branch looks like:
And this is on the merge commit:
I believe what happens here is that the evaluation counts as "cached" coming out of nix-eval-jobs, which prompts buildbot-nix to not build & then not schedule the following postBuildSteps either.
What I think would be neat is if we could:
postBuildSteps
, either by default or via a configurable attributedoStepIf
buildbot attribute, so steps can decide for themselves based on the properties of the build whether they are supposed to run or not.The text was updated successfully, but these errors were encountered: