Better tool for Prow job debugging.

[erain]

This is a feature request for better Prow debugging tooling / documents.

Currently, the only way to debug a Prow job is to use the docker image, however some of the kubetest images, e.g. kubekins-e2e, have non-trivial setup (its entrypoint, $GOPATH, etc.). It is lack of clear guidance for Prow users on how to start debugging.

It would be great to have tooling for such purpose. Just some ideas:

1. local Prow: to have a Prow job running locally inside a docker container.
2. debug image: the image contains everything (minus security key), and could "just run".

Also many to have documents on how to debug Prow job in general. Currently I come up with something like the following to mount every dependent source repo into the image, but I guess there is something better:

docker run -it --entrypoint "" --name kubetest-$RANDOM \
  -v $GOPATH/src/repo-1:/workspace/repo1 \
  -v $GOPATH/src/repo-2:/workspace/repo2 \
  -v $GOPATH/src/k8s.io/test-infra:/workspace/test-infra \
  -v $HOME/.ssh:/workspace/.ssh/ \
 kubekins-e2e:<version-tag> /bin/bash
...
scenarios/kubernetes_e2e.py --check-leaked-resources --env ...

Thanks!

[stevekuznetsov]

There was an issue tracking this somewhere -- a tool to output the docker run command to debug locally. @BenTheElder do you remember where that was?

[BenTheElder]

#6590

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale