Monitor pods for crashes

[BlueskyFR]

Is your feature request related to a problem? Please describe.
I would like to be able to monitor a certain pod or deployment stack for crashes for say 10 minutes when running helmwave up, and rollback to the previous version if a crash happens.

Describe the solution you'd like
A monitor would actively listen for crashes and trigger a rollback if the crash count is > threshold.

Describe alternatives you've considered
I really don't know since I'm new to Helm, it would require a custom script currently I think.

[zhilyaev]

KISS https://kubernetes.io/docs/tasks/configure-pod-container/configure-liveness-readiness-startup-probes with atomic flag.
You can find more information in Helm and Kubernetes community.

For cases with more complexity such as external metrics, you can use helmwave feature https://docs.helmwave.app/0.41.x/examples/monitors/