Mix automatic retry with manual retry

As a follow up to this post: Ordering of automatic retry attributes

I’d like to do something similar, but after the automatic retry on -1, I’d like to fallback to the default behavior for hard failing but allowing manual retries. Is this possible?

Hey @cwawrzonek

Thanks for the message

As seen in the older post

        - exit_status: -1
          limit: 2
        - exit_status: "*"
          limit: 1

It will fail after the limit for the automatic retries, you should be able to manually retry the last job.


Got it, thank you!

Follow up feature request, can you provide a way to set a global retry policy on all steps in a pipeline? I’d like all my steps to retry on agent lost, but don’t want to have to add this to every step

Hey @cwawrzonek

Thanks for the follow up.

At the moment the retry is a step-level feature, i know it might be a bit daunting to specify this for each, but we encourage users to use YAML Anchor to avoid having to do the manual work of updating all the steps.

I can raise this with the product team, but not sure if this will be added to the roadmap for now


Got it, thanks for the quick response!