Automatic retry logic

[PRIHLOP]

Hello!

• Vote on this issue by adding a 👍 reaction
• If you want to implement this feature, comment to let us know (we'll work with you on design, scheduling, etc.)

Issue details

I have an issue with no configurable retries to connect to Kubernetes API. If the cluster is unavailable at the moment of apply action or if some network issue appears pulumi-kubernetes does not try to retry apply action request. But, I think this option should be configurable to make the provider more flexible.

If automatic retry logic does not exist for query/idempotent requests, retry logic should be added. If retry logic does exist, perhaps its behavior could be exposed as parameters on the provider (i.e. # of retry attempts, backoff base, backoff duration)

[rquitales]

We currently do not attempt any retries if a create/update/delete request fails. This is a good suggestion and we'll add this to our backlog for consideration. Thanks!