Feature Request: max_retries option for provider block

[kbruner]

Requesting support for a max_retries option for the kubernetes provider.

Sample use case: When creating a cluster via terraform, I also want to populate some ConfigMaps with terraform outputs. It does take a few minutes for the kubernetes API to come up, so being able to increase the retry limit so the resources can live in the same .tf files as the cluster infrastructure and succeed on the same terraform apply.

[paultyng]

This seems like the upstream progressive apply issue: hashicorp/terraform#4149

You cannot currently chain together a provider's config with the output of a resource (including a resource creating that cluster).

[rchernobelskiy]

Is there a way to have it retry if the api server has a momentary hiccup?

(Just as an aside, the progressive apply works fine with aws/eks)