Hey, Question about the behavior of certain rollout strategies with a termination grace period. Specifically - does it wait for the actual termination “success” or just for the termination process start (which with grace might be lone)
Recreate
With the semi equivalent for recreate (for a single pod):
type: RollingUpdate
rollingUpdate:
maxUnavailable: 100%
maxSurge: 0
I know that in certain cases I’ll a stateful set to ensure final termination before pod recreate- but I want to understand the common case behavior and if there is any difference in the behavior between those cases
I’ve checked the docs but it’s not clear to me
Cluster information:
Kubernetes version: v1.28.9-eks-036c24b
Cloud being used: aws
Installation method:
Host OS:
CNI and version:
CRI and version: