pulumi-kubernetes-operator
pulumi-kubernetes-operator copied to clipboard
Option to retain or delete the workspace after successful sync
The stack controller provisions a workspace on demand, and it is fine to delete the workspace after sync was successful. Obviously the user is trading off performance/efficiency.
Similar to the "reclaim policy" of a PVC. https://kubernetes.io/docs/concepts/storage/persistent-volumes/#reclaim-policy
For the retain case, we should consider making the workspace pods be "burstable" so that they consume minimal resources at rest but are allowed to consume unlimited memory resources during a deployment operation.
For example:
kind: Stack
spec:
workspaceReclaimPolicy: retain|delete
Added to epic https://github.com/pulumi/pulumi-kubernetes-operator/issues/586
- scale to zero would be a great feature, not everyone cares about the amount of time it runs vs more the cost.
This is really required, i plan to spin up quiet a few stacks for ephemeral environments and it just eats unnecessary resources