rds DBInstance storage autoscaling causes allocatedStorage drift

[saatvik-chugh]

What happened?

When RDS storage autoscaling is enabled by setting maxAllocatedStorage, AWS can increase the actual allocated storage for the DBInstance. After that happens, provider-aws appears to detect drift against the original allocatedStorage value and tries to reconcile the resource back to that lower value.

This creates reconciliation drift/noise and can cause failed updates because allocated storage cannot be decreased in RDS.

This was previously reported in #2255, but that issue was closed as stale/not planned. Opening a new issue because this is still affecting current usage.

How can we reproduce it?

1. Create an RDS DBInstance with allocatedStorage and maxAllocatedStorage set.
2. Allow RDS storage autoscaling to increase the allocated storage.
3. Observe provider-aws reconciliation attempting to restore the original allocatedStorage value.

Conceptual example:

apiVersion: rds.aws.crossplane.io/v1alpha1
kind: DBInstance
spec:
  forProvider:
    region: us-east-1
    allocatedStorage: 100
    maxAllocatedStorage: 500

What environment did it happen in?

Provider version: v0.56.0 / v0.58.0
Crossplane version:
Kubernetes version:
AWS region:
Engine:
Instance class:

Expected behavior

When storage autoscaling is enabled, provider-aws should not try to decrease allocatedStorage back to the original spec value after AWS increases it. It should either treat AWS-increased allocated storage as late-initialized/observed state or otherwise avoid reconciling storage downward.

[github-actions]

Crossplane does not currently have enough maintainers to address every issue and pull request. This issue has been automatically marked as stale because it has had no activity in the last 90 days. It will be closed in 14 days if no further activity occurs. Leaving a comment starting with /fresh will mark this issue as not stale.