-
Notifications
You must be signed in to change notification settings - Fork 1.4k
Description
What would you like to be added (User Story)?
As a Cluster API operator,
When I run clusterctl move --to-directory /tmp/backup-directory
on a namespace that contains a paused Cluster,
I want the Cluster to remain paused after the move,
So that I don’t unintentionally trigger reconciliation or cause disruptive changes on a Cluster that was deliberately paused.
Detailed Description
Currently, when running clusterctl move --to-directory
, a Cluster that was originally paused (spec.paused: true) ends up unpaused at the end of the move operation in the management cluster. This behavior can unintentionally trigger reconciliation during backup, restore, or migration workflows where the operator’s intent was to keep the Cluster inactive.
The expected behavior is that clusterctl move --to-directory
preserves the original spec.paused value, ensuring the Cluster remains in the same operational state before and after the move. This avoids surprises and provides safer handling of paused Clusters during move operations.
This change should only be applied to the ToDirectory
workflow. The Move
and FromDirectory
commands serve different use cases and should not be modified.
Anything else you would like to add?
No response
Label(s) to be applied
/kind feature