You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We've had cases when users manually deleted pods/statefulsets/etc of a PG cluster only to found out them to be re-created later. This leads to users' confusion and extra debugging overhead for maintenance teams. We should experiment with k8s finalizers available since k8s v1.11 to prevent such deletions. Things to watch out are (a) properly handling the deletion timestamp (is set on the first delete request ), (b) effectively informing users about why their deletions of operator managed resources do not succeed and (c) enabling members of admin teams manually delete misbehaving resources.
A more sophisticated option is to use k8s admission webhooks.
The text was updated successfully, but these errors were encountered:
We've had cases when users manually deleted pods/statefulsets/etc of a PG cluster only to found out them to be re-created later. This leads to users' confusion and extra debugging overhead for maintenance teams. We should experiment with k8s finalizers available since k8s v1.11 to prevent such deletions. Things to watch out are (a) properly handling the deletion timestamp (is set on the first delete request ), (b) effectively informing users about why their deletions of operator managed resources do not succeed and (c) enabling members of admin teams manually delete misbehaving resources.
A more sophisticated option is to use k8s admission webhooks.
The text was updated successfully, but these errors were encountered: