Perhaps this is related. I just recently attempted to restore a backup without creating a new MongoDB cluster. This restore most definitely worked: shortly before the scheduled hourly backup, I added some nonsense into the database. After initiating the “Restore from a backup” functionality from the Everest UI, the nonsense data were removed (as expected). However, that restore process is still active as well.
Name: restore-yjx
Namespace: cym-test
Labels: <none>
Annotations: <none>
API Version: psmdb.percona.com/v1
Kind: PerconaServerMongoDBRestore
Metadata:
Creation Timestamp: 2024-03-27T13:03:56Z
Generation: 1
Owner References:
API Version: everest.percona.com/v1alpha1
Block Owner Deletion: true
Controller: true
Kind: DatabaseClusterRestore
Name: restore-yjx
UID: 5445b571-6d06-4daf-b48c-b32d8263b56d
Resource Version: 146951060
UID: 2f51674e-d5c6-4ee5-9709-07cd9915f74e
Spec:
Backup Name: cron-test-mongo-1-20240327120000-x4j9x
Cluster Name: test-mongo-1
Status:
Pbm Name: 2024-03-27T13:04:28.714968651Z
State: running
Events: <none>
I connected directly to the MongoDB cluster within one of its mongod
containers to verify that the nonsense data were removed. I was able to disconnect and reconnect to that cluster without issue both before and after the restore process “succeeded.”