Related issues
LEP
Test Normal Case (use nfs)
Given Longhorn cluster with 3 worker nodes.
And Create a volume and attach it.
And Write large data to the volume.
And Create the first backup from the volume.
And Write small data to the volume.
And Create a snapshot.
Then Delete the first backup and create second backup from the snapshot at the same time.
Verify The second backup will be in Pending state with message including waiting for backup to be deleted.
Verify After the first backup is deleted, second backup should be in progress and complete in the end.
Test Backup Delete Error Case (use nfs)
Given Longhorn cluster with 3 worker nodes.
And Create a volume and attach it.
And Write large data to the volume.
And Create the first backup from the volume.
And Write small data to the volume.
And Create a snapshot.
And Connect to the backupstore pod and make the backup.cfg file immutable by running the command $ chattr +i backups/backup_backup-*.cfg.
Then Delete the first backup and create second backup from the snapshot at the same time.
Verify The first backup will repeatedly enter the Deleting and Error states as it attempts to retry the deletion. When in the Error state, an error message related to permissions will be displayed.
Verify The second backup will be InProgress when the first backup is in Error state. The second backup should be complete after awhile.
Verify After make the config mutable by running the command $ chattr -i backups/backup_backup-*.cfg, after a while, the second backup should be in Deleting again and should be deleted successfully.
Test Controller Crashes Case (use nfs)
Given Longhorn cluster with 3 worker nodes.
And Create a volume and attach it.
And Write large data to the volume.
And Create the first backup from the volume.
And Connect to the backupstore pod and make the backup.cfg file immutable by running the command $ chattr +i backups/backup_backup-*.cfg.
Then Delete the first backup.
Verify The first backup will repeatedly enter the Deleting and Error states as it attempts to retry the deletion. When in the Error state, an error message related to permissions will be displayed.
Then When the backup is in Deleting state, delete the longhorn-manager pod which is the owner of the backup. (by checking Backup.Status.OwnerID)
Verify After the longhorn manager pod is recreated, the backup should turn into Error state with message No deletion in progress record, retry the deletion command.
Then Make the config mutable by running the command $ chattr -i backups/backup_backup-*.cfg.
Verify The backup should be in Deleting again and should be deleted successfully.