Disk-level snapshot restore fails if the original disk is detached from the instance

This issue occurs if you are performing a disk-level snapshot restore to the same location.

When you trigger a disk-level snapshot restore to the same location, NetBackup first detaches the existing original disk from the instance, creates a new volume from the disk snapshot, and then attaches the new volume to the instance. The original disk is automatically deleted after the restore operation is successful.

However, if the original disk whose snapshot is being restored is manually detached from the instance before the restore is triggered, the restore operation fails.

You may see the following message on the NetBackup UI:

Request failed unexpectedly: [Errno 17] File exists: '/<app.diskmount>'

The NetBackup coordinator logs contain messages similar to the following:

flexsnap.coordinator: INFO - configid : <app.snapshotID> status changed to
 {u'status': u'failed', u'discovered_time': <time>, u'errmsg': u'
Could not connect to <application> server localhost:27017: 
[Errno 111]Connection refused'}

Workaround:

If the restore has already failed in the environment, you may have to manually perform a disk cleanup first and then trigger the restore job again.

In general, if you want to detach the original application disks from the instance, use the following process for restore: