Before I log a support call I wanted to ask if anyone has had experience with this issue.
I have a Storage Policy with the following copies configured:
- Primary (Snap) Copy
- SnapVault to another SVM on the same cluster
- SnapMirror to an SVM on a different cluster
- Tape (Backup) Copy
On some NAS volumes/subclients, I get an error when trying to run the backup (tape) copy:
NDMP Server reported the following error: [DATA: Operation terminated: backup configuration failure (2) on RPC error: ERROR: BAD ENV OP (for /ncww_v07_backup/CC_nspprod01_SP_22_Copy_86_ncba_v04_cifs_ncba_v04_vol_s_drive02_Mirror/.snapshot/snapmirrorSE.66c81f94-5].
In the NasBackup.log I see the full name of the snapshot that the backup copy job is trying to access:
The trouble is, that snapshot doesn’t actually exist on the SnapMirror destination volume:
ncww::> vol snap show -vserver ncww_v07_backup -volume CC_nspprod01_SP_22_Copy_86_ncba_v04_cifs_ncba_v04_vol_s_drive02_Mirror -snapshot snapmirrorSE.66*
(volume snapshot show)
There are no entries matching your query.
For the other subclients/volumes where tape backups *do* work , the Snapshot names in NasBackup.log are generally something like “SP_2_xxxx”. For the ones that fail, SnapProtect seems to be looking for non-existant snapshots with a longer name, prefixed by "snapmirrorSE.xxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx.”.
The environment is SnapProtect 10 SP6 with Clustered ONTAP 8.2.1P1. I’m using Cluster-Aware-Backup with node-scope NDMP disabled on the clusters and SVM tunnelling configured in SnapProtect.
I have tried re-running the previous SnapVault/SnapMirror Aux copies in the storage policy, but they say there is no data to be copied.
The volumes affected have multiple qtrees, and I have found mentions of a bug/limitation titled "A SnapMirror tape backup from a SnapVault auxiliary copy fails if the volume has multiple qtrees” - but this bug was supposed to be fixed in SP6.
Any help or suggestions would be appreciated.
(Also cross-posting to xdl-snapprotect@netapp)