Issue With NAS Path

I'm hoping someone else has seen this weird behavior with NAS paths and can shed some light on the issue. What I am currently experiencing is constant errors issuing to the actlog of my library controller instance caused by a failed mount/dismount for an NDMP backup. The TSM instance has been rebooted and the path put offline but I am still getting these errors:

05/15/2008 14:51:18 ANR9999D pvrremot.c(488): ThreadId <32> The PVR remote
descriptor is NULL.(SESSION: 24492)
05/15/2008 14:51:18 ANR9999D ThreadId <32> issued message 9999 from: (SESSION:
24492)
05/15/2008 14:51:18 ANR9999D ThreadId <32> 0x0000000100025bf4 outDiagf
(SESSION: 24492)
05/15/2008 14:51:18 ANR9999D ThreadId <32> 0x00000001004cd380
PvrRemoteOpenDrive (SESSION: 24492)
05/15/2008 14:51:18 ANR9999D ThreadId <32> 0x000000010025193c
PvrOpenTapeDrive@AF144_48 (SESSION: 24492)
05/15/2008 14:51:18 ANR9999D ThreadId <32> 0x0000000100245a0c
PvrOpenTapeDrive (SESSION: 24492)
05/15/2008 14:51:18 ANR9999D ThreadId <32> 0x0000000100245930 PvrOpenDrive
(SESSION: 24492)
05/15/2008 14:51:18 ANR9999D ThreadId <32> 0x00000001002a0150 MgrGetIdleVol
(SESSION: 24492)
05/15/2008 14:51:18 ANR9999D ThreadId <32> 0x00000001002a0e78 PerformDismount
(SESSION: 24492)
05/15/2008 14:51:18 ANR9999D ThreadId <32> 0x00000001002a0850
PerformDismountNow (SESSION: 24492)
05/15/2008 14:51:18 ANR9999D ThreadId <32> 0x00000001002a0d24 DismountThread
(SESSION: 24492)
05/15/2008 14:51:18 ANR9999D ThreadId <32> 0x000000010000fe5c StartThread
(SESSION: 24492)
05/15/2008 14:51:18 ANR9999D ThreadId <32> 0x090000000040f448 _pthread_body
(SESSION: 24492)

The problem is that support wants me to reboot my Powderhorn, which is just not feasible. ACSLS is not showing anything in the drive and that the drive is available so I don't see how it's a library/ACSLS issue.

NDMP Question

I have a customer who wants to have a copy of a NetApp volume. That's not a problem but my question is this, is an NDMP dump initiated by TSM capable of being restored without a TSM server? Since TSM labeled the tape is there any issue with trying to restore the data directly from a NetApp with a direct attached drive? Also when running an NDMP dump directly from the NetApp and not from TSM is the TOC/Catalogue stored on the same tape with the data? My thinking is that the NDMP dump even though initiated by TSM is still just an NDMP dump. My concern is the TSM labeling and if it would cause any readability issues for a NetApp based restore?

UPDATE!!!

I contacted Tivoli support and a NDMP backup is fully independent of TSM. I figured it was but wanted to make sure. The customer wanted us to backup a NetApp volume and send him the tape to hold indefinitely. Now I know I can just send him one of the older NetApp backups from the time period requested without needing to do anything special.

EMC & NDMP

Does anyone know the answer to this question? I have an EMC Clariion and it looks like the easiest way to backup some problem file systems is by NDMP. Does EMC Clariion support NDMP? Is there any specific configuration that does or do they all support it natively? Does it need a Celerra NAS head? Any information from EMC knowledgeable people is appreciated.

TSM V5.4 Enhancements

TSM V5.4 Enhancements Selected for Beta Account Validation

Availability:
Cluster Failover for TSM HSM for Windows (available October, 2006)

Capability:
TSM Express to TSM Enterprise upgrade                  
Backupset Enhancements

• Generate to a specified point-in-time


• Improve  tracking,


• Stack multiple nodes' data on a single set volumes,


• Enable the client to display contents and allow selection of individual files for restore,


• Support image data,


• And more

Compatibility:
Support for Windows Vista (available November, 2006)
Support for Intel based Macintosh systems

Interoperability:
Using TSM Server as an NDMP Tape Server (Filer to Server to backup)
Create Media for offsite vaulting of NDMP Generate Data        

Performance:
Reconciliation for TSM HSM for Windows  (available October, 2006)

Scalability:
Improved BA client memory usage

• Note: Using the disk cache setting will require additional disk space. The disk cache file can exceed 4 gigabytes for a very large filesystem incremental backup. Therefore, if the filesystem does not support large files, the user will need to enable large file support for the filesystem or use the diskcachelocation setting to point to a filesystem that can support large files.

Collocation of active data

• Enhancement provides a new type of storage pool for storing only active versions of backup client data.        

Usability:
Administrator Center Enhancements Part 3

• Update administrator password, cmd line applet, AUDIT LIBRARY/VOLUME, QUERY NODEDATA/MEDIA, MOVE NODEDATA/MEDIA

UPDATE on NDMP Problem!

Support has identified the NDMP problem exists in TSM 5.3.3 if the ONTAP version is lower than 7.1. If you have NetApps make sure your ONTAP version is at or higher than version 7.1 before you upgrade your TSM server to 5.3.3 or higher.

TSM 5.3.3 NDMP Issue!

We discovered a problem with TSM 5.3.3 and failed NDMP backups. It seemed to be occurring on large volumes. After investigating the issue support was called and through further investigation it was discovered that there is a bug in the implementation of TSM’s use of the NMDP 4 protocol. This bug will cause some NDMP backups to fail when the backup spans volumes (i.e. uses more than one tape to do the backup of the volume). The suggestion is to either stay at version 5.3.2 or to use the following option in the dsmserv.opt file:

(Note: there is no problem with data integrity of backups that show completed)

FORCENDMPVERSION3    YES

This option will use the older NDMP protocol and should resolve the failures until a fix is released.