SQL restore timeout after around 2hours

[Zahid_Haseeb]

i have an environment of two Netbackup Servers 6.5 version and OS is win2003......Netbackup Server1 is at primary site and Netbackup Server2 is at DR site. i can sucessfully restore the catalog from my primary site to secondary site(DR). But when i try to restore the SQl database Tape at my Secondary DR site from the Tape of my primary location, while backup restoration after around 2 hours my backup restoration gets fail with the below error

9/8/2009 2:45:51 PM - begin Restore
9/8/2009 2:45:57 PM - 1 images required
9/8/2009 2:45:57 PM - media 0000L4 required
9/8/2009 2:46:03 PM - restoring image sql-cluster_1252188468
9/8/2009 2:46:08 PM - requesting resource 0000L4
9/8/2009 2:46:08 PM - granted resource 0000L4
9/8/2009 2:46:08 PM - granted resource IBM.HHLTOGEN4.000
9/8/2009 2:46:10 PM - started process bptm (4692)
9/8/2009 2:46:10 PM - mounting 0000L4
9/8/2009 2:46:14 PM - mounted; mount time: 00:00:04
9/8/2009 2:46:15 PM - positioning 0000L4 to file 93
9/8/2009 2:46:17 PM - connected
9/8/2009 2:47:48 PM - positioned 0000L4; position time: 00:01:33
9/8/2009 2:47:48 PM - begin reading
9/8/2009 4:46:17 PM - Error bpbrm(pid=4660) socket read failed, An existing connection was forcibly closed by the remote host.  (10054)
9/8/2009 4:46:19 PM - Error bptm(pid=4644) The following files/folders were not restored:       
9/8/2009 4:46:19 PM - Error bptm(pid=4644) UTF - /Server-AB.MSSQL7.SQL-CLUSTER.db.DB-DotNet.~.7.001of001.20090906030615..C          
9/8/2009 4:46:19 PM - Error bptm(pid=4644) UTF - /Server-AB.MSSQL7.SQL-CLUSTER.db.DB-DotNet.~.7.001of001.20090906030615..C          
9/8/2009 4:46:19 PM - Error bptm(pid=4644) UTF - /Server-AB.MSSQL7.SQL-CLUSTER.db.DB-DotNet.~.7.001of001.20090906030615..C          
9/8/2009 4:46:19 PM - Error bptm(pid=4644) UTF - /Server-AB.MSSQL7.SQL-CLUSTER.db.DB-DotNet.~.7.001of001.20090906030615..C          
9/8/2009 4:46:19 PM - Error bptm(pid=4644) UTF - /Server-AB.MSSQL7.SQL-CLUSTER.db.DB-DotNet.~.7.001of001.20090906030615..C          
9/8/2009 4:46:19 PM - Error bptm(pid=4644) UTF - /Server-AB.MSSQL7.SQL-CLUSTER.db.DB-DotNet.~.7.001of001.20090906030615..C          
9/8/2009 4:46:19 PM - Error bptm(pid=4644) UTF - /Server-AB.MSSQL7.SQL-CLUSTER.db.DB-DotNet.~.7.001of001.20090906030615..C          
9/8/2009 4:46:19 PM - Error bptm(pid=4644) UTF - /Server-AB.MSSQL7.SQL-CLUSTER.db.DB-DotNet.~.7.001of001.20090906030615..C          
9/8/2009 4:46:19 PM - Error bptm(pid=4644) UTF - /Server-AB.MSSQL7.SQL-CLUSTER.db.DB-DotNet.~.7.001of001.20090906030615..C          
9/8/2009 4:46:19 PM - Error bptm(pid=4644) UTF - /Server-AB.MSSQL7.SQL-CLUSTER.db.DB-DotNet.~.7.001of001.20090906030615..C          
9/8/2009 4:46:19 PM - Error bptm(pid=4644) more than 10 files were not restored, remaining ones are shown in the progress log.
9/8/2009 4:46:20 PM - restored image sql-cluster_1252188468 - (file read failed(13)); restore time 02:00:17
9/8/2009 4:46:28 PM - end Restore; elapsed time: 02:00:37
the restore failed to recover the requested files(5)

[Andy_Welburn · Accepted solution]

You may have to look at increasing your CLIENT_READ_TIMEOUT setting.

DOCUMENTATION: Restores of large Microsoft SQL server databases using the Veritas NetBackup (tm) for...

[Marianne]

Agree with Andy - make timeout at least 4 hours. Remember to drop down again after restore.
Also have a look at this section in the NBU for SQL manual:

Instant data file initialization
When you restore a database, filegroup, or database file, SQL Server zeroes the
file space before beginning the restore operation. This can slow the total
recovery time by as much as a factor of 2. To eliminate file initialization, run the
MSSQLSERVER service under a Windows account that has been assigned the
SE_MANAGE_VOLUME_NAME. For more information about eliminating file
initialization, see the SQL Server 2005 and Windows 2003 documentation.