You have not given us anything to work with....
Can you show us all text in Details tab of failing job?
Did you create dbclient log on the destination client and increased logging level to 3?
Have you checked NBU for SQL manual regarding large DB restore?
Look for: "About minimizing timeout failures on large SQL Server database restores"
Have you set the Instant data file initialization ? If not, that's probably a good start. It is described in the NBU SQL admin guide as far as I remember.
Is there any firewalls between the netbackup servers and the SQL server ? Most firewalls closes the connection if it has been idle for while.
Other than that, use the bphdb and dbclient logs, think they are describe in troubleshooting section of NBU SQL admin guide.
Don't get me started on SQL. I gave up on it and now we backup to disk and I backup the disk files for recovery purposes.
We were only able to have success restoring if we single thread the recovery - which takes too long.
Sorry NetBackup, IMHO your SQL recovery does not meet our needs.