I am using an Exabyte 220 Autoloader, with 2 Mammoth drives, partitioned into 10 partitions using the Veritas software. Server is NT4, sp6a, 2GB RAM, BE v8.5 build 3572. I run two concurrent jobs nightly at 9:00pm, all jobs complete successfully except the very last one on Friday. The jobs are set this way due to amount of data, and heavy traffic on the server. The job that is failing backsup about 58GB of data then fails. I replaced the .pvl file a week ago, the Friday after I did that the job ran successfully. Now it is failing again. The jobs are full backups, just selective through the job definition. Due to the nature of our business this is best for us, no incremental or differential. This started occurring right after the upgrade from v8.0 to v8.5. Veritas support has not been able to figure this one out either. Any help or suggestions would be appreciated!!