Virtual TapeServer 8.0 Release Notes

Resolved Issues and Enhancements | 17
General operations
 Could not create jobs because the process required for jobs was not running.
When a Stacked Export job was the first job to run, a process (RPC Server) that required a
particular class to exist for resolution failed in the previous release. To account for that,
the web interface is preloaded in the job scheduler process and ensures the process has all
classes it needs. (TL-6251)
 Job process stopped after logrotate and no jobs could start until you rebooted.
Log rotation was causing a restart of processes and causing jobs to fail. Rath logs are now
rotated only when they reach 10MB and 10 files are retained. When the number exceeds
10, the oldest is deleted. This is different from other logs whose rotation is being managed
by logrotate. (TL-6743)
 FC Abort task set was not supported. During communication failure processing, VTS
now responds to aborted FC command sets when requested. (TL-6769, EAR 7333, EAR
7850, EAR 7793)
 Read Position with no tape in drive gave check condition. Previously, the Read
Position command returned a check condition if there was no media in the drive. Now, it
returns the proper response that indicates there is no tape in the drive rather than the
check condition. (TL-6640, EAR 7333)
 New emulation type 3580-TD4 replaces the deprecated ULT3580-TD4 type.
Standard emulation code is now used for a new TD4 emulation, providing better
emulation compatibility. (TL-6689)
 Failure indicated host with initiator ID of all 1s if initiator aborted a SCSI
command. During a sequence of SCSI commands, if the initiator aborts a command, VTS
no longer incorrectly indicates a failure on a command due to a SCSI reservation in place
by a host with an initiator ID of all 1s, host=0xFFFFFFFFFFFFFFFF. (TL-6778, EAR
7793)
 Jobs may have failed if run immediately after TapeServer startup. If you ran a job
right after starting the system, an error may have resulted if TapeServer had not
completely started. This no longer occurs. (TL-6420)
 TapeServer startup performance was slow. Because rescan-scsi-bus ran to remove
and re-add all devices during startup, startup may have taken a long time to complete if
there were multiple devices attached to VTS. This has been fixed. (TL-6224)
 Could not import pools if pool name included non-alphanumeric characters.
Importing a pool failed if the pool name contained a non-alphanumeric character. This
has been fixed. (TL-6200, EAR 7346, EAR 7233, EAR 7418)
 VTD could no longer be accessed if doing backup to a reserved VTD and another
host had a reservation conflict on this VTD. Commands received while another host
was writing to a VTD it had reserved may have resulted in a segfault in the system
messages log and the loss of access to the VTD. This has been fixed. (TL-6634, TL-6123,
EAR 7278, EAR 7361)
 Tape operations may have failed. After log rotate, XMLRPC may not have restarted
correctly causing some operations to be unusable (mount, replicate, or export). This has
been fixed. (TL-6093, EAR 7241, EAR 7282, EAR 7284)
 Required Log Pages for Ultrium were not supported. NetBackup requests Log Page
0x31 but VTS returned check condition 5/24/00 “Invalid Field in CDB”. This has been
fixed. Now, a generic “Tape Capacity for Log Sense 0x31” response is returned. (TL-5941,
EAR 7065)