Virtual TapeServer 8.3 Release Notes

Known Issues | 44
 Restoring individual files from export that spanned files using CA ARCserve
fails. (TL-6517)
WORKAROUND: Restore the entire backup to obtain the single file.
 TSM running on a host system fails to label tapes when the VTDs are Ultrium.
Though the drives are detected and associated with the library, when it tries to load the
tapes for labeling, an error is displayed indicating it cannot label the volume. (TL-4670)
AutoCopy, Instant DR, and Replication
 Error reported when replication jobs are unable to communicate with the
target system. A message similar to the following may be seen in the job log when
replication jobs are unable to communicate to the target system: “2012/06/04 12:23:07
[16392] ERROR> source_locator.rb:157 send_request - Received 'NoMethodError' while
connecting with '[e-mail]' for file, '/VAULT01/
M6000MAG/DT0005': private method `chomp' called for nil:NilClass”. (TL-7988)
WORKAROUND: Enable or reconfigure the replication target.
 If Instant DR fails it may leave a piece behind in the pool. If Instant DR fails, a file
with the .out extension may be left behind. These files are harmless and can be ignored.
(TL-8046)
 Replications should be kept to less than 400 virtual tapes at a time. Attempting to
replicate over 400 virtual tapes at one time will result in replication jobs that fail.
Additionally, the error message “Too many open files - Failed to initialize inotify: the user
limit on the total number of inotify instances has been reached.” will be seen in the job log.
(TL-8082)
 Instant Disaster Recovery will copy a virtual tape regardless of its locked state.
The user is expected to make sure that the virtual tapes that are to be copied by the
Instant DR feature are not currently in use during the copy actions. If a file is being
modified when Instant DR performs the copy of the file, the copied file may not be usable.
Therefore, it is highly recommended that Instant DR jobs be run during a time that there
is little to no data being written to the VTS. (TL-8117)
 Unusable hostname displayed on Hosted Data page if hostname cannot be
resolved. An appliance with an unresolved hostname has the name
“ip_address.confluence-networks.com” on the Hosted Data page. (TL-7595)
 Replicate jobs that use WAN Acceleration delete the file on the target before
transferring the file. The file is saved with a partial suffix until the transfer is
completed. Until the transfer is completed, the file will not be usable for tape
operations.(TL-7272)
 Replicate jobs with a large number of cartridges can cause failures. (TL-7372)
WORKAROUND: Do not replicate more than 200 cartridges per job.
 Ethernet interface becomes unresponsive during replication. When performing
many replications, it is possible that the Ethernet interface can become unresponsive.
This was discovered with 400 file replications. However, with 200 file replications in a
single job, the issue did not occur. (TL-7367)
WORKAROUND: Keep the number of replications per job at or below 200.