Backup tapes marked as full - they are not

[woodings]

We have a Scalar I2000 Robot with 6 LT03 tape drives and 200 tapes - this is connected via fibre to Unix Networker server - O/S Solaris 10, we also have 2 Windows 2000 dedicated storage nodes , currently we only backup 10 hosts.
Version of Networker is 7.3.2 ( we are to upgrade to 7.3.3 shortly).We have utilised dynamic drive sharing - thus all 6 drives can be seen by all three nodes. We also have three tape pools - one for each node.

The backup tapes are being marked as full when they have as little as 5GB and as much as 500GB of data on them. Could anyone give any indication as to why.

Also we have three LT03 backup tapes that state on Networker GUI that there is 1718GB of data backed up to a specific tape, but they are LT03 tape 400/800 . How can this be??

 

[605]

If NW can not recover from a write error it will mark the tapes as full. more details will be available in the daemon.log file.

 

[woodings]

There are no error messages or write errors in /var/nsr/logs/daemon.log.
In /var/adm/messages when backups are running we get these sort of messages - Aug 29 08:00:41 dmbcs8 root: [ID 702911 daemon.notice] NetWorker media: (warning
) /dev/rmt/1cbn moving: fsf 13: drive status is The end of data was reached
Aug 29 08:00:53 dmbcs8 root: [ID 702911 daemon.notice] NetWorker media: (info) f
sf 13: drive status is The end of data was reached
Aug 29 18:05:34 dmbcs8 root: [ID 702911 daemon.notice] NetWorker media: (warning
) /dev/rmt/5cbn moving: fsf 10: drive status is The end of data was reached
Aug 29 18:05:46 dmbcs8 root: [ID 702911 daemon.notice] NetWorker media: (info) f
sf 10: drive status is The end of data was reached
Aug 29 19:09:36 dmbcs8 root: [ID 702911 daemon.notice] NetWorker media: (warning
) /dev/rmt/4cbn moving: fsf 20: drive status is The end of data was reached
Aug 29 19:09:52 dmbcs8 root: [ID 702911 daemon.notice] NetWorker media: (info) f
sf 20: drive status is The end of data was reached
Aug 29 20:01:18 dmbcs8 root: [ID 702911 daemon.notice] NetWorker media: (warning
) rd=dmbccorp2:\\.\Tape3 moving: fsf 92: drive status is The end of data was rea
ched
Aug 29 20:01:56 dmbcs8 root: [ID 702911 daemon.notice] NetWorker media: (warning
) rd=dmbccorp2:\\.\Tape3 moving: fsf 92: drive status is The end of data was rea
ched
Aug 29 20:01:56 dmbcs8 root: [ID 702911 daemon.notice] NetWorker media: (info) f
sf 92: drive status is The end of data was reached
Aug 30 04:01:44 dmbcs8 root: [ID 702911 daemon.notice] NetWorker media: (warning
) /dev/rmt/5cbn moving: fsf 61: drive status is The drive is not ready and the r
eason is not known of messages.

The example I have given is of a backup which was only 22GB in size but marked a LT03 tape as full - there were no other backups on this tape

 

[605]

It looks that NW found a logical EOT during positioning.
This could point to a hardware problem. It could also be that you might have overwritten a piece of tape which NW did not become aware, for instance due to SCSI Bus resets or no tape path isolation.

Please have a look at the DDS guidelines which you can fine in the Admin Guide.