[Bacula-users] how long to move to end of data? - solved

[Dan L. <da...@la...>]

This problem is solved.  I discovered the issue *as* I was typing up the 
email.   But since it is written, it might as well go into the archives.

The problem, worded in different ways for searching:

* tape takes too long to get to end
* fast forward on tape takes too long

The issue I noticed this today:

12-Dec 14:35 kraken-sd JobId 41700: Volume "ETU014" previously written, 
moving to end of data.
12-Dec 14:40 kraken-sd JobId 41700: Ready to append to end of Volume 
"ETU014" at file=11.

The issue: it takes 5 minutes to get to the end of the tape.  The drive 
in question is a Digital DLT 7000 model TZ89N-AV.  Pictures of the drive 
appear here:

   http://www.freebsddiary.org/tape-library-replacing-drive.php

The device resource for this drive is:

Device {
   Name                    = "DTL01"
   Description             = "Digital DLT MiniLibrary - first drive"
   Media Type              = DLT
   Archive Device          = /dev/nsa0

   Autochanger             = yes
   Drive Index             = 0

   Offline On Unmount      = no
   Hardware End of Medium  = no
   BSF at EOM              = yes
   Backward Space Record   = no
   Fast Forward Space File = no
   TWO EOF                 = yes

   Spool Directory         = /home/bacula/spooling-lib
   Maximum Spool Size      = 11759496889
   Maximum Job Spool Size  = 11759496889    # 10GB
}

The OS is FreeBSD 8.2 and I'm running Bacula 5.0.3

After changing 'Fast Forward Space File' from no, to yes, things went 
much faster.  The next job was much faster:

12-Dec 15:17 kraken-sd JobId 41703: Volume "ETU014" previously written, 
moving to end of data.
12-Dec 15:18 kraken-sd JobId 41703: Ready to append to end of Volume 
"ETU014" at file=13.

:)

I love it when I discover the solution *while* typing up the email.  ;)

-- 
Dan Langille - http://langille.org/