> Does anyone know what is the format for entries files.f file has?

You don't want to parse those manually.  Use bpflist instead:

# bpflist -l -client buckeye -backupid buckeye_1062219777 -d 01/01/2001


> I am trying to look information using which my script can findout
> what file is on what tape.

Here's a file from the output of the above:

233 0 28 46 33063 1 0 0 774 /var/spool/lpd/delhi/lock.pr 33152 lp lp 0
1059385969 1062219839 1062219839 

Fifth field indicates that this file starts 33063 blocks (16531.5 KB)
into the backup image.  Now I go look at buckeye_1062219777_INCR and
check out the fragments:

FRAGMENT 1 1 16000 0 2 13 48 QZS321 dilbert 32768 213062 1061662562 1
0 *NULL* 1064898177 0 65539 0 0 
FRAGMENT 1 2 608 0 2 13 48 QZS322 dilbert 32768 213062 1061662562 1 0
*NULL* 1064898177 0 65539 0 0 

Using the fourth field to determine the size of each fragment, I can
conclude that said file is in fragment 2, which is on QZS322.

Of course, that's only where the file STARTS.  It could of course
reside in multiple fragments and multiple tapes.  Assume for the
moment we're not using multiplexing.  Using size of file (bpflist
field 14) to list all needed fragments is left as an exercise for the
reader.  

Now, if mpx is in use, I think you can only find the start.  We can't
use the same method in the exercise above because there could be other
data in same fragments from other streams.  Maybe one could look at
all fragments in all other streams which use same tape(s), using the
offset in the FRAGMENT line and determine how much of what was where.
I don't think even that would work.

-- 
Larry Kingery 



       Computers are not intelligent. They only think they are.