(oops!  Apologies to Marc Schaefer - I sent this to him, rather than the list!)

Marc SCHAEFER <[EMAIL PROTECTED]> wrote:
> your problem is in the SCSI recovery. Diff scsi.c from both versions, and
> put back the | in the case switch.

Could this error be related to a tape problem I have?  IO errors on reading
the end of a tar file...?

First the hardware:
Machine 1:  DELL dual CPU
Linux nebula.mpn.com 2.0.33 #5 Tue May 12 13:04:22 BST 1998 i686 unknown
AHA1542 controller,
Host: scsi0 Channel: 00 Id: 05 Lun: 00
  Vendor: ARCHIVE  Model: Python 28454-XXX Rev: 4ASB
  Type:  Sequential-Access                ANSI SCSI revision: 02
(This is an old Sun DDS-1 tape drive)
 
Machine 2: Old DX2-66
Linux womble.mew.co.uk 2.0.34 #1 Wed Jul 1 22:55:20 BST 1998 i486
AHA1542 controller,
Host: scsi0 Channel: 00 Id: 00 Lun: 00
  Vendor: HP      Model: C1533A          Rev: 9406
  Type:  Sequential-Access                ANSI SCSI revision: 02
(This is a HP 6000 tape drive - DDS-2)
 
 
Now the problem.
 
On machine 1 I do:
  mt -f /dev/st0 setdensity 0
  mt -f /dev/st0 setblk 64
  tar cvbf 126 /dev/st0 directory
 
This writes the tape perfect.
 
On machine 2 I do:
  mt -f /dev/st0 setblk 64
  tar xvf /dev/st0
 
This reads the whole tape excect the last few data block.  I can read 600Mb of
data easily (if that's on the tape) _except_ the last 100 Kb!
 
The error I see is
st0: Error with sense data: Current error st09:00: sns = f0  3
ASC=11 ASCQ= 0
Raw sense data:0xf0 0x00 0x03 0x00 0x00 0x02 0x00 0x0e 0x00 0x00 0x00 0x00
0x11 0x00 0x00 0x00 
 
Now that says "read error".  Ugh.  I get exactly the same read error with a
different HP6000 drive.  But an old WangDAT3100 on the same system can read
the tape perfectly (or it could until it stopped recognising tapes at all and
just ejects them).
 
The HP6000 then flashes the "needs cleaning" light, which goes out if I put a
cleaning tape in... until I try another tape.  I've tried two different HP6000
drives.
 
This is tape independent and happens around 90% of the time, depending on what
I record on the tape (it's different every time).
 
My "fix" at the moment is to save three 1Mb padding files afterwards
("tar cvbf .... directory padding").  The last tape I restored
restored resulted in

% ls -l padding/
total 2984
-rw-r--r--  1 root    wheel    1048576 Apr  1 18:22 xxx
-rw-r--r--  1 root    wheel    1048576 Apr  1 18:22 xxx2
-rw-r--r--  1 root    wheel      943104 Aug  8 00:58 xxx3
 
xxx3 is exactly 103Kb short!
 
Any advice?
 
Thanks!!!!
 
rgds
Stephen

-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to [EMAIL PROTECTED]

Reply via email to