The following (just scripted, with commentary added) is a little long
and rather un-good.


samar 9# script
Script started, file is typescript

Position us at the start of the next volume. This volume has been
relabeled after setting the block size on the tape (last week).


samar 1# /usr/local/sbin/amcheck samar
Amanda Tape Server Host Check
-----------------------------
ERROR: holding disk /usr5/dumps/amanda: statfs: No such file or directory
amcheck-server: slot 7: date X        label SAMAR08 (exact label match)
NOTE: skipping tape-writable test
Tape SAMAR08 label ok
Server check took 4.311 seconds

Amanda Backup Client Hosts Check
--------------------------------
Client check: 1 host checked in 0.168 seconds, 0 problems found

(brought to you by Amanda 2.4.4p1-20030716)

Find out the current block size. I has expected 32768, it isn't.

samar 2# mt -f /dev/sdlt2 blksize

 Recommended tape I/O size: 131072 bytes (256 512-byte blocks)
 Minimum block size: 4 byte(s)
 Maximum block size: 16777212 bytes
 Current block size: Variable

Get the current label.

We failed unless we specified the blocksize of 32768.

samar 5# mt -f /dev/sdlt2 rewind

samar 2# dd if=/dev/sdlt2 of=./scratch
Read error: Invalid argument
0+0 records in
0+0 records out

samar 3# dd if=/dev/sdlt2 of=./scratch bs=32768
1+0 records in
1+0 records out

Many null characters removed from cat output.

samar 6# dd if=/dev/sdlt2 of=./scratch bs=32768
1+0 records in
1+0 records out
samar 7# cat -evt scratch | more
AMANDA: TAPESTART DATE X TAPE SAMAR08$
^L$
^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@

Set the blocksize to the drive recommended size.

samar 8# mt -f /dev/sdlt2 setblksz 131072


samar 9# dd if=./scratch of=/dev/sdlt2 obs 131072
Bad argument: "obs"
samar 10# dd if=./scratch of=/dev/sdlt2 obs=131072
Write error: Invalid argument
64+0 records in
0+0 records out
samar 11# mt -f /dev/sdlt2 blksize

 Recommended tape I/O size: 131072 bytes (256 512-byte blocks)
 Minimum block size: 4 byte(s)
 Maximum block size: 16777212 bytes
 Current block size: 131072 byte(s)
samar 12# dd if=/dev/sdlt2 of=./scratch bs=131072
0+0 records in
0+0 records out

samar 13# mt -f /dev/sdlt2 rewind

samar 14# dd if=/dev/sdlt2 of=./scratch bs=131072
0+0 records in
0+0 records out
samar 15# !cat
cat -evt scratch | more

samar 16# mt -f /dev/sdlt2 rewind

samar 17# dd if=/dev/sdlt2 of=./scratch bs=131072
0+0 records in
0+0 records out

samar 18# /usr/local/sbin/amcheck samar
Amanda Tape Server Host Check
-----------------------------
ERROR: holding disk /usr5/dumps/amanda: statfs: No such file or directory
amcheck-server: slot 7: reading label: Invalid argument
amcheck-server: slot 8: reading label: Invalid argument

Progressed this way through the rest of the tapes.

- use amtape to reset us to slot 7.

Probing the drive I find a block size of 32768 again.

samar 25# mt -f /dev/sdlt2 setblksz 131072

samar 26# /usr/local/sbin/amlabel -f samar SAMAR08
labeling tape in slot 7 (/dev/sdlt2):
rewinding, reading label, reading label: Invalid argument
rewinding, writing label SAMAR08
amlabel: writing label: Invalid argument

Have we hit some sort of amanda buffer size limit ?
Something that didn't error during the write but did 
during the read ?

Are the new drivers in the OS failing to properly handle the drive ?

My money is on an OS problem.

Is anyone successfully using SDLT 320 drives on an SGI/IRIX system ?
Note, I have the required APD drivers installed and configured.

I'm gonna open a call with SGI.

Jan 10 10:58:54 3D:samar TapeCheckError[234379]: 
|$(409)tps1d4,SDLT320,,pid=411237,read,Aborted command,IDE message received
Jan 10 10:58:54 6D:samar tps1d4[234379]: Command aborted by target -- 
Jan 10 10:58:54 3D:samar tps1d4[234379]: Aborted command

---
   Brian R Cuttler                 brian.cuttler AT wadsworth DOT org
   Computer Systems Support        (v) 518 486-1697
   Wadsworth Center                (f) 518 473-6384
   NYS Department of Health        Help Desk 518 473-0773