HowToFixComputers.com

Christopher Tidy · Guest

Hi all,

I did a bit more investigating on the SCSI error problem. I am 95%
sure all the disks are good. The two 36 GB disks I just bought with a
guarantee, and the two 9 GB disks I've been using for over two years
with no problems. None of them show any grown defects. I don't think
it can be a cabling problem like Tim suggested, because the disks
connect directly to the motherboard.

Then I had a thought. I realised that each time I got a SCSI error
shown in /var/adm/messages, it corresponded to a time when I was
writing a CD using my ancient PlexWriter PX-R412Ci in a Sun SCSI box.
It's on the same SCSI controller as the two hard disks. So I changed
the hard disks, putting back my old disks and tried writing some CDs.

With Fujitsu 36 GB MAN3367MC disks (my new disks) and Solaris 10, I
get the following error:

Feb 2 01:31:05 einstein scsi: WARNING: /sbus@1f,0/SUNW,fas@e,8800000
(fas0):
Feb 2 01:31:05 einstein Connected command timeout for Target
5.0
Feb 2 01:31:05 einstein scsi: WARNING: /sbus@1f,0/SUNW,fas@e,8800000
(fas0):
Feb 2 01:31:05 einstein Target 5 reverting to async. mode
Feb 2 01:31:08 einstein scsi: WARNING:
/sbus@1f,0/SUNW,fas@e,8800000/sd@0,0 (sd3):
Feb 2 01:31:08 einstein Error for Command:
write(10)
Error Level: Retryable
Feb 2 01:31:08 einstein scsi: Requested Block:
44731672
Error Block: 44731672
Feb 2 01:31:08 einstein scsi: Vendor:
COMPAQ
Serial Number: UFA1P270FM2E
Feb 2 01:31:08 einstein scsi: Sense Key: Unit Attention
Feb 2 01:31:08 einstein scsi: ASC: 0x29 (scsi bus reset occurred),
ASCQ:
0x2, FRU: 0x0

With Fujitsu 9 GB MAG3091LC disks (my old disks) and Solaris 9 I get
the following error:

Feb 5 11:10:03 cdt22 scsi: [ID 365881 kern.info] fas:
5.0: cdb=[ 0x44 0x0 0x0
0x0 0x0 0x0 0x0 0x0 0x8 0x0 ]
Feb 5 11:10:03 cdt22 scsi: [ID 107833 kern.warning]
WARNING: /sbus@1f,0/SUNW,fa
s@e,8800000 (fas0):
Feb 5 11:10:03 cdt22 Connected command timeout for
Target 5.0
Feb 5 11:10:03 cdt22 scsi: [ID 107833 kern.warning]
WARNING: /sbus@1f,0/SUNW,fa
s@e,8800000 (fas0):
Feb 5 11:10:03 cdt22 Target 5 reducing sync.
transfer rate
Feb 5 11:10:03 cdt22 scsi: [ID 107833 kern.warning]
WARNING: /sbus@1f,0/SUNW,fa
s@e,8800000/sd@0,0 (sd22):
Feb 5 11:10:03 cdt22 SCSI transport failed: reason
'reset': retrying command
Feb 5 11:28:54 cdt22 scsi: [ID 365881 kern.info] fas:
5.0: cdb=[ 0x44 0x0 0x0
0x0 0x0 0x0 0x0 0x0 0x8 0x0 ]
Feb 5 11:28:54 cdt22 scsi: [ID 107833 kern.warning]
WARNING: /sbus@1f,0/SUNW,fa
s@e,8800000 (fas0):
Feb 5 11:28:54 cdt22 Connected command timeout for
Target 5.0
Feb 5 11:28:54 cdt22 scsi: [ID 107833 kern.warning]
WARNING: /sbus@1f,0/SUNW,fa
s@e,8800000 (fas0):
Feb 5 11:28:54 cdt22 Target 5 reverting to async. mode

The error messages aren't quite the same, but they both mention Target
5, which is my CD writer. Now I haven't used the CD writer in quite a
while. Maybe the CD writer or the Sun SCSI box is faulty? Or perhaps
the cable which connects the CD writer to the computer? But the CDs
which it writes are okay, so to me a faulty cable seems unlikely.

It seems to have really scr**ed up my file systems on the new disks.
Even fsck can't repair them to its satisfaction. When running fsck I
get the following error in /var/adm/messages:

Feb 5 12:59:35 einstein pseudo: [ID 129642 kern.info] pseudo-device:
wrsmd15
Feb 5 12:59:35 einstein genunix: [ID 936769 kern.info] wrsmd15 is /
pseudo/wrsmd@15

Any idea what that last error means?

Does anyone have an opinion on whether the symptoms fit a faulty
external CD writer? Suggestions would be appreciated.

Best wishes,

Chris Tidy

Richard B. gilbert · Guest

Christopher Tidy wrote: