It's a hardware problem. SCSI timeouts and transport failures almost always
mean a problem somewhere on the bus. Could be a bad cable, terminator,
adaptor or tape drive. Replace them in order of least to most expensive.

If it were a kernel problem, you never would have been able to back up in
the first place.

Or you could just call Cybernetics for support. Their gear can be
troublesome, but they're very forthcoming with support, in my experience...


"mei" wrote in message
om...
Hi all,
I got a problem on backup server again. I have read the manaul and
search on the web for a long time, but couldn't find any information.
I hope any of you can give me some kind of suggestion.

The backup acted weird last Sat while I was replacing the tape. At
first,
I could not remove the volume index out of the pool. After I fixed
this problem and tried to label the tape. It gave me the error on each
tape, said that bad file number. I have successfully finished labling
three tapes. At the fourth tape, it took more than half an hour, but
was still not done. So cancelled the operation of labeling. The
problem happened during the cancelling. The scsi error occurrs, had
the error message: Read open error, I/O error. The autochange stops in
the middle of the CY-TL8-2015 tape library subsystem. The light is
green for the autochanger. Then I opened the door, found the tape is
still mounted. After close the door, the light of the autochager stay
in red. Then I checked the log file , have some error messages like
this:

Sep 27 17:32:17 beverly unix: SCSI transport failed: reason
'timeout': givingup^M^M
Sep 27 17:32:17 beverly root: Solstice Backup media: (warning)
/dev/rmt/1mbn reading: I/O error

Sep 27 17:45:03 beverly unix: Disconnected command timeout for
Target 1.0
Sep 27 17:45:04 beverly unix: NOTICE: lus_intr(1.1.0): transport
failure (timeout)
Sep 27 17:46:05 beverly unix: NOTICE: lus_intr(1.1.0): transport
failure (incomplete)


The backup server is running SUNOS 5.6 with raid 5 external disks and
the autochager driver is CYBERNETICS CY-TL8-2015

Someone indicated that this may be the kernel error.
I have checked the /kernel/drv/st.config. Nothing was wrong there.

I don't know if I can turn off the power of the autochanger and
restart it.


Thanks so much!

Mei