[818] in linux-scsi channel archive
SCSI problem I think??
daemon@ATHENA.MIT.EDU (scsi@ns.cococo.net)
Sat Oct 19 17:15:04 1996
Date: Sat, 19 Oct 1996 17:06:56 -0400 (EDT)
From: scsi@ns.cococo.net
To: linux-scsi@vger.rutgers.edu
In-Reply-To: <Pine.LNX.3.91.961019165335.26943A-100000@ns.cococo.net>
Hey all,
I have recently started receiving a stream of messages on my
Linux Box, it is just a stream of what appears to be hex digits enclosed
inside < >. I am running kernel 2.0.18, and have been for several weeks
without any problems at all. About 4 days ago I got a frantic call, that
said the server was down, and it was just scrolling garbage on the
screen. When I went to investigate it was the above symptom I described.
I had no choice, but to power off the unit. When I powered it back up
when it got to part about resetting scsi bus, right after the reset the
same symptom started again, an endless stream of of hex digits enclosed
in < > symbols started. I finally turned off the machine and left it off
for about 2 minutes and it booted up, with many disk errros, which was
not unexpected, because of the power down I did. The system came back up
and ran for another 2 days, then it sort of locked up, wouldn't do name
serving. It is our name server and mail server and www server, so I
rebooted. When it came to the resetting SCSI bus on the boot up, the
stream of endless hex digits started again. I turned it off for a couple
of minutes and all was fine, no file system problems at all, because I
was able to reboot I guess. It is an adaptec 2940 controller with a 2
gig Seagate ST32151N drive and a Sanyo CD CRD-254S . Has anyone experienced
anything similar and if anywhere, where would you suggest I start to try
to find the culprit.
Sorry I can't supply any more details about the error messages,
but they are just flying by at the the speed of light, and can't stop
them other than power off.
By the way the system has been running for the last 3 days
without any more problems, but I feel sure it is going to happen again.
The machine has been running since June without anything like
this happening, it was started with kernel 1.3.100 and has been upgraded
thru the kernel releases, just stopped at 2.0.18, the only reason it is
running 2.0.18 is, thought I would wait for a few more kernel releases
before upgrading the kernel again.
Later
Kelley
