SCSI madness

[Bill Campbell]

On Thu, Jan 13, 2005, Net Llama! wrote:
>On Thu, 13 Jan 2005, A. Khattri wrote:
>> On Thu, 13 Jan 2005, Net Llama! wrote:
>>
>> > I've been tearing my hair out for the past 4 days trying to deal with SCSI
>> > errors on a box runnign FC3.
>> >
>> > For the past week or so, whenever I boot up, see the following error on
>> > the console:
>> > sym0: SCSI parity error detected: SCR1=132 DBC=50000000 SBCL=0
>> >
>> > When the above error occurs, the box functions just fine, but i fear for
>> > my data.  In fact, sometimes, i don't get the error above, and the
>> > box boots up fine, and then as soon as the login prompt appears, i start
>> > seeing SCSI reset/abort errors instead.  When the reset/abort errors show
>> > up then i can't do anything, all input fails to register and the box can
>> > only be reset.
>> >
>> > FWIW, i'm using an old symbios 8951 SCSI controller.  I've replaced the
>> > SCSI drive, the SCSI cable, and now the controller, and yet the fscking
>> > error persists.
>> >
>> > I'm at a loss as to how this error can persist when i've replaced every
>> > component on the chain.
>> >
>> > Anyone have any ideas?
>>
>>
>> Did some Googling and this might be a flaky Symbios driver in the 2.6
>> kernel:
>>
>> http://lists.suse.com/archive/suse-linux-e/2004-Oct/1745.html
>
>Hrmmm,maybe, but i've been using a 2.6.x kernel on this box for over a
>year now, and this problem started about a week ago.

That sounds like a hardware or cabling/termination problem, and given that
it's SCSI, the first thing I would check is that all the cables are firmly
seated.  Removing and replacing the cables may help to clean the contacts.
Active terminators are far superior to depending on the internal
termination of the SCSI devices.

Reset/abort errors may well indicate a failing hard drive.  I would suspect
the electronics on the HD rather than the media.  Overheating may well be a
problem.

Bill
--
INTERNET:   bill at Celestial.COM  Bill Campbell; Celestial Software LLC
UUCP:               camco!bill  PO Box 820; 6641 E. Mercer Way
FAX:            (206) 232-9186  Mercer Island, WA 98040-0820; (206) 236-1676
URL: http://www.celestial.com/

``There is nothing as stupid as an educated man if you get him off the
thing he was educated in.''
    Will Rogers