Re: 'parity error' in a SCSI board

[<dsavage peaknet net>]

On Wed, April 30, 2003 12:42, Oliver Schulze wrote:
> Hi,
> I'm running RH9 and noted this error en /var/log/messages:
>
> "kernel: (scsi:A:6:0): parity error detected in Status phase.
> SEQADDR(0x7f) SCSIRATE(0x88)"
>
> The SEQADDR change with this values: 0x9e, 0xa5, 0x82
>
> The Adaptec card is: AHA-2940 U/UW/D AIC-7881U
>
> The mobo is an ASUS P4S-533-MX running a Pentium 4 1700Mhz
>
> Anyone has seen this message? I want to know if this is a critical data
> loss error.
> I also have seen this error during install of RH8.0 in the same machine.

Oliver,

I'll assume your cabling and terminations are OK.

Is this a one-time error, or at least an infrequent one?  If so, it's
probably a random data error on Channel A, SCSI ID=6, LUN=0 (a tape
drive?).  I wouldn't worry about it unless you start seeing multiple
errors. In that case you could have a cable problem or a device that needs
maintenance.

You could check the parity configurations of your Adaptec card and device
#6 to make sure they're the same. If you disable parity on device #6
(usually via a jumper), parity will still be generated but any errors will
not be transmitted down the cable to the host adapter.

--Doc Savage
  Fairview Heights, IL