Re: More info on system crash..
Drew Eckhardt (drew@poohsticks.org)
Tue, 04 Jul 1995 00:44:36 -0600
In message <Pine.LNX.3.91.950627140241.9800A-100000@main.cccis.sfos.ro>, gafton
@cccis.sfos.ro writes:
>On Mon, 26 Jun 1995, Dave Wreski wrote:
>
>> Jun 26 08:27:06 gossamer linux: scsi0 : resetting for second half of retries
>.
>> Jun 26 08:27:06 gossamer linux: SCSI disk error : host 0 id 0 lun 0 return c
>ode
>> = 27070000
>> Jun 26 08:27:07 gossamer linux: scsidisk I/O error: dev 0803, sector 376932
>> Jun 26 08:27:07 gossamer linux: IO error syncing ext2 inode [0803:0000b968]
>> Jun 26 08:27:07 gossamer linux: IO error syncing ext2 inode [0803:0000b968]
>
> This was happening to me with a damaged HD both on a NCR-810 and
>AHA1542CF. The NCR complained about resetting ... and locked the machine
>rock solid, while Adaptec was giving me those messages all the time if I
>tryed to dd /dev/sda to null (because the Adaptec geometry differs from
>NCR and I couldn't imagine how to mount the root fs from /dev/sda).
For future reference, the geometries are irrelevant as far as Linux is
concerned. Linux just cares about the absolute sector location, which
(unless you changed your partition table) stays the same irregardless
of what mapping is being used.