[401] in linux-scsi channel archive
Re: What's this error message?
daemon@ATHENA.MIT.EDU (Drew Eckhardt)
Wed Jul 19 03:48:31 1995
To: schweikh@itosun.ito.uni-stuttgart.de (Jens Schweikhardt)
cc: submit-linux-dev-scsi@ratatosk.yggdrasil.com
In-reply-to: Your message of "18 Jul 1995 09:26:51 GMT."
<3ufuor$1a4l@info4.rus.uni-stuttgart.de>
Date: Tue, 18 Jul 1995 21:10:24 -0600
From: Drew Eckhardt <drew@poohsticks.org>
In message <3ufuor$1a4l@info4.rus.uni-stuttgart.de>, schweikh@itosun.ito.uni-st
uttgart.de writes:
>Hello *,
>
>every 10 to 20 hours I get the following scsi error.
>At the same time, something has gone heavily wrong,
>i.e. the Xserver crashed or bash did weird stuff.
>
> SCSI disk error: host 0 id 0 lun 0 return code = 28000002
> extra data not valid current error sd801: sense key Unit Attention
> Scsidisk I/O error: dev 0801, sector 1032210
> kernel panic: EXT2-fs panic (device 8/1): ext2_read_inode:
> unable to read i-node block - inode=129050, block=516105
>
>(sector, block and inode numbers vary)
>
>Question: who is responsible, my controller (Adaptec 1542CF)
>or my drive (Segate ST 13200N, 1GB)?
>I'd appreciate any explanation on what this message means
The drive told us something was wrong by returning a CHECK CONDITION
status on the last read command we sent it. The Adaptec firmware decided
that this wasn't the best thing, and asked the drive for more specifics
using a REQUEST SENSE command. The drive told us that there was a
UNIT ATTENTION condition pending, which usually happens on powerup,
after a SCSI bus reset, after a media change, etc (although usually it will
TELL you the exact cause in the extended sense information).
Given the age of the drive (not very old), and absence of any extended
sense information that would indicate a bus reset or something else, I'd
suspect a firmware problem.
