Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 26 May 1997 14:06:38 +0200
From:      Stefan Esser <se@FreeBSD.ORG>
To:        Andrew Wilson <andrew@aaaaaaaa.demon.co.uk>
Cc:        stable@FreeBSD.ORG
Subject:   Re: NCR controller problems since latest 2.2 upgrade?
Message-ID:  <19970526140638.60197@x14.mi.uni-koeln.de>
In-Reply-To: <199705241201.NAA00229@aaaaaaaa.demon.co.uk>; from Andrew Wilson on Sat, May 24, 1997 at 01:01:06PM %2B0100
References:  <199705241201.NAA00229@aaaaaaaa.demon.co.uk>

next in thread | previous in thread | raw e-mail | index | archive | help
On May 24, Andrew Wilson <andrew@aaaaaaaa.demon.co.uk> wrote:
> Hi,
> 
> 	I used cvsup to grab the latest 2.2-STABLE on Friday 
> night, ran 'make world' and almost immediately began seeing problems
> reading my SCSI disk.
> 
> The several errors on bootup report as:
> 
> sd0(ncr0:0:0): MEDIUM ERROR info:62686 asc:11,0 Unrecovered read error, retries 3

Well, a MEDIUM ERROR is just that: A medium error ...

> A subsequent reboot reported no errors.  The error seems to appear
> erratically.
> 
> QUESTIONS:
> 
> 1.	is this problem a result of bad software in the kernel?
> 2.	does this mean the NCR controller is toast?
> 3.	does this mean the disk is trashed?
> 4.	any recommendations?

Do you have automatic bad block replacement enabled for that 
drive ? (You definitely should !)

See below for instructions ...

> (ncr0:0:0): "Quantum XP32150 576D" type 0 fixed SCSI 2

Ahh, a decent drive!
Mine used to print exactly the same message, until 
I updated the drive's firmware, and now it says:

sd0: <Quantum XP32150 L915> type 0 fixed SCSI 2


Now the instructions for turning on bad block replacement:

# scsi -m 1 -P 3 -f /dev/rsd0 -e

This calls the editor defined in your EDITOR environment 
variable, and you should change the buffer to read:

AWRE (Auto Write Reallocation Enbld):  1 
ARRE (Auto Read Reallocation Enbld):  1 
TB (Transfer Block):  0 
RC (Read Continuous):  0 
EER (Enable Early Recovery):  0 
PER (Post Error):  1 
DTE (Disable Transfer on Error):  0 
DCR (Disable Correction):  0 
Read Retry Count:  8 
Correction Span:  80 
Head Offset Count:  0 
Data Strobe Offset Count:  0 
Write Retry Count:  8 
Recovery Time Limit:  0 

The AWRE and ARRE enable bad block replacement, if set to 1.
You will have to scan the disk (for example by using "dd" on 
the raw device), and this should take care of the replacement,
or will at least assign alternate sectors. If the sector can't
be reconstructed, then a "forced error" will be introduced in
the replacement sector, which will vanish as soon as you write
new data into it the next time. If you continue to get read
errors, then you may need to delete the file containing the
defective sector, and copy it back from an installation medium
or backup.

Regards, STefan



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19970526140638.60197>