Subject: Re: low-level error conditions
To: Jukka Marin <firstname.lastname@example.org>
From: Phil Knaack <email@example.com>
Date: 01/29/1997 13:21:15
>Why do some error messages go to the dmesg buffer (is that a 'kernel message
>buffer'?)? Some examples of these are "arp - address XXXX overwritten by.."
>and some SCSI messages. Yeah, and the serial port silo overflow messages.
>I find this annoying because the system configuration information gets lost
>sooner or later and I know no other way of getting it back but a reboot.
Well, I can't answer the "why" except to hazzard a guess that
it's historic; but, I can suggest a solution to the last part.
On all my machines, I've taken to putting
rm -f /var/log/dmesg-log
dmesg > /var/log/dmesg-log
in my rc scripts, so as to save boot messages. This has proven handy
at work in many cases because I work a lot with users' floppies, often
with bad sectors; this also overflows the kernel buffer. (I've also got
local mods which determine who the machine is by sed'ing the MAC addr
out of the dmesg and determine the rc actions to take; however its a
really really gross hack, not worth mentioning except in acknowledgement
of its existence. :)
Phillip F Knaack
Systems Administrator, Information Development for Extension Audiences (IDEA)
Iowa State University Extension