Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 16 Jan 2005 19:26:38 +0200
From:      "Reko Turja" <reko.turja@liukuma.net>
To:        <freebsd-questions@freebsd.org>
Subject:   Re: help logging a crash
Message-ID:  <026901c4fbf0$895490c0$92a7cb52@rekon>
References:  <6.2.0.14.2.20050116103645.00becec8@cheyenne.wixb.com> <A480AC9E-67DE-11D9-8D05-000D9333E43C@secure-computing.net> <6.2.0.14.2.20050116105704.00c08218@cheyenne.wixb.com> <025001c4fbef$01942ca0$92a7cb52@rekon> <6.2.0.14.2.20050116111733.00c20ed8@cheyenne.wixb.com>

next in thread | previous in thread | raw e-mail | index | archive | help
> for the moment, I only did the kernel. I dont want to spend alot of 
> time on this if its still unstable.
>
> Looks like as long as I add DDB to the kernel and make it, I should 
> get some crash dump ?

Yes, but you'll need to define the dump device and place to save the 
dump from the swap in the next reboot. In addition

makeoptions DEBUG=-g is needed in the kernel config as well (symbols!) 
and
options DDB_UNATTENDED can be a good idea.

I think that separate kernel and userland is a bad idea, especially now, 
when you have 5.3_STABLE kernel, and 5.3_RELENG userland. There are 
small and subtle changes in both which make the correct userland/kernel 
combination vital and following the correct procedure of 
compiling/installing them.

-Reko 



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?026901c4fbf0$895490c0$92a7cb52>