Re: [TCLUG:6646] server crash

[Peter Lukas <peter@math.umn.edu>]

This can often happen when logging has nowhere to go.  For example, if the
logs are being dumped to the root partition (really bad idea), the machine
will eventually fill up and run out of space for any additional
operations.   Also, Linux isn't totally bulletproof and can be hardened by
much more than `rpm -Uvh kern*.rpm.`  Read up on the LDP
(usually /usr/doc or ftp://sunsite.unc.edu/pub/Linux/docs) for more
information on Linux/Unix server performance tuning.  RedHat ships a
multipurpose kernel that isn't as robust as one you could build yourself
by hand.

Peter Lukas

On Mon, 21 Jun 1999, Mike Hicks wrote:

> > Was this machine on a publicly accessible network? There are many DOS's
> > for 2.0.35, some of which may have results similar to what you describe.
> > My recommendation would be to upgrade to 2.2.10 and see what happens.
> 
> It might be better to go to 2.0.36 instead, since 2.2 changes several
> things..  Besides, it's pretty easy to upgrade to..
> 
> Download the appropriate RPMs (if you're using RPMs)
> upgrade the packages with 'rpm -Uvh <file>.rpm'
> edit /etc/lilo.conf to point to the new kernel image
> run /sbin/lilo
> reboot (and cross your fingers ;-)
> 
> Anyway, just had to pipe in...
> 
> Mike Hicks
> 
> (missing .sig -- ack!)
> 
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: tclug-list-unsubscribe@mn-linux.org
> For additional commands, e-mail: tclug-list-help@mn-linux.org
> 
>