> You might try noflushd -- it replaces bdflush, with reasonable ideas
> of how a laptop should use things.  You can also change the settings
> for update in your /etc/init's.  Even when you get these things taken
> care of, you may find that it spins up more often than you want.
> 
> I just got this from a guy in Australia, and it's a good procedure to
> be a bloodhound and find those processes that are spinning you up:
> 
> "After each disk spin up run the following in
> /var/log
> 
> ls -lt *
> ls -lt */*
> ls -lt */*/*
> 
> and so on looking for files that have just been updated."
> 
> He said he thought he also had to make some adjustment to exim
> (Debian) -- but you might see if your MTA has something to mess with.
> Apache, too.
> 
> Keep in touch about it, though.  I feel I'm on the right track with
> it, but haven't completely nailed it.  Maybe if we finish it, we could
> either add it to the Battery-Powered HOWTO (have you looked at that?
> Sounds like you know what it covers) or maybe write up a Truly-Fixed
> Disk HOWTO (w/o using a nailgun.

yea, I did

$ls -lRt /var/log | grep "`date +'%b %e %k:%M'`"

and only saw /var/log/messages

thats how I knew it was syslogd

	marc