I would assume that since in the first sentence of the e-mail it was mentioned that it is a busy web server, that could be the problem. If an inefficiently written PHP/Perl, etc script is being run with each visitor, that could easily explain the load. Checking the access logs to find out the rate of requests could help you decide if this is the problem. The other good explanation is many connections being opened to the server and that combined with KeepAlive and host name look ups (if enabled) could be causing your heavy usage problems. Optimization could help you, there are plenty of Apache Optimization HOWTOs out there. Good luck Andy Schmid wrote: > It sounds like he knows what is eating up the CPU (httpd) so top would > be kind of useless. You could try using a couple of tools such as lsof > or strace (one of my favorite), as well as look through your apache logs > to see what is being accessed. This all might give you a clue to what > is causing the problem. Also, if the load is due to a high IO wait, > then this may hint to a hardware issue. > > Another thing you could try, see if you can downgrade the version of > apache installed. Perhaps the high load is caused by a bug (its > happened to me before). > > Andy > > Jima wrote: > > >> On Thu, 22 Mar 2007, admin at lctn.org wrote: >> >> >> >>> I have a pretty busy Centos server running httpd-2.0.52-22.ent.centos4. >>> Httpd is running between 50-98%. How can I diagnose what is causing it use >>> so much of the CPU? >>> >>> >>> >> Try `top`. >> >> Jima >> >> _______________________________________________ >> TCLUG Mailing List - Minneapolis/St. Paul, Minnesota >> tclug-list at mn-linux.org >> http://mailman.mn-linux.org/mailman/listinfo/tclug-list >> >> >> > >