when a server goes into the red zone and you sit there trying to find whats causing this overload, whats the best and fastest way? (top really doesnt tell you anything)

load average: 134.46, 267.87, 161.64

ps aux usually does the trick for me.

Wow.. those are high load averages. top should tell you *something*. (Are any processes taking up a large percent of CPU?) You can also look at a ps -aux as mentioned to see what processes are running. You may find an abnormally large number of httpd processes or something like that. If the server is still responding fine despite the high load, you may have zombie processes, in which case you will want to figure out what is causing them to go into that state.

It seems they found out that its mysql causing the problem, is there anyway to find out what db and user is causing the out of controlness?

Yes... log into your MySQL command-line interface and enter the command:

mysql> SHOW PROCESSLIST;

You will need to log in to mysql as root or a user with the "process" privilege.

You should also be able to do this from myphpadmin if you use that.

Doug

Jasen also said that the conf file for mysql had 2,000 processes avail, so he changed it to 50, so far, servers been OK

root@wolf [/home]# uptime
5:48pm up 1:19, 1 user, load average: 0.34, 0.24, 0.31