server backs up.
Linux OS: CentOS 4.1 (I think)
Kernel was recently updated (I'll find out what to) by another guy.
Latest of DirectAdmin.
There must be something wrong somewhere in a script or generally. At random moments, the server gets incredibly backed up, MySQL spawns way too many chld processes, (MySQL 4.0 was the problem, upgraded to 4.1 didnt help - just customers complaining) and same with httpd and exim (for some reason).
SSH etc is just unreachable,
every single site loads extremely slowly, or with a non-standard browser timeout, not at all. I have to just wait, and wait, and wait, until SSH logins in, to restart httpd, mysql and exim to fix it. Then it starts again later on.
Tips on to find out whats wrong with it?
I was in 'top' when it happened once, and I noticed a huge i/o backlog on the HD.. A sign of it going out? But it works fine normally.. Guess when some traffic hits?
Appreciated.
