Plesk 9.2.1: processor pegs out

[Highland]

I just upgraded my CentOS4 box to 9.2.1 (used yum). My problem is the server quickly pegs itself out now. I've had spikes as high as 95.0 (seems to level at around 80-ish) and the server is largely unresponsive. I have to use the 1and1 panel to reset the box but that only restores functionality for a couple of minutes as it quickly returns to pegging out. I can't figure out any one particular cause. I've tried stopping clamd, qmail and named and no luck. Next would be trying mysql but I don't want to kill it off if I don't have to but it's the only other high processor usage program running.

Any other ideas?

[Highland]

I think it was the qmail scanner that was doing it. Removed it and so far no spikes

New problem, tho: my Plesk sessions are timing out super fast. Any ideas?

[Highland]

psmon appears to have found part of the problem. It's emailing me this

Code: Select all

Spawned 'httpsd' with '/sbin/service psa start'

[Highland]

Couldn't find it from other searches. Finally found the psmon issue from this thread
http://www.atomicrocketturtle.com/forum ... f=3&t=2872