[cobalt-users] Whats Causing The Overload

["ISEE Multimedia" <mail@xxxxxxxxxxxxxxxxxxx>]

Hi All.

I really need some help in sorting this one out.

I am running a Raq4 256 and it seems to be dying on me nearly every day now.

I have some questions and information which should help you to help me. I
have searched and done as much as i could find in the archives and on
goodle, and im a little wiser but not much. Any pointers, help you can give
me would be much appreciated.

1. Log rotation / analog / split logs seems to take forever to complete, and
i would say this is a factor to the CPU / MEM usage. Is there a  way to
check what time the cron for this starts? as I think its not starting at
4:00am and how do i change the time to say start at 1am.

I've pasted the output of /var/log/httpd below minus empty files.

-rw-r--r--   1 root     root       289499 Dec 19 11:46 access
-rw-r--r--   1 root     root      2572724 Dec 19 11:33 access.1.gz
-rw-------   1 root     root            0 Dec 13 18:46 access.save
-rw-r--r--   1 root     root     31644310 Dec 19 11:45 adm_access
-rw-r--r--   1 root     root      5973508 Dec 19 11:45 adm_error
-rw-r--r--   1 root     root       231989 Dec 19 11:45 error
-rw-r--r--   1 root     root        99701 Dec 18 12:37 error.1.gz
-rw-r--r--   1 root     root       863803 Dec 19 09:52 fpexec_log

2. I have been monitoring 'top' this morning and seen some worrying things.
The load average at one point reached 99 !!! There were 300+ processes
running at one point.

It seems that something is adding to the processes but not releasing them,
also the memory usage was going up and up and not clearing down.

I have made some screenshots of various things for you if this helps in
working out whats going on.

http://www.isee-multimedia.com/cobalt/Untitled-1.gif  This is from top - PID
573 was running between 40 and 99% CPU usage.

http://www.isee-multimedia.com/cobalt/Untitled-2.gif This is ps -aux showing
PID 573

http://www.isee-multimedia.com/cobalt/Untitled-3.gif This is from procinfo
you can see i tried to kill the pid but i think i missed something.

I don't know if this helps you at all, but it would be nice to find out what
is causing all the processes, and the memory usage. In the end i did an
httpd restart and at the moment all seems to be OK.

Thank you all for your help in advance.

Kind Regards.

Mark Priest
ISEE Multimedia