Runaway Apache Processes - Second Victim
am 04.02.2010 20:24:58 von John ThomasI have been using Apache for several years without trouble on CentOS 4
originally, CentOS 5 now. Most of my applications are php/mysql and
mostly Drupal.
FYI, I have run into a problem that sounds identical to Dan Bunyards
thread here:
http://markmail.org/thread/pnmwmalsfe6c7j6c
Everything runs fine, then all of a sudden I have 12 or so Apache
processes all trying to use 100% of the CPU and the system slows to a
crawl. I can replicate the issue by running:
ab -n 100 -c 25 http://example.com/
The problem seems to be getting worse (today, a hacker sent a bunch of
urls to my server looking for phpmyadmin and that took me down). I have
a very low traffic server with personal websites only. I have a busy
day when Google crawls my system. I have about 1.5GB of ram.
List of some rpms I use:
httpd-2.2.3-31.el5.centos.2
php-5.3.1-1.el5.remi
php-mysql-5.3.1-1.el5.remi
php-mysql-5.3.1-1.el5.remi
I watch htop when it happens and the server seems to process everything
fine for a while with httpd and mysqld using a good amount of cpu, then
mysqld use slows and many more httpd processes start using some cpu.
Then about 12 or so (never counted) httpd procesess use the cpu about
equally and the server is nearly unresponsive. The memory was not
exhausted.
I have tried many, many things in my configuration hoping I could figure
it out. I do not know how to run straces or other developer type
things, but I can probably figure it out if you give me some hints.
--
Sincerely,
John Thomas
------------------------------------------------------------ ---------
The official User-To-User support forum of the Apache HTTP Server Project.
See
To unsubscribe, e-mail: users-unsubscribe@httpd.apache.org
" from the digest: users-digest-unsubscribe@httpd.apache.org
For additional commands, e-mail: users-help@httpd.apache.org