This message is broken into two parts because the MySQL list blocked it as being too
big!!!
Okay, here is the problem: everyone once in a while, say once or twice a day, my web
server running RH 7.2, PHP 4.0.6 and MySQL, will seem to freeze (from a web users
point of view). Upon SSH'ing into the server, and running "top" I see the system load
averages in the 90s (!!!) and all I see for processes on the list is mysqld. The CPU
usage is at 100%, and the top mysqld process ususally is taking 14% or so, then 12%
then a lot of 10%, and on down until it goes off the list.
Below, I have included a copy of the "top" status during one of the afore mentioned
problem times:
3:17pm up 23 min, 1 user, load average: 87.18, 54.78, 24.56
282 processes: 203 sleeping, 79 running, 0 zombie, 0 stopped
CPU0 states: 96.14% user, 3.0% system, 0.0% nice, 0.0% idle
CPU1 states: 95.5% user, 4.9% system, 0.0% nice, 0.0% idle
Mem: 1026964K av, 757988K used, 268976K free, 56K shrd, 8116K buff
Swap: 2048248K av, 0K used, 2048248K free 312316K cached
PID USER PRI NI SIZE RSS SHARE STAT %CPU %MEM TIME COMMAND
4784 mysql 18 0 28472 27M 2236 R 15.5 2.7 0:02 mysqld
4944 mysql 18 0 26316 25M 2236 R 7.6 2.5 0:00 mysqld
4940 mysql 19 0 26316 25M 2236 S 7.2 2.5 0:00 mysqld
4942 mysql 17 0 26316 25M 2236 S 7.2 2.5 0:00 mysqld
5049 mysql 19 0 24948 24M 2236 R 3.9 2.4 0:00 mysqld
4810 mysql 20 0 26316 25M 2236 R 3.1 2.5 0:01 mysqld
4808 mysql 20 0 26316 25M 2236 R 3.0 2.5 0:01 mysqld
4806 mysql 17 0 26316 25M 2236 R 2.8 2.5 0:01 mysqld
4727 mysql 20 0 28472 27M 2236 R 2.7 2.7 0:02 mysqld
4751 mysql 20 0 28472 27M 2236 R 2.7 2.7 0:02 mysqld
4759 mysql 20 0 28472 27M 2236 R 2.7 2.7 0:02 mysqld
4794 mysql 20 0 26468 25M 2236 R 2.7 2.5 0:01 mysqld
4799 mysql 14 0 26204 25M 2236 R 2.7 2.5 0:01 mysqld
4804 mysql 15 0 26316 25M 2236 R 2.7 2.5 0:01 mysqld
4823 mysql 20 0 26316 25M 2236 R 2.7 2.5 0:01 mysqld
4868 mysql 20 0 26316 25M 2236 R 2.7 2.5 0:01 mysqld
4798 mysql 14 0 26156 25M 2236 R 2.6 2.5 0:01 mysqld
4802 mysql 19 0 26316 25M 2236 R 2.5 2.5 0:01 mysqld
4866 mysql 20 0 26316 25M 2236 R 2.5 2.5 0:01 mysqld
4963 mysql 20 0 26316 25M 2236 R 2.5 2.5 0:01 mysqld
4762 mysql 16 0 28472 27M 2236 S 2.4 2.7 0:02 mysqld
4795 mysql 19 0 26580 25M 2236 R 2.4 2.5 0:01 mysqld
4821 mysql 18 0 26316 25M 2236 S 2.4 2.5 0:01 mysqld
4957 mysql 20 0 26316 25M 2236 R 2.4 2.5 0:01 mysqld
4997 mysql 20 0 26316 25M 2236 R 2.4 2.5 0:00 mysqld
4769 mysql 18 0 28472 27M 2236 S 2.3 2.7 0:02 mysqld
And here is my "my.cnf" file:
[mysqld]
datadir=/var/lib/mysql
socket=/var/lib/mysql/mysql.sock
set-variable = max_connections=300
[mysql.server]
user=mysql
basedir=/var/lib
[safe_mysqld]
err-log=/var/log/mysqld.log
pid-file=/var/run/mysqld/mysqld.pid
When I do an "/etc/rc.d/init.d/mysqld restart" the server restarts, though a bit
slowly, and I sometimes have to manually kill a few mysqld processes, but returns to
normal. I have tried all kinds of things to figure out what is going on here. I
raised the max_connections from 100 to 300, and edited the following httpd.conf
settings:
MinSpareServers 15
MaxSpareServers 30
StartServers 15
MaxClients 150
MaxRequestsPerChild 1000
Ryan Shrout
Owner - Amdmb.com
http://www.amdmb.com/
[EMAIL PROTECTED]
