Reduce number of Apache processes
It depends which mpm module (mpm_worker/mpm_prefork/mpm_event) your apache is using. If you are unsure, then post the output of cat /etc/apache2/mods-enabled/mpm*.conf
which reveals the number of StartServers/MinSpareServers/MaxSpareServers, MaxRequestWorkers and maybe ThreadsPerChild and ThreadLimit. If the filenames should be different in your linux distro, then post an output of your enabled modules ls -l /etc/apache2/mods-enabled
.
It should look like something like this and explains itself (Debian, Apache2.4):
root@debian:/# cat /etc/apache2/mods-enabled/mpm*.conf
# prefork MPM
# StartServers: number of server processes to start
# MinSpareServers: minimum number of server processes which are kept spare
# MaxSpareServers: maximum number of server processes which are kept spare
# MaxRequestWorkers: maximum number of server processes allowed to start
# MaxConnectionsPerChild: maximum number of requests a server process serves
<IfModule mpm_prefork_module>
StartServers 5
MinSpareServers 5
MaxSpareServers 10
# same as MaxClients in Apache 2.2
MaxRequestWorkers 150
MaxConnectionsPerChild 0
</IfModule>
# vim: syntax=apache ts=4 sw=4 sts=4 sr noet
See Reducing Apache Memory usage and Average Process Size Value
The documentation says (as described in the link above):
You can, and should, control the MaxClients setting so that your server does not spawn so many children that it starts swapping. The procedure for doing this is simple: determine the size of your average Apache process, by looking at your process list via a tool such as top, and divide this into your total available memory, leaving some room for other processes. https://httpd.apache.org/docs/2.2/misc/perf-tuning.html
Example:
Tasks: 207 total, 1 running, 206 sleeping, 0 stopped, 0 zombie
%Cpu(s): 2.4 us, 0.8 sy, 0.0 ni, 96.7 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st
KiB Mem : 16307588 total, 14360744 free, 1188636 used, 758208 buff/cache
KiB Swap: 0 total, 0 free, 0 used. 14686936 avail Mem
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
6567 www-data 20 0 296028 15288 5616 S 0.0 0.1 0:00.07 apache2
6569 www-data 20 0 296040 15360 5676 S 0.0 0.1 0:00.08 apache2
6571 www-data 20 0 295996 15200 5676 S 0.0 0.1 0:00.07 apache2
6572 www-data 20 0 296028 15348 5676 S 0.0 0.1 0:00.08 apache2
6573 www-data 20 0 296040 15356 5676 S 0.0 0.1 0:00.07 apache2
Running the cool script from the linked page above gives me:
root@debian:~# ps -ylC apache2 | awk '{x += $8;y += 1} END {print "Apache Memory Usage (MB): "x/1024; print "Average Process Size (MB): "x/((y-1)*1024)}'
Apache Memory Usage (MB): 100.711
Average Process Size (MB): 16.7852
Note: "Average Process Size" is the "RES" value when you run top
.
To determine MaxClients
(aka MaxRequestWorkers
), I need to calculate:
Maxclients=X/Y where
X=Max. Available Memory Reserved for Apache
Y=Average Process Size