I am hoping someone has experienced this before.
The installation is a binary install of Tomcat and a binary install of mod_jk
1.2.6 connector.
We have been running performance tests on this install and Tomcat is very, very
cpu intensive topping out at 55% of the cpu's on the box. I have never
experienced this before and I am having a hard time tracking down the problem.
The application does lookups and updates to an Oracle 9i database and it does
use its own homegrown connection pool. But the poor performance seems to be on
the web side with server threads climbing up into the four hundreds at times.
I am including portions of my config files hoping that someone will spot
something that I am doing wrong.
Httpd.conf:
Timeout 300
KeepAlive On
MaxKeepAliveRequests 500
KeepAliveTimeout 15
StartServers 2
MaxClients 400
MinSpareThreads 150
MaxSpareThreads 300
ThreadsPerChild 25
MaxRequestsPerChild 10000
AcceptMutex fcntl
Tomcat Server.xml:
<!-- Define an AJP 1.3 Connector on port 8009 -->
<Connector port="8011"
enableLookups="false" redirectPort="8443" protocol="AJP/1.3"
maxThreads="500" minSpareThreads="75" maxSpareThreads="250"/>
<Host name="localhost" appBase="webapps"
unpackWARs="true" autoDeploy="true"
xmlValidation="false" xmlNamespaceAware="false">
Tomcat workers.properties:
worker.ajp13w.type=ajp13
worker.ajp13w.host=somehost
worker.ajp13w.port=8011
worker.ajp13w.cachesize=600
worker.ajp13w.cache_timeout=600
Can someone tell the correlation of the worker.cachesize and an htptd.conf
directive?
Thanks in advance.
-SB
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
