What I observe is:
a) A certain amount of free memory is required to instantiate a web application
b) Concurrent requests are handled by separate HTTP Processor threads
c) The maxSpareThreads attribute of the HTTP Connector is not honored until GC
And when I:
1) Deploy three "helloworld" Turbine applications
2) Use JMeter to hit the first webapp with 100 concurrent threads
3) Subsequently hit the second and third webapps with the same test plan
I will get OutOfMemory Exceptions because there is no free memory to instantiate
the 3rd webapp, as the idle HTTP Connectors have consumed all available memory.
So I ask, "Am I understanding this correctly? Is this known and specified
behavior?"
---
using apache-tomcat-5.5.20 out-of-the-box (tar.gz)
with jdk1.5.0_09 and starting it up with:
catalina.sh jpda start -Dcom.sun.management.jmxremote -Xmx30m -Xms30m
sample apps created as:
maven -Dturbine.app.name=helloworld turbine:setup
---------------------------------------------------------------------
To start a new topic, e-mail: users@tomcat.apache.org
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
