Component Factory shutting down pre-maturely

I am trying to invoke mediator (7.4) sequences from the .NET web page
using the http access to the mediator servletportal. When I execute these
sequence multiple times (no specific order), the component factory shuts
down pre-maturely, and I see the following messages in the logs:

[slf5s.start]25 Mar 2005 17:36:01,270[slf5s.DATE]Finalizer[slf5s.THREAD]DEBUG[slf5s.PRIORITY]xbd[slf5s.CATEGORY]- Shutting down … Total is 0[slf5s.MESSAGE]
[slf5s.start]25 Mar 2005 17:36:01,270[slf5s.DATE]Finalizer[slf5s.THREAD]DEBUG[slf5s.PRIORITY]xbd[slf5s.CATEGORY]- Shutting down … Total task is 0[slf5s.MESSAGE]
[slf5s.start]25 Mar 2005 17:36:01,270[slf5s.DATE]Finalizer[slf5s.THREAD]DEBUG[slf5s.PRIORITY]xbd[slf5s.CATEGORY]- Thread Pool finalizing …[slf5s.MESSAGE]
[slf5s.start]25 Mar 2005 17:36:01,270[slf5s.DATE]Finalizer[slf5s.THREAD]DEBUG[slf5s.PRIORITY]xbd[slf5s.CATEGORY]- Shutting down … Total is 0[slf5s.MESSAGE]
[slf5s.start]25 Mar 2005 17:36:01,270[slf5s.DATE]Finalizer[slf5s.THREAD]DEBUG[slf5s.PRIORITY]xbd[slf5s.CATEGORY]- Shutting down … Total task is 0[slf5s.MESSAGE]

This happenes no matter which sequence I execute. Any Ideas??

Regards,
Sacheen Punde
Technical Services, Reston
DefaultComponentFactory.xbd.log (804 Bytes)

Hi Sacheen,

First, I would suggest that you turn on DEBUG level logging from the SMH configuration panel. That might provide some clues.

Also, what is the hardware/software environment you are running?

Mike

Mike,

Sorry, I should have been more specific… I actually have the
log level set to debug, and that is when I see these messages in the
logs, otherwise I do not see this.

I am running Mediator 7.4 on Windows XP Pro. machine. I have set the
JVM parameters to 128M initial and 256M max heap size. I also increased
the max. thread setting to 50. But when I monitor the runtime stats
during the sequence execution, the memory or the thread usage
does not get anywhere near the max values. Tomcat 5, on which
mediator is running, also has similar JVM parameters.

Another important observation is that, after I see these messages,
and I get timed out from the web app, the next time I am able to
execute the same sequence w/o any problems (without having to start
the component factory, so looks like the component factory is not
actually shut down). I experienced this multiple times.

Also, whenever I get timed out, mediator does not even show the first
log message in the sequence, which means it is not getting into the
sequence execution, so chances of something inside teh sequence
causing this are very low…

Regards,
Sacheen.