[jacorb-developer] Client connection threads are never released

shectman1 shectman1 at llnl.gov
Tue Feb 12 00:48:18 CET 2013


On 02/11/2013 03:38 PM, Stout, Eric A. wrote:
> We have a mixed-platform, mixed-language, mixed-ORB system of about 1500 processes.  A few Java JacORB processes are clients to all, or virtually all, of the processes in the system.  I have set jacorb.connection.client.max_receptor_threads=2500, so the system starts up fine, and runs for a few days.  But processes are occasionally stopped and restarted, causing the all-process clients to make new connections and consume more threads.  Eventually, I end up with an error like:
>
>
I don't know if it will help, but...
we run jacorb clients against TAO corba servers.
We also have a problem with the client side having problems when the server goes away.

I have a jacorb client side interceptor, that when a connection fails, interprets the ior, and

1) Find the poa for this object (TAO specific kludge)
2) tell the (TAO) imr to activate the poa (start the server)
3) get the port number of the poa from the imr
4) replace the port number in the ior
5) throw a ForwardRequest with the new ior


Its not the nicest code in the world, but it works.
It should cause you to theoretically not lose threads, since
reconnections are automatic.

I have no idea how your constantly changing implname would complicate things, but I can't
imagine it gonna make your life easy.


-- 

Robert M. Shectman

National Atmospheric Release Advisory Center (NARAC)
Lawrence Livermore National Laboratory

shectman at llnl.gov
(925)423-6714



More information about the jacorb-developer mailing list