Re: NIST SIP clean up?

Kalman,

Please try the example in

src/tests/load/concurrency

There is something quite wrong here because 16 calls/min is awfully 
slow. Its never been *that* bad.
I just tried on my laptop and got 77 CAPS. Check to see if you are 
getting slowed down by DNS lookup.



Regards

Ranga

Kalman Lau wrote:

> Is there a clean-up procedure needed when using NIST?
>
> From the caller-end, I'm just issuing INVITE messages and playing 
> along when the call is answered.  Calls are answered by a single 
> thread (different process) that also drops the call immediately after 
> answering (INVITE / TRYING / RINGING / OK / ACK / BYE / OK).  This 
> simple test-case is (serially) handling just 16 calls per min.
>
> But within 20 minutes, I'm seeing an increased memory consumption 
> around 30 MB according to perfmon's "private bytes" counter for the 
> (Java) process that's answering calls.  From JConsole, on the other 
> hand, there are no matching heap or non-heap memory increases (just 
> the usual GC).
>
> I've tried setting CACHE_SERVER_CONNECTIONS and 
> CACHE_CLIENT_CONNECTIONS to both "true" and "false"; set and unset 
> THREAD_POOL_SIZE; and I've tried against the current NIST SIP as well 
> as the older impl for JAIN SIP 1.1.  Same result so far...
>
> Is there a trivial clean-up that I'm completely missing, perhaps at 
> the transaction- or dialog- or even at the message-level?
>
> Comments very much appreciated.  Thanks.
>
> Kalman
>
>------------------------------------------------------------------------
>
>_______________________________________________
>nist-sip mailing list
>[email protected]
>http://www-x.antd.nist.gov/mailman/listinfo/nist-sip
>  
>


-- 
M. Ranganathan 

Advanced Networking Technologies Division,
National Institute of Standards and Technology (NIST),
100 Bureau Drive, Stop 8920, Gaithersburg, MD 20899. 
tel:301 975 3664 , fax:301 590 0932 http://w3.antd.nist.gov/
Advanced Networking Technologies For the People!