[omniORB] Double function calls

Masaaki Sekiya sekiya@nagoya.ydc.co.jp
Thu, 14 Mar 2002 10:46:26 JST


Hi David.
I've posted a same issue as yours.
http://www.uk.research.att.com/omniORB/archives/2000-12/0146.html
It referres a reason of Double function calls. ( but not solve )

Masaki Sekiya

> Hi,
> 
> I'm trying to build resilience into my OmniORB system and one of the things
> that I have done is to reduce the scan granularity to 1 and the client side
> call timeout period to a small value, experimenting with values of 10 or
> less.  The idea is that I should be able to detect a computer failure
> quickly rather than having pings taking 30 or more seconds to return.
> 
> One of the side effects of this is that when I make a call to a CORBA server
> on the local machine that may take several seconds to return, the client is
> getting an exception whilst the server is actually working - this I accept I
> have to work around by getting my timings right (unless anyone else has a
> more cunning way of doing things!).  What I am seeing though is that the
> server gets the distributed function called for a second time immediately
> after it has returned from the first attempt.  I can't see anything in the
> documentation to explain this or to prevent it.
> 
> Any ideas anyone???
> 
> Thanks
> 
> David