[omniORB] reliability

Julien Teisseire zzteisseire at echo.fr
Wed Apr 20 10:12:36 BST 2005


Thanks for your answer,

today, I am catching everyting (ORB_init, resolve_initial_references,
resolve, ...).
When I start my program (client side), I do init, connection to
NameService, and I get the reference of my obj.

During the execution of my program, I have to use object function.

when I do it while server is down, I get :terminate called after
throwing an instance of 'CORBA::TRANSIENT'

And I did not enter in my try/catch exception management (in order to
retry or whatever else).

Also, the ::_narrow(obj); proc. seams to leave "internally" in case of
problem, without providing a way to catch problem ?

I hope to be more clear.
Regards and thanks,
Julien

Le mercredi 20 avril 2005 à 08:01 +0200, Wernke zur Borg a écrit :
> > 
> > Could someone tell me why client program exits when my server stops ?
> >
> 
> Most probably there is a bug in your program. Are you catching exceptions?
> 
> >
> > I need to know how to retry connect to server after loosing it.
> >
> 
> I guess you have to restart your program. If you manage to keep it up just
> re-invoke your method. Connection handling is ORB internal.
>  
> > It seams there isn't any exception management when server is
> > disconnected from the network (or crashed).
> 
> Yes, there is. Are you catching exceptions?
> 
> > How to make client application reliable in all that cases ?
> 
> Put a try-catch block around all method invocations on remote objects.
> 
> Robustness and reliability does not come for free. You would have to become
> more specific if this does not help.
> 
> Regards, Wernke
> 
> 
> _______________________________________________
> omniORB-list mailing list
> omniORB-list at omniorb-support.com
> http://www.omniorb-support.com/mailman/listinfo/omniorb-list




More information about the omniORB-list mailing list