[omniORB] "Unexpected" exception at initialization

[Sai-Lai Lo]

omniConnectionBroken is indeed an internal exception that shouldn't be leak
out to the application space. I'm pretty sure this is the case for all but
the initialisation code you stumble on. I'll look into plugging that hole.

By the way, isn't it a bad practice to frequently use -ORBpoa_iiop_port
when the port no. has in fact been actively used by another process? Renzo
reported earlier on this list that on NT, reuse a port that is still
actively used by another process can result in a strange failure mode
instead of the friendly "address already in use" error.

Sai-Lai


>>>>> Chris Newbold writes:

> We're using omniORB 3.0, patched up through bug #11. A frequent user
> error for us is to specifity a port with -ORBpoa_iiop_port which
> happens to already be in use. When this happens, we get an "unexpected"
> exception when attempting to activate the root POA. I found this
> odd at first, since initialization is trapped it a try...catch which
> catches CORBA::Exception and omniORB::fatalException.

> After poking around a little, I found that an exception of type
> omniConnectionBroken is thrown when the incoming rope cannot open
> the specified socket. However, omniConnectionBroken is neither a
> CORBA execption nor an omniORB::fatalException.

> At the very least, this is rather annoying, since omniConnectionBroken
> is not documented. However, it also seems to violate the specified list
> of exceptions for POAManager::activate.





-- 
Sai-Lai Lo                                   S.Lo@uk.research.att.com
AT&T Laboratories Cambridge           WWW:   http://www.uk.research.att.com 
24a Trumpington Street                Tel:   +44 1223 343000
Cambridge CB2 1QA                     Fax:   +44 1223 313542
ENGLAND