[omniORB] Naming Server

Hai Wang wangh@Teltec.DCU.IE
Tue, 29 Jun 1999 22:29:26 +0100


Hi Rebecca
	Thanks for your help, Yes, you are right, my server died already
while the client is running.

Cheers!
Hai
To:             	wangh@Teltec.DCU.IE
Date sent:      	Tue, 29 Jun 1999 11:21:27 -0500
Subject:        	Re: [omniORB] Naming Server

> 
> 
> Not sure if you're on the omniorb-list.  I had sent this suggestion
> out the other day...
> --- becs
> 
> 
> ---------------------- Forwarded by Rebecca A Sanford/GDIS/GDYN on 06/29/99
> 11:21 ---------------------------
> 
> From: Rebecca A Sanford on 06/28/99 09:57
> 
> To:   omniorb-list@uk.research.att.com
> cc:
> Subject:  Re: [omniORB] Naming Server  (Document link not converted)
> 
> Is it possible that your server application has already finished
> executing since you used the non-blocking call on impl_is_ready?
> This would account for the COMM_FAILURE.
> 
> Using boa->impl_is_ready(0,1) tells the ORB that the BOA call is
> non-blocking.  The main thread continues after impl_is_ready.
> (Refer to discussion about "line 5" in paragraph 2.7.2, Object
> Initialization in the omniORB User's Guide.)
> 
> You also did not note if the client/server/Name Service were
> colocated.  The ORBInitialHost and ORBInitialPort entries in the
> omniORB.cfg will need to be set if the Name Service is on a different
> machine.
> 
> You might also want to run with -ORBtracelevel 20 in order to gain
> more insight as what's going on.
> 
> Hey, it's a stab in the dark...
> --- becs
> 
> 
> 
> 
> 
> From: Hai Wang <wangh@Teltec.DCU.IE> on 06/28/99 10:24 GMT
> 
> Please respond to Hai Wang <wangh@Teltec.DCU.IE>
> 
> To:   djr@uk.research.att.com
> cc:   omniorb-list@uk.research.att.com (bcc: Rebecca A Sanford/GDIS/GDYN)
> Subject:  Re: [omniORB] Naming Server
> 
> 
> 
> 
> Hi David,
>      Thanks for your reply, I have compared "IOR"s by using both catior and
> stringified version. By using stringified version, the result is different as I
> guess. In catior case, the IP addresses for IORs are the same but the port
> number is different. Thats why the call to the function of object reference
> threw a COMM_Failure out. I really don't know where the problem could be.
> 
> Once again, thanks for your kind help!
> 
> Hai
> 
> > Date: Mon, 28 Jun 1999 09:16:02 +0100 (GMT)
> > From: David Riddoch <djr@uk.research.att.com>
> > X-Sender: djr@pitanga
> > To: Hai Wang <wangh@teltec.dcu.ie>
> > cc: omniorb-list@uk.research.att.com
> > Subject: Re: [omniORB] Naming Server
> > MIME-Version: 1.0
> >
> > On Sun, 27 Jun 1999, Hai Wang wrote:
> >
> > >
> > > Hi  all,
> > >  I have got a serious problem working with naming server. In
> > > server side, I tried to produce an object of my application and
> > > register it with a name in naming server, for the purpose of
> > > comparison I print out the "ior" of the object.
> > >  In client, I tried to get the object reference and print out its "ior",
> > > an exception of COMM_Failure is thrown when the client tried to run
> > > the function in the object reference. I compared the two "ior"s, they
> > > are different.
> >
> > How did you compare the IORs?  Using catior, or by looking at the
> > stringified versions?  The stringified versions may differ, but the output
> > of catior is what matters.
> >
> > David
> >
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
>