AW: [omniORB] omniNames + bidir = Detected GIOP 1.2 protocol error ininput message

[Matej Kenda]

Hi,

I have run into this problem for a few times as well. It happens very
rarely. We use C++, not Python.

I'm curious why is it necessary to specify the GIOP protocol version
when communicating with omniNames? What triggers this error to be
reported?

Thanks,

Matej

On Wed, 2004-04-28 at 19:38, Scott Robertson wrote:
> That did the trick. Thanks allot!
> 
> On Wed, 2004-04-28 at 00:52, Barthel Marco (MPA/DS) wrote:
> > Hi,
> > 
> > I don't know if this is a bug or if it was intended.
> > 
> > We had the same problem. A solution to this problem is the following:
> > 
> > Either copy the ROOT-Context-IOR from the omninames console window and use it
> > 
> > or
> > 
> > provide the right GIOP-Version (1.2) when using
> > 
> > -ORBInitRef NameService=corbaname::1.2@<your-omninames-hostname
> > 
> > on the commandline.
> > 
> > Hope that helps.
> > 
> > -marco
> > 
> > > -----Ursprüngliche Nachricht-----
> > > Von: omniorb-list-bounces at omniorb-support.com
> > > [mailto:omniorb-list-bounces at omniorb-support.com]Im Auftrag von Scott
> > > Robertson
> > > Gesendet: Dienstag, 27. April 2004 23:44
> > > An: omniORB List
> > > Betreff: [omniORB] omniNames + bidir = Detected GIOP 1.2 
> > > protocol error
> > > ininput message
> > > 
> > > 
> > > I'm working with the latest copy of omniORB from CVS. The one that has
> > > the fix so that python can use BiDirectional IOP. And it now does
> > > bidirectional calls  with a simple example. However, if I add the
> > > settings to my real program, my client gets the following error from
> > > omniNames
> > > 
> > > omniORB: From endpoint: giop:tcp:192.168.0.80:49000. Detected GIOP 1.2
> > > protocol error in input message
> > > 
> > > If I disable bidir support by changing the client's transport rule to 
> > > clientTransportRule = * unix,tcp
> > > 
> > > I don't have a problem, but if I turn it on I do.
> > > 
> > > I tried setting the serverTransportRule in /etc/omniORB.cfg to see if
> > > that had any effect, but I get the same result.
> > > 
> > > Anyone else run into this? Or is it due to python again?
> > > 
-- 
Matej Kenda, Lead Engineer
HERMES SoftLab (www.hermes-softlab.com)
Erjavčeva 2, 5000 Nova Gorica, Slovenia