Re: ORBIT, GNOME, and security

[Michael Rumpf <michael rumpfonline de>]

Hi,

Ryan Lovett wrote:
> 
> On Tue, Sep 19, 2000 at 08:28:46PM +0200, Michael Rumpf wrote:
> > If you know the name of the server on which the NameSerice is running it
> > should be enough to start your client with the following command line:
> >
> > myclient -ORBInitRef NameService=corbaloc://nshost/NameService
> >
> > The client application can use just the call to
> > resolve_initial_references("NameService"); to get a valid object reference to
> > the Naming Service and from then everything works as before....
> 
> Hello,
>  Is this supported by the stable branch? I tried this with a small client (a
> derived example of Henning-Vinoski 3-1) and
> 
>   name_srv = CORBA_ORB_resolve_initial_references(orb, "NameService", &ev);
> 
> returns (struct CORBA_Object_struct *) 0x0 which causes a later invocation of
> CosNaming_NamingContext_bind to segfault (no system exception raised by the
> above resolve). My server/client do work fine with -ORBNamingIOR=...
>  I found <http://cvs.gnome.org/lxr/source/ORBit/test/run_via_ns> which tests
> invocations like the one you suggested. (run_via_ns is not included with 0.5.3)
>  In any case, I'm interested in your suggestion since I'd like to remove IOR
> string exchanges from the picture.

no the partial INS implementation didn't find its way into the stable branch (yet).
You need to check out the CVS Head version of ORBit !

See the end of the first question in the FAQ for a link to the ORBit CVS tree: http://orbit-resource.sourceforge.net/faq.html#corba

It would be great if you could inform me whether your tests are successful and which features of Kennards implementation are actually working.
A small sample showing how to use it would be even better.... ;-) 

> Ryan
> 
> PS- You had suggested Henning-Vinoski to me in answer to a question of mine
> which suggested the Event service. The book has proven to be extremely helpful,
> thanks!

Great ;-)

Michael