Bioperl: Re: Bio::CorbaServer Updated

[Jason Stajich]

I'll see if I can test the server with java clients later in the week.
I am halfway done with my Bio::CorbaClient additions.  Maybe will get them
done by the end of the week.

[more comments below]

On Tue, 9 May 2000, Brad Chapman wrote:
> 
> Hi Jason;
>     Sorry to be so slow in getting back with you, I just finished up 
> my finals and all that fun stuff...
>  
> > Brad, give it a try with some of your python clients when you have a
> > chance.  I'll be trying it out on the java clients as well later this
> > week.
> 
> I fixed up the python test client for the perl servers and ran it 
> against the four perl servers with the following results:
> 
> For the Fnorb python ORB implementation, everything runs smoothly for 
> all the servers! Yay :-)
> 
> For the omniORB python ORB implementation, everything does not go so 
> smoothly :-< I get a perl seg fault and core dump on all the servers 
> which return objects. Looking at it in gdb, and checking out the 
> differences between the omniORB generated stub code and the Fnorb stub 
> code, it looks like the problem is that omniORB calls 
> corba_object.is_a() whenever it recieves an object (to verify it) and 
> apparently CORBA::ORBit doesn't like this since it is segfaulting 
> somewhere during this call. I can also get a seg-fault if I call 
> ._is_a() explictly from Fnorb, so it doesn't appear to be a omniORB 
> specific problem, but rather an is_a() calling problem. I'm not 
> positive if this is a known problem or not so I sent a quick e-mail to 
> the author to see about this...

I'll look into the is_a problem - thanks for digging some to see where the
crash might be coming from.  
 
> One other minor point--could we modify seqdbsrv.pl so it spits out an 
> ior file with a unique name (like seqdbsrv.ior or something). I just 
> wrote one client which distinguishes between the servers based on the 
> ior file names (I could distinguish based on calling is_a(), if that 
> worked...) so it would be helpful if seqdbsrv.pl and simpleseq.pl 
> generated ior files with different names.

Certainly, I'll take a look this afternoon and see what I can do.

> If you want to play with the python clients I'll be updating cvs later 
> today. I'm going to be rearranging the directory structure and giving 
> it a nice install and other jazz like that.

Just send me a note when you've updated.
I'm glad to hear that we are MOSTLY interoperable - I'll see if I can
track down those other problems.  

>     Thanks again for working on this Jason and Ewan.
> 
> Brad
> 

Jason Stajich
Duke University Medical Center
Center for Human Genetics
jason@chg.mc.duke.edu
(919)684-1806 (office)
(919)684-2275 (fax)
http://wwwchg.mc.duke.edu/

=========== Bioperl Project Mailing List Message Footer =======
Project URL: http://bio.perl.org/
For info about how to (un)subscribe, where messages are archived, etc:
http://www.techfak.uni-bielefeld.de/bcd/Perl/Bio/vsns-bcd-perl.html
====================================================================