[Paraview] Paraview 3.0.2 crashes when trying to connect to a server. A bug?

[Marc Baaden]

(Recall: the problem was for paraview 3 client that crashes when 
connecting to any other server than builtin)

Ok, I have some more info.

I tried the same (copied the compiled binaries) on another machine. 
Here it works fine. I can also connect from this machine to the machine 
that I tried previously and that crashes when the client is run locally
(now only running the pvserver).

So the problem seems to be with the paraview.app client running on the 
machine that I tested first. The one thing that is different on this 
machine is that it is behind a firewall and has a proxy set up. Could 
that have something to do with it?

I can try and get some more info on the crash (stack trace or the like) 
next week, but for now I can only access the machine remotely and thus 
not start the client. [how do you do stack traces on the mac with an 
XXX.app executable?] The only 'debug' output I had so far was a segmentation 
fault, when trying to run either of
1) pvTestDriver --run-test-init=/opt/paraview-cvs/Applications/Client/Testing/Cxx/ManualConnectCS.xml
2) pvTestDriver --run-test-init=/opt/paraview-cvs/Applications/Client/Testing/Cxx/ManualConnectCRSDS.xml

Some other 'issues' or observations:
- when I start pvserver with eg -np 4, I only have 3 pvserver tasks 
running at about 99%, and occasionnally a 4th at ~20%. Is that to be 
expected?
- there seem to be some artefacts for eg Volume Rendering or Points 
representation (presumably at the boundaries of the rendered sub-images).
Can these artefacts be avoided or corrected for?
- is it possible to 'reload' sessions that were saved previously with 
the builtin server using another server setting? If I try and do so for 
eg a saved state file, it closes the server connection and starts up a 
builtin server..

Thanks very much in advance,
Marc

>>> "Moreland, Kenneth" said:
 >> Yes, it really should be as simple as that.  Can you give more
 >> information about the crash?  Was there any output before it happened?
 >> Can you give us a stack trace?
 >> -Ken
-- 
 Dr. Marc Baaden  - Institut de Biologie Physico-Chimique, Paris
 mailto:baaden at smplinux.de      -      http://www.baaden.ibpc.fr
 FAX: +33 15841 5026  -  Tel: +33 15841 5176  ou  +33 609 843217