[Paraview] Crash with Parallel Pvserver, Off-screen rendering

Gregory, David C CIV NSWCDL G24 david.c.gregory at navy.mil
Tue Oct 17 15:38:45 EDT 2006


Unfortunately, I can't post the errors or the CMakeCache.txt.  I'm
hoping to have another machine to run on in a week or so.
 
However, in this particular case, there are no "X" errors.  There is a
vtk-paraview error regarding a failed server connection, and then the
application abruptly quits.
 
However, I'm a little confused.  If I'm using a Mesa enabled server,
with off screen enabled at compile time and on the command line, should
I need an X server?
 
-David

	-----Original Message-----
	From: Berk Geveci [mailto:berk.geveci at gmail.com] 
	Sent: Monday, October 16, 2006 11:05 AM
	To: Gregory, David C CIV NSWCDL G24
	Cc: ParaView
	Subject: Re: [Paraview] Crash with Parallel Pvserver, Off-screen
rendering
	
	
	It sounds like a configuration problem. Can you post the X
errors? I am guessing the server does not have access to an X display or
one that supports GLX. Also, can you post your CMakeCache.txt?
	
	-berk
	
	
	On 10/13/06, Gregory, David C CIV NSWCDL G24
<david.c.gregory at navy.mil> wrote: 

		Hi-
		
		I'm running a CVS snapshot from 20 Sept of Paraview 2.5,
compiled with
		mpich-gm, Mesa-6.5.1 and off-screen rendering enabled in
cmake.  I also
		have a version compiled with mesa and off-screen but
without mpi. 
		
		I start paraview as follows:
		
		/paraview-cvs-mesa-nompi/bin/pvclient -rc
		
		/paraview-cvs-mesa-mpi/bin/pvserver -rc -ch=headnode
		--use-offscreen-rendering
		
		If I un-check the "compositing" box in the GUI, which as
I understand 
		things, disables compositing and does all rendering on
the head node,
		paraview runs fine.
		
		However, if I turn compositing on, the SERVERS crash
immediately if any
		data is loaded for display, and I get a message from the
client which 
		informs me that connection to the server has been lost.
There are no
		X-related errors in stdout or stderr - only a message
which gives an
		error in the vtk server connector.
		
		Is this likely a bug or a configuration problem? 
		
		-David
		
		David C. Gregory
		NSWC Dahlgren
		Weapons Lethality and Effectiveness Branch - G24
		17320 Dahlgren Road
		Dahlgren, VA  22448
		540-653-0424
		_______________________________________________
		ParaView mailing list
		ParaView at paraview.org
		http://www.paraview.org/mailman/listinfo/paraview
		


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://public.kitware.com/pipermail/paraview/attachments/20061017/afdc6420/attachment.htm


More information about the ParaView mailing list