[Paraview] Paralel mpi run of Text rendering

[Eelco van Vliet]

Dear paraview fellows,

I am trying to carry out a parallel rendering using paraview on our cluster.
I got it almost running except for one issue: rendering of Text fails.
Hopefully someone can advise me.

I followed the approach of the tuturial with the Sphere example (
http://www.cscs.ch/fileadmin/user_upload/customers/users_entry_point/HORUS_Doc/PvPython.pdf
)

So I added the renderview and ProcessIdScalars to all the Show() commands,
e.g.:

*RenderView1 = sm.CreateRenderView()
SurfaceVectors1 = SurfaceVectors(SelectInputVectors = ['POINTS', 'U'])
DataRepresentation5  = Show(ProcessIdScalars(SurfaceVectors1),RenderView1)
*
This works fine (also in parallel on 8 cores), except for the show of plain
Text with Text

There I had to remove the ProcessIdScalars function, so I do:


t=Text(Text=timestr)
Show(t,RenderView1,Color=[0.0,0.0,0.0],Position=[0.01,0.95],FontSize=15)

Now it works, even if I run it on 1 core on the cluster.
However, if I run it on 8 cores, I get the error message


*Process id: 0 >> ERROR: In
/scratch/ParaView-3.8.1/VTK/Filtering/vtkDemandDrivenPipeline.cxx, line 827
vtkCompositeDataPipeline (0x5fc3b80): Input for connection index 0 on input
port index 0 for algorithm vtkTransmitPolyDataPiece(0x5fc1ab0) is of type
vtkTable, but a vtkPolyData is required.*

Apperantly it is not allowed, it need information on the process Id I
suppose. However, ProcessIdScalars does not work. Is there anybody who would
know a solution?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.paraview.org/pipermail/paraview/attachments/20110121/708adc45/attachment.htm>