Reading Multiblock VTK file which inclueds named blocks, gets broken dataset.

ParaView Support
1

Hello.
I have lots multiblock dataset files and tried visualize those dataset with ParaView 5.10.1.
And I’ve met a problem.

Dataset readed from Multiblock VTK file which includes named blocks, will be broken on parallel pvserver of ParaView 5.10.x .

  1. Run pvserver in parallel.
    $ mpirun -n 2 pvserver
  2. Connect paraview to pvserver.
  3. Read test.vtm. (unzip attached testing file test.zip (17.2 KB))

test.vtm has 5 blocks.
In 2 pvserver processes case, first 3 blocks was succeeded to read and last 2 blocks was failed.
It seems reading is succeeded only blocks mapped on head process (rank 0).

Have I mistake to build paraview?
However, prebuilded binary distributed from paraview.org has this problem, too.
And ParaView 5.9.1 or older can read normally.

regards,

2

Hi @earth2001y,

This was a known issue https://gitlab.kitware.com/paraview/paraview/-/issues/21419 which is currently being fixed in VTK/ParaView.

Best regards,

Julien

3

Hi @jfausty ,

Thank you for your information about issue.
I expect to the issue will be fixed early.

thanks,

4

Hi @earth2001y,

The issue should now be fixed in the ParaView master branch.

Kind regards,

Julien

1 Like