# Half of ProcessIdScalars values are false for a vti file

**URL:** https://discourse.paraview.org/t/half-of-processidscalars-values-are-false-for-a-vti-file/9367
**Category:** Development
**Tags:** mpi
**Created:** [April 6, 2022, 8:49am UTC](https://discourse.paraview.org/t/half-of-processidscalars-values-are-false-for-a-vti-file/9367 "2022-04-06T08:49:43Z")
**Posts on this page:** 1
**Page:** 1

<div class="post-metadata">

### Author: ![eve-le-guillou](https://discourse.paraview.org/user_avatar/discourse.paraview.org/eve-le-guillou/32/11047_2.png) [@eve-le-guillou](https://discourse.paraview.org/u/eve-le-guillou)
#### Post date: [April 6, 2022, 8:49am UTC](https://discourse.paraview.org/t/half-of-processidscalars-values-are-false-for-a-vti-file/9367/1 "2022-04-06T08:49:43Z")

</div>

Hi all,

I have issues with using the ProcessIdScalars filter with the GhostCellsGenerator for a Uniform Rectilinear Grid.

I first open this [data](https://discourse.paraview.org/uploads/short-url/AjM9WuYnPCMxWPuTnyGAjJv4Izs.vti) (3.4 KB) that I produced in built in mode using the Wavelet and the GenerateGlobalIds filters.

Here is my pipeline, that I execute using MPI with pvserver:

```
volume_global_ids > ProcessIdScalars > GhostCellsGenerator

```

All options are at their default values.

Now, if I open the Spreadsheet View and look at the point of Global Id 35 (line 112 to 115) in the screenshot below), I can see that it is shown on 4 processes, and for 3 processes it appears as a ghost point (which is what it is supposed to do).

 ![Screenshot from 2022-04-06 10-41-37](https://discourse.paraview.org/uploads/default/original/2X/7/7675a6e8a0d17b925e99a5b4569c636c30736622.png)

However, the ProcessId values are not correct: they should all be equal to 2, as it is the only process on which the point is not a ghost, but on processes 0 and 4, the value of ProcessId is 0.

Could this be related to the fact that vti files are automatically distributed and duplicated some points on the border? In my example, the point 35 is indeed duplicated on both process 0 and process 2, before calling any other filter.

Many thanks