# "Generate Process Ids" doesn't work

**URL:** https://discourse.paraview.org/t/generate-process-ids-doesnt-work/15639
**Category:** ParaView Support
**Created:** [October 25, 2024, 6:22pm UTC](https://discourse.paraview.org/t/generate-process-ids-doesnt-work/15639 "2024-10-25T18:22:03Z")
**Posts on this page:** 5
**Page:** 1

<div class="post-metadata">

### Author: ![ANON1](https://discourse.paraview.org/user_avatar/discourse.paraview.org/anon1/32/3919_2.png) [@ANON1](https://discourse.paraview.org/u/ANON1)
#### Post date: [October 25, 2024, 6:22pm UTC](https://discourse.paraview.org/t/generate-process-ids-doesnt-work/15639/1 "2024-10-25T18:22:03Z")

</div>

Hello,

I am using Paraview 5.12.0 in the client/server mode. (Client on Windows 11), and pvserver (running in parallel with 24 CPUs) on a Linux cluster.

To confirm that pvserver is running in parallel I can color my dataset using the filter `Process Id Scalars` and it works correctly, and it shows that I am using 24 processes, as you can see below:

 ![image](https://discourse.paraview.org/uploads/default/original/2X/e/e53f22a63aefcd74f839d388dd9e0afb6534da88.png)

But, I get this warning:

```auto
Warning: In vtkSMProxy.cxx, line 887
vtkSMSourceProxy (000001E1B1654BC0): Proxy (filters, ProcessIdScalars) has been deprecated in ParaView 5.12 and will be removed by ParaView 5.14. 
        This has been replaced by 'GenerateProcessIds'. Please consider
        using that instead.
      

```

However, when I try to use the suggested filter `Generate Process Ids`, The datasets doesn’t get colored as shown below:

 ![image](https://discourse.paraview.org/uploads/default/original/2X/8/8452504a8132160107339935c515b4c264420ce9.png)

Am I missing something with this filter or is it a bug?

Thank you

---

<div class="post-metadata">

### Author: ![cory.quammen](https://discourse.paraview.org/user_avatar/discourse.paraview.org/cory.quammen/32/11193_2.png) [@cory.quammen](https://discourse.paraview.org/u/cory.quammen)
#### Post date: [October 27, 2024, 7:30pm UTC](https://discourse.paraview.org/t/generate-process-ids-doesnt-work/15639/2 "2024-10-27T19:30:22Z")

</div>

I’m reproducing this in ParaView Nightly build from October 27, 2024.

Strange, the generated data in arrays “PointProcessIds” and “CellProcessIds” appears to be correct, but coloring by these arrays is not working. Coloring by the arrays named “vtkProcessId” in the representation for (both point and data array) _does_ work.

---

<div class="post-metadata">

### Author: ![cory.quammen](https://discourse.paraview.org/user_avatar/discourse.paraview.org/cory.quammen/32/11193_2.png) [@cory.quammen](https://discourse.paraview.org/u/cory.quammen)
#### Post date: [October 27, 2024, 7:34pm UTC](https://discourse.paraview.org/t/generate-process-ids-doesnt-work/15639/3 "2024-10-27T19:34:02Z")

</div>

Interestingly, if I apply a **Calculator** filter after the **Generate Process Ids** filter, coloring by the output array works. Note that the output of **Generate Process Ids** is `long long`, while outputting any other type from the **Calculator** makes it work okay. I wonder if there is a problem coloring by such data arrays.

---

<div class="post-metadata">

### Author: ![cory.quammen](https://discourse.paraview.org/user_avatar/discourse.paraview.org/cory.quammen/32/11193_2.png) [@cory.quammen](https://discourse.paraview.org/u/cory.quammen)
#### Post date: [October 27, 2024, 7:56pm UTC](https://discourse.paraview.org/t/generate-process-ids-doesnt-work/15639/4 "2024-10-27T19:56:30Z")

</div>

Reported a bug here: [https://gitlab.kitware.com/paraview/paraview/-/issues/22785](https://gitlab.kitware.com/paraview/paraview/-/issues/22785)

---

<div class="post-metadata">

### Author: ![mwestphal](https://discourse.paraview.org/user_avatar/discourse.paraview.org/mwestphal/32/17_2.png) [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [October 28, 2024, 7:57am UTC](https://discourse.paraview.org/t/generate-process-ids-doesnt-work/15639/5 "2024-10-28T07:57:07Z")

</div>

Another workaround is to force remote rendering instead of using LocalRendering. (RemoteRenderThreshold)

With a bigger dataset, this issue would not happen as RemoteRendering would kick off.

More details here: [https://gitlab.kitware.com/paraview/paraview/-/issues/22285](https://gitlab.kitware.com/paraview/paraview/-/issues/22285)
