Learning Forum

[1985184216]

Dear Ansys community,

I am using Fluent 2026 R1 with the GPU solver enabled, running on a parallel cluster (MPI). The simulation runs perfectly fine without any monitors. However, when I define a point pressure monitor (Report Definition → Point → Pressure) before initialization, the calculation crashes shortly after starting, with the following error messages:

Abort(17) on node 3 (rank 3 in comm 0): Fatal error in PMPI_Waitall: See the MPI_ERROR field in MPI_Status for the error code
Abort(17) on node 2 (rank 2 in comm 0): Fatal error in PMPI_Waitall: See the MPI_ERROR field in MPI_Status for the error code
Error in the fluent (fl) process : ended abnormally

The crash is reproducible and only occurs when the point monitor is active. I have tried different monitor report intervals, but the issue persists.

My questions are:

1. Does the GPU solver in 2026 R1 officially support point monitors (e.g., pressure at a specific coordinate)? If not, where can I find the official list of supported monitor types for the GPU solver?

2. Are there any workarounds to monitor a local pressure value while still using GPU acceleration? 

3. Apart from switching back to the CPU solver, are there any recommended MPI settings or environment variables that might prevent this crash (e.g., changing the MPI implementation, disabling certain BTLs, or adjusting the monitor frequency)?

I am aware that GPU solver support is evolving, and I have seen mentions of "volume and mass monitors" being added in 2024 R2, but I couldn't find clear documentation on point monitors for 2026 R1. Any guidance, official reference, or personal experience would be greatly appreciated.

Thank you in advance for your help!