Particle tracer deadlock

I’ve been trying to use the Particle Tracer filter in ParaView 6.0. I am in client/server running on 1 node with 8 MPI ranks. After running a few steps (5-25), pvserver deadlocks. (At least, it becomes unresponsive but the executables are still running.)

I suspect this is related to the problem reported in Particle tracer crashing. Based on the discussion there, I tried backing out to ParaView 5.12, and it worked with no problem. However, it looks like @mwestphal’s fix for the crash for PV 6.0 is not fixing this deadlock.

Unfortunately, this dataset is too big to try to post somewhere. However, if someone has access to OLCF I can point you to the files.

The issue I fixed was indeed not a deadlock but a crash, and I dont see how it would relate to deadlocking.

Also I did not specifcally test for distributed computing, apart from the existing test being run by CI using 4 cores.

So I dont doubt that there can be issues in distributed in this filter, can you try reproducing with a tay dataset you could share with me ?