bstrap_proxy error running Fluent in parallel on multiple nodes

- November 8, 2024 at 1:49 am
  
  jedicker
  Subscriber
  
  I'm trying to run a parallel Fluent job that spans multiple nodes on a Linux cluster. The job works fine when run on a single node, but a multi-node job fails with the error "unable to run bstrap_proxy on (hostname)."
  The cluster is running Linux (RHEL 9.4) with Slurm 24.05. I've tried Ansys 2024R1 and 2023R2. They both fail in the same way.
  I strongly suspect the issue is related to Slurm 24.05. Are there any workarounds?
  I saw another issue about AnsysEM that provided a patch for the HFSS slurm_srun_wrapper.sh script. Perhaps a similar issue affects fluent.
- December 9, 2024 at 8:32 am
  
  Christian Woiczinski
  Subscriber
  
  Facing similar problem when trying to run CFX spread over multiple nodes.
- December 13, 2024 at 4:56 pm
  
  George Karnos
  Ansys Employee
  
  Hi,
  Please see if the following post helps
  Fluent fails with Intel MPI protocol on 2 nodes