ANSYS Fluent v19 Buffering for file scan… fluent process could not be started

    • oobiols

      Hello everybody and thanks for reading.

      I installed ANSYS 19 in my 32-core machine. I want to use Fluent to solve Fluid Dynamics cases. I want to use multi-block grids, and I am having no issues when solving my own benchmark cases (Compressible flow around a 2D Cylinder (1block), Compressible flow around a 2D Cylinder (2 blocks), Compressible flow around an airfoil (8 blocks) ) as test cases. These cases work fine.

      However, I eventually want to solve a larger case. This larger case contains a grid with 248 blocks, and the entire case file's size is 17GB. When I try to read the case, I get the following messages in the console:

      "Multicore SMT processors detected. Processor affinity set!

      "Reading case ..."

      "buffering for file scan..."

      "MPI Application rank 0 killed before MPI_Finalize() with signal 9. /fluent installation directory/ : line 3191: 31467 Killed. $EXE_CMD $CX_FLAGS $FL_FLAGS "$HOST_FUNCTION. The fluent process could not be started"

      There is no other message in the console. Before sending this discussion I tried hard to figure the issue out with suggestions from other posts, but none could help me figure it out. All I could conclude - and probably I am wrong - is that it may be either a RAM memory issue or a permission on the number of compute nodes I am allowed to use due to my license. 

      Help is much appreciated. Thank you very much

      Some specifications of my machine are:

      Total RAM: 62GB

      Available RAM: 60GB,

      Available buff/cache: 60 GB

      CPU model name: Intel(R) Xeon(R) CPU E5-2630 v3 @ 2.40GHz





    • Rob
      Ansys Employee

      On the student bundle you're capped at 512k cells, and possibly 4 cores. Exceeding these should give a licence error. Running out of RAM can give that error, or a malloc error, and a 17GB case file is huge.  If you try reading it into Fluent in standalone mode what happens? I also suggest monitoring RAM usage via Task Manager (Windows) or "top" (LINUX) to see if that's the cause. 

    • oobiols

      Thanks! Yes, it was a RAM memory problem. I could try a larger RAM memory and it could handle the case. I really appreciate your help. Monitoring RAM usage also helped. Thanks!!

Viewing 2 reply threads
  • You must be logged in to reply to this topic.