MPI_Abort fails with TMI fabric

[rhnobes]

I am finding that MPI_Abort is not killing processes on remote nodes when I_MPI_FABRICS=shm:tmi. The attached PBS job works correctly when the default fabric  (shm:dapl) is used, but fails to abort cleanly when shm:tmi is used. Any help with overcoming this problem would be much appreciated.
 

[Artem_R_Intel1]

Hello,

Could you please try the same scenario with latest Intel MPI Library 5.1.1? It looks like known already fixed issue.

[rhnobes]

Dear Artem,

I apologise for the delay in replying. Thank you very much for your advice - I can confirm that the problem has been fixed in Intel MPI 5.1.1.

Best wishes,

Ross