I installed intel MPI on windows 7 x64 and executed "> mpiexec -n 4 program.exe", it seemed to be running fine for about 30h and was using the appropriate resources expected. However, the process suddenly exited with about 10h remaining in the computation with the following error stack:
[mpiexec@Simulation-PC] ..\hydra\pm\pmiserv_cb.c (773): connection to proxy 0 at host Simulation-PC failed
[mpiexec@Simulation-PC] ..\hydra\tools\demux\demux_select.c (100): callback returned error status
[mpiexec@Simulation-PC] ..\hydra\pm\pmiserv\pmiserv_pmci.c (501): error waiting for event
[mpiexec@Simulation-PC] ..\hydra\ui\mpich\mpiexec.c (1059): process manager error waiting for completion
I cant seem to find anyone else in these forums with these errors, so hopefully you can provide some insight. I am running this on a single-socket intel i7-5960x processor with 8-cores and it is not part of a cluster.
Unfortunately this diagnostic message is quite common and may happen in different problem situations.
Could you please provide more details about your runs:
- Is the problem reproducible or at least observed regularly?
- Which version of Intel MPI Library do you use?
- Do you use any specific Intel MPI environment variables (please specify it if any)?