Hi:
I am running a parallel MPI code on a Intel Pentium
based Linux Cluster using 16processors. The code runs
fine for a long time before aborting with the
following error messages:
MPI_Waitany: process in local group is dead (rank 14,
MPI_COMM_WORLD)
MPI_Waitany: process in local group is dead (rank 15,
MPI_COMM_WORLD)
MPI_Waitany: process in local group is dead (rank 6,
MPI_COMM_WORLD)
MPI_Waitany: process in local group is dead (rank 7,
MPI_COMM_WORLD)
MPI_Waitany: process in local group is dead (rank 4,
MPI_COMM_WORLD)
MPI_Waitany: process in local group is dead (rank 5,
MPI_COMM_WORLD)
MPI_Waitany: process in local group is dead (rank 13,
MPI_COMM_WORLD)
MPI_Waitany: process in local group is dead (rank 3,
MPI_COMM_WORLD)
MPI_Waitany: process in local group is dead (rank 8,
MPI_COMM_WORLD)
MPI_Waitany: process in local group is dead (rank 9,
MPI_COMM_WORLD)
MPI_Waitany: process in local group is dead (rank 12,
MPI_COMM_WORLD)
Rank (8, MPI_COMM_WORLD): Call stack within LAM:
Rank (8, MPI_COMM_WORLD): - MPI_Waitany()
Rank (8, MPI_COMM_WORLD): - MPI_Waitall()
Rank (8, MPI_COMM_WORLD): - MPI_Alltoall()
Rank (8, MPI_COMM_WORLD): - main()
The above message, i.e., Rank(8, .......) also appears
for all the processors except a couple of them.
I also have a call to MPI_ALLREDUCE. Is this where the
error is coming.
Thanks
Sarma
=====
Sarma
===========================================================
Chesindhi podhu, Cheyyandhi raadhu
===========================================================
R. Laxmi Narasimha Sarma
Dept. of Mechanical Engineering
University of Illinois, Urbana-Champaign
__________________________________
Do you Yahoo!?
Yahoo! Mail Address AutoComplete - You start. We finish.
http://promotions.yahoo.com/new_mail
|
