flredecomp fails with newer version of parmetis
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Fluidity |
Confirmed
|
High
|
Jon Hill |
Bug Description
In build of fluidity and tools linked with petsc-dev (which includes a newer version of parmetis: parmetis-4.0.2-p3), flredecomp fails to upscale from a single process to multiple processes. The following comes from the output of running the channel_
mpiexec -np 4 ../../bin/
PARMETIS ERROR: The sum of tpwgts for constraint #0 is not 1.0
PARMETIS ERROR: The sum of tpwgts for constraint #0 is not 1.0
PARMETIS ERROR: The sum of tpwgts for constraint #0 is not 1.0
PARMETIS ERROR: The sum of tpwgts for constraint #0 is not 1.0
../../bin/
../../bin/
../../bin/
../../bin/
../../bin/
../../bin/
/lib/x86_
....
-------
MPI_ABORT was invoked on rank 2 in communicator MPI_COMM_WORLD
with errorcode 16.
NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI processes.
You may or may not see output from other processes, depending on
exactly when Open MPI kills them.
-------
*** FLUIDITY ERROR ***
Source location: (Zoltan_
Error message: After load balancing process would have an empty partition.
Backtrace will follow if it is available:
Use addr2line -e <binary> <address> to decipher.
Error is terminal.
....
I presume this is to do with the newer version of parmetis that's being linked into petsc-dev.
Changed in fluidity: | |
importance: | Low → High |
I'll take a look at this