Partitioning an exodus mesh with node sets results in strange entries in the partitioned file.
For instance, the attached mesh has lots of node sets for testing Amanzi's mesh capabilities. ncdumping the .exo file shows, for instance, node set 11:
node_ns11 = 33, 34, 50, 49, 35, 51, 36, 52 ;
After partitioning via:
mpiexec -n 2 meshconvert --classify=1 --partition=1 --partition-method=0
we see:
ncdump hex_3x3x3_sets.par.2.0 -->
node_ns11 = 0, 0, 0, 26, 34, 25, 33 ;
ncdump hex_3x3x3_sets.par.2.1 -->
node_ns11 = 18, 17, 33, 34, 22, 37, 0, 0 ;
It is a bit unclear what is going wrong here.
hex_3x3x3_sets.exo.txt
Partitioning an exodus mesh with node sets results in strange entries in the partitioned file.
For instance, the attached mesh has lots of node sets for testing Amanzi's mesh capabilities. ncdumping the .exo file shows, for instance, node set 11:
node_ns11 = 33, 34, 50, 49, 35, 51, 36, 52 ;After partitioning via:
mpiexec -n 2 meshconvert --classify=1 --partition=1 --partition-method=0we see:
It is a bit unclear what is going wrong here.
hex_3x3x3_sets.exo.txt