Håvard Berland
Håvard Berland
#9036 presents an alternative solution to the underlying problem.
Superseded by #9036
See https://github.com/equinor/ert/pull/8790, it is no (longer?) true that only SSH errors give 255.
It does not look like it is currently a problem with cluster failures in LSF in our tests, so maybe hold that until it is needed.
Wait until Python LSF driver has been stable something like a month.
Squash the commits? The explanation under "Approach" should also be in the git commit message.
nitpick: First line in git commit message should be less than 50 chars, and the remaining lines should be formatted at 72 chars.
It might make sense to keep the publisher sentinel in place, to avoid rare corner cases where all events are "sent" but not yet added to the queue, and then...
Proof of concept: ``` $ /usr/lib64/openmpi/bin/mpicxx -fopenmp -std=c++17 -o omp_mpi omp_mpi.c -lgomp $ cat runme.sh /usr/lib64/openmpi/bin/mpirun -np 8 ./omp_mpi $ time bash runme.sh I'm thread 2 out of 10 on...
Experimenting with OMP_NUM_THREADS and the `-np` option, it seems we can only detect the case when `-np` is more than 1.