documentation for aborting parallel PyMeep jobs (#1184)

* documentation for aborting parallel PyMeep jobs

* fix

* revert to -np (instead of -n)

* Update Parallel_Meep.md

Co-authored-by: Steven G. Johnson <stevenj@mit.edu>

doc/docs/Parallel_Meep.md

@@ -40,6 +40,11 @@ For a potential improvement in [load balancing](FAQ.md#should-i-expect-linear-sp
 
 In general, you cannot run Meep interactively on multiple processors.
 
+**Warning:** when running a parallel PyMeep job, the failure of any one MPI process may cause the simulation to deadlock and not abort. This is due to a [behavior of `mpi4py`](https://mpi4py.readthedocs.io/en/stable/mpi4py.run.html). To avoid having to manually kill all the remaining processes, a simple solution is to load the `mpi4py` module (for versions 3.0+) on the `mpirun` command line:
+```sh
+mpirun -np 4 python -m mpi4py foo.py
+```
+
 ### Different Forms of Parallelization
 
 Parallel Meep works by taking your simulation and dividing the cell among the MPI processes. This is the only way of parallelizing a single simulation and enables simulating very large problems.

scheme/Makefile.am

@@ -56,7 +56,7 @@ meep-enums.scm: meep_enum_renames.i
 ##############################################################################
 
 # what is printed out when invoking your program with --version:
-VERSION_STRING = "Meep @VERSION@, Copyright (C) 2005-2020 Massachusetts Insitute of Technology."
+VERSION_STRING = "Meep @VERSION@, Copyright (C) 2005-2020 Massachusetts Institute of Technology."
 
 MY_DEFS = -DHAVE_CTL_HOOKS=1 -DHAVE_CTL_EXPORT_HOOK=1