Lisandro Dalcin

> I'll note that this is consistent with the centuries old practice with errata in printed books, for better or worse. We are talking about a 1K pages document that...

My fixes related to the use of the random number generator have been extremely useful in production for getting reproducible partitions in parallel runs.

@prj- I'll start by submitting a PR with my random number generator fixes. These should go to the GKLib repo. @karypis I do not see any repository for ParMETIS. Is...

Is this ever going to happen? Or Am I supposed to keep opening issues over and over again when broken things get merged into main?

@hppritcha You may want to take a look at this one.

@hppritcha Could you please rebase this PR to latest main? That way, I can give it a try with mpi4py testsuite.

@hppritcha FYI, all mpi4py tests [passed](https://github.com/mpi4py/mpi4py-testing/actions/runs/3100525240/jobs/5020903830). In the singleton case, the unsupported operation message is printed to stdout [[link]](https://github.com/mpi4py/mpi4py-testing/actions/runs/3100525240/jobs/5020903830#step:17:576).

@hppritcha Could you please rebase this PR to latest main?

@hppritcha When running in singleton init mode, `MPI_Group_from_session_pset` now fails with error class `MPI_ERR_OTHER`. Perhaps `MPI_ERR_UNSUPPORTED_OPERATION` would be more appropriate and in line with other recent changes from you?

@hppritcha I'm not sure what to say. mpi4py's singleton, np=1, and np=2 are passing. However, for np=3, I got a failure in a [test case](https://github.com/mpi4py/mpi4py/blob/master/test/test_comm.py#L46) that basically calls `MPI_Comm_dup` and...