A whole swathe of people have been made aware of the issues that have arisen as a result of a researcher here looking to run PISM, which sits on top of PETSc, which sits on top of OpenMPI.
I am happy to be able to inform you that the problems we were seeing would seem to have been arising down at the OpenMPI level. If I remove any acknowledgement of IPv6 within the OpenMPI code, then both the PETSc examples and PISM application have been seen to be running upon my initial 8-processor parallel environment when submitted as an Sun Grid Engine job. I guess this means that the PISM and PETSc guys can "stand easy" whilst the OpenMPI community needs to follow up on why there's a "addr.sa_len=0" creeping through the interface inspection code (upon NetBSD at least) when it passes thru the various IPv6 stanzas. Thanks for all the feedback on this from all the quarters, Kevin -- Kevin M. Buckley Room: CO327 School of Engineering and Phone: +64 4 463 5971 Computer Science Victoria University of Wellington New Zealand
