> This is extremely tricky to do. How do you know which environment > variables to forward (foo in this case) and which not to (hostname). > SLURM has a better chance, since it's linux only and generally only run on > tightly controlled clusters. But there's a whole variety of things that > shouldn't be forwarded and that list differs from OS to OS. > > I believe we toyed around with the "right thing" in LAM and early on with > OPen MPI and decided that it was too hard to meet expected behavior.
It's an easy decision for Catamount, so the whole user environment gets sent out to the compute nodes. Anybody care to guess how big that can be when a (goofball) user invokes job launch from within a Makefile? -Ron
