According to this:
On Feb 12, 2013, at 2:13 PM, Cody Permann <email@example.com> wrote:
> We have a number of tests that have started failing that use several hundred subdomains. I haven't dived into this at all yet, but on the surface it looks like we might need to call additional API functions or something in our ExodusII_IO wrappers.
> Just an FYI - I'll let you know when I get a chance to investigate.
I need to increase some parameter bounds in netcdf - I'm doing that and will commit a fix if it still works.