On Mon, 16 Feb 2004, Neil Storer wrote:
> I tried running a LAM-MPI program on a couple of of our IBM Nighthawk2
> nodes and got the error message:
>
> -----------------------------------------------------------------------------
> The boot SSI rsh module found that your local host is not in the
> hostfile "./hostlist".
> [snipped]
Excellent diagnosis -- thanks!
You're right -- we had a fixed sized buffer where we shouldn't have. I
actually extrapolated your solution and made lamnet.c use
dynamically-sized buffers so that it will always guarantee to have enough
memory to get all interfaces from ioctl(). This has been fixed both at
the CVS HEAD and on the 7.0 branch; it will be released as part of 7.0.5.
I've attached a new lamnet.c for those who care. :-)
--
{+} Jeff Squyres
{+} jsquyres_at_[hidden]
{+} http://www.lam-mpi.org/
|
