On Feb 2, 2006, at 3:31 PM, Nuno Subtil wrote:
> Recently I have installed LAM 7.1.1 on a 40-node cluster of Intel
> machines running Debian. Since all machines are booting over the
> network, I seem to have hit the /tmp-on-NFS bug which causes tkill
> to hang upon issuing a lamhalt command.
>
> This issue is listed as corrected for the next LAM version. Is
> there a way to get a patch for this (and only this) bug for LAM
> 7.1.1? I would like to patch the Debian package sources and rebuild
> the package for the cluster.
This specific fix looks like it was committed at r9886. This command
gives the diff:
svn di -r 9885:9986 http://sourcehaven.osl.iu.edu/svn/trillium/trunk/
tools/tkill/tkill.c
--
{+} Jeff Squyres
{+} The Open MPI Project
{+} http://www.open-mpi.org/
|
