Christian Simon ha scritto:
> Hi everyone.
>
> I have a cluster with G5 xserve cluster nodes. System is MacOSX server
> 10.3.9,
> the switch is a GigE Asante (G52400W); lam-mpi version 7.1.1 was
> compiled with xlf/xlc.
>
> Everything works fine with the default network settings for the nodes.
>
> As soon as I try larger MTU than 1500, the lam-mpi environment fails to
> start:
> lamboot hangs forever.
>
> I guess -and I hope- it is not, strictly speaking, a lam-mpi problem.
> Did I miss something in the FAQ ? Any suggestion ?
> --
> Christian
>
> _______________________________________________
> This list is archived at http://www.lam-mpi.org/MailArchives/lam/
>
My suggestion:
you should make sure that all the the network cards in your subnet are set to
that MTU value and -very important- that the switch can support it. A simple way
to verify is using ping within 2 hosts with a packet size higher than 1500 (it
can be specified by an option), and notice whether it gets an answer from the
other host. Your network may apparently work "at rest" but hang as soon as some
network-hungry application tries to send long packets. It is likely not to be a
LAM-MPI problem.
Davide
--
__________________________________________________________
Davide Cesari ARPA-Servizio Idro Meteorologico __
www http://www.arpa.emr.it/sim ---
Address: ARPA-SIM, Viale Silvani 6, 40122 Bologna, Italy
__________________________________________________________
|
