LAM/MPI logo

LAM/MPI General User's Mailing List Archives

  |   Home   |   Download   |   Documentation   |   FAQ   |   all just in this list

From: Brian W. Barrett (brbarret_at_[hidden])
Date: 2003-07-25 11:59:48


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Thursday, July 24, 2003, at 01:46 PM, Shahram Tehranian wrote:

> Basically I am wondering if it's possible to bring a remote deamon up
> separately once it has failed without doing a complete lamboot. I am
> developing a fault tolerant master-worker MPI-application. Currently
> my program can go on if the process or lam deamon fails on a worker
> node. Would it be possible to bring the remote deamon up separately
> and reconnect the worker process to the master without having to bring
> down the entire application.

This is possible, with some constraints. You have to start the new
daemon from a node with a daemon already running (using whatever boot
mechanism you used to start the original lam daemons). The command you
want is 'lamgrow' - see the lamgrow(1) manpage for more information.

Hope this helps,

Brian

- --
   Brian Barrett
   LAM/MPI developer and all around nice guy
   Have a LAM/MPI day: http://www.lam-mpi.org/
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.2 (Darwin)

iD8DBQE/IWIF3TvSMqaebW4RAjRVAJ99C+5KIR3HiHkUtffKLQP9Kv+fkQCg+xhd
LSniNL3GBozzNv3ruZXDqVE=
=Jj4o
-----END PGP SIGNATURE-----