mpdboot error using mpich2

[harold]

Hi all. I am new here so first lemme say hello!

Anyway, when I try to run mpdboot i get the error:

mpdboot_node1 (handle_mpd_output 392): failed to handshake with mpd on
node2; recvd output={}


Here is where I am:


- i have nfs set up
- user name 'mpid' on both nodes with home mounted at /nfs/home/mpid
- password-less ssh on both machines (tested and works fine)
- path is now PATH=$PATH:/nfs/mpich/bin

Then i have a mpd.hosts file that looks like the following:

node1
node2

where node1 and node2 are hostnames to my machines


mpd runs fine from both machines when i execute it myself... but when
i try it using mpdboot it errors:


[***@node1 ~]$ mpdboot -v -n 2 -f mpd.hosts
running mpdallexit on node1
LAUNCHED mpd on node1 via
RUNNING: mpd on node1
LAUNCHED mpd on node2 via node1
mpdboot_node1 (handle_mpd_output 392): failed to handshake with mpd on
node2; recvd output={}





any suggestions?

thanks in advance,
harold

[harold]

Found the problem. It was stupid. basically a bad /etc/hosts file.

I had something like

127.0.0.1 localhost
127.0.1.1 node0


192.168.3.120 node0
.....


When I should have just had:

127.0.0.1 localhost

192.168.3.120 node0

In the first case it never gets to the outside ip address in the name
resoultion (duh)


harold