<br>First thank you all a lot for providing help here. <br><br>I have two Linux SMP nodes, say n1 and n2. <br><br>n1 has Linux 2.1.18 with gcc 4.1.2 and n2 has Linux 2.1.20 with gcc 4.1.1 running. Both are on NFS and shh-accesible.<br>
<br>I got MPICH2 installed and running. <i>mpdringtest </i>with both nodes on the ring runs fine. <br><br>When I schedule tasks on same node, they run fine.<br><br>However, when I schedule tasks across both nodes, with n2 as the master node, I get the following message on n1:<br>
<br>mpd_uncaught_except_tb handling:<br> exceptions.KeyError: 'process_mapping'<br> /usr/local/bin/mpd 1354 do_mpdrun<br> msg['process_mapping'][lorank] = self.myHost<br> /usr/local/bin/mpd 984 handle_lhs_input<br>
self.do_mpdrun(msg)<br> /usr/local/bin/mpdlib.py 780 handle_active_streams<br> handler(stream,*args)<br> /usr/local/bin/mpd 301 runmainloop<br> rv = self.streamHandler.handle_active_streams(timeout=8.0)<br>
/usr/local/bin/mpd 270 run<br> self.runmainloop()<br> /usr/local/bin/mpd 1643 ?<br> mpd.run()<br>n1-wulf.myhost.org_mpdman_1 (run 287): invalid msg from lhs; expecting ringsize got: {}<br><br><br>
After doing <i>mpdallexit</i>, n2 shows the following message:<br><br><a href="http://mpiexec_n2-wulf.myhost.org">mpiexec_n2-wulf.myhost.org</a> (mpiexec 377): no msg recvd from mpd when expecting ack of request<br>--------------------<br>
<br>My question: Do I need exactly same OS version on both n1 and n2, or same gcc version on both, or I may have some other installation problem?<br><br>Thanks in advance,<br>- Rajnish.<br><br><br>