Neil,
Thank you for your quick response. I have tried the ping command and it
works OK. I tried the traceroute and it seems to
not to work. (1 * * * etc.) Do you have any idea why this is happening?
(What is the TTCP? Sorry i am a newbie)
What should i check?
Gkikas
PS: I am currently testing the MPICH1.2.5.2 and i am running a big job in
order to test the stability of the cluster.
Would this interrupt with the traceroute? The ping seems to work just fine.
When i tested the LAM i was not running
the MPICH. Should i try traceroute without the MPICH1.2.5.2 runing??? I
think it should work fine just as the ping.
RE:
>
Gkikas,
Have tried running a TCP/IP test (e.g. TTCP, ping, traceroute) between the
various nodes, to make sure that you don't have an intermittent network
problem. This should indicate whether or not the problem is really a LAM
issue or a hardware on.
Regards
Neil Storer
|
