One simple way to do so from the command line is (zsh approach, modify
accordingly for other shells):
for node in node1 node2 node3 node4; do
rsh $node print "node $node: \$(uptime)"
done
where node1, node2,... are the names for the nodes you want to status .
We use this on our Linux computational clusters to monitor how busy they
are (mind you, they are inaccessible to pubic access, so rsh is not a
security risk for us). The script is also readily adaptable to other
commands, such as w for monitoring who is logged in.
If you have a compute cluster and a numerical naming convention such as
n1, n2,...n64 for the nodes, then the first line can be modified to be
for node in n{1..64}; do
where the {1..64} would specify a range of values, in this instance 1
through 64. The variable node then become n1, n2,...n64.
Don Jones
>Is anyone aware of a utility for monitoring the CPU usage of each
>processor in an Xserve cluster?
>
>It'd be nice to have the ability to see "at a glance" how busy the
>cluster is ...
>
>Kerry
>_______________________________________________
>scitech mailing list | email@hidden
>Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/
scitech
>Do not post admin requests to the list. They will be ignored.
_______________________________________________
scitech mailing list | email@hidden
Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/scitech
Do not post admin requests to the list. They will be ignored.
