Hi all,
This is my first mail to the list.
My problem is that we want to monitor a cluster of computers consiting on:
1 front-end node: 2NICS
10 computing nodes with 2 NICS too.
I have managed to put all the information together on a web page using the
web front-end but I think it is not like it should. My /etg/gmetad.conf
looks like (On the server node):
data_source "dual" 192.168.1.2
data_source "dual2" 192.168.1.1
data_source "dual3" 192.168.1.3
data_source "dual4" 192.168.1.4
data_source "dual5" 192.168.1.5
data_source "dual6" 192.168.1.6
data_source "dual7" 192.168.1.7
data_source "dual8" 192.168.1.8
data_source "dual9" 192.168.1.9
data_source "dual10" 192.168.1.10
/etc/gmond.conf for sever node is:
name "dual"
location "0,0,0"
and for node1 is:
name "dual"
mcast_if eth1
mcast_ttl 3
Is it ok the data_source structure? I think it is not but I can not get it
running other way but this. What should I do to get a whole view of my
cluster like some web pages show?:
http://birdie.isi.edu/ganglia/
Another question:
What about the CGT Grid appearing on the example? It appears Grid on mine
but I don't even need it, just a cluster.
Thanks
