trouble with CPU utilization via net-snmp

Hi,

I'm trying to monitor a Linux device (NOT redhat) running net-snmp. If I set a ucd-snmp based monitor, I don't have error, but still 0 as result value (while the real server upload is : 15:23:52 up 82 days, 5:16, 2 users, load average: 0.06, 0.06, 0.09)

Since it's not a redhat, I can't use the CLI based monitor, which parse top output (and since it's not the same output in debian's top..)

by the way, why you don't simply use /proc/loadavg

and btw, you run the command : <code># top b n 1 | grep 'CPU[0-9]* states:'<code>
but I don't understand how you can have the cpu usage using this way (I ran this command
in rhel3)

Thanks