If I do a "show system status" command on a vEdge the output lists a number of CPU values :
Load average: 1 minute: 0.11, 5 minutes: 29, 15 minutes: 38
Processes: 241 total
CPU allocation: 32 total, 3 control, 29 data, 1 tcpd
CPU states: 11.00% user, 10.10% system, 78.90% idle
Memory usage: 2973024K total, 752796K used, 1865932K free
65348K buffers, 288948K cache
All these values are also available through SNMP. What's the difference between, say, "1 minute load average" and adding "CPU state user" to "CPU state system" ? and if I'm trying to monitor one variable to give a good, representative feel for system load, so I can spot when the vEdge is getting overworked, which variable should I monitor ?
I'd be interested in a 'theoretical' answer, or feedback from anyone running a Viptela network - what do you monitor ? Or what does vManage show ?
Thanks !
