Create charts which aggregate information for all servers (or choose
specific), I *think* this could be done with a massive CDEF in the RRD
graphs, but I am not sure if there is a limit.
Total Disk Used
Total Disk Avail
Total Network Bytes/In
Total Network Bytes/Out
Total Mem Used
Total Mem Avail
CPU is real tough. One way would be to try and inventory all the
procs and use CSPEC info to determine 'capacity' between procs. That
would be real hard.
Or, create a chart with the median and average CPU utilization.
Scott Walters
-PacketPusher