So I use a lot of reports and alerts to try to sniff out chassis's (which are usually core devices or devices that have a lot of users on them) that are being bandwidth, buffer, cpu, mem starved etc etc.
As I am sure many of you know each blade on a chassis shares resources with the other blades on the backblane, each blade contains port groups which share resources and bandwidth. So if you have a group of 8 ports and 4 are super busy and the other 4 do nothing but the first four use all the buffers then all 8 will drop traffic.
Right now I keep an eye on this by mostly watching discards on ports.
But to do this better and take this a step further I would think it would be great if there was a canned Cisco Chassis view/console that could take into account several factors for the health of a chassis like mem, cpu , discards maybe even syslog or log errors.
I think this would be basically the same type of direction you guys are going in with VM stuff but a one stop shop for chassis health, stats, and status would be awesome. Especially considering a chassis(s) will generally be by far some of the most important network gear in a network, especially large ones.
