I have an application that consumes most of the available WAN bandwidth during off-peak hours. Occasionally it runs on into peak hours and severely degrades WAN performance. The WAN router is SP managed, so I can not perform rate-limiting there.
Traffic from this application comes from a server connected to a WS-X6148A-TX port on the Cat and exits on a port on the same card to the SP managed router. The connection to the router is "routed" in that it is on a separate /30 Vlan than the user LAN traffic.
What I would like to do is rate-limit this traffic during prime time. What is the best way to enforce the rate limit? With a policer on the ingress interface? Some other method? Is there any way to make this time sensitive? If I were using IOS rather than CatOS, would this be easier?