A few QoS related questions. Any assistance appreciated.
I have a client with various remote sites served with different mixes of interfaces and speeds. We've set up some policies on WAN egress ports, however we left with a few nagging questions:
How is the bandwidth calculated on Gig, Multilink, and Serial subinterfaces? Via bandwidth statements, aggregate interface speeds (as in Multilink), the default interface speed, or in cases like a Serial subinterface is it the main Serial interface's speed?
For serial subinterfaces can a policy be applied to a subinterface alone?
We had a policy running on 6Mb/s Serial interface and noticed that the default class was dropping even though the interface wasn't hitting 100% utilization. I was told the interface never went much beyond 40%, yet we were seeing drops in the default class. Why would this be?
Thanks in advance,