This one time, at band camp, Dinesh Nair said:

> > just given a greater weight?  and how is that weight determined in
> > relation to define queues?
>
> if you have 3 queues going into a pipe with weights 20, 30 and 50
> respectively, then queue 1 would get 20% of the bandwidth, queue 2 would
> get 30% and queue 3 would get 50%. essential, a queue's bandwidth
> percentage is determined by the queue weight divided by the sum of all
> weights going to the same pipe. in effect, the weight is a ratio, not a
> priority.

i understand how queue weighting works.  my question was what happens if
you have a) rules based on queues for a given pipe, and b) rules based
directly on that pipe, all operating at the same time.  what happens?  how
is the weighting arbitrated between queue-based rules and pipe-based
rules?

jj