an another question please
if i enable q1 PQ (30% for example) but if there is no trafic for the PQ;
the bandwidth of other queues q2/q3/q4 is limited at 70%
or the bandwidth reserved PQ can be used by the others queues?

i think, the answer will be goog for ingress queues too ?

Thanks

but for the ingress queues, i don’t understand the notion of 2 queues when there is a priority queue.
Doc Cisco says, by default 90% q2 and 10% q1 whith a priority 10% for q2.
if q2 is priority with 10%, why q2 is too 10% in the SRR process ?

Thanks you

Shaping is always active, no matter whether the interface is congested or not. Thus, it performs function similar to egress policing on the older 3550 models.

Just a quick question - when there is no congestion present on the interface, the srr-queue bandwidth shaped command would not limit traffic that matches a queue that is shaped, correct?

Thank you!

Sort of But version 5 diagrams will probably have extra pictures of medieval castles, dragons, knights and fairies

Obviously, shared mode weight does not affect PQ behavior. Up until the moment when I first verified PQ functionality, I was sure that “shaped” weight does affect PQ sending rate. However, as you can see (and run a verification yourself just to be sure), shaped weight has no effect on PQ either. Actually, this is what they briefly mention in the DocCD:

”
When you configure this command [priority-queue out], the SRR weight and queue size ratios are affected because there is one fewer queue participating in SRR. This means that weight1 in the srr-queue bandwidth shape or the srr-queue bandwidth share command is ignored (not used in the ratio calculation).
”

I’ll try to make another post on ingress queueing in policing in near future and investigate how ingress PQ actually works.

In the quick questions and answers section you say that it is not possible to restrict the PQ sending rate using “shaped” weight.

However in the CCIE Routing and Switching Exam Certification Guide (Third Edition, page 462) it is written that when the PQ still has many frames to send, but queues 2, 3, and 4 are empty, in shared mode the PQ would send at full line rate while in shaped mode the switch would simply not service the PQ part of the time so that its overall rate would be the bandwidth configured for that queue.
According to the book it looks like it is possible to limit the PQ sending rate configuring a shaped weight but your verification scenario demonstrates that it is not true.

Have I understood wrong what is written in the certification guide?