02-07-2008 08:19 AM - edited 03-03-2019 08:36 PM
i have policy applied onto ethernet subinterface, and this is result of one class. How is it possible to have two nubers of packets different? 622673 vs 2720!?!?!
Class-map: VOICE (match-all)
622673 packets, 46077802 bytes
5 minute offered rate 0 bps, drop rate 0 bps
Match: ip dscp ef (46)
Queueing
Strict Priority
Output Queue: Conversation 40
Bandwidth 33 (%)
Bandwidth 168 (kbps) Burst 4200 (Bytes)
(pkts matched/bytes matched) 2720/201280
(total drops/bytes drops) 0/0
02-07-2008 09:10 AM
If I recall correctly, the first number is all packets that matched the class, the second number is those that actually software queued.
02-07-2008 09:22 AM
so ... is this ok or not ok. On my policies that I have on frame relay links .... these two numbers are the same.
02-07-2008 09:38 AM
Ok or not okay? Well it usually depends whether it adversely impacts your traffic. It's somewhat like asking whether a .25% drop rate is ok or not.
Also, with just these numbers, we don't know whether an occasional voice packet was queued or a cluster of such queued. Again, given just a drop count over total packets, we can calculate a percentage but also don't know whether just 1 packet fell of the end 10x or whether 10 packets fell off 1x, etc.
As to your frame relay links showing the numbers the same, also assuming I'm correct in what they represent and that the IOS you're using is counting properly, it could be the difference between the example you posted has little background traffic and the latter lots and lots of background traffic. Level of background traffic should have little impact to LLQ traffic. What differs, with little background traffic, LLQ packets often finds free interface and transmits without queuing. With lots of background traffic, LLQ fines interface busy sending a packet, queues, but goes next.
Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: