CBWFQ packet Drops

From: Brennan_Murphy@NAI.com
Date: Tue Sep 10 2002 - 11:33:56 GMT-3


I am having trouble understanding when/why CBWFQ drops packets.
I've got 5054 packet drops on a PVC but they started incrementing
even when the 5 minute offered rate was well below the shape
adaptive command. See config below.

More info:
This is on a 7505, CBWFQ, IP CEF Disti is on. Frame relay links

Any insights are appreciated.

class-map match-all gold
  match ip precedence 2
!
!
policy-map CBWFQ
  class gold
    bandwidth percent 30
    fair-queue
policy-map 256CIR
  class class-default
    shape average 768000 3072 3072
    shape adaptive 384000
   service-policy CBWFQ

rt1#sh pol int hss0/0/0.3
 Hssi0/0/0.3: DLCI 320 -

  Service-policy output: 256CIR (1231)

    Class-map: class-default (match-any) (1251/0)
      124803 packets, 109774228 bytes
      5 minute offered rate 417000 bps, drop rate 0 bps
      Match: any (1255)
      queue size 0, queue limit 17
      packets output 134504, packet drops 5054
      tail/random drops 5054, no buffer drops 0, other drops 0
      Shape: cir 768000, Bc 3072, Be 3072
        lower bound cir 384000, adapt to fecn 0
        output bytes 82134895, shape rate 437000 bps

      Service-policy : CBWFQ (1233)

        Class-map: gold (match-all) (1235/2)
          0 packets, 0 bytes
          5 minute offered rate 0 bps, drop rate 0 bps
          Match: ip precedence 2 (1239)
          queue size 0, queue limit 5
          packets output 0, packet drops 0
          tail/random drops 0, no buffer drops 0, other drops 0
          Bandwidth: 30%, kbps 230
          Fair-queue: per-flow queue limit 2

        Class-map: class-default (match-any) (1245/0)
          124803 packets, 109774228 bytes
          5 minute offered rate 417000 bps, drop rate 0 bps
          Match: any (1249)
          queue size 0, queue limit 11
          packets output 134504, packet drops 5054
          tail/random drops 5054, no buffer drops 0, other drops 0

-BM



This archive was generated by hypermail 2.1.4 : Mon Oct 07 2002 - 07:43:49 GMT-3