cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
1066
Views
0
Helpful
12
Replies

QoS on 7500

Daniel Graham
Level 1
Level 1

Having some issues with QoS on 7500. Policy is applied to a serial interface. IP dCEF is enabled globally.

Class class-default has no matches.

Any ideas?

When looking at output of show policy-map int:

     Class-map: ef (match-all)
      37827 packets, 9819852 bytes
      5 minute offered rate 0 bps, drop rate 0 bps
      Match: access-group 100
      Priority: kbps 1024, burst bytes 25600, b/w exceed drops: 0

    Class-map: cs3 (match-all)
      0 packets, 0 bytes
      5 minute offered rate 0 bps, drop rate 0 bps
      Match: access-group 101
      queue size 0, queue limit 32
      packets output 353, packet drops 0
      tail/random drops 0, no buffer drops 0, other drops 0
      Bandwidth: kbps 128, weight 8.33

    Class-map: class-default (match-any)
      0 packets, 0 bytes
      5 minute offered rate 0 bps, drop rate 0 bps
      Match: any
      queue size 0, queue limit 96
      packets output 464091, packet drops 87
      tail/random drops 87, no buffer drops 0, other drops 0
      Fair-queue: per-flow queue limit 24

Cisco Internetwork Operating System Software
IOS (tm) RSP Software (RSP-JK9O3SV-M), Version 12.3(10), RELEASE SOFTWARE (fc3)
Copyright (c) 1986-2004 by cisco Systems, Inc.
Compiled Mon 16-Aug-04 20:44 by kellythw
Image text-base: 0x40010AFC, data-base: 0x423C6000

ROM: System Bootstrap, Version 11.1(8)CA1, EARLY DEPLOYMENT RELEASE SOFTWARE (fc1)
BOOTLDR: RSP Software (RSP-BOOT-M), Version 11.2(12a)P, RELEASE SOFTWARE (fc1)

DSL1 uptime is 1 year, 20 weeks, 6 days, 12 hours, 17 minutes
System returned to ROM by reload at 05:54:41 UTC Tue Aug 12 2008
System restarted at 05:58:08 UTC Tue Aug 12 2008
System image file is "slot0:rsp-jk9o3sv-mz.123-10.bin"
Last reload reason: Reload command

RSP4 (R5000) processor with 262144K/2072K bytes of memory.
R5000 CPU at 200MHz, Implementation 35, Rev 2.1, 512KB L2 Cache
Last reset from power-on
G.703/E1 software, Version 1.0.
G.703/JT2 software, Version 1.0.
X.25 software, Version 3.0.0.
SuperLAT software (copyright 1990 by Meridian Technology Corp).
Bridging software.
TN3270 Emulation software.
Primary Rate ISDN software, Version 1.1.
Chassis Interface.
1 VIP2 controller (1 ATM).
3 VIP2 R5K controllers (2 FastEthernet)(2 ATM)(1 Channelized T3).
2 FastEthernet/IEEE 802.3 interface(s)
28 Serial network interface(s)
3 ATM network interface(s)
1 Channelized T3 port(s)
123K bytes of non-volatile configuration memory.

If

cisco

2 Accepted Solutions

Accepted Solutions

Hello Dan,

have you introduced MPLS in your network, is MPLS running on the interface?

May you post the configuration of the interface and of the ACLs?

Hope to help

Giuseppe

View solution in original post

Hello Dan,

the problem is clear.

try to do the following:

int ser 6/0/0:9

no service out t1-esp-qos-policy

service out t1-esp-qos-policy

I've found some QoS related bugs that suggest this action as an attempt to have CEF to rebuild the correct tables

Hope to help

Giuseppe

View solution in original post

12 Replies 12

Giuseppe Larosa
Hall of Fame
Hall of Fame

Hello Dan,

>> 5 minute offered rate 0 bps, drop rate 0 bps
      Match: any
      queue size 0, queue limit 96
   >>>>   packets output 464091, packet drops 87

there are no current matches but there have been matches in the past.

Check if invoked ACLs 100 and 101 contain a permit ip any any on the traffic classes if so the default class is empty.

Hope to help

Giuseppe

ACL's are ok.

There is always traffic on this interface, but always show 0 matches.

If there have been matches in the past, from what I see it will still display them.

Hello Dan,

have you introduced MPLS in your network, is MPLS running on the interface?

May you post the configuration of the interface and of the ACLs?

Hope to help

Giuseppe

Ok, I didnt mean to click the correct answer button, not sure how to reverse that. Sorry.

No MPLS running.

Here is the output:

interface Serial6/0/0:9
description xxxx

bandwidth 1536
ip address 207.241.135.41 255.255.255.252
no ip redirects
service-policy output t1-esp-qos-policy
no cdp enable
no clns route-cache

Extended IP access list 100
    10 permit esp any any dscp ef (1982822 matches)


Extended IP access list 101
    10 permit esp any any dscp cs3 (10024 matches)

Hello Dan,

thanks for your kind remarks don't worry about this mistake.

if from the router you try to ping the other side what happens?

I'm starting to think you are facing a bug, but the above basic test should help.

Use an extended ping to send 1000 packets then post sh policy-map int Serial6/0/0:9

it looks like that ESP traffic should go on the defined traffic classes, there is a crypto map applied to the device interface?

Hope to help

Giuseppe

ping


Protocol [ip]:
Target IP address: XXX

Repeat count [5]: 1000
Datagram size [100]:
Timeout in seconds [2]:
Extended commands [n]:
Sweep range of sizes [n]:
Type escape sequence to abort.
Sending 1000, 100-byte ICMP Echos to XXX, timeout is 2 seconds:
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!
Success rate is 100 percent (1000/1000), round-trip min/avg/max = 12/13/36 ms

show policy-map int ser 6/0/0:9

Serial6/0/0:9

  Service-policy output: t1-esp-qos-policy

    queue stats for all priority classes:
      queue size 0, queue limit 256
      packets output 595017, packet drops 0
      tail/random drops 0, no buffer drops 0, other drops 0

    Class-map: ef (match-all)
      595040 packets, 154453008 bytes
      5 minute offered rate 0 bps, drop rate 0 bps
      Match: access-group 100
      Priority: kbps 1024, burst bytes 25600, b/w exceed drops: 0

    Class-map: cs3 (match-all)
      0 packets, 0 bytes
      5 minute offered rate 0 bps, drop rate 0 bps
      Match: access-group 101
      queue size 0, queue limit 32
      packets output 10586, packet drops 0
      tail/random drops 0, no buffer drops 0, other drops 0
      Bandwidth: kbps 128, weight 8.33

    Class-map: class-default (match-any)
      0 packets, 0 bytes
      5 minute offered rate 0 bps, drop rate 0 bps
      Match: any
      queue size 0, queue limit 96
      packets output 6949624, packet drops 567
      tail/random drops 567, no buffer drops 0, other drops 0
      Fair-queue: per-flow queue limit 24

Hello Dan,

the problem is clear.

try to do the following:

int ser 6/0/0:9

no service out t1-esp-qos-policy

service out t1-esp-qos-policy

I've found some QoS related bugs that suggest this action as an attempt to have CEF to rebuild the correct tables

Hope to help

Giuseppe

Thats actually one if the first things I tried. I was thinking of downing the interface and then remove and re-add the policy.

I removed the policy and waited around 15 minutes this time, re-created and now:

Class-map: class-default (match-any)
      169084 packets, 167193699 bytes
      5 minute offered rate 558000 bps, drop rate 0 bps
      Match: any
      queue size 1, queue limit 96
      packets output 169470, packet drops 2
      tail/random drops 2, no buffer drops 0, other drops 0
      Fair-queue: per-flow queue limit 24

Will test call quality and hopefully it is set.

Thanks!

Hello Dan,

good news waiting some minutes was the key to give CEF time to remove the old wrong object so new object was created successfully.

Hope to help

Giuseppe

Aparently it has stoped working again. As none of the statisticas are incrementing.

Hello Dan,

I apologize for my late answer.

I would consider to try with another IOS image this looks like a systematic problem.

I guess you have already done this.

Hope to help

Giuseppe

Getting Started

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: