2801 2x ADSL2+ WIC MLPPP QOS issue

Unanswered Question

I am having trouble with our 2801 QOS after changing over to a PPP multilink.

QoS setup is as follows.


policy-map IBC-QOS-ORIGINAL

class IBC-VOIP

priority percent 30

class IBC-GOLD

bandwidth percent 30

class class-default

fair-queue 4096

random-detect


Each ATM is as follows


interface ATM0/1/0

description FNN0264921537

bandwidth 850

no ip address

load-interval 30

no atm ilmi-keepalive

bundle-enable

dsl noise-margin 3

dsl bitswap both

pvc 8/35

vbr-nrt 850 850

tx-ring-limit 3

encapsulation aal5mux ppp dialer

dialer pool-member 1


Dialer is as follows


interface Dialer1

bandwidth 1850

ip address negotiated

ip mtu 1460

ip flow ingress

encapsulation ppp

load-interval 30

dialer pool 1

ppp authentication chap callin

ppp chap hostname xxxxxxx

ppp chap password 7 xxxxxxx

ppp multilink

ppp multilink interleave

ppp multilink fragment delay 20

ppp multilink queue depth qos 3

service-policy output IBC-QOS-ORIGINAL



I am running 12.4(24)T1


show ppp multilink displays


Virtual-Access4

Bundle name: SydneySGBP

Remote Username: SydneySGBP

Remote Endpoint Discriminator: [1] SydneySGBP

Local Username: [email protected]

Local Endpoint Discriminator: [1] [email protected]

Bundle up for 1d18h, total bandwidth 3400, load 5/255

Receive buffer limit 24384 bytes, frag timeout 1000 ms

Interleaving enabled

Dialer interface is Dialer1

0/0 fragments/bytes in reassembly list

933 lost fragments, 64095 reordered

0/0 discarded fragments/bytes, 0 lost received

0x5EA704 received sequence, 0x579C2E sent sequence

Member links: 2 (max not set, min not set)

Vi3, since 1d18h, 4250 weight, 1496 frag size

PPPoATM link, ATM PVC 8/35 on ATM0/3/0

Packets in ATM PVC Holdq: 0, Particles in ATM PVC Tx Ring: 0

Vi2, since 1d18h, 4250 weight, 1496 frag size

PPPoATM link, ATM PVC 8/35 on ATM0/1/0

Packets in ATM PVC Holdq: 0, Particles in ATM PVC Tx Ring: 0

No inactive multilink interfaces


The bundle bandwidth concerns me ? If I change the dialer bandwidth the QOS policies break with the percentages of bandwidth going all wrong.


I also get the virtual-access queueing displaying output drops on the Voice priority class when I am sure that there is not that many calls as we use g729.


Any help or advice anybody can give me is most appreciated.


BTW this is the show policy map on the dialer if it helps.


show policy-map interface dialer 1

Dialer1


Service-policy output: IBC-QOS-ORIGINAL


queue stats for all priority classes:

queue limit 64 packets

(queue depth/total drops/no-buffer drops) 0/0/0

(pkts output/bytes output) 0/0


Class-map: IBC-VOIP (match-any)

1106655 packets, 61662210 bytes

30 second offered rate 0 bps, drop rate 0 bps

Match: dscp ef (46)

1100759 packets, 61185446 bytes

30 second rate 0 bps

Match: ip precedence 5

0 packets, 0 bytes

30 second rate 0 bps

Match: access-group name nodephone

0 packets, 0 bytes

30 second rate 0 bps

Match: access-group name Asterisk

5896 packets, 476764 bytes

30 second rate 0 bps

Priority: 30% (555 kbps), burst bytes 13850, b/w exceed drops: 0


Class-map: IBC-GOLD (match-any)

476057 packets, 86197097 bytes

30 second offered rate 2000 bps, drop rate 0 bps

Match: protocol citrix

113921 packets, 20388798 bytes

30 second rate 0 bps

Match: access-group name CitrixRDP

362135 packets, 65807735 bytes

30 second rate 2000 bps

Match: dscp af11 (10)

1 packets, 564 bytes

30 second rate 0 bps

Match: ip precedence 3

0 packets, 0 bytes

30 second rate 0 bps

Match: access-group name ibc-management

0 packets, 0 bytes

30 second rate 0 bps

Queueing

queue limit 64 packets

(queue depth/total drops/no-buffer

  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 0 (0 ratings)
Loading.
Joseph W. Doherty Thu, 07/09/2009 - 04:29
User Badges:
  • Super Bronze, 10000 points or more

%&^%^&&%$#^%& - don't you just hate a long (unsaved) response post failure?


Sorry, can't recall all I had just written, but a condensed version would be:


There might be issues using this "T" variant, either not working correctly or QoS features implementation difference starting with 12.4(20)T. Slight modification to your policy and/or fallback to an older IOS (e.g. 12.4(15)T9 or 12.4) might help. More information on newer HQF QoS:


http://www.cisco.com/en/US/prod/collateral/iosswrel/ps6537/ps6558/white_paper_c11-481499.html

and

http://www.cisco.com/en/US/docs/ios/qos/configuration/guide/qos_frhqf_support_ps6441_TSD_Products_Configuration_Guide_Chapter.html#wp1052220

Actions

This Discussion