cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
932
Views
0
Helpful
3
Replies

QoS Configuration for Voice in ME3400EG

GQM
Level 1
Level 1

Hi All

I am looking some help in configuring the QoS on my Hub locations which have Cisco ME3400-EG switches. The WAN link is 1.2Gig Microwave link. Right now there is no QoS on the ME3400 and we are having issues with voice quality. our vendor has configured QoS on the Microwave with the following settings:

LAN QoS Settings (on vendor side which is microwave)

Voice Media             DSCP (46)                L2 Priority (6)

Voice Signalling        DSCP (26)                L2 Priority (3)

 

Now i have to configure my end (ME3400) to prioritize the voice traffic. Looking for the sample QoS configuration for this. Any help in this regard will be highly appreciated. 

Thank you. 

3 Replies 3

Francesco Molino
VIP Alumni
VIP Alumni

Hi

 

You have different ways in configuring qos. 

First i would suggest to take a look on documentation guide to know what are limitations on your platform with your ios version.

Then you can take a look on this Cisco live doc: https://clnv.s3.amazonaws.com/2017/usa/pdf/BRKCRS-2501.pdf

 

The goal is to classify your traffic and then apply some priority bandwidth, tagging...

 

Does your ip phones are tagging all packets sent out? If you don't know, you can span a phone port and see if you have any dscp value. In the doc you'll see how to configure a class-map matching a tag and calling it in a policy map.

 

If they're not making the packet, you'll need some acls to catch this specific traffic and mark it with the tag ef for voice or cs3 for signaling. But for that you'll need to know the traffic flow. Voice communication is based on rtp udp ports for communication and these are udp 16384 to 32767 and you can specify the source and destination if you know them. The simplest one as example, because i don't know your voice subnets would be:

access-list 100 permit udp any any range 16384 32767

 

Again this is an example to show how to build an acl to be able to classify the voice traffic using acl.

 

The the class-map and policy map would look like:

Class-map match-all VOICE

 match ip dscp ef --> if packets are tagged already.

 match access-group 100 --> if your acl is 100

 

Then the policy-map

Policy-map VOICE

 class VOICE

   Set ip dscp ef --> if packet isn't marked

   priority

    police xxxx --> Max bandwidth in kbps

 

For cs3 traffic it could in the same class-map and then you'll user match-any instead of match-all or it can be a dedicated one. If dedicated one, you can't assign it into the priority queue because only one can be there in the policy-map unless the version allows you to have different level of priority (1 and 2).

 

Take a look on the doc I've referred to and you'll understand + don't forget the configuration guide of your specific release for limitations.


Thanks
Francesco
PS: Please don't forget to rate and select as validated answer if this answered your question

Hi Francesco 

Thank you for your kind reply. i am going through the documentation and will get back if i have any query. I really appreciate your help.

 

Thanks again.

 

Best Regards

 

Ok no pb.

Thanks
Francesco
PS: Please don't forget to rate and select as validated answer if this answered your question