Per-VLAN Policy Map QoS and mls qos trust dscp--Catalyst 2960XRs
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
08-11-2017 02:19 PM - edited 03-08-2019 11:42 AM
I am working on a switch configuration and it would appear that the previous engineer made a mistake on the opposite sides of the trunk ports for the core and access switches' Etherchannel member ports, or else I am simply not familiar with QoS sufficiently enough to properly interpret the old configuration. Would someone please be able to clarify these examples:
There is a policy-map defined for QoS and applied at the VLAN level.
Etherchannel is currently setup between switches and will be on the new ones, as well.
Core side of trunk on switchports:
mls qos vlan-based
Access switch side of trunk on switchports:
mls qos trust dscp
Do both sides need to be defined with both for this to function properly? Is it even necessary to explicitly trust dscp on an Etherchannel member port or regular trunk port?
I would be greatly appreciative of any corrections or improvements on this configuration.
Here is an approximation of the old config:
Core
interface Port-channel1
switchport trunk encapsulation dot1q
switchport mode trunk
interface Gig1/0/1
description *** Etherchannel Member to Stack 1 ***
switchport trunk encapsulation dot1q
switchport mode trunk
srr-queue bandwidth share 1 60 10 29
srr-queue bandwidth shape 16 0 0 0
priority-queue out
mls qos vlan-based
channel-group 1 mode active
Access
interface Port-channel1
switchport trunk encapsulation dot1q
switchport mode trunk
interface GigabitEthernet1/0/1
description *** Etherchannel Member to Core ***
switchport trunk encapsulation dot1q
switchport mode trunk
srr-queue bandwidth share 1 60 10 29
srr-queue bandwidth shape 16 0 0 0
priority-queue out
mls qos trust dscp
channel-group 1 mode active
And the new:
Core
interface Port-channel1
switchport mode trunk
interface TenGigabitEthernet1/0/1
description *** Etherchannel Member to Stack 1 ***
switchport mode trunk
srr-queue bandwidth share 1 60 10 29
srr-queue bandwidth shape 16 0 0 0
priority-queue out
mls qos trust dscp
mls qos vlan-based
channel-group 1 mode active
Access
interface Port-channel1
switchport mode trunk
interface TenGigabitEthernet1/0/1
description *** Etherchannel Member to Core ***
switchport mode trunk
srr-queue bandwidth share 1 60 10 29
srr-queue bandwidth shape 16 0 0 0
priority-queue out
mls qos trust dscp
mls qos vlan-based
channel-group 1 mode active
- Labels:
-
LAN Switching
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
08-11-2017 03:56 PM
Hello
My understanding ...:
srr-queue bandwidth share 1 60 10 29 - Share the bw on % of the value specified
srr-queue bandwidth shape 16 0 0 0 - Overrides share mode, so in this case queue 1 in share mode is render ineffective , but the remaining 3 queues will be shared based on their value %
So the above reads
Shape 16 queue 1 = 1/16th of the interface ( 100mb link that's 6mb)
Shared queue 2/3/4 = 74mb remaining ratio: 6:1:3 ( if my maths are correct)
priority-queue out - This is a priority queue and then renders both the shape and shared values ineffective plus it will eat up as much BW as it desires and will stave the other queues
mls qos trust dscp < trust dscp traffic
mls qos vlan-based < - Enables vlan QOS on the trunks carrying the vlans with service policy's applied to the SVi of that vlan
res
Paul
Please rate and mark as an accepted solution if you have found any of the information provided useful.
This then could assist others on these forums to find a valuable answer and broadens the community’s global network.
Kind Regards
Paul
