QinQ Configuration example

This configuration example uses four HP switches to establish a QinQ tunnel through the provider network.

QinQ configuration example

QinQ configuration example

The design parameters are as follows:

  • The provider edge bridge and the provider core bridge are configured in svlan mode.

  • Each customer is associated with a single S-VLAN connecting two separate sites: customer A's VLANs (C-VLANs 1-10) are associated with S-VLAN 100; and customer B's VLANs (C-VLANs 1-20) are associated with S-VLAN 200.


[NOTE: ]

NOTE:

  • The VLANs of customers A and B can overlap: this will not result in intermixing of customer frames in the provider cloud because the S-VLANs associated with each customer are different.

  • Core devices are not mandatory to establish a QinQ tunnel. For example, two edge-bridges can be connected directly to create a provider bridge network.

  • The relationship between S-VLANs and C-VIDs is typically one to many. An alternative configuration might associate a single customer's C-VIDs with more than one S-VLAN. Such a configuration would most likely be used to tunnel distinct C-VIDs through various S-VLANs, but seldom be used to send the same C-VID through multiple S-VLANs.


Configuration example: Edge Switch 1

Configuration example: Edge Switch 1

At the end of the configuration, the following settings will apply:

  • All customer A site traffic received on port A1 will be associated with S-VLAN 100. This is independent of the C-VLAN tag information that the customer frames may carry.

  • All customer B Site 1 traffic will be associated with S-VLAN 200 and be switched out to the core (uplinks A3, A4) with the S-VLAN tag-id of 200.

  • The frame size will increase by 4 since ports A3 and A4 are tagged members of S-VLAN 100 and 200.

To configure the switch, follow these steps:

  1. Enable QinQ:

    Edge l(config)#: qinq svlan tag-type 88a8
    
  2. Reboot the box with the configuration saved to transfer into svlan bridge mode.


    [NOTE: ]

    NOTE: A reboot is required for the QinQ enable command to take effect.


  3. Configure S-VLANs and ports connected to the customer network.

    Edge1(config)#: svlan 100
    Edge1(svlan-100)#: untagged A1
    Edge1(svlan-100)#: exit
    Edge1(config)#: int A1 qinq port-type customer-network
    Edge1(config)#: svlan 200
    Edge1(svlan-200)#: untagged A2
    Edge1(svlan-200)#: exit
    Edge1(config)#: int A2 qinq port-type customer-network
    

    [NOTE: ]

    NOTE: Customer A is assigned S-VLAN 100 and customer B is assigned S-VLAN 200. However, the same customer can be associated with more than one SVLAN. Also, interfaces A1 and A2 are configured as customer network ports because they are linked to customer bridges.


  4. Configure the provider ports leading to the core of the provider network.

    Edge1(config)#: svlan 100 tagged A3, A4
    Edge1(config)#: svlan 200 tagged A3, A4
    Edge1(config)#: interface A3,A4 qinq port-type provider-network
    

    [NOTE: ]

    NOTE: As recommended by IEEE 802.1ad specification, uplink ports should generally be configured as tagged ports for S-VLANs that are used to carry customer traffic. However, this is not a mandatory requirement on HP switches—S-VLANs that are used for internal provider network use (not carrying customer traffic but for management of the provider network devices) can have untagged port memberships.


QinQ Configuration example: provider Edge 2 switch

The configuration details for the Edge 2 switch mirrors the configuration for the Edge 1 switch. All customer traffic received on port A1 from customer A's site 2 will be associated with S-VLAN 100. Similarly, all customer B's site 2 traffic will be associated with S-VLAN 200.

To configure the switch, follow these steps:

  1. Enable QinQ:

    Edge 2(config)#: qinq svlan tag-type 88a8
    
  2. Reboot the box with the configuration saved to transfer into S-VLAN bridge mode.

  3. Configure S-VLANs and customer ports connected to the customer network.

    Edge2(config)#: svlan 100
    Edge2(svlan-100)#: untagged A1
    Edge2(svlan-100)#: exit
    Edge2(config)#: int A1 qinq port-type customer-network
    Edge2(config)#: svlan 200
    Edge2(svlan-200)#: untagged A2
    Edge2(svlan-200)#: exit
    Edge2(config)#: int A2 qinq port-type customer-network
    
  4. Configure the provider ports leading to the core of the provider network.

    Edge1(config)#: svlan 100 tagged A3, A4
    Edge1(config)#: svlan 200 tagged A3, A4
    Edge1(config)#: interface A3,A4 qinq port-type provider-network
    

Configuring example: provider core 1 switch

Configuration example: Core 1 Switch

Configuration example: Core 1 Switch

To configure the Core 1 switch:

  1. Enable QinQ:

    Core l(config)#: qinq svlan tag-type 88a8
    
  2. Reboot the box with the configuration saved to transfer into svlan bridge mode.

  3. Configure S-VLANs and port assignments.

    Core 1(config)#: svlan 100
    Core 1(svlan-100)#: tagged A1, A2
    Core 1(svlan-100)#: exit
    Core 1(config)#: svlan 200
    Core 1(svlan-200)#: tagged A1, A2
    Core 1(svlan-200)#: exit
    Core 1(config)#: interface A1,A2 qinq port-type provider-network
    

[NOTE: ]

NOTE: The S-VLAN configuration for the core devices is based on what VLANs the edge devices (Edge 1 and 2) can send. Per the 802.1ad specification, all ports carrying customer traffic will be tagged on the VLAN that the port carries customer frames on.


To configure the Core 2 switch:

  1. Enable QinQ:

    Core 2(config)#: qinq svlan tag-type 88a8
    
  2. Reboot the box with the configuration saved to transfer into svlan bridge mode.

  3. Configure S-VLANs and port assignments.

    Core 2(config)#: svlan 100
    Core 2(svlan-100)#: tagged A1, A2
    Core 2(svlan-100)#: exit
    Core 2(config)#: svlan 200
    Core 2(svlan-100)#: tagged A1, A2
    Core 2(svlan-100)#: exit
    Core 2(config)#: interface A1,A2 qinq port-type provider-network
    

Verifing the configuration

After the edge and core switch configurations are completed, QinQ operations can begin. To verify operations, it should be possible to assign IP-addresses to customer A or B devices in site 1 and site 2 and ping them. If everything has been configured correctly, traffic will flow through the provider network cloud and reach the other site seamlessly.