Configuring Traffic Steering

This section describes how to configure traffic steering to recurse a route to an SR-MPLS TE Policy so that traffic can be forwarded through the path specified by the SR-MPLS TE Policy.

Usage Scenario

After an SR-MPLS TE Policy is configured, traffic needs to be steered into the policy for forwarding. This process is called traffic steering. Currently, SR-MPLS TE Policies can be used for various routes and services, such as BGP and static routes as well as BGP4+ 6PE, BGP L3VPN, and EVPN services. This section describes how to use tunnel policies to recurse services to SR-MPLS TE Policies.

Pre-configuration Tasks

Before configuring traffic steering, complete the following tasks:

  • Configure BGP routes, static routes, BGP4+ 6PE services, BGP L3VPN services, BGP L3VPNv6 services, and EVPN services correctly.

  • Configure an IP prefix list and a tunnel policy if you want to restrict the route to be recursed to the specified SR-MPLS TE Policy.

Procedure

  1. Configure a tunnel policy.

    Use either of the following procedures based on the traffic steering mode you select:

    • Color-based traffic steering

      1. Run system-view

        The system view is displayed.

      2. Run tunnel-policy policy-name

        A tunnel policy is created and the tunnel policy view is displayed.

      3. (Optional) Run description description-information

        Description information is configured for the tunnel policy.

      4. Run tunnel select-seq sr-te-policy load-balance-number load-balance-number unmix

        The tunnel selection sequence and number of tunnels for load balancing are configured.

      5. Run commit

        The configuration is committed.

      6. Run quit

        Exit the tunnel policy view.

    • DSCP-based traffic steering

      1. Run system-view

        The system view is displayed.

      2. Run segment-routing

        The Segment Routing view is displayed.

      3. Run sr-te-policy group group-value

        An SR-MPLS TE Policy group is created and the SR-MPLS TE Policy group view is displayed.

      4. Run endpoint ipv4-address

        An endpoint is configured for the SR-MPLS TE Policy group.

      5. Run color color-value match dscp { ipv4 | ipv6 } { { dscp-value1 [ to dscp-value2 ] } &<1-64> | default }

        A rule is configured for the mapping between the color values of SR-MPLS TE Policies in an SR-MPLS TE Policy group and the DSCP values of packets.

        Each SR-MPLS TE Policy in the SR-MPLS TE Policy group has its own color attribute. To configure mappings between the color values of SR-MPLS TE Policies in an SR-MPLS TE Policy group and the DSCP values of packets to be steered into the SR-MPLS TE Policies, run the color match dscp command. In this way, DSCP values, color values, and SR-MPLS TE Policies are associated within the SR-MPLS TE Policy group. Based on the association, packets with the specified DSCP values can be steered into SR-MPLS TE Policies with the mapping color values.

        When using the color match dscp command, pay attention to the following points:
        1. The mappings between color and DSCP values can be separately configured for an IPv4 or IPv6 address family. However, in the same IPv4 or IPv6 address family, each DSCP value can be associated with only one SR-MPLS TE Policy. Only up SR-MPLS TE Policies can be associated with DSCP values.

        2. You can run the color color-value match dscp { ipv4 | ipv6 } default command to specify a default SR-MPLS TE Policy for an IPv4 or IPv6 address family. When a DSCP value is not associated with any SR-MPLS TE Policy in an SR-MPLS TE Policy group, the packets with this DSCP value are forwarded through the default SR-MPLS TE Policy. In an SR-MPLS TE Policy group, only one default SR-MPLS TE Policy can be specified for an IPv4 or IPv6 address family.

        3. In scenarios where no default SR-MPLS TE Policy is specified for an IPv4 or IPv6 address family in an SR-MPLS TE Policy group:
          1. If the mappings between color and DSCP values are configured in the group and only some of the DSCP values are associated with SR-MPLS TE Policies, the packets with DSCP values that are not associated with SR-MPLS TE Policies are forwarded through the SR-MPLS TE Policy associated with the smallest DSCP value in the address family.
          2. If no DSCP value is associated with an SR-MPLS TE Policy in the group, for example, the mappings between color and DSCP values are not configured or DSCP values are not successfully associated with SR-MPLS TE Policies, the default SR-MPLS TE Policy specified for the other type of address family (IPv4 or IPv6) is used to forward packets. If no default SR-MPLS TE Policy is specified for the other type of address family, packets are forwarded through the SR-MPLS TE Policy associated with the smallest DSCP value in the local address family.
      6. Run quit

        Return to the SR view.

      7. Run quit

        Return to the system view.

      8. Run tunnel-policy policy-name

        A tunnel policy is created and the tunnel policy view is displayed.

      9. (Optional) Run description description-information

        Description information is configured for the tunnel policy.

      10. Run tunnel binding destination dest-ip-address sr-te-policy group sr-te-policy-group-id [ ignore-destination-check ] [ down-switch ]

        A tunnel binding policy is configured to bind the destination IP address and SR-MPLS TE Policy group.

        The ignore-destination-check keyword is used to disable the function to check the consistency between the destination IP address specified using destination dest-ip-address and the endpoint of the corresponding SR-MPLS TE Policy.

      11. Run commit

        The configuration is committed.

      12. Run quit

        Exit the tunnel policy view.

  2. Configure a route or service to be recursed to an SR-MPLS TE Policy.
    • Configure a non-labeled public BGP route to be recursed to an SR-MPLS TE Policy.

      For details about how to configure a non-labeled public BGP route, see Configuring Basic BGP Functions.

      1. Run route recursive-lookup tunnel [ ip-prefix ip-prefix-name ] [ tunnel-policy policy-name ]

        The function to recurse a non-labeled public BGP route to an SR-MPLS TE Policy is enabled.

      2. Run commit

        The configuration is committed.

    • Configure a static route to be recursed to an SR-MPLS TE Policy.

      For details about how to configure a static route, see Configuring IPv4 Static Routes.

      The color attribute cannot be added to static routes. Therefore, static routes support only DSCP-based traffic steering to SR-MPLS TE Policies, not color-based traffic steering to SR-MPLS TE Policies.

      1. Run ip route-static recursive-lookup tunnel [ ip-prefix ip-prefix-name ] [ tunnel-policy policy-name ]

        The function to recurse a static route to an SR-MPLS TE Policy for MPLS forwarding is enabled.

      2. Run commit

        The configuration is committed.

    • Configure a BGP L3VPN service to be recursed to an SR-MPLS TE Policy.

      For details about how to configure a BGP L3VPN service, see Configuring a Basic BGP/MPLS IP VPN.

      1. Run ip vpn-instance vpn-instance-name

        The VPN instance view is displayed.

      2. Run ipv4-family

        The VPN instance IPv4 address family view is displayed.

      3. Run tnl-policy policy-name

        A tunnel policy is applied to the VPN instance IPv4 address family.

      4. (Optional) Run default-color color-value

        The default color used by the L3VPN service to recurse to the SR-MPLS TE Policy tunnel is set.

        After a remote VPN route is leaked to the local VPN instance, if the route does not carry the color extended community attribute, the default color value is used for SR-MPLS TE Policy recursion.

      5. Run commit

        The configuration is committed.

    • Configure a BGP L3VPNv6 service to be recursed to an SR-MPLS TE Policy.

      For details about how to configure a BGP L3VPNv6 service, see Configuring a Basic BGP/MPLS IPv6 VPN.

      1. Run ip vpn-instance vpn-instance-name

        The VPN instance view is displayed.

      2. Run ipv6-family

        The VPN instance IPv6 address family view is displayed.

      3. Run tnl-policy policy-name

        A tunnel policy is applied to the VPN instance IPv6 address family.

      4. (Optional) Run default-color color-value

        The default color used by the L3VPNv6 service to recurse to the SR-MPLS TE Policy tunnel is set.

        After a remote VPN route is leaked to the local VPN instance, if the route does not carry the color extended community attribute, the default color value is used for SR-MPLS TE Policy recursion

      5. Run commit

        The configuration is committed.

    • Configure a BGP4+ 6PE service to be recursed to an SR-MPLS TE Policy.

      For details about how to configure a BGP4+ 6PE service, see Configuring BGP4+ 6PE.

      1. Run bgp { as-number-plain | as-number-dot }

        The BGP view is displayed.

      2. Run ipv6-family unicast

        The BGP IPv6 unicast address family view is displayed.

      3. Run peer ipv4-address enable

        A 6PE peer is enabled.

      4. Run peer ipv4-address tnl-policy tnl-policy-name

        A tunnel policy is applied to the 6PE peer.

      5. Run commit

        The configuration is committed.

    • Configure an EVPN service to be recursed to an SR-MPLS TE Policy.

      For details about how to configure an EVPN service, see Configuring BD-EVPN Functions.

      To apply a tunnel policy to an EVPN L3VPN instance, perform the following steps:
      1. Run ip vpn-instance vpn-instance-name

        The VPN instance view is displayed.

      2. Run ipv4-family or ipv6-family

        The VPN instance IPv4/IPv6 address family view is displayed.

      3. Run tnl-policy policy-name evpn

        A tunnel policy is applied to the EVPN L3VPN instance.

      4. (Optional) Run default-color color-value evpn

        The default color used by the EVPN L3VPN service to recurse to the SR-MPLS TE Policy tunnel is set.

        After a remote EVPN route is leaked to the local VPN instance, if the route does not carry the color extended community attribute, the default color value is used for SR-MPLS TE Policy recursion.

      5. Run commit

        The configuration is committed.

      To apply a tunnel policy to a BD EVPN instance, perform the following steps:
      1. Run evpn vpn-instance vpn-instance-name bd-mode

        The BD EVPN instance view is displayed.

      2. Run tnl-policy policy-name

        A tunnel policy is applied to the BD EVPN instance.

      3. (Optional) Run default-color color-value

        The default color used by the EVPN service to recurse to the SR-MPLS TE Policy tunnel is set.

        After a remote EVPN route is leaked to the local EVPN instance, if the route does not carry the color extended community attribute, the default color value is used for SR-MPLS TE Policy recursion.

      4. Run commit

        The configuration is committed.

      To apply a tunnel policy to an EVPN instance that works in EVPN VPWS mode, perform the following steps:
      1. Run evpn vpn-instance vpn-instance-name vpws

        The view of the EVPN instance that works in EVPN VPWS mode is displayed.

      2. Run tnl-policy policy-name

        A tunnel policy is applied to the EVPN instance that works in EVPN VPWS mode.

      3. (Optional) Run default-color color-value

        The default color used by the EVPN service to recurse to the SR-MPLS TE Policy is set.

        After a remote EVPN route is leaked to the local EVPN instance, if the route does not carry the color extended community attribute, the default color value is used for SR-MPLS TE Policy recursion.

      4. Run commit

        The configuration is committed.

      To apply a tunnel policy to a basic EVPN instance, perform the following steps:
      1. Run evpn vpn-instance vpn-instance-name

        The EVPN instance view is displayed.

      2. Run tnl-policy policy-name

        A tunnel policy is applied to the basic EVPN instance.

      3. Run commit

        The configuration is committed.

Copyright © Huawei Technologies Co., Ltd.
Copyright © Huawei Technologies Co., Ltd.
< Previous topic Next topic >