Configure dual-stack protocol-based evolution so that VPWS accessing L3VPN over MPLS can evolve to VPWS accessing EVPN over SRv6 when IPv4 and IPv6 networks coexist.
Procedure
- Configure priority-based route selection on ASGs and RSGs so that the ASGs and RSGs preferentially select SRv6 EVPN routes.
- Run system-view
The system view is displayed.
- Run bgp as-number
The BGP view is displayed.
- Run l2vpn-family evpn
The BGP-EVPN address family view is displayed.
- Run peer peerIpv6Addr high-priority or peer peerGroupName high-priority
The EVPN routes learned from an IPv6 peer or peer group are granted the high priority for route selection.
After this step is performed, the device preferentially selects the SRv6 EVPN routes learned from the specified IPv6 peer or peer group among routes with the same prefix.
- Run commit
The configuration is committed.
- Configure route-policies on each RR to filter out VPNv4 or VPNv6 routes.
After completing the preceding step, configure the import and export route-policies on each RR to filter out received VPNv4 or VPNv6 routes and reflect only EVPN routes, reducing memory consumption.
- Run system-view
The system view is displayed.
- Run route-policy RR_IN permit node node
An import route-policy and a permit node are configured to add a specified community attribute (aa:nn) to VPNv4 or VPNv6 routes to be accepted by the RR.
- Run apply community aa:nn
A community attribute (aa:nn) is set for routes.
- Run quit
Return to the system view.
- Run ip community-filter cfIndex permit aa:nn
A community filter is created.
The community filter is used to match the VPNv4 or VPNv6 routes against the specified community attribute (aa:nn), which is further used to apply an export route-policy.
- Run route-policy RR_OUT deny node node1
An export route-policy and a deny node are configured to filter out the VPNv4 or VPNv6 routes with a specified community attribute to be advertised by the RR.
- Run if-match community-filter cfIndex
A rule for matching routes against the specified community filter is created.
- Run quit
Return to the system view.
- Run route-policy RR_OUT permit node node2
Another node is configured in the export route-policy to allow unmatched routes to be advertised.
- Run quit
Return to the system view.
- Run bgp as-number
The BGP view is displayed.
- Run ipv4-family vpnv4 or ipv6-family vpnv6
The BGP-VPNv4 or BGP-VPNv6 address family view is displayed.
- Run peer ipv4-address route-policy RR_IN import or peer group-name route-policy RR_IN import
The route-policy RR_IN is applied to the RR to filter VPNv4 or VPNv6 routes to be accepted from a specified peer or peer group.
- Run peer ipv4-address route-policy RR_OUT export or peer group-name route-policy RR_OUT export
The route-policy RR_OUT is applied to the RR to filter out VPNv4 or VPNv6 routes so that the routes are not advertised to a specified peer or peer group.
- Run commit
The configuration is committed.
- After the evolution is complete, delete BGP VPNv4 or VPNv6 peer relationships from the RRs, ASGs, and RSGs.
- Run system-view
The system view is displayed.
- Run bgp as-number
The BGP view is displayed.
- Run ipv4-family vpnv4 or ipv6-family vpnv6
The BGP-VPNv4 or BGP-VPNv6 address family view is displayed.
- Run undo peer ipv4-address enable or undo peer group-name enable
The specified VPNv4 or VPNv6 peer or peer group is deleted.
- Run commit
The configuration is committed.