Controlling the Advertisement of BGP Routes
Context
In most cases, a BGP routing table has a large number of routes. Transmitting them will intensify the load of a device. To address this problem, configure the device to advertise only desired routes. Multiple routes to the same destination may exist and traverse different ASs. Routes to be advertised need to be filtered in order to direct routes to specific ASs.
Procedure
- Configure a BGP device to advertise routes to all peers or peer groups.
You can configure a BGP device to filter routes to be advertised.
- Perform either of the following operations to configure the BGP device to advertise routes to all peers or peer groups:
- To filter routes based on an ACL, run the filter-policy { acl-number | acl-name acl-name } export [ protocol [ process-id ] ] or filter-policy { acl6-number | acl6-name acl6-name } export [ protocol [ process-id ] ] command.
- To filter routes based on an IP prefix list, run the filter-policy ip-prefix ip-prefix-name export [ protocol [ process-id ] ] or filter-policy ipv6-prefix ipv6-prefix-name export [ protocol [ process-id ] ] command.
If an ACL has been referenced in the filter-policy command but no VPN instance is specified in the ACL rule, BGP will filter routes including public and private network routes in all address families. If a VPN instance is specified in the ACL rule, only the data traffic from the VPN instance will be filtered, and no route of this VPN instance will be filtered.
When an ACL is configured in the filter-policy command to deny all packets, the default route will not be filtered out after the peer default-route-advertise command is configured to advertise the default route to BGP peers or peer groups.
- Perform either of the following operations to configure the BGP device to advertise routes to all peers or peer groups:
- Configure a BGP device to advertise routes to a specific peer or peer group.
- Perform any of the following operations to configure the BGP device to advertise routes to a specific peer or peer group:
To filter routes based on an ACL, run the peer { group-name | ipv4-address | ipv6-address } filter-policy { acl-number | acl-name acl-name | acl6-number | acl6-name acl6-name } export command.
To filter routes based on an IP prefix list, run the peer { ipv4-address | group-name } ip-prefix ip-prefix-name export or peer { group-name | ipv6-address } ipv6-prefix ipv6-prefix-name export command.
To filter routes based on an AS_Path filter, run the peer { ipv4-address | group-name | ipv6-address } as-path-filter { as-path-filter-number | as-path-filter-name } export command.
To filter routes based on a route-policy, run the peer { ipv4-address | group-name | ipv6-address } route-policy route-policy-name export command.
The routing policy applied in the peer route-policy export command does not support a specific interface as one matching rule. That is, the routing policy does not support the if-match interface command.
- Perform any of the following operations to configure the BGP device to advertise routes to a specific peer or peer group:
