Configuring BGP Connectivity
You can configure your Traffic IP Groups to use RHI as the IP distribution mode, using BGP as the routing protocol. To enable RHI for your Traffic IP addresses, use the Services > Traffic IP Groups page.
The settings in System > Fault Tolerance > BGP Route Health Injection control BGP connectivity between Virtual Traffic Managers in your cluster and the neighboring routing infrastructure:
- bgp!enabled: Whether BGP Route Health Injection is enabled.
- bgp!as_number: The Autonomous System (AS) in which your Virtual Traffic Manager cluster operates. Use a decimal value to represent the AS. This number applies to all Virtual Traffic Managers in your cluster (or each Configuration Location within a Multi-Site Manager cluster; see Multi-Site Cluster Management).
Configuring BGP Router IDs
By default, each Virtual Traffic Manager in the cluster joins the BGP routing domain using the address and subnet by which it is connected to its IPv4 default gateway. To override this behavior, you can set an alternate ID for each cluster member individually using flipper!bgp_router_id.
Managing BGP Neighbors
Unlike with OSPFv2, the Virtual Traffic Manager is unable to automatically discover its BGP neighbors. To enable your cluster to establish BGP sessions, you must first define the neighboring routers and then create applicable mappings between each cluster member and the desired neighbor.
To manage the BGP neighbors you can use in this mapping, click Manage BGP Neighbors. Use this page to define a new neighboring BGP enabled router, and to modify the parameters for existing neighbor definitions.
For each new router, input the following BGP connection parameters:
Field | Description |
|---|---|
BGP Neighbor name | An identifying name |
IP Address | The IPv4 address of the neighbor |
AS Number | The AS number for this neighbor. This might be the same AS as your proposed corresponding Virtual Traffic Manager (in which case the session uses iBGP), or it might be a different AS altogether (in which case the session uses eBGP). |
Keepalive | The interval, in seconds, between health message transmission. |
Holdtime | The minimum interval, in seconds, that must elapse after the last received keepalive before this neighbor assumes the mapped Virtual Traffic Manager is no longer active. |
Advertisement Interval | The interval, in seconds, that must elapse between advertisements made to this neighbor. |
Authentication Password | (optional) The per-BGP session password that your mapped Virtual Traffic Manager must use with this neighbor. Linux kernel versions earlier than 2.6.20, or any kernel built without the relevant option (CONFIG_TCP_MD5SIG) enabled, might not support this setting. For further clarification, see your system administrator or support provider. |
After you have defined your BGP neighbors, click back to the System > Fault Tolerance page to create the mappings to your individual Virtual Traffic Managers.
Before you can use BGP, you must configure the neighboring BGP routers to establish sessions with the selected Virtual Traffic Managers. This effectively corresponds with the neighbor definitions you have just created at the Virtual Traffic Manager end of the connection.
Click Update to apply your mappings and configuration. The settings on this page are replicated across all members of the Virtual Traffic Manager cluster.
The Virtual Traffic Manager reports all router peering and connection failures in the event log. The applet in the Admin UI and the Diagnose page both reflect the current status.