BGP
You can use the BGP template for all vEdge Cloud and vEdge router devices.
To configure the BGP routing protocol using vManage templates:
- Create a BGP feature template to configure BGP parameters, as described in this article. BGP can be used for service-side routing, to provide reachability to networks at the local site, and it can be used for transport-side routing, to enable communication between the vEdge router and other Viptela devices when the router is not directly connected to the WAN cloud. Create separate BGP templates for the two BGP routing types.
- Create a VPN feature template to configure VPN parameters for either service-side BGP routing (in any VPN other than VPN 0 or VPN 512) or transport-side BGP routing (in VPN 0). See the Configuration ► Templates ► VPN help topic.
- Create a device template that incorporates both the BGP and VPN feature templates. See the Configuration ► Templates help topic.
Navigate to the Template Screen
- In vManage NMS, select the Configuration ► Templates screen.
- From the Templates title bar, select Feature.
- Click Add Template.
- In the left pane, select one or more devices. The right pane displays the available templates for the selected devices.
- Select the BGP template.
The right pane displays the BGP template form.
- The top of the form contains fields for naming the template.
- The bottom contains fields for defining parameters applicable to that template.
- A drop-down menu to the left of each parameter field defines the scope of the parameter. When you first open a feature template form, for each parameter that has a default value, the scope is set to Default. To edit a parameter field, change the scope to Global or Device Specific. Note that if a parameter's scope is Device Specific, you cannot enter a value for it in the feature template. Instead, you enter a value when you attach the template to a device
- A plus sign (+) is displayed to the right when you can add multiple entries for the same parameter.
Minimum BGP Configuration
The following parameters are required (unless otherwise indicated) to configure BGP on a vEdge router:
| Step | Parameter Name | Description |
|---|---|---|
| 1. | Template Name | Enter a name for the template. It can be up to 128 characters and can contain only alphanumeric characters. |
| 2. | Description (Template) | Enter a description for the template. It can be up to 2048 characters and can contain only alphanumeric characters. |
| 3. | Shutdown | Ensure that No is selected, to enable BGP. |
| 4. | AS number | Enter the local AS number. |
| 5. | Router ID (optional) | Enter the BGP router ID, in decimal four-part dotten notation. |
| 6. | Internal Routes Distance (optional) | Enter a value to apply as the BGP route administrative distance for routes coming from one AS into another. |
| 7. | Local Routes Distance (optional) | Specify the BGP route administrative distance for routes within the local AS. By default, a route received locally from BGP is preferred over a route received from OMP. Range: 0 through 255 Default: 0 |
| 8. | External Routes Distance (optional) | Specify the BGP route administrative distance for routes learned from other sites in the overlay network. Range: 0 through 255 Default: 0 |
| 9. | Redistribute Protocols | Under the IPv4 Unicast Address Family tab, select the desired values. See following section. |
| 10. | Neighbors | Under the Neighbor tab, configure the address of the remote BGP peer and the remote AS number. |
| 11. | Save | Click Save to save the feature template. |
For service-side BGP, you might want to configure OMP to advertise to the vSmart controller any BGP routes that the vEdge router learns. By default, a vEdge router advertises to OMP both the connected routes on the vEdge router and the static routes that are configured on the vEdge router, but it does not advertise BGP external routes learned by the vEdge router. You configure this route advertisement in the OMP template for vEdge routers or vEdge software. See the OMP help topic.
For transport-side BGP, you must also configure a physical interface and a loopback interface in VPN 0. In addition, you should create a policy for BGP to advertise the loopback interface address to its neighbors, and apply the policy in the BGP instance or to a specific neighbor. See the Configuring Unicast Overlay Routing article for your software release.
CLI equivalent:
vpn vpn-id router bgp local-as-number distance external number internal number local number router-id ip-address [no] shutdown
Configure the IPv4 Unicast Address Family
To configure global BGP address family information, select the IPv4 Unicast Address Family tab. To display the fields in a section, click the arrow (>) on the right side of the title bar, and then click the plus sign (+).
| Section | Parameter Name | Description |
|---|---|---|
| Maximum paths | Specify the maximum number of parallel IBGP paths that can be installed into a route table to enable IBGP multipath load sharing. Range: 0 to 32 | |
| Address Family | Enter the BGP IPv4 unicast address family. Currently, only IPv4 is supported. | |
| Redistribute | ||
| Protocol | Select the protocols from which to redistribute routes into BGP, for all BGP sessions. Options are Connected, NAT, OMP, OSPF, and Static. At a minimum, select the following:
| |
| Route Policy | Enter the name of the route policy to apply to redistributed routes. | |
| Network | ||
| Network | Enter a network prefix, in the format of prefix/length, to be advertised by BGP. | |
| Aggregate Address | ||
| Prefix | Enter the prefix of the addresses to aggregate for all BGP sessions, in the format prefix/length. | |
| AS Set | Click On to generate set path information for the aggregated prefixes. | |
| Summary Only | Click On to to filter out more specific routes from BGP updates. | |
To add another entry, click the plus sign (+).
To remove an entry, click the trash icon to the right of the entry.
CLI equivalent:
vpn vpn-id router bgp local-as-number address-family ipv4-unicast aggregate-address prefix/length [as-set] [summary-only] maximum-paths paths number network prefix/length redistribute (connected | nat | omp | ospf | static)
Configure Neighbors
To configure a neighbor, select the Neighbor tab and click the plus sign (+):
| Parameter Name | Description |
|---|---|
| Address | Specify the IP address of the BGP neighbor. |
| Description | Enter a description of the BGP neighbor. |
| Remote AS | Enter the AS number of the remote BGP peer. |
| BGP Neighbor | Click No to set this IP address to not be a BGP neighbor. |
To configure parameters for the neighbor, click the More link. The Neighbor popup window opens:
| Parameter | Description |
|---|---|
| Address family | Select the appropriate address family. Currently, the Viptela software supports only the BGP IPv4 unicast address family. |
| Maximum Number of Prefixes | Specify the maximum number of prefixes that can be received from the neighbor. Range: 1 through 4294967295 If you specify a maximum number of prefixes, you can also configure:
You can specify either a restart interval or a warning only. |
| Route Policy | Specify policies to apply to prefixes received from or sent to the neighbor:
|
| Shutdown | Click On to enable the connection to the BGP neighbor. |
To configure advanced parameters for the neighbor, click Advanced Options in the Neighbor popup window:
| Parameter Name | Description |
|---|---|
| Next Hop Self | Click On to configure the router to be the next hop for routes advertised to the BGP neighbor. |
| Send Community | Click On to send the local router's BGP community attribute to the BGP neighbor. |
| Send Extended Community | Click On to send the local router's BGP extended community attribute to the BGP neighbor. |
| Negotiate Capability | Click On to allow the BGP session to learn about the BGP extensions that are supported by the neighbor. |
| Source Interface Address | Enter the IP address of a specific interface of the neighbor that BGP is to use for the TCP connection to the neighbor. |
| Source Interface Name | Enter the name of a specific interface of the neighbor that BGP is to use for the TCP connection to the neighbor, in the format geport/slot. |
| EBGP Multihop | Set the time to live (TTL) for BGP connections to external peers. Range: 0 to 255 Default: 1 |
| Password | Enter a password to use to generate an MD5 message digest. Configuring the password enables MD5 authentication on the TCP connection with the BGP peer. The password is case-sensitive and can be up to 25 characters long. It can contain any alphanumeric characters, including spaces. The first character cannot be a number. |
| Keepalive Time | Specify the frequency at which keepalive messages are advertised to a BGP peer. These messages indicate to the peer that the local router is still active and should be considered to be available. Specify the keepalive time for the neighbor, to override the global keepalive time. Range: 0 through 65535 seconds Default: 60 seconds (one-third the hold-time value) |
| Hold Time | Specify the interval after not receiving a keepalive message that the local BGP session considers its peer to be unavailable. The local router then terminates the BGP session to that peer. Specify the hold time for the neighbor, to override the global hold time. Range: 0 through 65535 seconds Default: 180 seconds (three times the keepalive timer) |
| Connection Retry Time | Specify the number of seconds between retries to establish a connection to a configured BGP neighbor peer that has gone down. Range: 0 through 65535 seconds Default: 30 seconds |
| Advertisement Interval | For the BGP neighbor, set the minimum route advertisement interval (MRAI) between when BGP routing update packets are sent to that neighbor. Range: 0 through 600 seconds Default: 5 seconds for IBGP route advertisements; 30 seconds for EBGP route advertisements |
CLI equivalent:
vpn vpn-id router bgp local-as-number neighbor ip-address address-family ipv4-unicast maximum-prefixes number [threshold] [restart minutes | warning-only] route-policy policy-name (in | out) capability-negotiate description string ebgp-multihop ttl next-hop-self password md5-digest-string remote-as remote-as-number send-community send-ext-community [no] shutdown timers advertisement-interval number connect-retry seconds holdtime seconds keepalive seconds update-source ip-address
Configure Advanced Parameters
To configure advanced parameters for BGP, click the Advanced tab:
| Parameter Name | Description |
|---|---|
| Hold Time | Specify the interval after not receiving a keepalive message that the local BGP session considers its peer to be unavailable. The local router then terminates the BGP session to that peer. This hold time is the global hold time. Range: 0 through 65535 seconds Default: 180 seconds (three times the keepalive timer) |
| Keepalive | Specify the frequency at which keepalive messages are advertised to a BGP peer. These messages indicate to the peer that the local router is still active and should be considered to be available. This keepalive time is the global keepalive time. Range: 0 through 65535 seconds Default: 60 seconds (one-third the hold-time value) |
| Compare MED | Click On to compare the router IDs among BGP paths to determine the active path. |
| Deterministic MED | Click On to compare MEDs from all routes received from the same AS regardless of when the route was received. |
| Missing MED as Worst | Click On to consider a path as the worst path if the path is missing a MED attribute. |
| Compare Router ID | Click On to always compare MEDs regardless of whether the peer ASs of the compared routes are the same. |
| Multipath Relax | Click On to have the BGP best-path process selct from routes in different in ASs. By default, when you are using BGP multipath, the BGP best path process selects from routes in the same AS to load-balance across multiple paths. |
CLI equivalent:
vpn vpn-id router bgp local-as-number best-path as-path multipath-relax compare-router-id med (always-compare | deterministic | missing-as-worst) timers holdtime seconds keepalive seconds
Release Information
Introduced in vManage NMS in Release 15.2.