Usage Scenario

By default, BGP advertises all preferred routes in the BGP routing table to neighbors. After the active-route-advertise is configured, only the routes preferred by BGP and also active at the routing management layer are advertised to neighbors.

Precautions

The active-route-advertise command and the routing-table rib-only command are mutually exclusive.

Usage Scenario

BGP route aggregation is classified into manual aggregation and automatic aggregation. The aggregate command can be used to configure manual aggregation. This command can aggregate routes in the local BGP routing table. Manually-aggregated routes take precedence over automatically-aggregated routes.

If the specific member routes of an aggregated route have different Origin attributes, the Origin attribute of the aggregated route uses the value in the order of incomplete, egp, and igp. The aggregated route will carry the community attribute of every specific route.

The parameter as-set is used to create an aggregated route whose AS_Path attribute contains AS_Path information of specific routes. Exercise caution when using this parameter if many AS_Path attributes need to be aggregated because frequent changes in routes may cause route flapping.

The parameter detail-suppressed is used to suppress the advertisement of specific routes. After detail-suppressed is configured, only aggregated routes are advertised. Aggregated routes carry the atomic-aggregate attribute, not the community attributes of specific routes.

The parameter suppress-policy is used to suppress the advertisement of specified routes. The if-match clause of route-policy can be used to filter the routes to be suppressed. This means that only the routes matching the policy will be suppressed, and the other routes will still be advertised. The peer route-policy command can also be used to filter out the routes not to be advertised to peers.

After the parameter origin-policy is used, only the routes matching route-policy are aggregated.

The parameter attribute-policy is used to set attributes for an aggregated route. If the AS_Path attribute is set in the policy using the apply as-path command and as-set is set in the aggregate command, the AS_Path attribute in the policy does not take effect. The peer route-policy command can also be used to set attributes for an aggregated route.

Precautions

If the undo aggregate command is run, the system will match routes base on the configured parameter attribute-policy, origin-policy, suppress-policy, as-set, or detail-suppressed. If the parameter is not configured, the system will fail to execute the undo aggregate command.

Usage Scenario

To save label resources on a 6PE device and avoid the problem that IPv6 routes cannot be advertised due to the shortage of labels, you can run the apply-label explicit-null command on the 6PE device.

Configuration Impact

After the apply-label explicit-null command is run on a 6PE device, all the IPv6 routes sent by the 6PE device to its peer share the explicit null label.

Precautions

If you run the apply-label explicit-null or undo apply-label explicit-null command after a 6PE peer relationship is established, temporary packet loss occurs.

Usage Scenario

By default, display commands such as the display bgp peer command display 4-byte AS numbers in dotted notation.

Assume that a 4-byte AS number in dotted notation is x.y. Following is the conversion relationship between an integral 4-byte AS number and a 4-byte AS number in dotted notation:

Integral 4-byte AS number = x x 65536 + y

For example, if a 4-byte AS number in dotted notation is 2.3, the corresponding integral 4-byte AS number is 131075 (2 x 65536 + 3).

Precautions

Usage Scenario

This command is used to restrict the maximum number of AS numbers in the AS_Path attribute only when the AS_Path attribute is reconstructed or the aggregated routes are generated.

Precautions

After the as-path-limit command is configured, a router checks whether the number of AS numbers in the AS-Path attribute of the incoming route exceeds the maximum value. If the number of AS numbers exceeds the maximum value, the local router will discard the route. Therefore, if the maximum number of AS numbers in the AS-Path attribute is set too small, routes will be lost.

Usage Scenario

After the bestroute as-path-ignore command is used, BGP does not compare the AS path lengths of routes. By default, the route with the shortest AS path is preferred.

Precautions

After the bestroute as-path-ignore command is run, the AS_Path attribute is not used as one of the BGP route selection rules.

Usage Scenario

On a BGP network, the BGP device always receives multiple routes with the same prefix but to different paths from neighbors. BGP must select the optimal route to the prefix to guide packet forwarding. By default, BGP compares the next-hop IGP route metric values of these routes and selects the route with the smallest metric value as the optimal route.

To customize route selection policies, you can run the bestroute igp-metric-ignore command to configure BGP to ignore the metric value of the next-hop IGP route when selecting the optimal route.

By default, BGP compares only MED values of routes from the same AS, not including sub-ASs in a confederation. To enable BGP to compare MED values of routes in a confederation when selecting the optimal route, run the bestroute med-confederation command.

After the bestroute med-confederation command is configured, BGP compares MED values only when AS_Path does not contain the external AS (AS that is not in the confederation) number. Otherwise, BGP does not compare MED values.

For example, ASs 65000, 65001, 65002, and 65004 belong the same confederation. Routes to the same destination are listed as follows:

• path1: AS_Path=65000 65004, med=2

• path2: AS_Path=65001 65004, med=3

• path3: AS_Path=65002 65004, med=4

• path4: AS_Path=65003 65004, med=1

After the bestroute med-confederation command is run, the AS_Path attributes of paths 1, 2, and 3 does not contain the numbers of ASs that belong to other confederations. Therefore, when selecting routes based on MED values, BGP compares the MED values of paths 1, 2, and 3 only. This is because the AS_Path attribute of path 4 contains the number of an AS that does not belong to this confederation.

Usage Scenario

The bestroute med-none-as-maximum command takes effect in BGP route selection and is used only when no MED is in the route attributes. When the MED is lost, the MED uses the maximum value 4294967295 if the bestroute med-none-as-maximum command is used. Otherwise, the MED uses the default value 0, which may cause a failure to select the optimal route.

Usage Scenario

By default, labeled BGP IPv4 unicast routes that are iterated to IP addresses participate in route selection. In non-RR networking, or RR networking where the BGP next hop is changed, BGP needs to check whether there is an LSP destined for the next hop. If such an LSP is not found, BGP needs to establish an LSP. If the default setting is adopted, a labeled BGP IPv4 unicast route may be selected prior to LSP establishment. As a result, services are incorrectly switched before LSP establishment, packet loss may occur, which affects service stability. The bestroute nexthop-resolved tunnel command can be run to allow route selection only after the labeled BGP IPv4 route is iterated to an LSP.

By default, the BGP VPNv4 routes can participate in route selection if the next hop iterates to an IP address. If the current device is an ASBR, it is required that the next hop of the preferred route can iterate to the LSP tunnel; if the ASBR is configured with load balancing, it is also required that next hop of the routes which are not preferred can iterate to the LSP tunnel so that these routes can participate in load balancing. Now the BGP VPNv4 routes can be set to participate in route selection when next hop iterates to the LSP tunnel.

Precautions

The bestroute nexthop-resolved ip and bestroute nexthop-resolved tunnel commands are mutually exclusive.

On a BGP network, after a router receives multiple routes with the same prefix but different paths from different peers, the router needs to select an optimal route from these routes to forward packets. By default, Cluster_List takes precedence over Router ID during BGP route selection. To enable Router ID to take precedence over Cluster_List during BGP route selection, run the bestroute routerid-prior-clusterlist command.

Usage Scenario

BGP is an inter-AS dynamic routing protocol. BGP running within an AS is called Internal BGP (IBGP), and BGP running between ASs is called External BGP (EBGP).

Whether to apply BGP to routing information transmission between ASs depends on the following conditions:

• If at least one of the following conditions is met, BGP can be used:

  • Data packets are allowed to be transmitted from one AS to another AS.

  • Multiple connections to external ISPs and Internet exist in ASs.

  • Data flows entering or leaving ASs must be controlled.

• If one of the following conditions is met, BGP does not need to be used:

  • Users are connected to only one ISP network.

  • The ISP does not need to provide Internet access services for users.

  • ASs adopt default routes between each other.

Precautions

After the bgp command is run, BGP is enabled.

Each device runs in only one AS; therefore, each device can be specified with only one local AS number.

Exercise caution when using the undo bgp command because it will delete all BGP configurations on a device.

If the BGP AS number is changed, the route calculation result of the local or remote OSPF VPN process may be affected. Because the tag value of the LSAs of the OSPF VPN process is calculated based on the BGP AS number, the local tag value of the OSPF VPN process and the tag value carried in the LSA advertised by the OSPF VPN process change. After receiving an LSA, the local or remote OSPF process checks the tag of the LSA. If the local tag of the OSPF process is the same as the tag of the received LSA, the LSA is not used for OSPF route calculation.

Usage Scenario

After the dynamic BGP peer function is enabled using the peer listen-net command, BGP listens to BGP connection requests from the network segment specified in the command and establish BGP peer relationships dynamically. If a large number of dynamic BGP peer sessions are established on the network segment, excessive system resources will be consumed. To prevent this problem, run the bgp dynamic-session-limit command to configure a maximum number for dynamic BGP peer sessions as required.

Precautions

This command does not apply to static BGP peer sessions.

Usage Scenario

By default, BGP checks the first AS number in the AS_Path list that is carried in the Update message sent by the EBGP peer. If only the first AS number indicates the AS where the EBGP peer locates, the Update message is accepted. Otherwise, the Update message is denied, and the EBGP connection goes Down.

Precautions

The check-first-as command is not listed in the configuration file.

After the undo check-first-as command is configured, loops have a greater chance to occur. Therefore, use the command with caution.

Follow-up Procedure

After the configuration is complete, run the refresh bgp command if you want to check the received routes again.

Usage Scenario

The command is used to control the MEDs to change the policy of BGP route selection. If the compare-different-as-med command is run, BGP will compare the MEDs of the routes from different ASs. If there are multiple reachable paths to the same destination, BGP prefers the route with the smallest MED.

Precautions

Do not use this command unless different ASs use the same IGP and route selection mode.

Usage Scenario

A large AS may contain a huge number of fully meshed IBGP peer relationships. In this situation, configure a confederation.

Divide the AS into multiple small sub-ASs and group them into a confederation. Then establish EBGP connections between sub-ASs and establish fully meshed IBGP connections within the same sub-AS. Some key attributes of routes, such as the next hop, MED, and local preference are not discarded when these routes pass through sub-ASs. This can reduce the number of fully meshed IBGP peer relationships in an AS and keep the integrity of the original AS.

Precautions

The confederation ID is equal to the AS number. An external AS must specify the confederation ID when specifying the AS number of the peer. All the sub-ASs in the same confederation must be configured with the same confederation ID that must be different from the number of any sub-AS.

Usage Scenario

To make nonstandard devices interwork, configure this command on all routers in a confederation.

Precautions

After the confederation ID is configured, running the confederation nonstandard command will disconnect the sessions between a router and its IBGP peers as well as its confederation EBGP peers. Then, new connections are re-established.

Usage Scenario

A large AS may contain a huge number of fully meshed IBGP peer relationships. In this situation, configure a confederation.

Prerequisites

The confederation id command specifies the confederation ID of each sub-AS. If the confederation ID is not configured, this configuration is invalid.

Precautions

The sub-ASs configured in this command belong to the same confederation, and each sub-AS uses fully meshed network.

If two PEs that are not RRs reside in two confederations, the PEs can advertise labeled routes carrying a label of all Fs to each other by default. However, non-Huawei devices may consider the label of all Fs invalid, and if these non-Huawei devices receive such labeled routes, BGP peer relationships may be disconnected. To prevent this problem, run the confederation route unicast-to-label disable command to prevent the PEs from advertising labeled routes carrying a label of all Fs.

Usage Scenario

If the suppression policy for route dampening is configured, routes preferentially match suppression parameters defined by route-policy in the command.

If no optional parameter is configured in the command, the default value of each parameter is used. The four parameters of the command are mutually dependent. If you configure one of the parameters, the other parameters also need to be configured in the command.

BGP route flap dampening measures the stability of a route by using a penalty value. The greater the penalty value, the less stable the route. Each time route flapping occurs, that is, when the router receives a Withdraw packet or an Update packet for a route, BGP adds a certain penalty value (1000) for the route.

When the penalty value of the route exceeds the suppression threshold, the route is suppressed. The router does not add the route to the IP routing table or advertise any Update packet to other BGP peers. BGP strips off the best flag of the route. If the route is marked with the d flag, the last packet received by the router is an Update packet; if the route is marked with the h flag, the last packet received by the router is a Withdraw packet. After the penalty value reaches a certain limit, it does not increase any more. The limit is called the penalty ceiling.

After the route is suppressed for a certain period, the penalty value is reduced by half. If the penalty value of a route marked with the d flag decreases to the reuse threshold, the d flag is removed, and the route becomes available and is selected preferentially. After that, the route can be added to the IP routing table and used to send Update packets to other BGP peers. If the penalty value of a route marked with the h flag decreases to 0, the route is deleted from the BGP routing table.

Configuration Impact

If the dampening command is run more than once, the latest configuration overrides the previous one.

After the dampening command is run, the system suppresses unstable routes. This means that the system does not add unstable routes to the BGP routing table or advertise them to other BGP peers.

Precautions

Note the following items when configuring BGP route flap dampening:

• The value of suppress must be greater than that of reuse and smaller than that of ceiling.

• If MaxSuppressTime obtained by using the formula of MaxSuppressTime = half-life-reach x 60 x (ln(ceiling/reuse)/ln(2)) is smaller than 1, suppression cannot be performed. You need to ensure that the value of MaxSuppressTime is equal to or greater than 1. This means that the value of ceiling/reuse must be great enough.

Creating a route-policy before it is referenced is recommended. By default, nonexistent route-policies cannot be referenced using the command. If the route-policy nonexistent-config-check disable command is run in the system view and a nonexistent route-policy is referenced using the current command, the configured dampening parameters apply to all routes; if no dampening parameters are configured, the default dampening parameters apply to the routes.

Usage Scenario

If the IPv4 unicast address family needs to be enabled by default for created BGP peers, the default ipv4-unicast command can be used to configure the default address family of BGP to the IPv4 unicast address family.

If the IPv4 unicast address family does not need to be enabled by default for created BGP peers, the undo default ipv4-unicast command can be used to disable the IPv4 unicast address family for all peers.

Precautions

After the undo default ipv4-unicast command is run, the peer enable command needs to be run if the created BGP peer needs to be enabled with the IPv4 unicast address family.

Usage Scenario

The Local_Pref attribute is used to determine an optimal route for traffic before traffic leaves an AS. When the router that runs BGP has multiple routes to the same destination, the router selects the route with the highest local preference as the optimal route.

Precautions

If the router is already configured with a default local preference for BGP routes, the configuration of a new default local preference will override the previous configuration and the new default local preference will overwrite the previous one.

The local preference is exchanged only between IBGP peers and is not advertised to other ASs.

Usage Scenario

The default med command is valid only for the routes imported by using the import-route (BGP) command and the routes summarized by BGP on the local device.

The default med command sets the default MED and controls route selection for the traffic entering an AS. If the other attributes of routes to the same destination are the same, BGP will select the route with the smallest MED as the optimal route.

Precautions

If a default MED is configured on a device, configuring a new default MED will override the previous configuration and the new default MED will overwrite the previous one.

The MED is transmitted between two neighboring ASs only. Devices in an AS do not advertise the received MED to peers in other ASs.

Usage Scenario

This configuration reduces the number of routes on the network and minimizes the routing table size. If the default-route imported command is run and a default route exists in the local IP routing table, the system imports the default route to the BGP routing table. After default routes are imported, the packets can be transmitted over the default route when no matched routing entry is found in the routing table, preventing packet loss.

Precautions

To import default routes, you need to run both the default-route imported command and the import-route (BGP) command. If only the import-route (BGP) command is used, default routes cannot be imported. In addition, the default-route imported command is used to import only the default routes that exist in the local routing table.

When a device needs to advertise default routes to a peer (group) and no default route exists in the local routing table, the peer default-route-advertise command needs to be used.

After the BGP deterministic-MED function is enabled, when an optimal route is to be selected from routes that are received from different ASs and carry the same prefix, routes are first grouped based on the leftmost AS number in the AS_Path attribute. Routes with the same leftmost AS number are grouped together, and after comparison, an optimal route is selected for the group. The group optimal route is then compared with optimal routes from other groups to determine the final optimal route. This mode of route selection ensures that the sequence in which routes are received is irrelevant to the route selection result.

If the BGP deterministic-MED function is disabled, routes are compared against each other according to the sequence in which they are received. In this manner, the sequence in which routes are received is relevant to the result of route selection.

If Route A1, Route A2, and Route B are received in turn, Route A1 and Route A2 are first compared. The leftmost AS number of Route A1 is the same as the leftmost AS number of Route A2, and therefore Route A1 is selected because its MED is smaller. After that, Route A1 and Route B are compared. Because the leftmost AS numbers of the two routes are different, the optimal route cannot be selected by comparing the MEDs of the two routes unless the compare-different-as-med command is configured. As a result, Route B is selected because its IGP metric is smaller.

If Route A2, Route B, and Route A1 are received in turn, Route A2 and Route B are first compared. Because leftmost AS number of Route A2 is different from the leftmost AS number of Route B, the optimal route cannot be selected by comparing the MEDs of the two routes unless the compare-different-as-med command is configured. As a result, Route A2 is selected because its IGP metric is smaller. After that, Route A2 and Route A1 are compared. The leftmost AS number of Route A1 is the same as the leftmost AS number of Route A2, and therefore Route A1 is selected because its MED is smaller.

Judging from the preceding route selection procedure, when the BGP deterministic-MED function is disabled, the sequence in which routes are received is relevant to the result of route selection. After the BGP deterministic-MED function is enabled, the sequence in which routes are received is no longer relevant to the result of route selection. Route A1 and Route A2 have the same leftmost AS number, Route A1 and Route A2 are compared first regardless of the sequence in which routes are received.

Usage Scenario

The BFD session set up by BGP helps BGP quickly discover faults between BGP peers and inform BGP to recalculate routes for packet forwarding.

The information about the BFD session of a specified BGP peer can be displayed by specifying different parameters.

• Run the display bgp bfd session vpnv4 vpn-instance vpn-instance-name peer ipv4-address command to display information about the BFD session of a specified BGP peer in a specified VRF.

• Run the display bgp bfd session peer ipv4-address command to display information about the BFD session of a specified BGP peer on the public network.
• Run the display bgp bfd session all command to display information about the BFD sessions of all BGP peers.

Prerequisites

The BFD session has been set up using the peer bfd enable command. If the BFD session has not been set up by BGP, no information is displayed after running the display bgp bfd session command.

Usage Scenario

When a BGP fault occurs, run the display bgp error command to display BGP error information. BGP error information includes peer error information, route error information, and resource threshold-crossing error information.

Precautions

If no BGP fault occurs, no information is displayed after running the display bgp error command.

Usage Scenario

When a BGP fault occurs, the display bgp error discard command can be run to check the information about discarded error packets. The errors include cluster ID conflicts and the threshold overflow of AS-Path attributes.

By specifying the peer parameter, you can check the information about the error packets discarded by a specified BGP peer.

Precautions

The display bgp error discard command can be used to check only the error routing. To check the error routing among BGP peers, run the display bgp error command.

Usage Scenario

Configuring BGP peer groups simplifies BGP network configuration and improves the route advertisement efficiency.

Precautions

BGP has multiple address families, each of which is configured independently. Information about peer groups in address families can be displayed by specifying different parameters.

If no parameter is specified, the display bgp group command displays default information about peer groups in the IPv4 unicast address family.

Usage Scenario

The BFD session set up by BGP4+ helps BGP4+ quickly discover faults between BGP4+ peers and inform BGP4+ to recalculate routes for packet forwarding.

The information about the BFD session of a specified peer can be displayed by specifying different parameters.

• Run the display bgp ipv6 bfd session vpnv6 vpn-instance vpn-instance-name peer ipv6-address command to display information about the BFD session of a specified peer in a VPNv6 instance.

• Run the display bgp ipv6 bfd session peer ipv6-address command to display information about the BFD session of a specified peer on the public network.
• Run the display bgp ipv6 bfd session all command to display information about the BFD sessions of all BGP4+ peers.

Prerequisites

The BFD session has been set up using the peer bfd enable command. If the BFD session has not been set up by BGP4+, no information is displayed after running the display bgp ipv6 bfd session command.

You can specify different parameters to view the specific routing information.

You can run the display bgp ipv6 routing-table time-range start-time end-time command to view BGP4+ routes that flap within the specified time period. For example, if service traffic is abnormal or CPU usage of the device remains high within a certain time period, you can run this command to check whether route flapping occurs within the specified time period. The faulty route can be viewed in the command output, facilitating fault location.

To display the statistics about BGP IPv6 routes, run display bgp ipv6 routing-table statistics command.

You can specify different parameters to view the specific routing information.

When BGP MDT routing table is displayed, if the length of the destination address mask of an IPv4 route is the same as that of its natural mask, the mask length is not displayed.

When the display bgp mdt brief command is used to display information about MDT, VPN instances are listed in the lexicographic order of the VPN instance name.

None

If the peer group is specified, the detailed information on the specified peer group is displayed.

If the peer group is not specified, the information on all MBGP peer groups is displayed.

To display the information about path attributes of MBGP, run display bgp multicast paths command.

To display the information about a specified MBGP peer, run display bgp multicast peer command.

If ip-address is not specified, this command will display all the information in the MBGP routing table.