Route faults: Answers to BGP questions about IP Route faults

Why?BGPCannot be aggregated with itRoutingWhich of the following network segments does the device establish a neighbor?

A: If BGP establishes a neighbor with the device of the network segment to which the aggregation route belongs, when the device or link of the target network segment fails, it cannot be sensed through the aggregation route, the device will still try to establish a neighbor relationship with the configured peer.

When a BGP peer is configured, the peer address cannot be obtained through the BGP aggregation route or the default route. Otherwise, the BGP peer neighbor relationship cannot be established even if the traffic can pass through.

Why do I need to specify a router-id when creating a BGP neighbor in an IPv6 network?

A: Generally, the router-id is not specified in the BGP view in the IPv4 network configuration. BGP automatically selects an interface IP address as the router-id. However, in an IPv6 network, the router-id is still 32-bit. If the router-id is not specified, no interface address is an IPv4 address, therefore, BGP cannot be successfully negotiated because it does not have a router-id. Therefore, we recommend that you specify a router-id when creating a BGP peer in an IPv6 network.

What are the differences between BGP refresh features for outbound and inbound route processing?

A: After a BGP connection is established, only the incremental route is sent. However, in some cases, when the routing policy is changed, the other party needs to resend the route to filter the route according to the new policy. Vrouters that support BGP refresh feature RFC 2918) can enable the changed BGP Route attributes to take effect without interrupting BGP neighbor sessions.

You do not need to make any configuration to change the BGP Route attribute of the outbound export. The change takes effect automatically.

To change the BGP Route attributes of inbound import, use the refresh bgp command in user mode to send an update request to the peer router. After this operation is performed, the changed attribute takes effect without interrupting the Session of the BGP peer.

What are the reasons for failed Establishment of BGP peers?

A: To establish a BGP peer, you must be able to use port 179 to establish a TCP session and correctly exchange Open packets. You can follow the steps below to check.

1. Check whether the AS number of the neighbor is correctly configured.

2. Check whether the IP address of the neighbor is correct.

3. If you use the Loopback interface, check whether the peer connect-interface command is configured. By default, the router uses the best local interface to establish a TCP connection, instead of using the Loopback interface.

4. If it is a physically non-directly connected EBGP neighbor, check whether peer ebgp-max-hop is configured.

5. run the ping command to check whether the TCP connection is normal. Because a vro may have multiple interfaces that can reach the peer end, therefore, use the extended ping-a ip-address command to specify the source ip address for sending the ping packet.

If the ping fails, run the "display ip routing-table" command to check whether there are available routes to neighbors in the routing table.

If ping is enabled, check whether the ACL list of TCP port 179 is disabled. If this parameter is set, port 179 is disabled.

What should I pay attention to when configuring BGP Route aggregation?

A: pay attention to the following in BGP Route aggregation configuration.

You must disable automatic aggregation of BGP routers because the automatically aggregated content is aggregated by class. For example, 61.0.0.0/24 is automatically aggregated and the result is 61.0.0.0/8. If an IGP aggregation route exists in the routing table, you can use the network command to directly introduce it to BGP for aggregation.

When configuring manual summarization, you must first configure a black hole route for the summary network segment and then publish it to BGP. Because first, the next hop of the summarized route is uncertain. Second, the router can first find the detailed route based on the longest matching principle. If no route is found, it indicates that the detailed route is unavailable, by summarizing the black hole routes, You can discard the inaccessibility routing information.

The detailed route is also published in the routing table through the aggregated route configured by the network. If you do not need to publish a detailed route, you can use the routing policy to disable it in the outbound direction. You can use the aggregate command to publish only the aggregated route information without publishing detailed routes, and generate a black hole route pointing to the Null Port.

Why does BGP publishing route fail?

A: When you publish a route in the BGP process, the route table of the device must exist. Otherwise, the published route does not take effect. Of course, if the routing table does not exist and needs to be released, you can configure a black hole route for the published network segment.