IP routing directs IP packet forwarding on routers. Based on the destination IP address in the packet, a router looks up a route for the packet in a routing table and forwards the packet to the next hop. Routes are path information used to direct IP packets.

Routing table

A RIB contains the global routing information and related information, including route recursion, route redistribution, and route extension information. The router selects optimal routes from the routing table and puts them into the FIB table. It uses the FIB table to forward packets. For more information about the FIB table, see Layer 3—IP Services Configuration Guide.

Route categories

Table 1 categorizes routes by different criteria.

Table 1 Route categories

Criterion	Categories
Origin	· Direct route—A direct route is discovered by the data link protocol on an interface, and is also called an interface route. · Static route—A static route is manually configured by an administrator. · Dynamic route—A dynamic route is dynamically discovered by a routing protocol.
Destination	· Network route—The destination is a network. The subnet mask is less than 32 bits. · Host route—The destination is a host. The subnet mask is 32 bits.
Whether the destination is directly connected	· Direct route—The destination is directly connected. · Indirect route—The destination is indirectly connected.

Dynamic routing protocols

Static routes work well in small, stable networks. They are easy to configure and require fewer system resources. However, in networks where topology changes occur frequently, a typical practice is to configure a dynamic routing protocol. Compared with static routing, a dynamic routing protocol is complicated to configure, requires more router resources, and consumes more network resources.

Dynamic routing protocols dynamically collect and report reachability information to adapt to topology changes. They are suitable for large networks.

Dynamic routing protocols can be classified by different criteria, as shown in Table 2.

Table 2 Categories of dynamic routing protocols

Criterion	Categories
Operation scope	· IGPs—Work within an AS. Examples include RIP, OSPF, and IS-IS. · EGPs—Work between ASs. The most popular EGP is BGP.
Routing algorithm	· Distance-vector protocols—Examples include RIP and BGP. BGP is also considered a path-vector protocol. · Link-state protocols—Examples include OSPF and IS-IS.
Destination address type	· Unicast routing protocols—Examples include RIP, OSPF, BGP, and IS-IS. · Multicast routing protocols—Examples include PIM-SM and PIM-DM.
IP version	· IPv4 routing protocols—Examples include RIP, OSPF, BGP, and IS-IS. · IPv6 routing protocols—Examples include RIPng, OSPFv3, IPv6 BGP, and IPv6 IS-IS.

An AS refers to a group of routers that use the same routing policy and work under the same administration.

Route preference

Routing protocols, including static and direct routing, each by default have a preference. If they find multiple routes to the same destination, the router selects the route with the highest preference as the optimal route.

The preference of a direct route is always 0 and cannot be changed. You can configure a preference for each static route and each dynamic routing protocol. The following table lists the route types and default preferences. The smaller the value, the higher the preference.

Table 3 Route types and default route preferences

Route type	Preference
Direct route	0
Multicast static route	1
OSPF	10
IS-IS	15
Unicast static route	60
RIP	100
OSPF ASE	150
OSPF NSSA	150
IBGP	255
EBGP	255
Unknown (route from an untrusted source)	256

Load sharing

A routing protocol might find multiple optimal equal-cost routes to the same destination. You can use these routes to implement equal-cost multi-path (ECMP) load sharing.

Static routing, IPv6 static routing, RIP, RIPng, OSPF, OSPFv3, BGP, IPv6 BGP, IS-IS, and IPv6 IS-IS support ECMP load sharing.

Route backup

Route backup can improve network availability. Among multiple routes to the same destination, the route with the highest priority is the primary route and others are secondary routes.

The router forwards matching packets through the primary route. When the primary route fails, the route with the highest preference among the secondary routes is selected to forward packets. When the primary route recovers, the router uses it to forward packets.

Route recursion

To use a BGP, static, or RIP route that has an indirectly connected next hop, a router must perform route recursion to find the output interface to reach the next hop.

Link-state routing protocols, such as OSPF and IS-IS, do not need route recursion, because they obtain directly connected next hops through route calculation.

The RIB records and saves route recursion information, including brief information about related routes, recursive paths, and recursion depth.

Route redistribution

Route redistribution enables routing protocols to learn routing information from each other. A dynamic routing protocol can redistribute routes from other routing protocols, including direct and static routing. For more information, see the respective chapters on those routing protocols in this configuration guide.

The RIB records redistribution relationships of routing protocols.

Extension attribute redistribution

Extension attribute redistribution enables routing protocols to learn route extension attributes from each other, including BGP extended community attributes, OSPF area IDs, route types, and router IDs.

The RIB records extended attributes of each routing protocol and redistribution relationships of different routing protocol extended attributes.

Setting the maximum lifetime for routes and labels in the RIB

About this task

Perform this task to prevent routes of a certain protocol from being aged out due to slow protocol convergence resulting from a large number of route entries or long GR period.

Restrictions and guidelines for setting the maximum lifetime for routes and labels in the RIB

The configuration takes effect at the next protocol or RIB process switchover.

Procedure (IPv4)

1. Enter system view.

system-view

2. Enter RIB view.

rib

3. Create the RIB IPv4 address family and enter its view.

address-family ipv4

4. Set the maximum lifetime for IPv4 routes and labels in the RIB.

protocol protocol [ instance instance-name ] lifetime seconds

By default, the maximum lifetime for routes and labels in the RIB is 480 seconds.

Procedure (IPv6)

1. Enter system view.

system-view

2. Enter RIB view.

rib

3. Create the RIB IPv6 address family and enter its view.

address-family ipv6

4. Set the maximum lifetime for IPv6 routes and labels in the RIB.

protocol protocol [ instance instance-name ] lifetime seconds

By default, the maximum lifetime for routes and labels in the RIB is 480 seconds.

Setting the maximum lifetime for routes in the FIB

About this task

When GR or NSR is disabled, FIB entries must be retained for some time after a protocol process switchover or RIB process switchover. When GR or NSR is enabled, FIB entries must be removed immediately after a protocol or RIB process switchover to avoid routing issues. Perform this task to meet such requirements.

Procedure (IPv4)

1. Enter system view.

system-view

2. Enter RIB view.

rib

3. Create the RIB IPv4 address family and enter its view.

address-family ipv4

4. Set the maximum lifetime for IPv4 routes in the FIB.

fib lifetime seconds

By default, the maximum lifetime for routes in the FIB is 600 seconds.

Procedure (IPv6)

1. Enter system view.

system-view

2. Enter RIB view.

rib

3. Create the RIB IPv6 address family and enter its view.

address-family ipv6

4. Set the maximum lifetime for IPv6 routes in the FIB.

fib lifetime seconds

By default, the maximum lifetime for routes in the FIB is 600 seconds.

Configuring RIB NSR

About this task

When an active/standby switchover occurs, nonstop routing (NSR) backs up routing information from the active process to the standby process to avoid routing flapping and ensure forwarding continuity.

RIB NSR provides faster route convergence than protocol NSR during an active/standby switchover.

Prerequisites for RIB NSR

Use this feature with protocol GR or NSR to avoid route timeouts and traffic interruption.

Procedure (IPv4)

1. Enter system view.

system-view

2. Enter RIB view.

rib

3. Create the RIB IPv4 address family and enter its view.

address-family ipv4

4. Enable IPv4 RIB NSR.

non-stop-routing

By default, RIB NSR is disabled.

Procedure (IPv6)

1. Enter system view.

system-view

2. Enter RIB view.

rib

3. Create the RIB IPv6 address family and enter its view.

address-family ipv6

4. Enable IPv6 RIB NSR.

non-stop-routing

By default, RIB NSR is disabled.

Configuring inter-protocol FRR

About this task

Inter-protocol fast reroute (FRR) enables fast rerouting between routes of different protocols. A backup next hop is automatically selected to reduce the service interruption time caused by unreachable next hops. When the next hop of the primary link fails, the traffic is redirected to the backup next hop.

Among the routes to the same destination in the RIB, a router adds the route with the highest preference to the FIB table. For example, if a static route and an OSPF route in the RIB have the same destination, the router adds the OSPF route to the FIB table by default. The next hop of the static route is selected as the backup next hop for the OSPF route. When the next hop of the OSPF route is unreachable, the backup next hop is used.

Restrictions and guidelines for inter-protocol FRR

This feature uses the next hop of a route from a different protocol as the backup next hop, which might cause loops.

Procedure (IPv4)

1. Enter system view.

system-view

2. Enter RIB view.

rib

3. Create the RIB IPv4 address family and enter its view.

address-family ipv4

4. Enable IPv4 RIB inter-protocol FRR.

inter-protocol fast-reroute [ vpn-instance vpn-instance-name ]

By default, inter-protocol FRR is disabled.

If you do not specify a VPN instance, inter-protocol FRR is enabled for the public network.

Procedure (IPv6)

1. Enter system view.

system-view

2. Enter RIB view.

rib

3. Create the RIB IPv6 address family and enter its view.

address-family ipv6

4. Enable IPv6 RIB inter-protocol FRR.

inter-protocol fast-reroute [ vpn-instance vpn-instance-name ]

By default, inter-protocol FRR is disabled.

If you do not specify a VPN instance, inter-protocol FRR is enabled for the public network.

Configuring BFD for primary link availability detection

About configuring BFD for primary link availability detection

This feature uses BFD to detect the availability of the primary link for inter-protocol FRR or ECMP routes. When the primary link fails, this feature enables the upper layer routing protocol to reallocate service traffic to the backup links or to the remaining ECMP links.

For inter-protocol FRR, the primary link is the route with the highest preference among the routes to the same destination. If BFD detects that the next hop of the primary link is invalid, it notifies the FIB of the invalid next hop. The FIB will use the backup routes to guide traffic forwarding.

For ECMP routes of a routing protocol, the primary link is each ECMP route. If BFD detects that the next hop of an ECMP route is invalid, it notifies the routing protocol of the invalid next hop. The routing protocol will reallocate the service traffic to the remaining ECMP links. When BFD detects that the next hops of all ECMP routes are invalid, BFD notifies the routing protocol to withdraw the routes. In addition, the device reselects optimal routes for the service traffic.

For more information about BFD, see High Availability Configuration Guide.

Restrictions and guidelines for configuring BFD for primary link availability detection

If you specify the ecmp-shared keyword when configuring FFR for a routing protocol, the device will use the LFA algorithm to calculate backup next hops for each ECMP route. The routes destined for the backup next hops are also added to the routing table as ECMP routes and the state of these routes is backup. BFD does not detect these backup ECMP routes.

Configuring BFD control packet mode

About this task

This mode uses BFD control packets to detect the status of a link bidirectionally at a millisecond level.

Restrictions and guidelines for BFD control packet mode

If you use BFD control packet mode at the local end, you must use this mode also at the peer end.

Procedure

1. Enter system view.

system-view

2. Enter RIB view.

rib

3. Create RIB IPv4 or IPv6 address family view and enter the view.

IPv4:

address-family ipv4

IPv6:

address-family ipv6

4. Enable BFD for primary link availability detection.

primary-path-detect bfd ctrl [ inter-protocol-frr | protocol-ecmp protocol ]

By default, BFD is disabled for primary link availability detection.

Configuring BFD echo packet mode

About this task

With BFD echo packet mode enabled, the output interface sends BFD echo packets to the destination device, and the device sends the packets back to test the link reachability.

Restrictions and guidelines

You do not need to configure BFD echo packet mode at the peer end.

Procedure

1. Enter system view.

system-view

2. Configure the source IP address of echo packets.

IPv4:

bfd echo-source-ip ip-address

By default, the source IPv4 address of echo packets is not configured.

To avoid network congestion caused by excessive ICMP redirect packets from the peer, make sure the source IPv4 address is not on the subnet of any interfaces on the device.

IPv6:

bfd echo-source-ipv6 ipv6-address

By default, the source IPv6 address of echo packets is not configured.

You must specify a global unicast address as the source IPv6 address of BFD echo packets.

For more information about the commands, see High Availability Command Reference.

3. Enter RIB view.

rib

4. Create RIB IPv4 or IPv6 address family view and enter the view.

IPv4:

address-family ipv4

IPv6:

address-family ipv6

5. Enable BFD for primary link availability detection.

primary-path-detect bfd echo [ inter-protocol-frr | protocol-ecmp protocol ]

By default, BFD is disabled for primary link availability detection.

Enabling route fast switchover

About this task

This feature applies to a device that provides the same physical output interface for large numbers of routes, including ECMP routes and primary/secondary routes. When a link failure occurs on the interface, the device must perform the following tasks before switching the traffic to another route:

1. Deletes all ARP or ND entries for the link.

2. Instructs the FIB to delete the associated FIB entries.

This procedure is time consuming and interrupts traffic for a long time. To resolve this issue, you can enable route fast switchover. This feature allows the device to instruct the FIB to delete the invalid FIB entries for route switchover first.

Procedure (IPv6)

1. Enter system view.

system-view

2. Enable IPv6 route fast switchover.

ipv6 route fast-switchover enable

By default, IPv6 route fast switchover is disabled.

Verifying and maintaining basic IP routing (IPv4)

Displaying IPv4 routing table information

Perform display tasks in any view.

· Display routing table information.

display ip routing-table [ all-vpn-instance | vpn-instance vpn-instance-name ] [ verbose ]

display ip routing-table [ all-routes ]

· Display information about routes permitted by an IPv4 basic ACL.

display ip routing-table [ vpn-instance vpn-instance-name ] acl ipv4-acl-number [ verbose ]

· Display information about routes to a specific destination address.

display ip routing-table [ vpn-instance vpn-instance-name ] ip-address [ mask-length | mask ] [ longer-match ] [ verbose ]

· Display information about routes to a range of destination addresses.

display ip routing-table [ vpn-instance vpn-instance-name ] ip-address1 to ip-address2 [ verbose ]

· Display information about routes permitted by an IP prefix list.

display ip routing-table [ vpn-instance vpn-instance-name ] prefix-list prefix-list-name [ verbose ]

· Display information about routes installed by a protocol.

display ip routing-table [ vpn-instance vpn-instance-name ] protocol protocol [ inactive | verbose ]

· Display brief routing table information, including maximum number of ECMP routes, maximum number of active routes, and number of remaining active routes.

display ip routing-table [ vpn-instance vpn-instance-name ] summary

Displaying IPv4 RIB information

Perform display tasks in any view.

· Display route attribute information in the RIB.

display rib attribute [ attribute-id ]

· Display RIB GR state information.

display rib graceful-restart

· Display next hop information in the RIB.

display rib nib [ self-originated ] [ nib-id ] [ verbose ]

display rib nib [ sub-nib nib-id ] [ verbose ]

display rib nib protocol protocol [ verbose ]

· Display next hop information for direct routes.

display route-direct nib [ nib-id ] [ verbose ]

Displaying and clearing IPv4 routing table statistics

To display IPv4 route or route prefix statistics, execute the following commands in any view:

· display ip routing-table [ all-routes | all-vpn-instance | vpn-instance vpn-instance-name ] [ prefix ] statistics

To clear route statistics, execute the following commands in user view:

· reset ip routing-table statistics protocol [ vpn-instance vpn-instance-name ] { protocol | all }

· reset ip routing-table [ all-routes | all-vpn-instance ] statistics protocol { protocol | all }

Verifying and maintaining basic IP routing (IPv6)

Displaying IPv6 routing table information

Perform display tasks in any view.

· Display IPv6 routing table information.

display ipv6 routing-table [ vpn-instance vpn-instance-name ] [ verbose ]

· Display information about routes to an IPv6 destination address.

display ipv6 routing-table [ vpn-instance vpn-instance-name ] ipv6-address [ prefix-length ] [ longer-match ] [ verbose ]

· Display information about routes permitted by an IPv6 basic ACL.

display ipv6 routing-table [ vpn-instance vpn-instance-name ] acl ipv6-acl-number [ verbose ]

· Display information about routes to a range of IPv6 destination addresses.

display ipv6 routing-table [ vpn-instance vpn-instance-name ] ipv6-address1 to ipv6-address2 [ verbose ]

· Display information about routes permitted by an IPv6 prefix list.

display ipv6 routing-table [ vpn-instance vpn-instance-name ] prefix-list prefix-list-name [ verbose ]

· Display information about routes installed by an IPv6 protocol.

display ipv6 routing-table [ vpn-instance vpn-instance-name ] protocol protocol [ inactive | verbose ]

· Display brief IPv6 routing table information, including maximum number of ECMP routes, maximum number of active routes, and number of remaining active routes.

display ipv6 routing-table [ vpn-instance vpn-instance-name ] summary

Displaying IPv6 RIB information

Perform display tasks in any view.

· Display route attribute information in the IPv6 RIB.

display ipv6 rib attribute [ attribute-id ]

· Display IPv6 RIB GR state information.

display ipv6 rib graceful-restart

· Display next hop information in the IPv6 RIB.

display ipv6 rib nib [ self-originated ] [ nib-id ] [ verbose ]

display ipv6 rib nib [ sub-nib nib-id ] [ verbose ]

display ipv6 rib nib protocol protocol [ verbose ]

· Display next hop information for IPv6 direct routes.

display ipv6 route-direct nib [ nib-id ] [ verbose ]

Displaying and clearing IPv6 routing table statistics

To display IPv6 route or route prefix statistics, execute the following command in any view:

display ipv6 routing-table [ all-routes | all-vpn-instance | vpn-instance vpn-instance-name ] [ prefix ] statistics

To clear IPv6 route statistics, execute the following commands in user view:

· reset ipv6 routing-table statistics protocol [ vpn-instance vpn-instance-name ] { protocol | all }

· reset ipv6 routing-table [ all-routes | all-vpn-instance ] statistics protocol { protocol | all }

09-Layer 3—IP Routing Configuration Guide

Dynamic routing protocols

Route preference

Load sharing

Extension attribute redistribution

About this task

Restrictions and guidelines for setting the maximum lifetime for routes and labels in the RIB

Procedure (IPv4)

Procedure (IPv6)

About this task

Procedure (IPv4)

Procedure (IPv6)

About this task

Prerequisites for RIB NSR

Procedure (IPv4)

Procedure (IPv6)

About this task

Restrictions and guidelines for inter-protocol FRR

Procedure (IPv4)

Procedure (IPv6)

About this task

Restrictions and guidelines for BFD control packet mode

Procedure

About this task

Restrictions and guidelines

Procedure

About this task

Procedure (IPv6)

Intelligent Terminal Products

Product Support Services

Technical Service Solutions

Resource Center

Policy

Online Help

Become a Partner

Partner Policy & Program

Global Learning

Partner Sales Resources

Service Business

News & Events

Contact Us