Contents

Configuring IP forwarding basic settings························································ 1

About FIB table······························································································································· 1

vSystem support for features··········································································································· 1

Enabling last hop holding················································································································ 2

Enabling last hop backup················································································································ 3

Display and maintenance commands for FIB table············································································ 3

Configuring load sharing··················································································· 4

About load sharing·························································································································· 4

Configuring load sharing mode········································································································ 4

Enabling IPv4 bandwidth-based load sharing···················································································· 4

Enabling local-first load sharing······································································································· 5

Configuring IP forwarding basic settings

About FIB table

A device uses the FIB table to make packet forwarding decisions.

A device selects optimal routes from the routing table, and puts them into the FIB table. Each FIB entry specifies the next hop IP address and output interface for packets destined for a specific subnet or host.

For more information about the routing table, see Layer 3—IP Routing Configuration Guide.

Use the display fib command to display the FIB table. The following example displays the entire FIB table.

<Sysname> display fib

 

Destination count: 4 FIB entry count: 4

 

Flag:

  U:Useable   G:Gateway   H:Host   B:Blackhole   D:Dynamic   S:Static

  R:Relay     F:FRR

 

Destination/Mask   Nexthop         Flag     OutInterface/Token       Label

10.2.0.0/16        10.2.1.1        U        GE1/0/1                  Null

10.2.1.1/32        127.0.0.1       UH       InLoop0                  Null

127.0.0.0/8        127.0.0.1       U        InLoop0                  Null

127.0.0.1/32       127.0.0.1       UH       InLoop0                  Null

A FIB entry includes the following items:

·     Destination—Destination IP address.

·     Mask—Network mask. The mask and the destination address identify the destination network. A logical AND operation between the destination address and the network mask yields the address of the destination network. For example, if the destination address is 192.168.1.40 and the mask 255.255.255.0, the address of the destination network is 192.168.1.0. A network mask includes a certain number of consecutive 1s. It can be expressed in dotted decimal format or by the number of the 1s.

·     Nexthop—IP address of the next hop.

·     Flag—Route flag.

·     OutInterface—Output interface.

·     Token—MPLS Label Switched Path index number.

·     Label—Inner label.

vSystem support for features

Non-default vSystems do not support the following features:

·     Enabling the VPN peer-based processing.

·     Enabling IPv4 bandwidth-based load sharing.

For information about the support of non-default vSystems for the commands, see basic IP forwarding command reference. For information about vSystem, see Virtual Technologies Configuration Guide.

Enabling last hop holding

About this task

Last hop holding implements symmetric routing.

When the interface enabled with this feature receives the first IP packet of a forward flow, this feature implements the following operations:

·     Obtains the forward flow information and last hop information of the packet.

·     Based on the information, creates a fast forwarding entry for the reverse flow.

When packets of the reverse flow arrive at the device, the device forwards those packets based on the entry.

As is shown in Figure 1, when the external server sends a request to the internal server, the packet travels through ISP 1 to Interface A on the device. The last hop holding feature on the device ensures that the reply packet follows the same route as the request packet back to ISP 1. If last hop holding is disabled, the reply packet might be sent out of Interface B or Interface C to the external network.

Figure 1 Last hop holding application

 

Restrictions and guidelines

This feature is based on fast forwarding entries. If the MAC address of a last hop changes on an Ethernet link, this feature can function correctly only after the fast forwarding entry is updated for the MAC address.

Procedure

1.     Enter system view.

system-view

2.     Enter Layer 3 Ethernet interface view or subinterface view.

interface interface-type { interface-number | interface-number.subnumber }

3.     Enable last hop holding.

ip last-hop hold

By default, last hop holding is disabled.

Enabling last hop backup

About this task

This feature enables the system to transmit the forward flow and reverse flow between the local device and a peer device over the same path in an IRF fabric.

With this feature configured, the IRF master device performs the following operations when receiving the first IP packet of a forward flow on an interface enabled with last hop holding:

1.     Saves the last hop information of the packet.

2.     Synchronizes the last hop information to subordinate devices in the IRF fabric.

The last hop information can be used for guiding the reverse flow when the flow arrives at the master device or is forwarded through a subordinate device.

This feature takes effect only when the session synchronization is enabled by using the session synchronization enable command. For more information about this command, see Security Command Reference.

This feature is also applicable to multi-module devices enabled with service backup. If this feature is enabled on such a device, a device module performs the following operations when receiving the first IP packet of a forward flow on an interface enabled with last hop holding:

1.     Saves the last hop information of the packet.

2.     Synchronizes the last hop information to other modules in the device.

The last hop information can be used for guiding the backward flow when the flow arrives at one of these modules.

For this feature to take effect on a multi-module device, you must also enable session flow redirection by using the session flow-redirect enable command. For more information about the session flow-redirect enable command, see Security Command Reference.

Procedure

1.     Enter system view.

system-view

2.     Enable last hop backup.

last-hop backup enable

By default, last hop backup is enabled.

Display and maintenance commands for FIB table

Execute display commands in any view.

 

Task	Command
Display FIB entries.	display fib [ vpn-instance vpn-instance-name ] [ ip-address [ mask | mask-length ] ]

‌

Configuring load sharing

About load sharing

If a routing protocol finds multiple equal-cost best routes to the same destination, the device forwards packets over the equal-cost routes to implement load sharing.

Configuring load sharing mode

About this task

In the per-flow load sharing mode, the device forwards flows over equal-cost routes. Packets of one flow travel along the same routes. You can configure the device to identify a flow based on the following criteria: source IP address, destination IP address, source port number, destination port number, IP protocol number, and ingress port.

In the per-packet load sharing mode, the device forwards packets over equal-cost routes.

Procedure

1.     Enter system view.

system-view

2.     Configure load sharing.

In standalone mode:

ip load-sharing mode { per-flow [ dest-ip | dest-port | ip-pro | src-ip | src-port ] * | per-packet } { global | slot slot-number [ cpu cpu-number ] }

In IRF mode:

ip load-sharing mode { per-flow [ dest-ip | dest-port | ip-pro | src-ip | src-port ] * | per-packet } { chassis chassis-number slot slot-number [ cpu cpu-number ] | global }

By default, the device performs per-flow load sharing.

Enabling IPv4 bandwidth-based load sharing

About this task

This feature load shares flow traffic among multiple output interfaces based on their load percentages. The device calculates the load percentage for each output interface in terms of the interface expected bandwidth.

Devices that run load sharing protocols, such as Locator/ID Separation Protocol (LISP), implement load sharing based on the ratios defined by these protocols.

Procedure

1.     Enter system view.

system-view

2.     Enable IPv4 bandwidth-based load sharing.

bandwidth-based-sharing

By default, the IPv4 bandwidth-based load sharing is disabled.

3.     (Optional.) Configure the expected bandwidth of the interface.

a.     Enter interface view.

interface interface-type interface-number

b.     Configure the expected bandwidth of the interface.

bandwidth bandwidth

By default, the expected bandwidth is the physical bandwidth of the interface.

Enabling local-first load sharing

About this task

Local-first load sharing distributes traffic preferentially across the output interfaces on the receiving IRF member device if output interfaces for multiple equal-cost routes are on different members. This feature enhances packets forwarding efficiency.

Procedure

1.     Enter system view.

system-view

2.     Enable local-first load sharing.

ip load-sharing local-first enable

By default, local-first load sharing is disabled.