User's Manual

Table Of Contents
453
Users Manual of GPL-8000
The ARP of the next-hop IP address does not exist, the system will send an ARP request and
temporarily designate the next-hop routing item as CPU. After the system receives the ARP response,
the system then update the next hop to the user-designating address. If the VLAN interface where the
next hop resides is found different from the configured interface during the ARP response, the next hop
of the route is designated as CPU. Users then need to correct the configuration.
If the next-hop interface or the interface protocol does not exist, the item will not be added to the
hardware subnet routing table.
Suppose a switch has the following routing items:
(1) 192.168.0.0/16 next hop 192.168.26.3/vlan1
(2) 192.168.20.0/24 next hop 192.168.26.1/vlan1
(3) 192.168.1.0/24 direct-connecting routing
(4) 192.168.26.0/24 direct-connecting routing
(5) 10.0.0.0/8 next hop 192.168.1.4/vlan2
(6) 0.0.0.0/0 next hop 192.168.1.6/vlan2
The destination subnet of route item 1 is the prefix of subnet 2, 3 and 4. Therefore,these items should be
added to the hardware subnet routing table together. Item 3 and 4 are direct-connecting routing and the next
hop is CPU.
The relative configuration is as follows:
ip exf 192.168.20.0 255.255.255.0 nexthop 192.168.26.1 vlan 1
ip exf 192.168.1.0 255.255.255.0 cpu
ip exf 192.168.26.0 255.255.255.0 cpu
ip exf 192.168.0.0 255.255.0.0 nexthop 192.168.26.3 vlan 1
ip exf 10.0.0.0 255.0.0.0 nexthop 192.168.1.4 vlan 2
ip exf 0.0.0.0 0.0.0.0 nexthop 192.168.1.6 vlan 2
IP-PBR Configuration
IP-PBR Configuration
IP-PBR realizes software PBR functions through the hardware of switch chip.
PBR stands for Policy Based Routing. PBR enables users to rely on a certain policy not on routing protocol for
routing. Software based PBR supports multiple policies and rules and also load balance. You can designate
the next hop’s IP address or port for those packets that are in line with policy. PBR supports load balance and
applies multiple next-hop IP addresses or ports on those policy-supported packets.
Only when the next-hop egress ARP designated by route map is already learned can IP-PBR regard that this
egress is valid and then the corresponding rule is effective. When a packet satisfies IP-PBR policy, the
hardware directly forwards this packet to the next-hop egress that the rule specifies. This process is finished