Unicast Reverse Path Forwarding


Unicast Reverse Path Forwarding

In IPv6, unicast RPF helps assure a router from DoS attacks from spoofed IPv6 host addresses.

When you configure IPv6 unicast RPF by arising the ipv6 verify unicast reverse-path command

on an interface, the router performs a recursive lookup in the IPv6 acquisition table to verify that the

packet came in on the actual interface. If this analysis passes, the packet in catechism is allowed

through; if not, the router drops it.

Cisco IOS Software gives you the advantage of defining a array of assurance boundary. This way, a router

can verify alone called antecedent IPv6 addresses in the unicast RPF check. To do this, configure an

access account on the router and alarm it with the ipv6 verify unicast reverse-path command.

In Example 20-4, the router will accomplish the RPF analysis on all IPv6 packets that access the router’s

Fast Ethernet 0/0 interface. The router will again bead packets that accommodated both of these conditions:

1. The RPF analysis fails.

2. The antecedent abode is aural the 2007::/64 range.

If either of these altitude is not met, the packet will be routed. If both altitude are met, the

router drops the packet.

Example 20-4 Unicast Reverse-Path Forwarding Configuration

HiramMaxim(config)# ipv6 access-list urpf

HiramMaxim(config-ipv6-acl)# abjure ipv6 2007::/64 any

HiramMaxim(config-ipv6-acl)# admittance ipv6 any any

HiramMaxim(config-ipv6-acl)# interface fa0/0

HiramMaxim(config-if)# ipv6 verify unicast reverse-path urpf

HiramMaxim(config-if)# end

HiramMaxim# ipv6 interface fa0/0

FastEthernet0/0 is up, band agreement is up

IPv6 is enabled, link-local abode is FE80::207:85FF:FE80:7208

No Virtual link-local address(es):

Global unicast address(es):

2002:192:168:1::1, subnet is 2002:192:168:1::/64

2002:192:168:2::1, subnet is 2002:192:168:2::/64 [ANY]

Joined accumulation address(es):

FF02::1

FF02::2

FF02::A

FF02::D

FF02::16

FF02::1:FF00:1

FF02::1:FF80:7208

MTU is 1500 bytes

ICMP absurdity letters bound to one every 100 milliseconds

ICMP redirects are enabled

Basic IPv6 Functionality Protocols 763

For added advice about how RPF checks work, see Chapter 16, “Introduction to IP

Multicasting.”

DNS

DNS for IPv6 is absolutely agnate to DNS for IPv4; it provides resolution of area names to IPv6

addresses. One key aberration is the name acclimated for DNS annal for IPv6 addresses. In IPv4, these

are accepted as A records; in IPv6, RFC 1886 cleverly agreement them AAAA records, because IPv6

addresses are four times best (in bits) than IPv4 addresses. RFC 1886 and RFC 2874 are both

IPv6 DNS extensions. RFC 2874 calls IPv6 abode annal A6 records. Today, RFC 1886 is most

commonly used; however, RFC 2874 expects to eventually anachronistic RFC 1886.

IPv6 DNS extensions additionally accommodate the changed lookup action of PTR records, which maps IPv6

addresses to host names.

CDP

Cisco Discovery Agreement provides all-encompassing advice about the agreement and

functionality of Cisco devices. Because of its extensibility, it should be no abruptness to you that

CDP additionally provides advice about Cisco IPv6 host configuration. To see IPv6 information

ICMP unreachables are sent

Input features: RPF

Unicast RPF access-list urpf

Process Switching:

0 analysis drops

0 suppressed analysis drops

CEF Switching:

0 analysis drops

0 suppressed analysis drops

ND DAD is enabled, cardinal of DAD attempts: 1

ND attainable time is 30000 milliseconds

ND advertised attainable time is 0 milliseconds

ND advertised retransmit breach is 0 milliseconds

ND router advertisements are beatific every 200 seconds

ND router advertisements alive for 1800 seconds

ND advertised absence router another is Medium

Hosts use stateless autoconfig for addresses.

Example 20-4 Unicast Reverse-Path Forwarding Agreement (Continued)

764 Chapter 20: IP Version 6

transmitted in CDP frames, you charge use the detail keyword for the appearance cdp neighbor

command, as apparent in Example 20-5.

DHCP

One another to changeless IPv6 addressing, namely stateless autoconfiguration, was covered earlier.

Another another additionally exists: stateful autoconfiguration. This is area DHCPv6 comes in.

DHCPv6 is defined in RFC 3315.

Two altitude can account a host to use DHCPv6:

■ The host is absolutely configured to use DHCPv6 based on an implementation-specific setting.

■ An IPv6 router advertises in its RA letters that it wants hosts to use DHCPv6 for

addressing. Routers do this by ambience the M banderole (Managed Abode Configuration) in RAs.

To use stateful autoconfiguration, a host sends a DHCP appeal to one of two acclaimed IPv6

multicast addresses on UDP anchorage 547:

■ FF02::1:2, all DHCP broadcast agents and servers

■ FF05::1:3, all DHCP servers

The DHCP server again provides the all-important agreement advice in acknowledgment to the host on

UDP anchorage 546. This advice can accommodate the aforementioned types of advice acclimated in an IPv4

network, but additionally it can accommodate advice for assorted subnets, depending on how the

DHCP server is configured.

To configure a Cisco router as a DHCPv6 server, you aboriginal configure a DHCP pool, aloof as in IPv4

DHCP. Then, you charge accurately accredit the DHCPv6 account application the ipv6 dhcp server poolname

interface command.

Example 20-5 IPv6 Advice Available from CDP Output

Rivers# appearance cdp neighbors detail

-------------------------

Device ID: Mantle

Entry address(es):

IP address: 10.7.7.6

IPv6 address: FE80::207:85FF:FE80:7208 (link-local)

IPv6 address: 2001::207:85FF:FE80:7208 (global unicast)

Platform: Cisco 1760, Capabilities: Router Switch

Interface: Serial0/0, Anchorage ID (outgoing port): Serial0/0

Holdtime : 159 sec

(output bare for brevity)