cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
714
Views
0
Helpful
1
Replies

IPv6 link-local addressing and NBMA route recursion

bmcgahan
Level 1
Level 1

Fellow professionals,

I am looking for a definitive answer as to why dynamically learned IPv6 prefix information recurses to the remote link-local address when the outgoing interface is an NBMA media vs the global unicast address. I've seen this behavior occur on Frame Relay, ATM, and ISDN, and would assume it to be true for other NBMA medias for which IPv6 support currently exists. The behavior on point-to-point medias such as HDLC and broadcast medias such as Ethernet is to recurse the next-hop to the assigned global unicast address.

The document Implementing Basic Connectivity for IPv6 says that

"Given that IPv6 supports multiple address types, and depending on which applications or protocols are configured on a point-to-multipoint interface, you may need to configure multiple explicit mappings between the IPv6 addresses of the interface and the PVC used to reach the addresses. For example, explicitly mapping both the link-local and global IPv6 address of a point-to-multipoint interface to the PVC that the interface terminates ensures that the Interior Gateway Protocol (IGP) configured on the interface forwards traffic to and from the PVC correctly."

What I am wondering is WHY is this the chosen behavior? The only thing that I could think of it to phase out the usage of global unicast on WAN links that are used simply for transit, kind of like a transit provider using RFC 1918 addresses for transit links that aren't advertised to other AS.

The problem however is that I can't seem to find ANY other reference to this behavior in the IPv6 RFCs or other Cisco documentation other than the short excerpt that "you may need to configure multiple explicit mappings", which is fairly vague on it's own.

So the question... why choose this behavior? What is the advantage of recursing to the link local address over the global unicast address? In what situation would it be advantageous NOT to enable a global unicast address on the interface?

Thanks,

Brian McGahan, CCIE #8593

bmcgahan@internetworkexpert.com

Internetwork Expert, Inc.

http://www.InternetworkExpert.com

Toll Free: 877-224-8987 x 705

Outside US: 775-826-4344 x 705

24/7 Support: http://forum.internetworkexpert.com

Live Chat: http://www.internetworkexpert.com/chat/

1 Reply 1

Harold Ritter
Cisco Employee
Cisco Employee

Brian,

The local link address is used not only for NBMA but also for all other media types as you can see in the following output:

r2#sh clns nei

System Id Interface SNPA State Holdtime Type Protocol

r1 Et0/0 aabb.cc00.8200 Up 25 L2 IS-IS

r1 Se2/0 DLCI 201 Up 21 L2 IS-IS

r3 Se3/0 *PPP* Up 26 L2 IS-IS

r3 Se2/0 DLCI 203 Up 9 L2 IS-IS

r2#sh ipv6 ro 2001:12ff::1

IPv6 Routing Table - 13 entries

Codes: C - Connected, L - Local, S - Static, R - RIP, B - BGP

U - Per-user Static route

I1 - ISIS L1, I2 - ISIS L2, IA - ISIS interarea, IS - ISIS summary

O - OSPF intra, OI - OSPF inter, OE1 - OSPF ext 1, OE2 - OSPF ext 2

ON1 - OSPF NSSA ext 1, ON2 - OSPF NSSA ext 2

I2 2001:12FF::1/128 [115/10]

via FE80::A8BB:CCFF:FE00:8200, Ethernet0/0

via FE80::6FF:FE75:4082, Serial2/0

r2#sh ipv6 ro 2001:12ff::3

IPv6 Routing Table - 13 entries

Codes: C - Connected, L - Local, S - Static, R - RIP, B - BGP

U - Per-user Static route

I1 - ISIS L1, I2 - ISIS L2, IA - ISIS interarea, IS - ISIS summary

O - OSPF intra, OI - OSPF inter, OE1 - OSPF ext 1, OE2 - OSPF ext 2

ON1 - OSPF NSSA ext 1, ON2 - OSPF NSSA ext 2

I2 2001:12FF::3/128 [115/10]

via FE80::6FF:FE75:4084, Serial2/0

via FE80::6FF:FE75:4084, Serial3/0

Hope this helps,

Harold Ritter
Sr Technical Leader
CCIE 4168 (R&S, SP)
harold@cisco.com
México móvil: +52 1 55 8312 4915
Cisco México
Paseo de la Reforma 222
Piso 19
Cuauhtémoc, Juárez
Ciudad de México, 06600
México