Frame-Relay

[Jean-Francois Gagnon]

Hi!

I'm trying to put up a lab in GNS3 with Ipv6 and OSPFv3

I need to put up Frame-Relay.

So I have R1 connected to FRS port 1 DLCi 101 mapped to port 10 DLCI 202 To R2

R1 connected to FRS port 1 DLCi 102 mapped to port 11 DLCI 203 To R3

Config R1:

interface Serial0/0
no ip address
encapsulation frame-relay
ipv6 address 123::1/64
ipv6 rip R1R3 enable
ipv6 rip R1R2 enable
clock rate 2000000
frame-relay map ipv6 123::2 101 broadcast
frame-relay map ipv6 123::3 102 broadcast
no frame-relay inverse-arp

R2::

interface Serial0/0
no ip address
encapsulation frame-relay
clock rate 2000000
!
interface Serial0/0.1 point-to-point
ipv6 address 123::2/64
ipv6 rip R1R2 enable
frame-relay interface-dlci 202

R3::

interface Serial0/0
no ip address
encapsulation frame-relay
clock rate 2000000
!
interface Serial0/0.1 point-to-point
ipv6 address 123::3/64
ipv6 rip R1R3 enable
frame-relay interface-dlci 203

They can ping each other

Problem is, I have R4 connected to Fa ports to both R2 and R3, and setup OSPF between them, so i can see all routes. But when I try to ping from R1 to R4 trafic does not form because of frame relay:

*Mar 1 00:21:45.167: Serial0/0:Encaps failed--no map entry link 79(IPV6).

How could I make this more dynamic so that traffic still gets through the Frame-Relay without explicitily mapping it to every IP addresses ?

Thanks!

[paul driver · Accepted solution]

Hello

---

@Jean-Francois Gagnon wrote:

How could I make this more dynamic so that traffic still gets through the Frame-Relay without explicitily mapping it to every IP addresses ?

Thanks!

---

Hello

configure P2P sub interfaces on each router instead-

hub
int ser 0/0
enacap fr

int ser 0/0.12 point-to-point

ipv6 address xxxx:x
frame-relay interface dlci 102

nt ser 0/0.13 point-to-point

ipv6 address xxxx:x
frame-relay interface dlci 102

spokes

int ser 0/0
enacap fr

int ser 0/0.21 point-to-point

ipv6 address xxxx:x
frame-relay interface dlci 201

res
Paul

[Jean-Francois Gagnon]

In my book, Hub use only one IP and still gets full connectivity to all

Problem is, I don't have the config so I don't know how they do it

[paul driver]

Hello

---

@Jean-Francois Gagnon wrote:

In my book, Hub use only one IP and still gets full connectivity to all

Problem is, I don't have the config so I don't know how they do it

---

Humm...Even in IPV6 i am on the understanding depending on implementing FR on the Hub rtr on a  main interface or sub-interface you will still require either dlci mappings to each spoke or multiple p2p sub interfaces to each spoke.-  but I could be totally wrong.

I am curious to what book you are reading - can you share please -

res
Paul

[Harold Ritter]

BonsoirJean-François,

You only need mapping for the next hop addresses, so in this case the link local address of R2 and R3, since you are using ripng to propagate the routes and that ripng uses the link local addresses rather than the global ipv6 addresses as the next hop address.

Regards,

[paul driver]

Hello Harold

---

@Harold Ritter wrote:

BonsoirJean-François,

You only need mapping for the next hop addresses, so in this case the link local address of R2 and R3, since you are using ripng to propagate the routes and that ripng uses the link local addresses rather than the global ipv6 addresses as the next hop address.

---

Can you clarify that please - i may have misinterpreted it  - when you say mappings your relating that to multipoint FR dlci mappings correct? if so my understanding is they then would still require mapping to/from hub and each spoke  apposed to P2P FR which doesn't map the dlci .

res
Paul

[Harold Ritter]

Hi Paul,

Yes, that is correct. My statement was about the configuration that Jean-François is currently using on the hub site and would not be required if he was also using P2P subinterfaces at the hub site. My response was not about FR best practices either, but was more to help him fix the issue in the current config and also to explain some differences that are likely to be encountered when implementing IPv6.

Regards,

[paul driver]

Hello Harold

Thank you - very much appreciated

kind regards.

[Jean-Francois Gagnon]

So if my topology is as follow:

                     R1

                /                  \

             F-R                  F-R      

             /                           \

             R2                         R3

             \                              /

               Fa0/0                   Fa0/0

                   \                       /

                            R4

I need mappings of the IPs of R4 also?

It just doesn't make sense to me, it seems like a lot of static things.

Thanks again for your help!

[paul driver]

Hello

For FR  and to limit DLCI mappings with R1 as you hub router , i would utilize FR P2P sub-interfaces

No you don't need mappings att all and especially for r4 as that isn't FR, you just need FR P2P statements on rtr1-2-3 and then advertise rtr 4 subnets in the IGP you are using

res
Paul

[Harold Ritter]

Bonjour Jean-François,

As I indicated in my previous reply, R1 only needs mappings for R2 and R3 next hop, but not for R4. In fact, any destination behind R2 and R3 will use the mappings for R2 and R3.

You should make sure though that the mappings are for the appropriate next hops. Do a "show ipv6 route" and check what the next hops are for the ipv6 address of R4 and make sure that you add those in the mapping statements.

Regards,

Harold

[Jean-Francois Gagnon · Accepted solution]

I finally found the issue.

Mappings and config was good, I just dforgot to map the link-local address to the multipoint interface.

Thanks for all!

[Harold Ritter · Accepted solution]

Bonjour Jean-François,

I had explicitly stated that in my last post.

"You only need mapping for the next hop addresses, so in this case the link local address of R2 and R3, since you are using ripng to propagate the routes and that ripng uses the link local addresses rather than the global ipv6 addresses as the next hop address."

Regards,

[Jean-Francois Gagnon]

Yeah I haven't seen that portion.

Sorry and thanks again!