ISATAP is for dual stack hosts, ISATAP has a 64 bit IPv6 prefix and a 64
bit host portion created in EUI format. It is intended for v6 transport
over a v4 infrastrucutre, so the routers adjacent to the v6 hosts do the
encapsulation/decapsulation of v6 packets in to v4 tunnels.
Intermediate routers do not need to be configured with ISATAP
With ISATAP you have to re-enable v6 router advertisements on the tunnel
interface to allow for client auto-configuration for handling multiple
hosts.

Have you tried to configure this and had problems?

Chris

-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of
Curt Girardin
Sent: Monday, September 12, 2005 8:33 PM
To: ccielab@groupstudy.com
Subject: IPv6 ISATAP Tunnels

Hi,

I'm having a really difficult time understanding ipv6 ISATAP tunnels;
although I understand IPv6 manual tunnels, IPv6 GRE tunnels, and Ipv6
6-to-4 auto tunnels.

I've read and re-read the cisco documentation and it seems very
high-level:

Few details on who is supposed to do encapsulation/decapsulation at each
end.
No details on how that embedded IPv4 address is used.  Is this the next
hop?  A tunnel endpoint?  Do all routers in-between also need to be
configured for ISATAP?
If the embedded IPv4 address is a tunnel endpoint, then how do you
handle multiple hosts at each site?
Is this really designed for dual-stack end-hosts?

Am I making this more complicated than it needs to be?  I've also read
the IETF draft, and that's not much help either.

If anyone has a better explanation of this type of tunnel (along with
ipv4-compatable), I would very much appreciate it.

Here is the documentation I've read thus far:

http://www.cisco.com/univercd/cc/td/doc/product/software/ios123/123cgcr/
ipv6_c/sa_tunv6.htm

http://www.ietf.org/internet-drafts/draft-ietf-ngtrans-isatap-24.txt

Thank you,

Curt

---


To add to Brians post... Make sure you filter out the routes which you
learn back from the hub you already know about.... IOW, manual
split-hozizon.

This is necessary because ipv6 split-horizon is a global proposition to
the router, not per link.

HTH

andy

-----Original Message-----
From: Brian McGahan [mailto:bmcgahan@internetworkexpert.com]
Sent: Thursday, September 22, 2005 9:47 AM
To: Easyman; GroupStudy CCIE-Lab
Subject: RE: RIPng on NBMA issue


Easyman,

       Split horizon for RIPng is controlled under the RIPng process
with the "split-horizon" command, not at the interface level like in
IPv4.


----


Why would you need to do this?  The metric will increment when the route
is received back automatically.

---

Subject: RE: IPV6 addressing
Stefan,

If you are given the whole address then use it, if not, add the eui-64
keyword at the end which will make up the rest of the address using part of
the mac.

Pretty straight forward really, for example if you have
2001:CC1E:192:10:1::254/128 this is the subnet and the host address and does
not require a supplement from the mac so you would not use the eui part.

If I had the address 2001:CC1E:192:10:1::/64 eui  then this is only the
subnet and would need supplementing with the mac to make up a full address.

Does that help you?

Regards

Lee.



-----Original Message-----
From: Stefan Grey [mailto:examplebrain@hotmail.com]
Sent: 12 October 2005 09:46
To: pv.ryan@gmail.com; ccielab@groupstudy.com
Subject: Re: IPV6 addressing

Sure, but how to understand from the task which address type to use??
ipv6 address 2080:1:1:43::3/64

or ipv6 address 2080:1:1:43:3/64 eui-64???

----

The rule is that you map the link-local address with the broadcast
keyword, and you map the global address sans the global keyword.

Routing protocols use the link-local address for adjacencies.  You only
need to map the global address if you want to be able to ping it.

-Dave

On Wed, 12 Oct 2005, Stefan Grey wrote:

> Hello,
> Question:
>
> When generally configuring IPv6 over frame relay should we just map the
> global unicast address of the remote side to the dlci. (Which is configured
> using ipv6add/64 command)

----
Subject: RE: IPV6 Question!
Ipv6 prefix-list DEFAULT ::/0 le 128

Same #$*&# Different Protocol

-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of
Victor Cappuccio
Sent: Thursday, October 27, 2005 4:53 PM
To: ccielab@groupstudy.com
Subject: IPV6 Question!

Hello all,

Sorry for this silly post, But there goes..

Is something similar to ip prefix-list DEFAULT 0.0.0.0/0 le 32 in IPV6 ?

My Question is because is you are required to filter some traffic and you
are not allowed to use the  ipv6 traffic-filter access-list-name {in | out}
and if running certain IGP like rip maybe the distribute-list could help me

---