From: Chris Riling (criling@gmail.com)
Date: Fri Nov 02 2007 - 20:16:20 ART
I'm still not sure that I entirely understand your question, but I'll take
another crack at it. Are you saying in a multipoint frame relay environment,
where you have frame relay map statements, you've seen some people add
mappings to link locals for things on the far end of the network (other
spokes), instead of just the other side of a link local segment? If this is
what you're asking, you're correct, there would be no need, for a couple of
different reasons:
1. First and foremost, IPv6 link local addresses are just that, local to
that particular link... they have no relevance on any other segment
2. In your topology, where R1 would be the hub in a multipoint environment,
all traffic would have to flow through the hub anyway, since there is no
connection between the spokes. This isn't to say that you couldn't make a
mapping to another spoke's global unicast address via the local DLCI to the
hub, though...
Make sense?
Anyone else? Am I missing something?
Thanks,
Chris
On 11/2/07, sirus MOGHADASIAN <cyrus.mgh@gmail.com> wrote:
>
> Thanks all for answering but I think u misunderstand my question,
>
> My qustion was not why we use link-local address in ipv6 , the question is
> why we need additional ipv6 link-local address,for example link-local
> address of R2 on R5?? I said I think if I map R1 link local address on R5
> ,every think will be correct but why in some workbooks do additional mapping
> (R2 link local mapping in R5?? according to the fact that there is no need
> that R5 and R2 become neighbor.
>
>
>
> Cyrus
>
> On 11/1/07, Chris Riling <criling@gmail.com> wrote:
>
> > Yep... think about it... the v6 routing protocols use it for next hop,
> > and
> > you could have two v6 global unicast addresses on each side of a p2p
> > link
> > that are in different subnets, and it *should* still work (haven't
> > gotten a
> > chance to lab it yet), because it uses link locals for communication -
> > it
> > took me a little while to get used to that... :)
> >
> > Chris
> >
> >
> > On 10/31/07, Joseph Saad < joseph.samir.saad@gmail.com> wrote:
> > >
> > > Also, to establish neighbors on NBMA networks, you'll need to refer to
> > the
> > > neighbor's ipv6 link-local address. Which is, by the way, an
> > > interface-level
> > > command.
> > >
> > > On 11/1/07, Chris Riling <criling@gmail.com> wrote:
> > > >
> > > > The v6 protocols seem to like to use the link local addresses for
> > next
> > > > hop,
> > > > so I'd make sure there's a way for resolve the link locals... Take a
> > > look
> > > > at
> > > > the v6 routing table when you get OSPF up and you'll see why you
> > need
> > > > it...
> > > >
> > > > Chris
> > > >
> > > > On 10/31/07, hadek.el-ayachi@nsn.com <hadek.el-ayachi@nsn.com >
> > wrote:
> > > > >
> > > > > You need this when running ospf in hub-and-spoke topology using
> > > > > broadcast or non-broacast network type. Match like ipv4.
> > > > >
> > > > >
> > > > > -----Original Message-----
> > > > > From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On
> > Behalf
> > > Of
> > > > > ext sirus MOGHADASIAN
> > > > > Sent: mercredi 31 octobre 2007 14:44
> > > > > To: Cisco certification
> > > > > Subject: is it necessary to map L3 to L2 add of IPv6 in
> > frame-relay
> > > > > spoke-to-spoke connection?
> > > > >
> > > > > Hi Gs,
> > > > >
> > > > > R5 ------R1---------R2
> > > > >
> > > > > R1 ,R2 ,R5 run frame-relay on their main serial interface.
> > > > >
> > > > > I ran Ipv6 and Ripng on them.
> > > > >
> > > > > I have v6 routes is each routers table
> > > > >
> > > > > R5:
> > > > >
> > > > > interface Serial1/0
> > > > >
> > > > > ipv6 address 2001:CC1E:1:125::5/64
> > > > > ipv6 address FE80::5 link-local
> > > > > ipv6 rip 1 enable
> > > > > frame-relay map ipv6 FE80::CE00:15FF:FE2C:0 501
> > broadcast frame-relay
> > > > > map ipv6 2001:CC1E:1:125::1 501 broadcast frame-relay map ipv6
> > > > > 2001:CC1E:1:125::2 501 broadcast
> > > > >
> > > > >
> > > > >
> > > > > R1 :
> > > > >
> > > > > interface Serial1/0
> > > > > ipv6 address 2001:CC1E:1:125::1/64
> > > > > ipv6 address FE80::1 link-local
> > > > > ipv6 rip 1 enable
> > > > > frame-relay map ipv6 FE80::CE01:15FF:FE2C:0 102
> > broadcast frame-relay
> > > > > map ipv6 FE80::CE04:15FF:FE2C:0 105 broadcast frame-relay map
> > ipv6
> > > > > 2001:CC1E:1:125::2 102 broadcast frame-relay map ipv6
> > > > > 2001:CC1E:1:125::5 105 broadcast
> > > > >
> > > > >
> > > > > R2 :
> > > > >
> > > > > interface Serial1/0
> > > > > ipv6 address 2001:CC1E:1:125::2/64
> > > > > ipv6 address FE80::2 link-local
> > > > > ipv6 rip 1 enable
> > > > > frame-relay map ipv6 FE80::CE00:15FF:FE2C:0 201
> > broadcast frame-relay
> > > > > map ipv6 2001:CC1E:1:125::1 201 broadcast frame-relay map ipv6
> > > > > 2001:CC1E:1:125::5 201 broadcast
> > > > >
> > > > >
> > > > >
> > > > > Question is : Is there any nedd to L3 to L2 link-local mapping for
> > R2
> > > on
> > > > > R5 s1/0 ??? I see this solution with L3 to L2 mapping in some
> > > > > workbooks.
> > > > >
> > > > > As I said every thing is right ,I think this L3->L2 mapping will
> > be
> > > > > redundant.
> > > > >
> > > > >
> > > > > any idea?
> > > > >
> > > > >
> > > > > Cyrus
> > > > >
> > > > >
> > >
> > _______________________________________________________________________
> > > > > Subscription information may be found at:
> > > > > http://www.groupstudy.com/list/CCIELab.html
> > > > >
> > > > >
> > >
> > _______________________________________________________________________
> > > > > Subscription information may be found at:
> > > > > http://www.groupstudy.com/list/CCIELab.html
> > > >
> > > >
> > _______________________________________________________________________
> > > > Subscription information may be found at:
> > > > http://www.groupstudy.com/list/CCIELab.html
> > >
> > >
> > _______________________________________________________________________
> > > Subscription information may be found at:
> > > http://www.groupstudy.com/list/CCIELab.html
> >
> > _______________________________________________________________________
> > Subscription information may be found at:
> > http://www.groupstudy.com/list/CCIELab.html
This archive was generated by hypermail 2.1.4 : Sat Dec 01 2007 - 06:37:27 ART