From: Xu, James (james.xu@xxxxxxx)
Date: Sun Jul 01 2001 - 13:55:18 GMT-3
Richard:
Thanks for the info.
James
-----Original Message-----
From: Richard Gallagher [mailto:rgallagh@cisco.com]
Sent: Saturday, June 30, 2001 5:30 AM
To: Xu, James; 'Richard Gallagher'; Xu, James; 'Monty.Majszak@Level3.com
'; tr1525@sbc.com; ccielab@groupstudy.com
Subject: Re: BGP Route-Reflectors and Next-Hop-Self
James,
No new commands, just making use of existing ones. You use an outbound
router
map the change the next hop:
neighbor x.x.x.x route-reflector-client
neighbor x.x.x.x route-map rr-out out
route-map rr-out
match ip adress 101
set ip next-hop x.x.x.x
I think you get the jist!!
Rich
On Jun 29, 8:40pm, Xu, James chatted about:
> Subject:RE: BGP Route-Reflectors and Next-Hop-Self
> Richard:
>
> Thanks for the info. Have you tried this new "knobs" yet? I didn't have
> 12.1.6 IOS in our lab yet. What is the command syntax?
>
> ==================================================================
> Release Notes
>
> Adds knobs to set/not-set the nexthop.
> 1. Ability in RR to set BGP attribute - outbound route map
> 2. Ability for a eBGP-multihop peer to propagate the nexthop unchanged.
> Note that CSCdr80335 should be paired with CSCdu02357.
>
===========================================================================
>
> James
>
> -----Original Message-----
> From: Richard Gallagher [mailto:rgallagh@cisco.com]
> Sent: Friday, June 29, 2001 2:35 PM
> To: Xu, James; 'Monty.Majszak@Level3.com '; tr1525@sbc.com;
> ccielab@groupstudy.com
> Subject: Re: BGP Route-Reflectors and Next-Hop-Self
>
>
> For those of you that want to change the next-hop on an RR IPG connection,
> this
> feature is available in 12.1(6)
>
> Check out CSCdr80335 on CCO.
>
> Rich
> CCIE#7211
>
> On Jun 29, 6:23pm, Xu, James chatted about:
> > Subject:RE: BGP Route-Reflectors and Next-Hop-Self
> > Monty:
> >
> > The route-reflector tells the IBGP client that the next hop for the EBGP
> > routes is ITS PEERING ADDRESS with the ibgp client, unless the command
> "nei
> > x.x.x.x update-source lo0" is used, in that case, the lo0 address will
be
> > used for the next-hop-address.
> >
> > James Xu
> >
> > -----Original Message-----
> > From: Monty.Majszak@Level3.com [mailto:Monty.Majszak@Level3.com]
> > Sent: Friday, June 29, 2001 9:19 AM
> > To: tr1525@sbc.com; ccielab@groupstudy.com
> > Subject: RE: BGP Route-Reflectors and Next-Hop-Self
> >
> >
> > Thanks Tim. Let me "try" to clarify what i'm trying to say. Can we agree
> > that any EBGP routes be default will have the next-hop attribute set to
> it's
> > IP address that the IBGP neighbor is peering with? That being said, the
> > next-hop-self statement is used within the IBGP because (as stated by
Bob
> > Dixon) the IGP withing your IBGP doesn't know how to reach the EBGP
> > neighbor's address(dang, gettin' funky w/ the acronyms). So the
> > route-reflector tells the clients that the next hop for the EBGP route
is
> > it's Loopback, which all the other internal neighbors know how to reach.
> Not
> > sure if that clears this up, doesn't hurt to try though....
> >
> >
> >
> > -----Original Message-----
> > From: RANGE, TIM (SBCSI) [mailto:tr1525@sbc.com]
> > Sent: Friday, June 29, 2001 7:05 AM
> > To: Majszak, Monty; ccielab@groupstudy.com
> > Subject: RE: BGP Route-Reflectors and Next-Hop-Self
> >
> >
> >
> > Well, I'd say you are somebody Monty... Maybe not just anybody! But
> > anyway...
> >
> > I follow and agree with all of your statements here... My concern is
that
> > under normal peering conditions in an iBGP mesh, internals are NOT
shared
> > with other iBGP peers. That is the whole point of a RR... Therefore, my
> only
> > question is whether or not an RR can modify the attributes of the routes
> > that he "Reflects"... I hope I have clarified the question here.
> >
> > Thoughts ???
> >
> > Thanx, Tim
> >
> > -----Original Message-----
> > From: Monty.Majszak@Level3.com [mailto:Monty.Majszak@Level3.com]
> > Sent: Friday, June 29, 2001 7:55 AM
> > To: RANGE, TIM (SBCSI); ccielab@groupstudy.com
> > Subject: RE: BGP Route-Reflectors and Next-Hop-Self
> >
> >
> > I'm still cracking up over my last sentence in my previous response
> "Correct
> > me if I'm anybody".
> >
> > By default, IBGP does not reset the "next-hop" attribute, EBGP does. So,
> if
> > you have a large network and an EBGP peer advertising routes to that
> > network, you'd see the next-hop as the peering address of the EBGP
> neighbor,
> > unless, if your using the "neighbor x.x.x.x next-hop-self" command on
your
> > IBGP, you'd see the next hop as the edge router peering with the EBGP
> > neighbor.
> >
> >
> >
> >
> > -----Original Message-----
> > From: RANGE, TIM (SBCSI) [mailto:tr1525@sbc.com]
> > Sent: Friday, June 29, 2001 6:41 AM
> > To: Majszak, Monty; ccielab@groupstudy.com
> > Subject: RE: BGP Route-Reflectors and Next-Hop-Self
> >
> >
> > Monty,
> >
> > I am actually seeing the following;
> >
> > iBGP routes on the RR are sent to RR-client with Next-Hop unaltered.
> > eBGP routes on the RR are sent to RR-client with Next-Hop altered.
> >
> > Tim
> >
> > -----Original Message-----
> > From: Monty.Majszak@Level3.com [mailto:Monty.Majszak@Level3.com]
> > Sent: Friday, June 29, 2001 7:26 AM
> > To: RANGE, TIM (SBCSI); ccielab@groupstudy.com
> > Subject: RE: BGP Route-Reflectors and Next-Hop-Self
> >
> >
> > With the below config, all routes advertised to 192.168.171.148 from
this
> > router would see the next hop for those routes as the Loopback of this
> > router.
> >
> > -Correct me if I'm anybody.
> >
> >
> >
> > -----Original Message-----
> > From: RANGE, TIM (SBCSI) [mailto:tr1525@sbc.com]
> > Sent: Friday, June 29, 2001 6:03 AM
> > To: ccielab
> > Subject: BGP Route-Reflectors and Next-Hop-Self
> >
> >
> > Group,
> >
> > I was hoping I could get some assistance on a BGP Route-Reflector
> > question...
> >
> > Is it SOP for the Route-Reflector to forward all internal routes with
the
> > Next-Hop address unaltered, even with the "next-hop-self" statement? In
> the
> > following implementation, the route-reflector client is receiving all
> > RR-internal routes with the next-hop address unaltered... All RR
external
> > routes are being received at the client with the next-hop-self set
> > accordingly. I am assuming that the RR is in essence a true reflector
and
> > will NOT modify the internal routes, hence next-hop address intact.
> Correct
> > assumption??
> >
> > Thanks in advance for the assistance...
> >
> > router bgp 64512
> > neighbor 192.168.171.148 remote-as 64512
> > neighbor 192.168.171.148 route-reflector-client
> > neighbor 192.168.171.148 next-hop-self
> >
> >
> >
> > Tim Range
> > (o)314-235-0324
> > (p)314-870-1996
> > **Please read:http://www.groupstudy.com/list/posting.html
> > **Please read:http://www.groupstudy.com/list/posting.html
> > **Please read:http://www.groupstudy.com/list/posting.html
> >-- End of waffle from Xu, James
>
>
>
> --
**Please read:http://www.groupstudy.com/list/posting.html
This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 10:31:28 GMT-3