From: Joel Amao (femmy79@hotmail.com)
Date: Fri Nov 23 2007 - 13:13:03 ART
Hello Gops,
the ability to use this command was changed in 12.3(14). The reason given as
far as i know was to avoid misconfiguration in the field, most especially the
creation of an extra update-group when the command is used (even though ebgp
changes the next-hop automatically).
You will be suprised that a lot of people still like doing things manually
even when they know the task is already being done automatically by IOS :-).
You can always use route-map to change this in your case or like i said
earlier, configure a mapping for all your spokes so that you have full
reachability.
thanks,
Joel Amao
CCIE#18128
Subject: RE: BGP Next-HopDate: Fri, 23 Nov 2007 21:19:09 +0530From:
gopal.gupta@hp.comTo: femmy79@hotmail.com; ccielab@groupstudy.com
Thanks for your inputs,
Till here its fine, but i would like to know why we cant we use neighbor
x.x.x.x next-hop-self command in this case, why to set next-hop manually; this
is waht my concern is.
Thanks
Gops
From: Joel Amao [mailto:femmy79@hotmail.com] Sent: Friday, November 23, 2007
21:14To: Gupta, Gopal (NWCC); Cisco certificationSubject: RE: BGP Next-Hop
Hello Gupta, what you are seeing is normal and expected. There is an exception
to egbg next hop modification in broadcast and NBMA mode called third-party
next hop modification. The normal rule regarding next-hop modification is
below; 1. If the route is locally originated set the NEXT-HOP to the
peering ip address (even if sent to an IBGP peer) 2. If the route is
propagated to an eBGP peer, set the NEXT-HOP to our peering ip address 3. If
the route is propagated to an iBGP peer, do not change the NEXT-HOP
**exception to this rule **When two peers are in a multiaccess network (Frame
Relay multipoint interface, Ethernet, etc) we dont change the next-hop.
thanks, Joel AmaoCCIE#18128 > Subject: BGP Next-Hop> Date: Fri, 23 Nov
2007 19:33:56 +0530> From: gopal.gupta@hp.com> To: ccielab@groupstudy.com> >
Hi Folks,> > I am facing issue with BGP..> > R1 , R2, R3 are spokes connected
to hub R5 throgh Frame relay .... all> with the major interface.> having
subnet 155.1.0.0/24 and all are in their own individual AS ; R1> is in AS1, R2
in 2, R3 in 3, R5 in 5.> > when routes originated by R2 are sent to R5 and
then R5 in turn sends> that route to R3 or R1, the next hop doesn't change
which should be> changed coz of EBGP connections and should not be changed coz
of> Frame-relay NBMA on same subnet.> > and if i try to set next-hop-self
under neighbor stement on R5 for> R1,R2&R3..i get an error message> >
R5(config-router)#neighbor 155.1.0.1 next-hop-self> %BGP: next-hop-self is
implied for an eBGP peer, there is no need to> configure it>
R5(config-router)#neighbor 155.1.0.2 next-hop-self> %BGP: next-hop-self is
implied for an eBGP peer, there is no need to> configure it>
R5(config-router)#neighbor 155.1.0.3 next-hop-self> %BGP: next-hop-self is
implied for an eBGP peer, there is no need to> configure it> > I know i can
set it manually throgh Route-map, is this the only way we> can have workaround
or i am missing something.....not sure.> > Any Comments please !!!!!!> > Best
Regards,> Gops> >
This archive was generated by hypermail 2.1.4 : Sat Dec 01 2007 - 06:37:31 ART