From: BUI, TIN T (SBCSI) (tb4565@xxxxxxx)
Date: Fri Jan 05 2001 - 19:32:14 GMT-3
This makes total sense as my understanding is that Router Reflectors (R1)
will advertise to Route-reflector clients and its IBGP peers. Eventhough R3
is not route-reflector client of R1, it will also receive route because it
is an IBGP peer of R1. This is put in place so that there is no restriction
that forces all IBGP routers (that are not route reflectors) in same AS to
be route reflector clients.
> Tin T. Bui
> Senior Network Manager
> Network Management Center
> SBC Services Inc.
> 7337 Trade Street, Rm 1110
> San Diego, Ca 92121
> Office #: 858-886-4644/858-886-4589
> Pager #: 858-494-0482
> Fax #: 858-549-4103
> Email: tb4565@sbc.com
>
-----Original Message-----
From: Connary, Julie Ann [mailto:jconnary@cisco.com]
Sent: Friday, January 05, 2001 2:00 PM
To: David C Prall
Cc: CCIE Lab groupstudy.com
Subject: Re: route-reflector question
Hi All,
I fully realize that route-reflectors are a replacement for fully-meshed
IBGP. Hence my question why
R3 should get the route at all since it is not internally peered to
R5 and R1 is not setup to reflect to it - although it seems to be doing
just that.
Sorry - long day - I'm going home now. If anyone has any ideas why R1 would
reflect an internal
router to another internal peer that is not a route-reflector client
---------
Julie Ann
At 04:41 PM 1/5/2001 -0500, David C Prall wrote:
>I believe this is confusion between internal and external routes. Internal
>BGP routers must be fully meshed, they only advertise their local networks
>to other internal neighbors. External BGP routes will be advertised to
other
>internal neighbors. The use of a route reflector allows internal routes to
>be advertised to other internal routers, thus relieving the need for a
fully
>meshed topology.
>
>David C Prall dcp@dcptech.com http://dcp.dcptech.com
>----- Original Message -----
>From: "Connary, Julie Ann" <jconnary@cisco.com>
>To: <ccielab@groupstudy.com>
>Sent: Friday, January 05, 2001 4:23 PM
>Subject: route-reflector question
>
>
> > Hi All,
> >
> > I have a frame-relay hub and spoke network with the hub as the
> > route-reflector server and the spokes as clients.
> >
> >
> > R3---------------EBGP ---------R4
> > | 137.20.34.3 137.20.34.4
> > | 137.20.1.3
> > |
> > IBGP
> > |
> > |
> > | 137.20.1.1 137.20.1.5 137.20..129.5 137.20.129.7
> > R1-----------------ibgp ---------R5----------EBGP ------------R7
> >
> >
> > R1 is the frame-relay hub with R3 and R5 as spokes.
> > R7 is initiating a route - 160.0.0.0/4. R4 is also intiating a route -
> > 172.168.2.0
> > I had setup R1 to route-reflect to R5, but not to R3. So I had expected
> > that R5 would see the
> > route from R4, but R3 should not see the route from R7 - as R1 should
not
> > reflect it - correct?
> >
> > On R3 I see the following:
> >
> > r3#show ip bgp
> > BGP table version is 33, local router ID is 137.20.31.65
> > Status codes: s suppressed, d damped, h history, * valid, > best, i -
>internal
> > Origin codes: i - IGP, e - EGP, ? - incomplete
> >
> > Network Next Hop Metric LocPrf Weight Path
> > *>i160.0.0.0/4 137.20.129.7 100 0 3 i
> > *> 172.168.2.0/24 137.20.34.4 0 0 1 ?
> >
> > r3#show ip bgp 160.0.0.0
> > BGP routing table entry for 160.0.0.0/4, version 33
> > Paths: (1 available, best #1)
> > 3, (aggregated by 3 170.10.10.1)
> > 137.20.129.7 (metric 84) from 137.20.1.1 (137.20.129.5)
> > Origin IGP, localpref 100, valid, internal, atomic-aggregate,
best
> > Originator : 137.20.129.5, Cluster list: 137.20.11.33
> >
> > With a BGP setup of:
> >
> > !
> > router bgp 2
> > neighbor 137.20.1.1 remote-as 2
> > neighbor 137.20.34.4 remote-as 1
> >
> >
> > On R1 I see:
> >
> > r1#show ip bgp
> > BGP table version is 25, local router ID is 137.20.11.33
> > Status codes: s suppressed, d damped, h history, * valid, > best, i -
>internal
> > Origin codes: i - IGP, e - EGP, ? - incomplete
> >
> > Network Next Hop Metric LocPrf Weight Path
> > *>i160.0.0.0/4 137.20.129.7 100 0 3 i
> > *>i172.168.2.0/24 137.20.34.4 0 100 0 1 ?
> >
> > r1#show ip bgp 160.0.0.0
> > BGP routing table entry for 160.0.0.0/4, version 25
> > Paths: (1 available, best #1, advertised over IBGP)
> > 3, (aggregated by 3 170.10.10.1), (Received from a RR-client)
> > 137.20.129.7 (metric 84) from 137.20.1.5 (137.20.129.5)
> > Origin IGP, localpref 100, valid, internal, atomic-aggregate,
best
> >
> > with a bgp setup of:
> >
> > router bgp 2
> > no synchronization
> > neighbor 137.20.1.3 remote-as 2
> > neighbor 137.20.1.5 remote-as 2
> > neighbor 137.20.1.5 route-reflector-client
> > neighbor 137.20.12.2 remote-as 2
> > neighbor 137.20.12.2 route-reflector-client
> >
> > (note that 137.20.1.3 is not a route-reflector-client.)
> >
> > On R5 I have:
> >
> > r5#show ip bgp
> > BGP table version is 3, local router ID is 137.20.129.5
> > Status codes: s suppressed, d damped, h history, * valid, > best, i -
>internal
> > Origin codes: i - IGP, e - EGP, ? - incomplete
> >
> > Network Next Hop Metric LocPrf Weight Path
> > *> 160.0.0.0/4 137.20.129.7 0 3 i
> > *>i172.168.2.0/24 137.20.34.4 0 100 0 1 ?
> > r5#show ip bgp 160.0.0.0
> > BGP routing table entry for 160.0.0.0/4, version 3
> > Paths: (1 available, best #1, advertised over IBGP)
> > 3, (aggregated by 3 170.10.10.1)
> > 137.20.129.7 from 137.20.129.7 (170.10.10.1)
> > Origin IGP, valid, external, atomic-aggregate, best
> >
> > with a setup of:
> >
> > router bgp 2
> > no synchronization
> > neighbor 137.20.1.1 remote-as 2
> > neighbor 137.20.129.7 remote-as 3
> > !
> >
> >
> >
> > any ideas why R3 sees this route? Is this correct?
> >
> > thanks,
> >
> > Julie Ann
> >
> >
> > ------------------------------------------------------------------------
> > Julie Ann Connary
> > | | Network Consulting Engineer
> > ||| ||| Federal Support Program
> > .|||||. .|||||. 13635 Dulles Technology
Drive,
> > Herndon VA 20171
> > .:|||||||||:.:|||||||||:. Pager: 1-888-642-0551
> > c i s c o S y s t e m s Email: jconnary@cisco.com
> >
> > ------------------------------------------------------------------------
This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 10:27:23 GMT-3