From: Marko Milivojevic (markom@vodafone.is)
Date: Wed Jun 06 2007 - 16:27:22 ART
Aha, you didn9t mention two BGP speakers in AS2. That complicates things a
little bit... Because announcemet from R1 to R2 is iBGP and if you wish to
redistribute that into OSPF, it is of course needed to configure BGP process
correctly. Note however that this is not very good idea.
From: CCIE GOAL <ccie.goal@gmail.com>
Date: Wed, 6 Jun 2007 14:39:21 -0300
To: Marko Milivojevic <markom@vodafone.is>
Cc: Cisco certification <ccielab@groupstudy.com>
Subject: Re: eBGP redistribution
No, I am not using confederations... it is a simple scenario
Follow the "show ip bgp" outputs.
AS1#sh ip bgp | b Net
Network Next Hop Metric LocPrf Weight Path
*> 10.10.0.0/24 <http://10.10.0.0/24> 2.2.2.2 <http://2.2.2.2>
0 2 3 i
*> 10.10.1.0/24 <http://10.10.1.0/24> 2.2.2.2 <http://2.2.2.2>
0 2 3 i
*> 10.10.2.0/24 <http://10.10.2.0/24> 2.2.2.2 <http://2.2.2.2>
0 2 3 i
*> 10.10.3.0/24 <http://10.10.3.0/24> 2.2.2.2 <http://2.2.2.2>
0 2 3 i
*> 10.10.4.0/24 <http://10.10.4.0/24> 2.2.2.2 <http://2.2.2.2>
0 2 3 i
*> 10.10.5.0/24 <http://10.10.5.0/24> 2.2.2.2 <http://2.2.2.2>
0 2 3 i
*> 10.10.6.0/24 <http://10.10.6.0/24> 2.2.2.2 <http://2.2.2.2>
0 2 3 i
AS1#
-----------------------------------------------------------------
AS2 (Router1)#sh ip bgp | b Net
Network Next Hop Metric LocPrf Weight Path
*> 10.10.0.0/24 <http://10.10.0.0/24> 3.3.3.3 <http://3.3.3.3>
0 0 3 i
*> 10.10.1.0/24 <http://10.10.1.0/24> 3.3.3.3 <http://3.3.3.3>
0 0 3 i
*> 10.10.2.0/24 <http://10.10.2.0/24> 3.3.3.3 <http://3.3.3.3>
0 0 3 i
*> 10.10.3.0/24 <http://10.10.3.0/24> 3.3.3.3 <http://3.3.3.3>
0 0 3 i
*> 10.10.4.0/24 <http://10.10.4.0/24> 3.3.3.3 <http://3.3.3.3>
0 0 3 i
*> 10.10.5.0/24 <http://10.10.5.0/24> 3.3.3.3 <http://3.3.3.3>
0 0 3 i
*> 10.10.6.0/24 <http://10.10.6.0/24> 3.3.3.3 <http://3.3.3.3>
0 0 3 i
AS2#
-----------------------------------------------------------------
AS2 (Router2)#sh ip bgp | b Net
Network Next Hop Metric LocPrf Weight Path
*> 10.10.0.0/24 <http://10.10.0.0/24> 3.3.3.3 <http://3.3.3.3>
0 0 3 i
*> 10.10.1.0/24 <http://10.10.1.0/24> 3.3.3.3 <http://3.3.3.3>
0 0 3 i
*> 10.10.2.0/24 <http://10.10.2.0/24> 3.3.3.3 <http://3.3.3.3>
0 0 3 i
*> 10.10.3.0/24 <http://10.10.3.0/24> 3.3.3.3 <http://3.3.3.3>
0 0 3 i
*> 10.10.4.0/24 <http://10.10.4.0/24> 3.3.3.3 <http://3.3.3.3>
0 0 3 i
*> 10.10.5.0/24 <http://10.10.5.0/24> 3.3.3.3 <http://3.3.3.3>
0 0 3 i
*> 10.10.6.0/24 <http://10.10.6.0/24> 3.3.3.3 <http://3.3.3.3>
0 0 3 i
AS2#
-----------------------------------------------------------------
AS3#sh ip bgp vpnv4 vrf AS3 | b Net
Network Next Hop Metric LocPrf Weight Path
Route Distinguisher: 2:1 (default for vrf AS3)
*> 10.10.0.0/24 <http://10.10.0.0/24> 0.0.0.0 <http://0.0.0.0>
0 32768 i
*> 10.10.1.0/24 <http://10.10.1.0/24> 0.0.0.0 <http://0.0.0.0>
0 32768 i
*> 10.10.2.0/24 <http://10.10.2.0/24> 0.0.0.0 <http://0.0.0.0>
0 32768 i
*> 10.10.3.0/24 <http://10.10.3.0/24> 0.0.0.0 <http://0.0.0.0>
0 32768 i
*> 10.10.4.0/24 <http://10.10.4.0/24> 0.0.0.0 <http://0.0.0.0>
0 32768 i
*> 10.10.5.0/24 <http://10.10.5.0/24> 0.0.0.0 <http://0.0.0.0>
0 32768 i
*> 10.10.6.0/24 <http://10.10.6.0/24> 0.0.0.0 <http://0.0.0.0>
0 32768 i
AS3#
-----------------------------------------------------------------
Regards,
Luciano
2007/6/6, Marko Milivojevic <markom@vodafone.is>:
>
> Redistribute-internal was my initial gut response to this problem, but then
I
> decided to write up the solution that is not using it. Are you using BGP
> confederations? Can you send us "sh ip bgp" output from all BGP speakers?
>
>
> ________________________________
>
> From: CCIE GOAL [mailto:ccie.goal@gmail.com]
> Sent: 6. jznm 2007 16:48
> To: Marko Milivojevic
> Cc: Cisco certification
> Subject: Re: eBGP redistribution
>
>
> Hey Marko, many thanks for the reply!
>
> Nice check list you have done, and I am conforming with it...
> Follow the anwers, and bellow is the Problem and Solution... now I am
looking
> for the reason...
>
>
-----------------------------------------------------------------------------
-
> ------------------------
> 1- AS2 and AS3 need to have directly connected peering
> - Otherwise you need to redistribute that routing protocol into AS2 OSPF
>
> A: Ok, they are directly connected.
>
> 2- AS2-AS3 border router in AS2 needs to redistribute prefixes learned from
> AS3 BGP into OSPF
>
> A: Ok, there is redistribution from BGP 2 into OSPF
>
> 3- AS1-AS2 need to have multihop eBGP peering
>
> A: Ok, It is configured...
>
> 4- AS2 OSPF router connected to AS1 needs to inject AS1 prefixes into OSPF
> - it may also filter routes propagating to AS2, as those routes would be
> learned in AS1 via BGP
> - it must not filter out AS2 BGP's peering address!
>
> A: Actually the OSPF is announcing a default-route to AS1 OSPF
domain...
>
-----------------------------------------------------------------------------
-
> ------------------------
>
>
> The Problem:
> The AS2 was receiving the prefixes from AS3 and then, for some reason, the
AS2
> is treating the AS3 prefixes as internal and not external...
>
> When I tried to redistribute the BGP 2 into OSPF the AS3 prefixes could not
be
> sent as they are treated as internal...
>
> The Solution:
> To solve this problem I had to add the "bgp redistribute-internal"...
>
> The "Why?":
> Why the AS3 prefixes is treated as internal in AS2?
> If someone knows, can please explain me that?
>
> Best Regards,
> Luciano
>
>
>
>
> 2007/6/6, Marko Milivojevic <markom@vodafone.is>:
>
>
> As usual, I may be wrong, but with the information provided, I think
> the text below will summarize what needs to be done to make this work.
>
> You will need to make sure few things:
>
> 1- AS2 and AS3 need to have directly connected peering
> - Otherwise you need to redistribute that routing protocol into AS2
> OSPF
> 2- AS2-AS3 border router in AS2 needs to redistribute prefixes
learned
> from AS3 BGP into OSPF
> 3- AS1-AS2 need to have multihop eBGP peering
> 4- AS2 OSPF router connected to AS1 needs to inject AS1 prefixes
into
> OSPF
> - it may also filter routes propagating to AS2, as those routes
would
> be learned in AS1 via BGP
> - it must not filter out AS2 BGP's peering address!
>
> Alternative approach to this problem is to use MPLS in AS2, but then
> again, I'm preparing for SP lab and that may cloud my judgement here :-)
>
> There may be other alternatives, but to present them, we may need
> better description of the problem, as well as "sh ip bgp" outputs from all
> involved routers, to begin with.
>
>
> -----Original Message-----
> From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf
Of
> CCIE GOAL
> Sent: 6. jznm 2007 15:42
> To: ccielab@groupstudy.com
> Subject: eBGP redistribution
>
> Hi guys,
>
> I ran into a problem with BGP, maybe it's simple to resolve but I
got
> stuck
> on it.
> People say that sometimes it is easier to spot the issue from
outside
> the
> picture...
>
> Scenario:
>
> [ BGP AS 1 ] -- (OSPF/not in bgp) -- [ BGP AS 2 ] -- [ BGP AS 3
]
>
> - AS 1 is peering to AS 2
> - AS 2 is peering to AS 3
>
> - Between AS 1 and AS 2; there is some devices in OSPF domain only
(Not
> BGP).
>
> - AS 3 is announcing some prefixes, let's say 10.10.X.0/24...
>
> These prefixes are pingable from AS 2 but not from AS 3 even the BGP
> table
> from AS 3 has these prefix as valid.
>
> The issue is that the OSPF domain do not know how to reach the
prefixes
> 10.10.X.0/24.
>
> I am not allowed to use static routes anywhere...
> I have tried to redistribute BGP 2 into OSPF but no success, it
seems
> that
> eBGP learned prefixes are not redistributed...
>
> Can somebody help me on this?
>
> Follow the configs:
>
>
> ---------------------------------------------------
> AS1#
> router bgp 1
> no synchronization
> neighbor 2.2.2.2 <http://2.2.2.2> remote-as 2
> no auto-summary
> !
> router ospf 1
> network 10.10.1.1 <http://10.10.1.1> 0.0.0.0 <http://0.0.0.0> area
1
>
> ---------------------------------------------------
> OSPF#
> router ospf 1
> network 10.10.1.2 <http://10.10.1.2> 0.0.0.0 <http://0.0.0.0> area
1
> network 10.10.2.2 <http://10.10.2.2> 0.0.0.0 <http://0.0.0.0> area
0
>
> ---------------------------------------------------
> AS2#
> router bgp 2
> no synchronization
> neighbor 1.1.1.1 <http://1.1.1.1> remote-as 1
> neighbor 3.3.3.3 <http://3.3.3.3> remote-as 3
> no auto-summary
> !
> router ospf 1
> network 10.10.2.1 <http://10.10.2.1> 0.0.0.0 <http://0.0.0.0> area
0
> ---------------------------------------------------
>
> AS3#
> router bgp 3
> no synchronization
> neighbor 2.2.2.2 <http://2.2.2.2> remote-as 2
> no auto-summary
> ---------------------------------------------------
>
>
> Thanks in advance,
> Luciano
>
>
This archive was generated by hypermail 2.1.4 : Sun Jul 01 2007 - 17:24:47 ART