From: Leigh Harrison (ccileigh@gmail.com)
Date: Sat Oct 29 2005 - 10:29:36 GMT-3
Danny,
Yup - that's right, as the AS is passed through.
Regardless of "no-prepend" (that's only for the benefit of nodes behind
r2) - it would still be shown to r1, as would AS 2
LH
Danny Cox wrote:
>Thanks Leigh
>
> that makes sense and is more or less what I understood. Taking your
>example, this would still mean that R1 would show as paths from R2 as
>'2 3 i' I think .. agreed?
>
>cheers
>Danny
>
>On 29/10/05, Leigh Harrison <ccileigh@gmail.com> wrote:
>
>
>>Danny,
>>
>>Use local-as prepend when you are lying to a neighbour about you AS
>>number, but don't want to pass that number to anyone.
>>
>>i.e:
>>
>>R1 -- bgp1-2 -- r2 -- ibgp3-3 -- r3
>>
>>
>>R1 peers with r2 using AS1 to AS2
>>R2 is actually running in AS3 and peers with r3 using ibgp
>>
>>If you just put the normal neighbour statement on r2 to r1, then they
>>will not form a relationship, as they disagree on AS numbers. If you
>>put in the "neighbor 1.2.3.4 local-as 2" command, then they will form
>>the adjacency.
>>
>>When any routes from R1 appear in the table of R3, they will have an AS
>>path of " 1 2 i "
>>
>>Putting the command "neighbor 1.2.3.4 local-as 2 no-prepend" in will
>>ensure that r3 will only have an AS path of " 1 i "
>>
>>LH
>>
>>Danny Cox wrote:
>>
>>
>>
>>>I've had an interesting time reading about local-as. The solutions to
>>>a lab I'm doing indicate that 'local-as 60 no-prepend' should be
>>>configured on one of the routers, R6. R6 is in AS 600.
>>>
>>>My reading of what 'local-as no-prepend' does, is that it will make
>>>sure that 60 (not 600) is not prepended to routes learned *by* R6 from
>>>its ebgp peers. It doesn't stop 600 being prepended to routes learned
>>>
>>>
>>>from R6 by its peers.
>>
>>
>>>Anyone agree with me here?
>>>
>>>Many thanks!
>>>
>>>Danny
>>>
>>>_______________________________________________________________________
>>>Subscription information may be found at:
>>>http://www.groupstudy.com/list/CCIELab.html
This archive was generated by hypermail 2.1.4 : Sun Nov 06 2005 - 22:00:55 GMT-3