From: Hobbs (deadheadblues@gmail.com)
Date: Sun Sep 21 2008 - 13:19:59 ART
yeah, thats sounds right. i forgot about R3 withdrawing its advertisement.
thats what i get for thinking, and not labbing
On Sun, Sep 21, 2008 at 9:57 AM, ccie2k4@gmail.com <ccie2k4@gmail.com>wrote:
> When R2 advertises the prefix to R3 with a higher local preference than R3
> will withdraw its update since the prefix received from R2 will become the
> best path.
>
> On a separate note in a outbound route-map the local preference on the
> local router does not get affected. Only the neighbor towards which the
> route-map has been applied will see the higher local preference. So say its
> R1 and R2 and R1 sends a higher local preference to R2 via an outbound
> route-map than R2 will see the higher local preference from R1. The best
> path will now change on R2 and it will withdraw its original advertisement.
>
> Atleast that's the way I think this is working. Please correct if
> incorrect. Thanks
>
> On Sat, Sep 20, 2008 at 11:41 PM, Hobbs <deadheadblues@gmail.com> wrote:
>
>> If the value is what it is supposed to be, then I guess it will work. I
>> was just bringing up this possibility:
>>
>> Suppose R2 and R3 are in AS65000 and each have a learned route through
>> AS100.
>> R2 sets local pref to 200
>> R3 sets local pref to 300.
>> R2 and R3 are ibgp neighbors so R2 prefers the path through R3.
>>
>> You want to prefer R2 and you think you can set outbound localpref on
>> R2->R3 to accomplish this.
>>
>> This wont affect R2's local pref, only R3's....right?
>>
>> So R3 will have higher local pref (400>300) pointing to R2
>> and R2 will have higher local pref (300>200) pointing to R3.
>>
>> So the packet bounces between R2-R3. This was my theory, I haven't tried
>> it.
>>
>>
>> On Sat, Sep 20, 2008 at 9:06 PM, ccie2k4@gmail.com <ccie2k4@gmail.com>wrote:
>>
>>> I tried using it by setting it in an outbound route-map and I could see
>>> no
>>> difference. does it seem incorrect?
>>>
>>>
>>> Blogs and organic groups at http://www.ccie.net
>>>
>>> _______________________________________________________________________
>>> Subscription information may be found at:
>>> http://www.groupstudy.com/list/CCIELab.html
Blogs and organic groups at http://www.ccie.net
This archive was generated by hypermail 2.1.4 : Sat Oct 04 2008 - 09:26:19 ART