something strange about bgp community

Song Li refresh.lsong at gmail.com
Thu Jan 8 02:31:38 UTC 2015


Thanks!

Because there is no standard syntax on the description of BGP community, 
I think the problem is hard to understand.

在 2015/1/7 23:25, joel jaeggli 写道:
> 2914:429 is ntt's do not advertise to any peer community
>
> bgp communities are transitive attributes, e.g. you can just pass them
> to peers unmolested.  so someone that's presumably not ntt ( e.g. the
> neighbor is digital ocean) is sending that commmunity to route views as
> part of their export.
>
> Their utility in routviews depends on context, when I see communities I
> tagged on my own routes in routviews for example I can tell what pop the
> announcement  originated from which is rather useful. other's like the
> one above do tell you something about the policy of somebody on the
> internet. you can also tell who strips communities...
>
> On 1/7/15 5:35 AM, Song Li wrote:
>> Hi everyone,
>>
>> Today when I check one route in Routeviews I find something strange as
>> follows:
>>
>> route-views>sh ip bgp 176.108.0.0
>> BGP routing table entry for 176.108.0.0/19, version 23405621
>> Paths: (33 available, best #28, table default)
>>    Not advertised to any peer
>>    Refresh Epoch 1
>>    202018 35320 35320 57800
>>      5.101.110.2 from 5.101.110.2 (5.101.110.2)
>>        Origin IGP, localpref 100, valid, external
>>        Community: 702:120 2914:429 20485:52990 20485:53990 20485:54040
>> 20485:54050 47541:10001
>>        rx pathid: 0, tx pathid: 0
>>
>> the AS-Path is "202018 35320 35320 57800" but the community is 702:120
>> 2914:429 20485:52990 ....
>>
>> According to RFC 1997, the community format is AA:NN and AA means the
>> AS#. Here, AS702, AS2914 and AS20485 do not appear in the AS-Path and
>> as a result they should not appear in the community. Could anybody
>> tell me what's the reason they do appear in the community of this route?
>>
>> Thanks!
>>
>
>


-- 
Song Li
Room 4-204, FIT Building,
Network Security,
Department of Electronic Engineering,
Tsinghua University, Beijing 100084, China
Tel:( +86) 010-62446440
E-mail: refresh.lsong at gmail.com



More information about the NANOG mailing list