At 10:44 PM +0000 7/31/02, nrf wrote:
>""Robert D. Cluett""  wrote in message
>[EMAIL PROTECTED]">news:[EMAIL PROTECTED]...
>>  I am just getting into BGP and I understand it like this.  An update can
>>  carry all or some of the attributes with it which define the route to the
>>  destination.  There is a pecking order of what attributes determine what
>>  route will be utilized.  So if you were to look inside of a BGP update
>>  packet you will see the mandatory attributes being AS-PATH, Next-Hop and
>>  Orign (these are mandatory so they will alway be in the packet).  The
>>  discretionary and optional attributes may or may not be there.  When
>>  updating the BGP table these attributes are analyzed and then the route
is
>>  chosen to the destination.  Remember that a router will send an update
>>  packet to you.  It is adding the AS that is passes through to the AS_PATH
>>  field.  This is one of the requirements in order to have a route to a
>>  destination.  Next_Hop (IP Address) is also a requirement so you know
>where
>>  to go next.  Origin defines what AS this update came from.
>
>Actually, this in incorrect.  The Origin attribute has nothing to do with
>what AS the route comes from.  It has to do with how the route in question
>was passed into BGP.
>
>The best reference for path-attributes is still, the one and only, RFC1771.
>Anybody who really really wants to understand BGP must read this.

Actually, RFC 1771 is getting a little dated -- it's best to go to 
the IETF page, navigate to the IDR working group, and get the most 
recent BGP specification draft (I've lost track...14? 15?).  This 
contains a lot of real-world practice that differs from 1771.

Personally, I recommend getting an understanding of RPSL, and _then_ 
studying BGP selection because now you know _why_ you're trying to do 
something.

 From my Wiley book, Building Service Provider Networks, here's a 
reasonably up-to-date reference.  Do remember that filters, 
communities, etc. can have
major effects on selection, but they take effect before the selection 
algorithm runs.   Unfortunately, it's formatted as a two-column 
table, which doesn't come across well in mail.  I think you'll be 
able to see the separation between rule and explanation, though.


Table 5:  BGP Route Selection
IETF-specified rule     Common industry extension
Discard any route whose next hop is unreachable.
        If the next hop is accessible, prefer the route with the 
highest weight. Weight is a Cisco-specific parameter that is not 
transmitted in BGP updates, but is a manually configured parameter 
local to the current router.

Prefer the route with the highest local preference attribute.

Prefer routes originated on the current router.
        A widely used criterion that is not part of the IETF 
specifications is to prefer the route with the shortest AS_PATH 
(i.e., the least number of AS in the path).

If routes have the same AS_PATH length, prefer Interior to Exterior 
to Incomplete origin.
       Cisco interprets confederation routes as interior, but the 
algorithm prefers confederation exterior to confederation interior.

Of paths with the same origin, prefer those with the lowest MED value.
      In the absence of a specific IETF interpretation, Cisco's 
default was to assume routes without an explicit MED have a MED of 
zero.  Recently, the IETF clarified the expected behavior, which is 
the opposite of Cisco's interpretation.  This clarification specifies 
that when presented with routes that either have MEDs or do not, the 
routes without MEDs should be less preferable than a route with an 
explicit MED of any value.
     Also, the BGP specification states that MEDs should be compared 
only between connections to the same adjacent AS.  There are 
applications such as multilateral exchange points, however, where it 
can be very useful to compare MEDs between multiple adjacent AS. See 
Chapter 12.

If the MEDs are equal, prefer routes with external rather than 
internal sources.          
      Some implementations reverse the order of this and the next step.

Prefer the path through the closest IGP neighbor (i.e., lowest IGP metric).

Otherwise, select the path with the lowest originating router ID.

     Since router IDs must be unique, this will be a final tie 
breaker. Router ID is selected by the same algorithm used to select 
the OSPF router ID.     The BGP specification does not support equal 
cost multipath, but several vendor implementations have proprietary 
extensions for doing so.

     If multiple paths are enabled, add the current route if both the 
current best route and the new route are external and come from the 
same adjacent AS.  Cisco supports up to 6 load-shared paths.

>
>>  I re-read the
>>  chapter 3 times and have it cold now!  I typically re-read and it clears
>>  everything up that I missed.
>>
>>  ""maine dude""  wrote in message
>>  [EMAIL PROTECTED]">news:[EMAIL PROTECTED]...
>>  > Hi  I'm having problems on understanding how attributes work. I
>understand
>>  > what they are there for but when questions come up I am unable to
relate
>>  the
>>  > question to what I understand when I read the book. Are there any
>simpler
>  > > guides available. I have the same problem with prefix-lists. -DJ




Message Posted at:
http://www.groupstudy.com/form/read.php?f=7&i=50333&t=50248
--------------------------------------------------
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]

Reply via email to