From: maureen schaar (maureen.schaar@gmail.com)
Date: Tue Apr 17 2007 - 04:51:06 ART
Hi,
In your last case, I see two paths from the same as. In this case you
don't have either deterministic-med or always-compare-med enabled.
BGP compares the med by default for paths received from the same as,
so the path with lowest med is chosen. You don't need
always-compare-med for this. So yes, this is what I would have
expected.
Both commands are really different: always-compare-med only influences
path selection for paths received from a different as. BGP
deterministic-med influences path selection for paths received from
the same AS by reordering the paths based on the AS (instead of order
they were received). This will force the same path to be chosen
regardless of the order in which the paths were received, which is
generally not desired. This is why cisco recommends enabling
deterministic-med in new installations.
If you read through the links you provided once more, it should become clear.
HTH
Maureen
On 4/16/07, ccie anees <ccieanees@yahoo.com> wrote:
> Hi,
>
> I don't understand even after reading the Cisco documents
> http://www.cisco.com/en/US/tech/tk365/technologies_tech_note09186a0080094934.shtml
> http://www.cisco.com/en/US/tech/tk365/technologies_tech_note09186a0080094925.shtml
>
> My topolgy
>
> ----------- R2 AS 65502
> |
> AS 65501 R1------ ---------- R3 AS 65503
> |
> ---------- R4 AS 65504
> From R1:
>
> router bgp 65501
> no synchronization
> bgp router-id 1.1.1.1
> bgp log-neighbor-changes
> neighbor 100.100.102.2 remote-as 65502
> neighbor 100.100.103.3 remote-as 65503
> neighbor 100.100.104.4 remote-as 65503
> no auto-summary
> !
>
> Case 1:
> **********
> R1#show ip bgp 200.200.200.0
> BGP routing table entry for 200.200.200.0/24, version 2
> Paths: (3 available, best #1, table Default-IP-Routing-Table)
> Advertised to update-groups:
> 1
> 65502
> 100.100.102.2 from 100.100.102.2 (2.2.2.2)
> Origin IGP, metric 100, localpref 100, valid, external, best
> 65503
> 100.100.104.4 from 100.100.104.4 (4.4.4.4)
> Origin IGP, metric 25, localpref 100, valid, external
> 65503
> 100.100.103.3 from 100.100.103.3 (3.3.3.3)
> Origin IGP, metric 50, localpref 100, valid, external
> R1#
> Case 1 is fine. It works as it suppose to. The route with the lowest router ID becomes the best.
>
> Case 2
> #######
> router bgp 65501
> bgp always-compare-med
> !
> R1#show ip bgp 200.200.200.0
> BGP routing table entry for 200.200.200.0/24, version 2
> Paths: (3 available, best #1, table Default-IP-Routing-Table)
> Advertised to update-groups:
> 1
> 65502
> 100.100.102.2 from 100.100.102.2 (2.2.2.2)
> Origin IGP, metric 100, localpref 100, valid, external
> 65503
> 100.100.104.4 from 100.100.104.4 (4.4.4.4)
> Origin IGP, metric 25, localpref 100, valid, external, best
> 65503
> 100.100.103.3 from 100.100.103.3 (3.3.3.3)
> Origin IGP, metric 50, localpref 100, valid, external
>
> Great., It works as it supopose to. But I have issues with the third one.
>
> Case 3:
> *********
> router bgp 65501
> no bgp always-compare-med
>
> R1#show ip bgp 200.200.200.0
> BGP routing table entry for 200.200.200.0/24, version 2
> Paths: (2 available, best #1, table Default-IP-Routing-Table)
> Advertised to update-groups:
>
> 65503
> 100.100.104.4 from 100.100.104.4 (4.4.4.4)
> Origin IGP, metric 25, localpref 100, valid, external, best
> 65503
> 100.100.103.3 from 100.100.103.3 (3.3.3.3)
> Origin IGP, metric 50, localpref 100, valid, external
>
> Here it is comparing the MED value. Is it the default behavior? Do I suppose to add bgp deterministic-med to have the BGP to use MED? how come it is choosing the best path based on MED without the bgp deterministic-med?
>
> Also If I configure bgp always-copare-med, I don;t need to use bgp deterministic-med. Am i correct?
>
> Thanks,
>
> Anees.
>
>
> ---------------------------------
> Ahhh...imagining that irresistible "new car" smell?
> Check outnew cars at Yahoo! Autos.
>
> ---------------------------------
> Ahhh...imagining that irresistible "new car" smell?
> Check outnew cars at Yahoo! Autos.
>
> _______________________________________________________________________
> Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html
This archive was generated by hypermail 2.1.4 : Tue May 01 2007 - 08:28:36 ART