Re: Quick Question.... - more help!!!

From: Donny MATEO (donny.mateo@xxxxxxxxxxxxxxxxxx)
Date: Tue Jul 09 2002 - 05:00:06 GMT-3


   
I remember reading somewhere in Halabi's Internet routing architecture for
this.
I'll quote " Inbound soft reconfiguration is a bit moe involved. All
inbound updates (the adj-RIB-In) rom the specified peer are stored in
memory unmodified. When a new policy is applied and activated, the stored
Adj-RIB-In is again presented to the Input Policy Engine"

In your case you are doing the "clear ip bgp * soft in) on router A and do
the changes in routerB. If you refer to the quote nothing will happen, it's
normal. because you are doing the modification of Output Policy Engine in
router B, instead of Input Policy Engine in routerA.
what you want is probably the capabilities known as BGP Route Refresh.

Hope it helps..
Donny

                      Hunt Lee

                      <huntl@webcentral To: 'Gyori Gabor' <Gabor.G
yori@lnx.hu>, "'Krzysztof.Horszczaruk@getronics.com'"
                      .com.au> <Krzysztof.Horszczaruk@getronic
s.com>
                      Sent by: cc: "'ccielab@groupstudy.c
om'" <ccielab@groupstudy.com>
                      nobody@groupstudy Subject: Quick Question.... -
more help!!!
                      .com

                      09-07-2002 15:12

                      Please respond to

                      Hunt Lee

Hello guys,

Thanks so much for the explanation. I managed the get the 2nd method
working with no problems (without "neighbor soft-reconfiguration" command).
Just a minor question though, for the 1st method (by using the "neighbor
x.x.x.x soft-reconfiguration inbound"), after I put that into RouterA, and
make the outgoing route-map changes on RouterB, when I do the "clear ip bgp
* soft in" on RouterA, it said that inbound soft reconfiguration has
started... but the new changes (with a higher MED) that i put in refused to
change... unless if I do a hard BGP reset (by clear ip bgp *).

Why is this?? Please help!!!

Best Regards,
Hunt Lee

RouterB#sh route-map

route-map test, permit, sequence 10

  Match clauses:

  Set clauses:

    metric 203

  Policy routing matches: 0 packets, 0 bytes

RouterB#

RouterB#sh ip bgp n

BGP neighbor is 10.1.1.1, remote AS 1, external link

BGP version 4, remote router ID 10.3.0.1

Password: BGP state = Established, up for 00:03:11
Last read 00:00:09, hold time is 180, keepalive interval is 60 seconds

Neighbor capabilities:

    Route refresh: advertised and received(new)

    Address family IPv4 Unicast: advertised and received

  Received 48 messages, 0 notifications, 0 in queue

  Sent 50 messages, 0 notifications, 0 in queue

  Route refresh request: received 4, sent 0

  Default minimum time between advertisement runs is 30 seconds

 For address family: IPv4 Unicast

  BGP table version 13, neighbor version 13

  Index 1, Offset 0, Mask 0x2

  Outbound path policy configured

  Route map for outgoing advertisements is test

  2 accepted prefixes consume 72 bytes

  Prefix advertised 14, suppressed 0, withdrawn 0

  Number of NLRIs in the update sent: max 2, min 0

  Connections established 3; dropped 2

  Last reset 00:03:54, due to Peer closed the session

 --More--

RouterA#sh ip bgp

BGP table version is 5, local router ID is 10.3.0.1

Status codes: s suppressed, d damped, h history, * valid, > best, i -
internal
Origin codes: i - IGP, e - EGP, ? - incomplete

   Network Next Hop Metric LocPrf Weight Path

*> 10.2.0.0/16 0.0.0.0 0 32768 i

*> 10.3.0.0/16 0.0.0.0 0 32768 i

*> 20.1.0.0/16 10.1.1.2 203 0 2 i

*> 20.2.0.0/16 10.1.1.2 203 0 2 i

RouterA#

And after I updated the outgoing route-map...

RouterB#sh route

route-map test, permit, sequence 10

  Match clauses:

  Set clauses:

    metric 1000

  Policy routing matches: 0 packets, 0 bytes

RouterB#

RouterA#clear ip bgp * soft in

03:11:26: BGP(0): start inbound soft reconfiguration for 10.1.1.2

03:11:26: BGP(0): process 20.1.0.0/16, next hop 0.127.245.126, metric 203
from 1
0.1.1.2

03:11:26: BGP(0): No inbound policy. Prefix 20.1.0.0/16 accepted
unconditionally
03:11:26: BGP(0): process 20.2.0.0/16, next hop 0.127.245.86, metric 203
from 10
.1.1.2

03:11:26: BGP(0): No inbound policy. Prefix 20.2.0.0/16 accepted
unconditionally
03:11:26: BGP(0): complete inbound soft reconfiguration, ran for 20ms

03:11:26: BGP(0): 10.1.1.2 computing updates, afi 0, neighbor version 5,
table v
ersion 7, starting at 0.0.0.0

03:11:26: BGP(0): 10.1.1.2 update run completed, afi 0, ran for 4ms,
neighbor ve
rsion 5, start version 7, throttled to 7

RouterA#sh ip bgp

BGP table version is 7, local router ID is 10.3.0.1

Status codes: s suppressed, d damped, h history, * valid, > best, i -
internal
Origin codes: i - IGP, e - EGP, ? - incomplete

   Network Next Hop Metric LocPrf Weight Path

*> 10.2.0.0/16 0.0.0.0 0 32768 i

*> 10.3.0.0/16 0.0.0.0 0 32768 i

*> 20.1.0.0/16 10.1.1.2 203 0 2 i

*> 20.2.0.0/16 10.1.1.2 203 0 2 i

RouterA#

-----Original Message-----
From: Gyori Gabor [mailto:Gabor.Gyori@lnx.hu]
Sent: Monday, 8 July 2002 11:53 PM
To: Horszczaruk Krzysztof; Harish DV/peakxv; Hunt Lee
Cc: ccielab@groupstudy.com
Subject: RE: Quick Question....

Hi Krzysztof !

You are absolutely right. I have tried the above second version (that you
explained first) and it works explicitly as You described.
The first one works with arbitrary peer router, the second need capable
router
(IOS 12.1 or above as doc say).

Thanks for explanation,

Gabor

> -----Original Message-----
> From: Horszczaruk Krzysztof
> [mailto:Krzysztof.Horszczaruk@getronics.com]
> Sent: Monday, July 08, 2002 3:31 PM
> To: Gyuri Gabor; Harish DV/peakxv; Hunt Lee
> Cc: ccielab@groupstudy.com
> Subject: RE: Quick Question....
>
>
> Hi Gyori Gabor,
>
> I think you are right and I am right too.
>
> in my understanding it works as follows:
>
> 1) when there is "neighbor soft-reconfiguration" configured
> on your router, there is no request sent to the neighbor, and
> all information is gathered from the local cache, buffer or
> table (how to name it ?)
>
> 2) when there is no "neighbor soft-reconfiguration"
> statement, then in case of "clear ip bgp * soft in" the route
> refresh request is sent to the neighbor and information is
> gatheret directly from the neighbor. of course it is possible
> only if the neighbor support this feature.
>
> the simple analogy is browsing online/offline.
>
> cisco support it starting from 12.0. I do not know if it is
> proprietary or not.
> you can determine if your neighbor is route-refresh-capable
> by "show ip bgp neighbors x.x.x.x"
>
> regards,
> Krzysztof Horszczaruk
> Senior Consultant, System Engineer
> Network Integration
>
> Getronics Polska Sp. z o.o.
> ul. Pulawska 352a
> 02-819 Warszawa
> http://www.getronics.com
> http://www.getronics.pl
>



This archive was generated by hypermail 2.1.4 : Sat Sep 07 2002 - 19:36:23 GMT-3