sh ip bgp neighbor received-routes

[arun]

Running BGP connection between my router and ISP. Have applied route-map in inbound direction blocking all the prefixes except default route.Everything works fine.

But when i check "sh ip bgp neighbor x.x.x.x received-routes" output, i could see many routes though they are not installed in BGP table. I beleive it should show only the default route since we blocked all other prefixes. is this something related to IOS version?

Thanks,

Arun

[amabdelh]

Hi Arun
Can you procide more details like show version, show ip bgp neighbor x, bgp config


Sent from Cisco Technical Support iPhone App

[arun]

Hi Amjad,

Here is the output from sh version..

Cisco IOS Software, c7600rsp72043_rp Software (c7600rsp72043_rp-ADVIPSERVICESK9-M), Version 12.2(33)SRB2, RELEASE SOFTWARE (fc1)

Technical Support: http://www.cisco.com/techsupport

Copyright (c) 1986-2007 by Cisco Systems, Inc.

Compiled Wed 10-Oct-07 08:10 by prod_rel_team

Config:

neighbor x.x.x.x remote-as yy

neighbor x.x.x.x version 4

neighbor x.x.x.x timers 15 45

neighbor x.x.x.x fall-over

neighbor x.x.x.x send-community

neighbor x.x.x.x soft-reconfiguration inbound

neighbor x.x.x.x route-map xxxx in

neighbor x.x.x.x route-map xxxx ou

Thanks,

Arun

[amabdelh]

Arun, I agree with Richard this is expected

The neighbor soft-reconfiguration inbound

command causes the router to store all received (inbound) routing policy updates without modification, for example, a duplicate table is stored in the memory for each peer. This method is memory-intensive and not recommended unless absolutely necessary

if you wants to see the routes after applying all the filters, do show ip bgp neighbor X routes

regards

[Richard Burts · Accepted solution]

Arun

Depending on the version of code that you are running, and that your neighbor is running this may be an expected behavior. It was done to enable a feature sometimes referred to as soft reset. It would allow you to change policy affecting routes to/from a neighbor without requiring a hard reset of the neighbor relationship. The way it worked is that the router keeps a copy of all prefixes learned from a neighbor without any modification (no filtering etc). Then it passes through this table applying your configured policy to build the BGP tables and the routing table. Then if you change your policy the router has only to pass through the table again applying the new policy. There would be no need to tear down the neighbor relationship.

Then there was an enhancement to the soft reset feature that did not require keeping a copy of all routes learned from a neighbor. It sounds like you or your neighbor may be running a version of code with the earlier implementation of the feature.

You can use this link to find more information about this feature.

http://www.cisco.com/en/US/products/ps6599/products_data_sheet09186a0080087b3a.html

HTH

Rick

[Richard Burts · Accepted solution]

Arun

I had not seen your post which includes the bgp configuration when I wrote my post. Now that I see it I notice this in the config

neighbor x.x.x.x soft-reconfiguration inbound

the soft-reconfiguration inbound is the command used in the earlier implementation of the soft reset feature. So I am convinced that what you are seeing is indeed an expected behavior in your version of code. You see all the advertised prefixes, without modification, which you did indeed receive. Then your policy filtered this table and resulted in only the default route.

HTH

Rick