BGP + 3640 + Refresh of the routing table

Unanswered Question
Jan 16th, 2009

I have a R7000 been neighbor witch a 3640 and like CPE a 2800 (Customer router, i dont have administration of this one)

sh version

Cisco Internetwork Operating System Software

IOS (tm) 3600 Software (C3640-IS-M), Version 12.1(19), RELEASE SOFTWARE (fc1)

Copyright (c) 1986-2003 by cisco Systems, Inc.

Compiled Mon 03-Mar-03 21:11 by kellythw

Image text-base: 0x60008940, data-base: 0x60EB2000

ROM: System Bootstrap, Version 11.1(20)AA2, EARLY DEPLOYMENT RELEASE SOFTWARE (fc1)

CPE(2800) uptime is 2 years, 11 weeks, 6 days, 5 hours, 8 minutes

System returned to ROM by reload at 13:07:15 ARG Thu Oct 26 2006

System restarted at 13:13:11 AR Thu Oct 26 2006

System image file is "flash:c3640-is-mz.121-19.bin"

cisco 3640 (R4700) processor (revision 0x00) with 49152K/16384K bytes of memory.

My problem is that i have the routing table continuisly updating on certain routes, the 50% aprox. I don't see logs of peer problems or big change's on the received prefixes

CPE(2800)#show ip bgp neighbors 161.190.x.x

BGP neighbor is 161.190.x.x, remote AS 13474, external link

Description: Router Cliente

BGP version 4, remote router ID 161.190.x.x

BGP state = Established, up for 1d06h

Last read 00:00:58, 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 1170614 messages, 0 notifications, 0 in queue

Sent 3783247 messages, 5 notifications, 0 in queue

Route refresh request: received 0, sent 0

Default minimum time between advertisement runs is 30 seconds

For address family: IPv4 Unicast

BGP table version 23064792, neighbor version 23064792

Index 1, Offset 0, Mask 0x2

Inbound soft reconfiguration allowed

NEXT_HOP is always this router

Inbound path policy configured

Outbound path policy configured

Route map for incoming advertisements is CLIENTE-IN

Route map for outgoing advertisements is CLIENTE-OUT

1 accepted prefixes consume 36 bytes

Prefix advertised 12031501, suppressed 530010, withdrawn 9837951

1 denied but saved prefixes consume 36 bytes

Number of NLRIs in the update sent: max 288, min 0

Connections established 42; dropped 41

Last reset 1d06h, due to Peer closed the session

What could the be the cause of the partial refresh of the routing table every one minute , reboot the router its a possible solution?

I have this problem too.
0 votes
  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 0 (0 ratings)
Loading.
Paolo Bevilacqua Fri, 01/16/2009 - 13:48

This usually happens when a router doesn't have enough memory to store all the prefixes sent, so at some point the connection is broken and the cycle repeats infinitely.

juan.pagani Sat, 01/17/2009 - 16:48

mmmm you`re probably right its there any table on this website with the specs of the routers.

Thanks.

Giuseppe Larosa Sat, 01/17/2009 - 23:54

Hello Juan,

you can use

sh proc mem | inc Free

sh ip bgp sum

it says how much memory is used by BGP

if you can do this on both ends you can see if the BGP table is too big as suggested by Paolo.

Hope to help

Giuseppe

juan.pagani Sun, 01/18/2009 - 16:24

Hi Giusepe:

I have apparently enough memory but the routing table it`s still refreshing

Total: 25825344, Used: 13343196, Free: 12482148

PID TTY Allocated Freed Holding Getbufs Retbufs Process

BGP router identifier 200.9.X.X, local AS number XXXX

BGP table version is 31758836, main routing table version 31758836

7338 network entries and 7344 paths using 976170 bytes of memory

390 BGP path attribute entries using 23400 bytes of memory

24 BGP rrinfo entries using 624 bytes of memory

304 BGP AS-PATH entries using 7760 bytes of memory

1 BGP community entries using 24 bytes of memory

0 BGP route-map cache entries using 0 bytes of memory

0 BGP filter-list cache entries using 0 bytes of memory

2 received paths for inbound soft reconfiguration

BGP activity 1717805/2199433 prefixes, 2224154/2216806 paths, scan interval 60 secs

SLDS.

milan.kulik Mon, 01/19/2009 - 01:28

Hi,

isn't there eBGP multihop involved wit a default route only?

The configuration guide says "The multihop session is not established if the only route to the multi-hop peer's address is the default route (0.0.0.0)."

But I remember a case the session was Up but the BGP table was refreshing every minute - exactly as in your case.

BR,

Milan

Mohamed Sobair Sat, 01/24/2009 - 13:12

Hello,

Its normal, Partial EBGP update should be exchanged every 30sec and every 5sec for IBGP.

Another way to have the updates manually done is by configuring BGP route refresh Capability.

HTH

Mohamed

Actions

This Discussion