BGP Help #2

Unanswered Question
May 6th, 2009
User Badges:

Nordick26 or Anyone,


Hope your out there..

Listen, I came across another BGP issue that doesn't make sense..

Not as complex as my other BGP question..


Anyways..


Router A ----- Router B ---- Our Network..


1. Router A is a stub

2. ISIS running between them

3. iBGP Peering


Router A has a x.x.x.x/19 to advertize to our network..

Router B receives the x.x.x.x/19 route via iBGP considers it as best and inserts it into the routing table.

Router B receives the x.x.x.x/19 network with the next-hop of Router A loopback Addess.


The problem is that Router B does not advertise the x.x.x.x/19 in it's BGP updates with our Network..

The rest of the network doesn't see the /19 network and is unreachable..


So, I remembered what you told me..


"Example: you want to advertise (to external peer) 10.1.1.0/24, which is a summary of 4 subnet's you are using in your AS (10.1.1.0/26; 64/26; 128/26; 192/26). You're receiving this 4 prefixis via iBGP, but in other BGP can advertise this out of you AS, you need to have 10.1.1.0/24 in your routing table. This is why you need to use the static route (pointing back to router you get the more specific route from, or pointing to Null0)."


So, I created a static route to that subnet on Router B pointing to Router A and it worked.. The network see it..

But, why wasn't it propagated throughout the network via BGP??

Router B had an entry in the routing table and considered it as best??


Why??

  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 0 (0 ratings)
Loading.
Laurent Aubert Wed, 05/06/2009 - 07:18
User Badges:
  • Cisco Employee,

Hi,


With and without the static route configured, could you post router B configuration and show ip bgp x.x.x.x/19 ?


Thanks


Laurent.

Mavrick25 Wed, 05/06/2009 - 07:32
User Badges:

Hello,


I can't post the config..

But the show ip bgp command I could..


show ip bgp x.x.x.x/19

BGP routing table entry for x.x.x.x/19, version 42556205

Paths: (2 available, best #1, table Default-IP-Routing-Table)

Advertised to update-groups:

1 3 4

Local

RouterA (metric 11) from RouterA (RouterA)

Origin IGP, metric 130, localpref 100, weight 32768, valid, sourced, best

Community: 82381832

Laurent Aubert Wed, 05/06/2009 - 09:45
User Badges:
  • Cisco Employee,

From this output, the prefix is advertised to group 1,3 and 4. Use the show ip bgp update-group to see if your router is part of one of those groups.


You can also use the debug ip bgp updates out cmd on routeur B and clear softly your BGP session with router B.


Laurent.

Giuseppe Larosa Wed, 05/06/2009 - 11:39
User Badges:
  • Super Silver, 17500 points or more
  • Hall of Fame,

    Founding Member

Hello Mav,


just two adds to what Laurent has already noted.


the route has a BGP community associated to it that can trigger a route filter in other routers.


Note1:

the BGP community is not showed using the bgp new-format if you have enable access I would recommend to add

conf t

ip bgp-community new-format


http://www.cisco.com/en/US/docs/ios/iproute/command/reference/irp_bgp2.html#wp1015944


this allows to see BGP communities as ASN:value making the config more readable.


note2:

also you can check with


sh ip bgp neigh advertised-routes


if your prefix is there


Hope to help

Giuseppe



Mavrick25 Wed, 05/06/2009 - 23:56
User Badges:

Giuseppe...


Grazie Mille..

It is a lot easier to read with the new-format command applied..


Thanks..


Giuseppe Larosa Thu, 05/07/2009 - 11:36
User Badges:
  • Super Silver, 17500 points or more
  • Hall of Fame,

    Founding Member

Hello Mav,

I would suggest you to use


sh ip bgp update-groups N


to identify to what peers is the route advertised then you need to verify on the other router.


As noted by Milan route filters can play a role here.

They can be outbound RB and inbound RC


Hope to help

Giuseppe


Mavrick25 Wed, 05/06/2009 - 23:55
User Badges:

Laurent,


Thank you for the assistance..

Listen..

I tried creating a route.. A test route that could help us out.. 20.20.20.0/30 subnet.. put it on Router A to simulate the issue I'm having..


Please take a look at the output..

RouterB#show ip bgp neighbors RouterA received-routes

BGP table version is 43689367, local router ID is RouterB Loopback

Status codes: s suppressed, d damped, h history, * valid, > best, i - internal,

S Stale

Origin codes: i - IGP, e - EGP, ? - incomplete


Network Next Hop Metric LocPrf Weight Path

*>i20.20.20.0/30 Router A 130 100 0 i


Total number of prefixes 1


RouterB1#show ip route 20.20.20.0

Routing entry for 20.20.20.0/30

Known via "bgp 448563", distance 200, metric 130, type internal

Last update from RouterA 16:54:20 ago

Routing Descriptor Blocks:

* RouterA, from RouterA, 16:54:20 ago

Route metric is 130, traffic share count is 1

AS Hops 0


RouterB#show ip bgp neighbors RouterC advertised-routes | i 20.20.20.0

RouterB# (nothing)



Considers the route via BGP and inserts it into the routing table but doesn't advertise the route...


Mav




milan.kulik Thu, 05/07/2009 - 04:05
User Badges:
  • Red, 2250 points or more

Hi,


is there eBGP running between routers B and C?

If yes, isn't any route-map, filter-list, prefix-list used to filter the updates sent to router C?

Can you provide "sh ip bgp nei RouterC" output from router B?


BR,

Milan


Mavrick25 Wed, 05/13/2009 - 03:54
User Badges:

Sorry for the late reply..

Trying to still figure out this problem...


The answer is no.. they are not running a eBGP relationship but a iBGP relationship..

The show command to follow:


RouterB#show ip bgp neighbors Router C

BGP neighbor is RouterC, remote AS xxxx, internal link

Description: FIRZ-NCORE-1

BGP version 4, remote router ID RouterB

BGP state = Established, up for 49w0d

Last read 00:00:00, last write 00:00:20, hold time is 180, keepalive interval is 60 seconds

Neighbor capabilities:

Route refresh: advertised and received(new)

Address family IPv4 Unicast: advertised and received

Graceful Restart Capabilty: advertised

Message statistics:

InQ depth is 0

OutQ depth is 0

Sent Rcvd

Opens: 26 24

Notifications: 23 2

Updates: 3737 19157612

Keepalives: 494634 494634

Route Refresh: 0 0

Total: 498401 19652272

Default minimum time between advertisement runs is 0 seconds


For address family: IPv4 Unicast

BGP table version 45189612, neighbor version 45189612/0

Output queue size : 0

Index 3, Offset 0, Mask 0x8

3 update-group member

Inbound soft reconfiguration allowed

NEXT_HOP is always this router

Community attribute sent to this neighbor

Sent Rcvd

Prefix activity: ---- ----

Prefixes Current: 370 8071 (Consumes 387408 bytes)

Prefixes Total: 6051 29163563

Implicit Withdraw: 620 21460343

Explicit Withdraw: 5323 7695149

Used as bestpath: n/a 7839

Used as multipath: n/a 0


Outbound Inbound

Local Policy Denied Prefixes: -------- -------

CLUSTER_LIST loop: n/a 6640

ORIGINATOR loop: n/a 263

Bestpath from this peer: 29074156 n/a

Bestpath from iBGP peer: 8259468 n/a

Total: 37333624 6903

Number of NLRIs in the update sent: max 341, min 0


Connections established 1; dropped 0

Last reset never

Connection state is ESTAB, I/O status: 1, unread input bytes: 0

Local host: 212.151.128.132, Local port: 179

Foreign host: 212.151.128.133, Foreign port: 21899


Enqueued packets for retransmit: 0, input: 0 mis-ordered: 0 (0 bytes)


Event Timers (current time is 0x11108B2014):

Timer Starts Wakeups Next

Retrans 498429 202 0x0

TimeWait 0 0 0x0

AckHold 10734169 7779444 0x0

SendWnd 0 0 0x0

KeepAlive 15 0 0x0

GiveUp 0 0 0x0

PmtuAger 0 0 0x0

DeadWait 0 0 0x0


iss: 663007513 snduna: 672655114 sndnxt: 672655114 sndwnd: 16327

irs: 3070273282 rcvnxt: 435114213 rcvwnd: 16198 delrcvwnd: 186


SRTT: 300 ms, RTTO: 303 ms, RTV: 3 ms, KRTT: 0 ms

minRTT: 0 ms, maxRTT: 368 ms, ACK hold: 200 ms

Flags: passive open, retransmission timeout, nagle, gen tcbs


Datagrams (max data segment is 536 bytes):

Rcvd: 19871123 (out of order: 41), with data: 19480260, total data bytes: 1659808226

Sent: 11157526 (retransmit: 202), with data: 498311, total data bytes: 9647600

Mavrick25 Wed, 05/13/2009 - 03:56
User Badges:

Sorry for the late reply..

Trying to still figure out this problem...


The answer is no.. they are not running a eBGP relationship but a iBGP relationship..

What kind of information are you searching for, regarding the output show ip bgp neigh command..


Please let me know.. I will provide it

milan.kulik Wed, 05/13/2009 - 04:37
User Badges:
  • Red, 2250 points or more


Hi,


so if I understand correctly there is router A peering by iBGP to router B and router B again peering by iBGP to router C?


Router B will not advertise the prefixes received from router A to router C then.


One of basic BGP rules is:

"Peering inside an AS must consist of a full mesh, because an iBGP speaker is not allowed to pass on prefixes learned from one iBGP peer to another iBGP peer."


Or am I mising something here?


BR,

Milan

Mavrick25 Fri, 05/15/2009 - 00:43
User Badges:

Milan,


I think you absolutely correct..

But when I go and present my finding to my team lead he thinks it's crap...


I thinking of performing a debug on router C to find out exactly what is going on..


something like a debug ip bgp update

what do you recommend..

Mav

Actions

This Discussion