RE: OSPF tracing next hop

From: Joseph Brunner (joe@affirmedsystems.com)
Date: Sun May 11 2008 - 01:43:07 ART


Good evening, (or afternoon, don't your GMT offset)

 

R5 is answering the traceroute packets via either equal cost path it has for
R4's loopback.

 

I would speculate on R5 if you did

 

R5#show ip route 150.1.4.4

 

150.1.4.4/32 via 155.1.45.4, outgoing S2/1

                        155.1.0.4, outgoing S2/0

 

This is perfectly normal behavior for ospf or any routing protocol.

 

Either you have disabled CEF and its doing ip destination doing load
balancing using the routing table, or you have cef enabled and the Frame
path is OLDER.

 

Here is one way you can handle this.

 

Local policy routing!

 

You can say ROUTER traffic (which this is for R5) must leave on the
interface it arrived.. we used this several years back at branch offices
with Two T-1 lines responded to monitoring pings from corporate BACK OUT THE
SAME interface they arrived. check this config.

 

 

ip local policy route-map ping-policy

 

route-map ping-policy permit 10

 match ip address 105

 set ip next-hop <qwest next hop ip>

!

route-map ping-policy permit 20

 match ip address 106

 set ip next-hop <mci next hop ip>

 

access-list 105 permit ip host 65.123.0.98 any

access-list 106 permit ip host 157.130.192.186 any

 

 

In your case you can match input interface in your local policy route-map
and do a set ip next-hop (must be this one for broadcast networks like
Ethernet), or set interface (since you are using p2p serial/frame links)

 

-Joe

 

  _____

From: Nitro Drops [mailto:nitrodrops@hotmail.com]
Sent: Saturday, May 10, 2008 11:45 PM
To: Joseph Brunner; ccielab@groupstudy.com
Subject: OSPF tracing next hop

 

Hi Joe and All,

Encountered some similar symptons now with OSPF with regards to tracing

R4 S2/1 (155.1.45.4/24) >> Serial >> (155.1.45.5/24)S2/1 R5
R4 S2/0 (155.1.0.4/24) >> Frame Relay >> (155.1.0.5/24) S2/0 R5
R4 Loopback 0 - 150.1.4.4
R5 Loopback 0 - 150.1.5.5

When i did a trace from R4 Loopback 0 to R5 Loopback 0

it shows

  1 155.1.45.5 100 msec
    155.1.0.5 184 msec
    155.1.45.5 92 msec

How come the trace is hitting R5 S2/1, to R5 S2/0 and back to R5 S2/1,
instead of R5 S2/1 and to R5 Looback?

I encountered the same problem the last time when practising EIGRP, posted
the same query, Joe B gave some good advises to disable IP CEF, which i have
done on all the serials interfaces, however i am still not seeing a direct
trace to R5 S2/1 and to R5 Looback.

Thanks for any kind replies.

Cheers
Nit

  _____

From: nitrodrops@hotmail.com
To: joe@affirmedsystems.com; ccielab@groupstudy.com
Subject: RE: EIGRP - Unequal-Cost LoadBalancing
Date: Tue, 15 Apr 2008 06:31:14 +0800

Thanks mate, it works now

it hits the next hop directly. and no more loops.

you are a legend.

cheers
Nit

> From: joe@affirmedsystems.com
> To: nitrodrops@hotmail.com; ccielab@groupstudy.com
> Subject: RE: EIGRP - Unequal-Cost LoadBalancing
> Date: Fri, 11 Apr 2008 22:18:24 -0400
>
> CEF which supersedes the routing table will cause the oldest route to be
> used.
>
> Disable cef
>
> No ip cef
>
> And then tell us what you found.
>
> -----Original Message-----
> From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of
> Nitro Drops
> Sent: Friday, April 11, 2008 9:53 PM
> To: ccielab@groupstudy.com
> Subject: EIGRP - Unequal-Cost LoadBalancing
>
> Guys,
>
> was practising this scenario.
>
> R4 S2/1 (155.1.45.4/24) >> 64K Serial 64K>> (155.1.45.5/24)S2/1 R5
> R4 S2/0 (155.1.0.4/24) >> 256 K Frame Relay 256K >> (155.1.0.5/24) S2/0 R5
> R4 Loopback 0 - 150.1.4.4
> R5 Loopback 0 - 150.1.5.5
>
> Variance calculated is 4.
>
> R5:
> router eigrp 100
> variance 4
>
>
>
>
>
>
>
>
>
>
>
>
>
> R5#sh ip eigrp neighbors
> IP-EIGRP neighbors for process 100
> H Address Interface Hold Uptime SRTT
> RTO Q Seq
> (sec) (ms) Cnt Num
> 1 155.1.0.4 Se2/0 150 00:42:28 189 1134 0 29
> 0 155.1.45.4 Se2/1 12 00:44:48 191 2280 0 30
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> R5#sh ip
> route
> Gateway of last resort is not set
> 155.1.0.0/24 is
> subnetted, 2 subnets
> C 155.1.0.0 is
> directly connected, Serial2/0
> C 155.1.45.0 is
> directly connected, Serial2/1
> 150.1.0.0/24 is
> subnetted, 2 subnets
> C 150.1.5.0 is
> directly connected, Loopback0
> D 150.1.4.0 [90/40640000] via 155.1.45.4,
> 00:00:46, Serial2/1
> [90/10639872] via 155.1.0.4,
> 00:00:46, Serial2/0
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> R5#trace 150.1.4.4
> Type escape sequence to abort.
> Tracing the route to 150.1.4.4
> 1 155.1.45.4 156
> msec
> 155.1.0.4 148 msec
> 155.1.45.4 28 msec
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> R5#trace 150.1.4.4
> Type escape sequence to abort.
> Tracing the route to 150.1.4.4
> 1 155.1.0.4 176 msec
> 155.1.45.4 128
> msec
> 155.1.0.4 152 msec
>
>
>
>
>
>
>
>
>
>
>
> R5#trace 150.1.4.4
> Type escape sequence to abort.
> Tracing the route to 150.1.4.4
> 1 155.1.0.4 168 msec
> 180 msec *
>
>
>
>
>
>
>
>
>
>
>
> R5#trace 150.1.4.4
> Type escape sequence to abort.
> Tracing the route to 150.1.4.4
> 1 155.1.0.4 152 msec
> 40 msec *
>
>
> I can see the packets being load-balanced when tracing from R4 to R5's
> loopback. Did 4 traces continuously and noticed the follows
>
> Q1.) For the 1st 2 traces, how come the packets reaches R4's S2/1
interface,
> it will go to R4's S2/0 interface again before hitting R4's loopback? The
> same goes for the 2nd trace, it will go to R4's S2/0 interface 1st,
followed
> by R4's S2/1 interface before reaching R4's loopback.
>
> Q2) For the 3rd and 4th traces, it hits the R4's loopback0 immediately as
> next hop, without hitting R4's S2/0 or S2/1.
>
>
> Appreciate any kind advice.
>
>
> Cheers
> Nit
>
>
>
>
>
>
>
>
> _________________________________________________________________
> Easily manage multiple email accounts with Windows Live Mail!
> http://www.get.live.com/wl/all
>
>
> Pass the CCIE in six weeks, Guaranteed!
> http://www.certscience.com/CCIE
> _______________________________________________________________________
> Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html
>
>
> Pass the CCIE in six weeks, Guaranteed!
> http://www.certscience.com/CCIE
> _______________________________________________________________________
> Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html
>
>
>

  _____

Express yourself and stay connected with the latest Windows Live Messenger!
Windows <http://www.get.live.com/wl/all> Live Messenger

 

  _____

Share your beautiful moments with Photo Gallery. Windows Live
<http://get.live.com/photogallery/overview> Photo Gallery



This archive was generated by hypermail 2.1.4 : Mon Jun 02 2008 - 06:59:16 ART