Re: trace route

jkimcv60 · ‎08-13-2002

I have 4 routers and 2 switches on the same subnet all connecting to a catalyst switch. Routers can ping to each other fine but can't trace to any other routers. It traces to catalyst switch and atm switch fine. I have EIGRP and BGP running on the routers. Has anyone come across bizzare situation like this? This only started happening after I introduced another router and unfortunately I can't take that router out. any suggestions or ideas will be greatly appreciated.

JK

Gilles Dufour · ‎08-13-2002

are they all in the same VLAN ?

When you say it fails, do you mean that you don't even see the first hop ?

Or does it fail after a few hops ?

If so, does it always fail at the same device ?

What type of router is the new one ?

Do you have any ACL or firewall device that could block icmp or udp ?

jkimcv60 · ‎08-13-2002

they are all in vlan 1 and this is what I get when tracing

1 * * *

2 * * *

3 * * *

4 * * *

5 * * *

6 * * *

I do not have an acl applied on the interface I"m tracing to. here's what I have on the interface.

interface GigabitEthernet0/0/0

description Connected to CAT4003

ip address x.x.x.x y.y.y.y

no ip redirects

no ip unreachables

load-interval 30

negotiation auto

bourse · ‎08-13-2002

It may be a routing issue. Are you using a routing protocol (EIGRP, OSPF, etc...)? Perhaps when you introduced the new router, you introduced a routing loop.

DALE FRANCIS · ‎08-13-2002

The fact you can ping indicates no particular L3 problem unless you notice extensive delay times.

Bear in mind Cisoc uses UDP for traceroute hence you could have and access-list that is being hit with no icmp unreachables conf on interfaces.

I would debug the trace by using debug icmp to see you you get anything back from remote router... Eg:

ldvd1mr01#trace 135.136.28.50

Type escape sequence to abort.

Tracing the route to 135.136.28.50

1 192.247.3.1 0 msec 4 msec 0 msec

2 135.136.28.50 0 msec 0 msec 0 msec

Router1#

3w5d: ICMP: time exeeded rcvd from 192.247.3.1

3w5d: ICMP: dst (192.247.3.17) port unreachable rcv from 135.136.28.50

DALE FRANCIS · ‎08-13-2002

One more test you could do...

I have written an extended access-list for a particular destination host, below is the result of the trace..

ldvd1mr01#debug ip packet 101 detail

IP packet debugging is on (detailed) for access list 101

ldvd1mr01#

ldvd1mr01#term mon

ldvd1mr01#trace 135.136.28.50

Type escape sequence to abort.

Tracing the route to 135.136.28.50

1 192.247.3.1 0 msec 4 msec 0 msec

2 135.136.28.50 4 msec 0 msec 0 msec

ldvd1mr01#

3w5d: datagramsize=60, IP 35950: s=192.247.3.17 (local), d=135.136.28.50, totlen 28,

fragment 0, fo 0, cef process switched

3w5d: UDP src=37431, dst=33434

3w5d: datagramsize=60, IP 35950: s=192.247.3.17 (local), d=135.136.28.50 (Vlan28), t

otlen 28, fragment 0, fo 0, sending

3w5d: UDP src=37431, dst=33434

3w5d: datagramsize=92, IP 35950: s=192.247.3.17 (local), d=135.136.28.50 (Vlan28), t

otlen 28, fragment 0, fo 0, sending full packet

3w5d: UDP src=37431, dst=33434

3w5d: datagramsize=60, IP 35952: s=192.247.3.17 (local), d=135.136.28.50, totlen 28,

fragment 0, fo 0, cef process switched

3w5d: UDP src=36720, dst=33435

3w5d: datagramsize=60, IP 35952: s=192.247.3.17 (local), d=135.136.28.50 (Vlan28), t

otlen 28, fragment 0, fo 0, sending

3w5d: UDP src=36720, dst=33435

3w5d: datagramsize=92, IP 35952: s=192.247.3.17 (local), d=135.136.28.50 (Vlan28), t

otlen 28, fragment 0, fo 0, sending full packet

3w5d: UDP src=36720, dst=33435

3w5d: datagramsize=70, IP 35955: s=192.247.3.17 (local), d=135.136.28.50, totlen 28,

fragment 0, fo 0, cef process switched

3w5d: UDP src=38482, dst=33436

3w5d: datagramsize=70, IP 35955: s=192.247.3.17 (local), d=135.136.28.50 (Vlan28), t

otlen 28, fragment 0, fo 0, sending

3w5d: UDP src=38482, dst=33436

3w5d: datagramsize=92, IP 35955: s=192.247.3.17 (local), d=135.136.28.50 (Vlan28), t

otlen 28, fragment 0, fo 0, sending full packet

3w5d: UDP src=38482, dst=33436

3w5d: datagramsize=70, IP 35957: s=192.247.3.17 (local), d=135.136.28.50, totlen 28,

fragment 0, fo 0, cef process switched

3w5d: UDP src=40994, dst=33437

3w5d: datagramsize=70, IP 35957: s=192.247.3.17 (local), d=135.136.28.50 (Vlan28), t

otlen 28, fragment 0, fo 0, sending

3w5d: UDP src=40994, dst=33437

3w5d: datagramsize=92, IP 35957: s=192.247.3.17 (local), d=135.136.28.50 (Vlan28), t

otlen 28, fragment 0, fo 0, sending full packet

3w5d: UDP src=40994, dst=33437

3w5d: datagramsize=70, IP 35959: s=192.247.3.17 (local), d=135.136.28.50, totlen 28,

fragment 0, fo 0, cef process switched

3w5d: UDP src=39424, dst=33438

3w5d: datagramsize=70, IP 35959: s=192.247.3.17 (local), d=135.136.28.50 (Vlan28), t

otlen 28, fragment 0, fo 0, sending

3w5d: UDP src=39424, dst=33438

3w5d: datagramsize=92, IP 35959: s=192.247.3.17 (local), d=135.136.28.50 (Vlan28), t

otlen 28, fragment 0, fo 0, sending full packet

3w5d: UDP src=39424, dst=33438

3w5d: datagramsize=70, IP 35961: s=192.247.3.17 (local), d=135.136.28.50, totlen 28,

fragment 0, fo 0, cef process switched

3w5d: UDP src=33030, dst=33439

3w5d: datagramsize=70, IP 35961: s=192.247.3.17 (local), d=135.136.28.50 (Vlan28), t

otlen 28, fragment 0, fo 0, sending

3w5d: UDP src=33030, dst=33439

3w5d: datagramsize=92, IP 35961: s=192.247.3.17 (local), d=135.136.28.50 (Vlan28), t

otlen 28, fragment 0, fo 0, sending full packet

3w5d: UDP src=33030, dst=33439

Gilles Dufour · ‎08-13-2002

do a 'sh ip route 135.136.28.50'.

What is the next-hop ip address ?

Can you ping it ?

can you do a trace to it ?

Could you check the config of this device and check ACL or sth preventing icmp messages.

Finally, if you don't see anything, put a sniffer on the vlan and see if this next-hop sends icmp messages back to the original router that issued the trace.

stan.beckmann · ‎08-13-2002

Can you ping all the intermediate routers/switches... maybe have you got the route back from the final router but not on the intermediate ones