[c-nsp] ospf mtu issue

[Shuvo Shuvo]

Hi

We have a customer who has 8 branches. 6 branches gre over 
ipsec tunnel is established with their headoffice without having any 
problem. Only problem is happening at 2 branches. recently we have 
implemented MPLS in our network and that customer all the branches are 
connected through MPLS vrf. 6 branches are connected through VLAN point 
to point ip with us and those branches are coming through layer2 
switches upto PE router. rest 2 branches is connected through L2TPV3 
router in between then through VLAN upto PE. the scenerio is

branch
 office ---->L2TPV3 Router fa0/1 xconnect------>L2TPV3 router 
Fa0/0 pseduowire-----> PE Router 1 VLAN----> PE Router2 VRF with 
same VLAN

L2TPV3 router 

pseudowire-class ethernet
 encapsulation
 l2tpv3
 ip local interface Loopback0
 ip pmtu

nterface 
FastEthernet0/1
 no ip address
 ip route-cache flow
 duplex 
auto
 speed auto
 no cdp enable
 xconnect 10.10.10.254 110 
pw-class ethernet
==========================================

Customer
 branch router

 interface Tunnel1
 description "headoffice"
 bandwidth
 1024
 ip address 111.2.1.6 255.255.255.252
 ip mtu 1452
 no ip
 route-cache cef
 no ip route-cache
 ip tcp adjust-mss 1360
 no
 ip mroute-cache
 tunnel source GigabitEthernet0/0
 tunnel 
destination 10.9.9.1
 tunnel path-mtu-discovery


customer 
headoffice

interface Tunnel21
 description "HO-BGRA-ALAP"
 ip
 address 111.2.1.5 255.255.255.252
 ip mtu 1452
 ip tcp adjust-mss
 1360
 ip ospf cost 10
 tunnel source 10.9.9.1
 tunnel 
destination 10.9.9.17
 tunnel path-mtu-discovery


problem 
is ospf neighbour is not getting up automatically if the router gets 
down. after clearing the ospf process tunnel is getting up. sometimes 
tunnel is getting down by giving the ospf too many retransmission error.
 before applying the MTU size we saw the debug ip ospf adjacency which 
was showing 

*Mar 14 17:03:00.493 BDT: OSPF: Send DBD to 
172.24.24.1 on Tunnel1 seq 0x1605 opt 0x52 flag 0x3 len 1452
*Mar 14 
17:03:00.493 BDT: OSPF: Retransmitting DBD to 172.24.24.1 on Tunnel1 
[25]
*Mar 14 17:03:05.502 BDT: OSPF: Killing nbr 172.24.24.1 on 
Tunnel1 due to excessive (25) retransmissions
*Mar 14 17:03:05.502 
BDT: OSPF: 172.24.24.1 address 172.21.1.9 on Tunnel1 is dead, state DOWN
*Mar
 14 17:03:05.502 BDT: %OSPF-5-ADJCHG: Process 100, Nbr 172.24.24.1 on 
Tunnel1 from EXCHANGE to DOWN, Neighbor Down: Too many retransmissions
*Mar
 14 17:04:13.666 BDT: OSPF: Rcv DBD from 172.24.24.1 on Tunnel1 seq 
0x1EA0 opt 0x52 flag 0x7 len 32  mtu 1476 state EXSTART
*Mar 14 
17:04:13.670 BDT: OSPF: First DBD and we are not SLAVE
*Mar 14 
17:04:13.698 BDT: OSPF: Rcv DBD from 172.24.24.1 on Tunnel1 seq 0xF5D 
opt 0x52 flag 0x2 len 1452  mtu 1476 state EXSTART

so we set ip 
MTU 1452 at branch office and headoffice tunnel. can someone help us 
what would be the MTU setting which will help to gets the ospf neighbour
 and tunnel up and running 		 	   		  
_________________________________________________________________
The New Busy is not the old busy. Search, chat and e-mail from your inbox.
http://www.windowslive.com/campaign/thenewbusy?ocid=PID27925::T:WLMTAGL:ON:WL:en-US:WM_HMP:032010_3