06-28-2007 03:14 AM - edited 03-03-2019 05:38 PM
Looking at the logs Grin router 1 and 2 go to exstart for most neighbours and then fails due to too many retransmits. However Glue routers at the same time go to loading and full. So one end is exstart and the other end is full. I didn't think this was possible.....Can anybody crack this..?
06-28-2007 03:27 AM
Hi,
This might be an MTU problem, please review the following document:
http://www.cisco.com/en/US/tech/tk365/technologies_tech_note09186a0080093f0d.shtml
HTH,
Mohammed Mahmoud.
06-28-2007 03:44 AM
Hi,
I agree with mohamed, u can try mtu-ignore command and try it. good problem you have posted.
Thanks and Regards,
Srinath
06-28-2007 04:52 AM
ThankU folks for the response...
I hve cfgured the mtu-ignore and that does not seem to be the problem :-( this is hw the config looks like on the core switch
--------------------------------
CONFIGURATION GRIN SIDE
interface Vlan101
description *** MAN_GRIN2GLUEVLAN ***
ip address 192.168.1.8 255.255.255.240
no ip redirects
no ip unreachables
ip ospf message-digest-key 1 md5 7 *******
ip ospf network point-to-multipoint
ip ospf cost 10
ip ospf mtu-ignore
!
interface Vlan102
description *** MAN_B2GRIN VLAN ***
ip address 192.168.1.33 255.255.255.240
no ip redirects
no ip unreachables
ip ospf message-digest-key 1 md5 7 *******
ip ospf network point-to-multipoint
ip ospf cost 10
ip ospf mtu-ignore
!
router ospf 100
router-id 10.10.255.5
log-adjacency-changes
area 100 authentication message-digest
redistribute static metric 50 subnets route-map FILTER-ROUTES
passive-interface default
no passive-interface Vlan101
no passive-interface Vlan102
network 192.168.5.0 0.0.0.15 area 100
network 192.168.1.32 0.0.0.15 area 100
06-28-2007 05:24 AM
From all the adjacent routers please type the following:
show ip os int bri
show system mtu
show ip os nei
and post the results here.
Did you try to isolate the problem and see if it's authentication related ?
06-28-2007 06:10 AM
Hi,
Agree with Edison, and please make sure that both routers don't have the same RID.
HTH,
Mohammed Mahmoud.
06-28-2007 07:04 AM
Hi,
if you can run a debug of ospf.
debug ip ospf events
and pass on the o/p it will be of great help.
I suggest to check the port status of the switches as well.
Thanks and Regards,
Srinath.M
Cable&Wireless
06-28-2007 08:04 AM
As discussed, why would you configure ospf as point to multipoint on metro ethernet? Suggest you change this to barodcast which is default on Ethernet?
06-28-2007 08:14 AM
Can anyone tell what is the significance of using "ip ospf network point-to-multipoint"..
06-28-2007 08:19 AM
I dont see any reason to enable the "ip ospf network point-to-multipoint". This will not make any difference on broadcast network.It will make a difference on Frame-relay or ATM PVC's..
-amit singh
06-28-2007 02:32 PM
"ip ospf network point-to-multipoint" configuration allows you to run OSPF without the need of a DR or BDR.
Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: