From: Jason Madsen (madsen.jason@gmail.com)
Date: Thu Oct 23 2008 - 23:07:21 ARST
I stand corrected again...it did work, but it was VERY slow. I was testing
via Dynamips btw...
Jason
On Thu, Oct 23, 2008 at 7:05 PM, Jason Madsen <madsen.jason@gmail.com>wrote:
> ok I take back part of what I said...with "ip ospf net non" an adjacency
> doesn't seem to form without using "neighbor" on both ends of the link.
> With "ip ospf net point-to-multi non" it didn't seem to matter either way.
>
> Jason
>
>
> On Thu, Oct 23, 2008 at 6:59 PM, Jason Madsen <madsen.jason@gmail.com>wrote:
>
>> Hi Group,
>>
>> To be officially correct, do we need to use the "neighbor" command on one
>> side of each link or on both sides of each link? From what I've seen OSPF
>> always seems to work just fine with "neighbor" on one end of the link only
>> and as a matter of face I've read more than one writeup stating that
>> configuring it on both ends of a link rather than just one can actually
>> cause problems in some scenarios.
>>
>> This is strictly an "approved in the lab" type question. As I stated,
>> I've never had issues with just configuring this command on one end of a
>> given link without any issues. I did find one somewhat vague statement in
>> the Command Reference that leads me to believe that in Cisco's eyes we are
>> to use this command on both ends of a link for it to be "correct". Here is
>> the statement:
>>
>> *"One neighbor entry must be included in the Cisco IOS software
>> configuration for each known nonbroadcast network neighbor*" (
>> http://www.cisco.com/en/US/docs/ios/iproute/command/reference/irp_osp2.html#wp1013124
>> ).
>>
>> I guess technically this statement would lead me to believe that it should
>> be on both ends. Anyone have any insight as to what would be "correct" in
>> a lab scenario?
>>
>> Thanks,
>> Jason
Blogs and organic groups at http://www.ccie.net
This archive was generated by hypermail 2.1.4 : Sat Nov 01 2008 - 15:35:22 ARST