Re: OSPF DR/BDR

Tony,

I totally agree with you at this point. My questions is about DR minimize the LSA replication. I saw that while I watched the INE video.

Thanks for the sharing!

--
Mohammad Mousa
CCIE #36990

On Mar 21, 2013, at 10:52 AM, "Tony Singh" <mothafungla_at_gmail.com> wrote:

isnt the alternative to have full mesh in which case would cause more
cpu/overhead with LSA's crashing into each other left/right/centre
after an SPF calculation? i.e all routers peering with each other on a
broadcast segment, not possible by design i think...

well described here

"OSPF routers on a multiaccess segment will elect a designated router
(DR) and backup designated router (BDR) with which all non-designated
routers will form an adjacency. This is to ensure that the number of
adjacencies maintained does not grow too large; a segment containing
ten routers would require 45 adjacencies to form a mesh, but only 17
when a DR and BDR are in place."

http://packetlife.net/blog/2008/jun/19/ospf-network-types/

network types in OSPF & NSSA variants are the most confusing thing IMO
and the hardest to maintain!

On 21 March 2013 13:02, Carlos G Mendioroz <tron_at_huapi.ba.ar> wrote:
> But an adjacency is more than send and forget! You have to take the whole
> system into consideration for doing a proper analysis.
>
> It is in the details that unmanaged communication breaks.
> -Carlos
>
> Mohammad Mousa @ 21/03/2013 09:21 -0300 dixit:
>
>> Carlos,
>>
>> This is exactly what I mean. If you have two routers or 100 on the LAN
>> segment. You have send the update once not twice with DR.
>>
>> I think with the existence of the DR, you can easily help other routers to
>> graph the network by generating type-2 Network LSA to provide a good input
>> for the SPT calculations. I mean one type-2 LSA describe all the segment is
>> better than X number of type one LSA to all connection. Do you agree with me
>> Carlos?
>>
>> --
>>
>> Mohammad Mousa
>> CCIE #36990
>>
>> On Mar 21, 2013, at 7:17 AM, "Carlos G Mendioroz" <tron_at_huapi.ba.ar>
>> wrote:
>>
>> Your perceived lack of advantage comes from paying only attention to the
>> number of packets sent in the network, and not to the whole cost of
>> mantaining an adjacency.
>>
>> From your point of view, having 2 routers or having 100 would cost the
>> same ? (1 tx per LSU message).
>>
>> -Carlos
>>
>> Mohammad Mousa @ 21/03/2013 03:46 -0300 dixit:
>>>
>>> Hello Folks,
>>>
>>> Appreciate if someone could clear up this point for me.While I'm
>>> reviewing the OSPF, I noticed this issue.
>>> In the ospf Net-type (broadcast,Non-broadcast). The advantages of using
>>> the DR/BDR are:
>>>
>>> 1-Minimize the adjacencies.
>>> 2-Minimize LSA replication.
>>>
>>> Suppose we have ethernet LAN segment between four routers R1,R2,R3,R4. R1
>>> is DR, R2 is BDR. I know drothers will stay in the 2-way states. When R3
>>> generate LSU it will go for 224.0.0.6 and then DR will replicate this to all
>>> multicast ospf routers 224.0.0.5. My question is suppose we don't have the
>>> DR. R3 will generate the LSU to 224.0.0.5 (All routers on that LAN segment).
>>> So using DR IS NOT giving me the benefit by reducing the LSA replication on
>>> the segment. Am I correct? Any thought will be highly appreciated!
>>>
>>> Thanks in advance
>>> --
>>> Mohammad
>>> CCIE #36990
>>>
>>>
>>> Blogs and organic groups at http://www.ccie.net
>>>
>>> _______________________________________________________________________
>>> Subscription information may be found at:
>>> http://www.groupstudy.com/list/CCIELab.html
>>
>>
>
> --
> Carlos G Mendioroz <tron_at_huapi.ba.ar> LW7 EQI Argentina
>
>
> Blogs and organic groups at http://www.ccie.net
>
> _______________________________________________________________________
> Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html

Blogs and organic groups at http://www.ccie.net
Received on Thu Mar 21 2013 - 10:58:21 ART