Re: OSPF Area0 Vs Non-Zero Area [ Anyone Please ]

From: Rob Clav <robclav_at_gmail.com>
Date: Sun, 7 Nov 2010 23:43:02 +0100

Bilal,
 At osfp there are a rule (same cost) that says:
-If you have the route thru area 0 you will use that one.
-If not, then you will go thru your same area
-If not, then you will go thru other area

Sorry, because my last email it's a bit confusing. I was trying to explain
you how the cost is calculate between one router to other router, not when
this cost is already calculated how is use it.
So, you RIGHT, if you have a shorter way thru inter or intra area route you
will use it. The question answered by me(expected) is why the cost seems
have no sense sometimes. Or how ospf must have area 0 connected and happens
if not.
hth,
RobClav

2010/11/7 Bilal Hansrod <bilal.hansrod_at_gmail.com>

> Thank you Rob for the reply- Sorry if I have created a confusion, but
> topology as below:
>
> R6 (A1) - Virtual Link - R1 (A0) (In a way R6 is also connected to A0, via
> Virtual Link.
> (A1)
> | R6 is connected to R4 in area 1
> (A1)
> R4
>
> The route is preferred over area 1 because it is shortest. This is default
> behaviour of Cisco OSPF implementation.
>
> So what I understand from your email is that it will never prefer route via
> non-zero area, meaning the specified route is always routed via area 0. Is
> this what you mean? but as per my understanding by default it can select
> route via non-zero area if that is the shortest and it does NOT have to
> choose route, routed via area 0 - If you want to change this behaviour and
> force to use route via area 0, then we need to disable capability transit.
>
> Thanks again -
>
> Regards,
>
> Bilal Hansrod
>
>
>
>
> On Sun, Nov 7, 2010 at 8:05 PM, Rob Clav <robclav_at_gmail.com> wrote:
>
>> Hi Bilal,
>>
>> the questions is, Why all the areas must be connected to Area 0, and what
>> happens if you lose the connectivity to area 0?
>> Basically, I recommend to you think about, how do you think ospf avoid
>> routing loops when dijkstra's algoritm is calculating the shortest path?
>> How other dynamics routing protocols as RIP or BGP do it? Maybe
>> Split-Horitzon?
>>
>> So the answer to your q|estions:
>> OSFP, don't use the shortest path from "R6" to "R1" thru area 1 because
>> doing that you can loop the network. The mechanism that OSPF has is use
area
>> 0 as a central point. So you are nearest of farest from area 0 non nearest
>> or farest from device A to B.
>> If you add the cost from router 6 to area 0 and later you add the cost
>> from router 1 to area 0, and inside the area 0 to the DR if you have it,
>> then you will have "total cost".
>>
>> You need to add your own cost, because the cost is calculate it, from
>> router 6 to the DR(itself) PLUS the other router R1 to the DR. At this
last
>> point is when you are adding your own cost, because the total cost is from
>> area 0 and to area 0. Non from router 6 to router 1.
>>
>> HTH,
>> Robclav
>>
>> --
>> Robert Clavero
>> CCIE RS/wr, CCNP, CCSP, CCSE NGX, SCSA 9, WLFES, BNP y JNCIA WX
>> blog:http://robclavbcn.blogspot.com
>>
>> web:http://www.kubsolutions.com
>>
>>
>

--
Robert Clavero
CCIE RS/wr, CCNP, CCSP, CCSE NGX, SCSA 9, WLFES, BNP y JNCIA WX
blog:http://robclavbcn.blogspot.com
 web:http://www.kubsolutions.com
Blogs and organic groups at http://www.ccie.net
Received on Sun Nov 07 2010 - 23:43:02 ART

This archive was generated by hypermail 2.2.0 : Sun Dec 05 2010 - 22:14:55 ART