From: Katson PN Yeung (kyeung@xxxxxxxxx)
Date: Fri May 03 2002 - 00:35:17 GMT-3
Hi Parry,
Yes, I know that, and I notice if I do a "area 0 range" at R3 in this
scenario:
(area123)-R3----(area0)----R5-(area456)
The summary route just flood to area 123 only, but not to area 456.
I just wonder why P.791 in Solie's is so "conclusive", and don't know if
this technique is "okay to use" in the lab environment.
And I am so curious to know, in case of classless->classful redistribution,
which kind of technique is preferred most. I can name the followings:
1. tunnel interface (may require creation an ip address that is not
specifiy on lab)
2. sec ip addr (may require creation an ip address that is not specifiy
on lab)
3. area x range (require ABR)
4. ospf summary (not work for redis connected at area 0 after 12.1)
5. piggy-back OSPF.
What are their degree of preference?
At 10:00 AM 5/3/2002, Chua, Parry wrote:
>My understand is that area x (include 0) range will work on any ABR, but
>you need to know the area concepts of OSPF. If R3 only in area 0, then it
>is not a ABR. If you allow to create a dummy area at R3, then it become a
>ABR, area 0 range will work.
>
>When you do area 0 range at a ABR, it will only summary into the other
>area where the ABR is attach. For example, if R3 has area 123 and area 0,
>area 0 range at R3 will has area 0 summary into area 123 only. If you do
>area 123 range at R3, it will pass the summary to area 0 and other areas
>except arae 123.
>
> > Parry Chua
> >
> >
>
>
>-----Original Message-----
>From: Katson PN Yeung [mailto:kyeung@hkcix.com]
>Sent: Friday, May 03, 2002 9:30 AM
>To: ccielab@groupstudy.com
>Subject: Solie's P.791 confusion.
>
>
>Dear group,
>
>I have a question about Solie's book P.791.
>
>On the 2nd paragraph from the top, it saids:
>"You cannot summarize Area 0 or the backbone area. All summaries are
>flooded into area 0 and then are flooded out from that point. Therefore
>Area 0 routes cannot be summarized"
>
>However, in the scenario such as below:
>
> R4
> |s0
> |
> --s0.1---| R1---|e0
>|e0-R3< /
> --s0.2---------<
> \
> R2---|e0
>
>Where R3s0.2, R1s0, and R2s0 forming OSPF area 0 (/28 subnet)
>- R1e0 area 1 (/24)
>- R2e0 area 2 (/24)
>- R3e0 being area 3 (/24 or whatever)
>- R3 s0.1 and R4 running igrp (/24)
>
>Isn't it easy (and convenient) to create a /24 summary route by "area 0
>range x.x.x.x 255.255.255.0" at R3, and then get it redistributed to igrp?
>I did they before since the redis connected/summary ASBR method doesn't
>work now.
>
>Solie P.791 makes me confused. Can someone clarifies if this "area 0
>range" method is "okay to use" in the real lab environment?
>
>Or just like someone saids, tunnel method, sec interface method? or even
>piggy-back OSPF
>method are the more preferable way to make a /24 summary route?
>
>Thanks a lot.
This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 10:58:49 GMT-3