Re: OSPF network statement

From: fwells12 (fwells12@xxxxxxxxxxx)
Date: Thu Dec 28 2000 - 19:57:38 GMT-3


   
Your answers were exactly what I came up with. Even though I can figure out
the supernets easily, I am still struggling to put them into good network
and area range commands.

You are right about the question being flaky. I should have said how do I
do this with the fewest commands. Thanks for pointing this out Brian and
Eric.

----- Original Message -----
From: Eric Lemmons <ericlemmons@yahoo.com>
To: <ccielab@groupstudy.com>
Cc: <ccielab@groupstudy.com>; 'Connary, Julie Ann' <jconnary@cisco.com>;
fwells12 <fwells12@hotmail.com>
Sent: Thursday, December 28, 2000 1:42 PM
Subject: RE: OSPF network statement

> Good point, Brian.
> I was thinking of "conserving" addresses, in the sense that the other
> addresses would be available to be used on the same router, if
> needed, and be put in different areas. Or, to be used with other
> routing protocols for that matter. If you had been given the address
> space in question, and there was a need to use some of that range
> with other routing protocols, that would be "conserving" addresses,
> in a way, I t hink.
> I agree, the question was sort of flaky :).
>
> --- Brian Hescock <bhescock@cisco.com> wrote:
> > The wildcard bits don't waste any ip addresses at all, all it
> > merely does
> > is turn on ospf on an interface. ospf takes the network mask from
> > the
> > interface, not your network statement. The only way to save
> > address space
> > is to use the most efficient mask on the interface that meets the
> > requirements.
> >
> > In summary, it was a bad question, the requirement wasn't really
> > clear
> > given the information we had.
> >
> > Brian
> >
> > On Thu, 28 Dec 2000, Eric Lemmons wrote:
> >
> > > As I understand the requirement asked for, "What is the correct
> > > command to insert these networks into the OSPF process in the
> > most
> > > efficient manner? By that, I mean conserving the maximum amount
> > of
> > > IP addresses", the correct answer would be the host route answer.
> > > Meaning, Julie Ann's answer. The requirement was not concerned
> > with
> > > minimizing the number of router config statements, but
> > "conserving
> > > the maximum amount of IP addresses". With the 0.0.0.0 mask, this
> > > leaves all of the other IP addresses available between the
> > 20,28,36,
> > > and 44 subnets, instead of "swallowing" them in the mask. For
> > > example, you couldn't put an IP address like 192.168.21.x into a
> > > different area, because it has been put into area 0 with the
> > larger
> > > statement that Earl suggests, "Network 192.168.0.0 0.0.63.255
> > area
> > > 0".
> > > I know this is all pretty picky :).
> > > But I think the individual 0.0.0.0 masks fits the requirements
> > the
> > > best.
> > >
> > > --- Brian Hescock <bhescock@cisco.com> wrote:
> > > > Earl's answer is probably the one you're looking for if we
> > > > understand your
> > > > requirement correctly. But if you aren't told you must only
> > have
> > > > one
> > > > network entry, I agree with the other person that said to use
> > the
> > > > ip
> > > > address off of the interface and 0.0.0.0 wildcard bits, as in:
> > > >
> > > > network 172.168.1.1 0.0.0.0 area 0
> > > >
> > > > There's less chance for error and it's obvious if the proctor
> > > > changes it
> > > > because the ip address will always be the ip address from the
> > > > interface
> > > > and the wildcard bits are always 0.0.0.0. This is also a
> > > > recommended
> > > > way to do it in a production network *IF* you don't have a lot
> > of
> > > > interfaces, it makes troubleshooting far easier.
> > > >
> > > > By the way, why such a huge network for the loopback, why not a
> > /24
> > > > or a /30 or /32? It's going to be advertised as a /32 anyway
> > if
> > > > you don't
> > > > change the ospf network type to something other than the
> > default.
> > > >
> > > > Brian
> > > >
> > > > On Thu, 28 Dec 2000, Earl Aboytes wrote:
> > > >
> > > > > I think this is the answer that you are looking for.
> > > > >
> > > > > Network 192.168.0.0 0.0.63.255 area 0
> > > > >
> > > > > This would put all networks 192.168.0.0 through 192.168.63.0
> > in
> > > > area 0
> > > > >
> > > > > Watch out that you don't have any other conflicts. To be
> > > > cautious I always
> > > > > make separate entries and make sure that my masks match my
> > > > wildcards.
> > > > >
> > > > > Earl Aboytes CCIE 6097
> > > > >
> > > > > PS. Sam@datastreet, if you read this post I want you to know
> > that
> > > > I think
> > > > > that it's pretty cool how involved you are in this CCIE
> > thing.
> > > > My dad can't
> > > > > even remember the meaning of the four letters C-C-I-E.
> > > > >
> > > > > -----Original Message-----
> > > > > From: Connary, Julie Ann [mailto:jconnary@cisco.com]
> > > > > Sent: Thursday, December 28, 2000 11:51 AM
> > > > > To: fwells12
> > > > > Cc: ccielab@groupstudy.com
> > > > > Subject: Re: OSPF network statement
> > > > >
> > > > > Since each loopback interface will be a host route, use 4
> > network
> > > > > statements like:
> > > > >
> > > > > network 192.168.20.2 0.0.0.0 area X
> > > > >
> > > > > Julie Ann
> > > > >
> > > > > At 11:17 AM 12/28/2000 -0800, fwells12 wrote:
> > > > > >I have the following IP addresses configured as loopback
> > > > interfaces on =
> > > > > >an OSPF router. What is the correct command to insert these
> > > > networks =
> > > > > >into the OSPF process in the most efficient manner? By
> > that, I
> > > > mean =
> > > > > >conserving the maximum amount of IP addresses.
> > > > > >
> > > > > >interface Loopback0
> > > > > > ip address 192.168.20.2 255.255.248.0
> > > > > >!
> > > > > >interface Loopback1
> > > > > > ip address 192.168.28.2 255.255.248.0
> > > > > >!
> > > > > >interface Loopback2
> > > > > > ip address 192.168.36.2 255.255.248.0
> > > > > >!
> > > > > >interface Loopback3
> > > > > > ip address 192.168.44.2 255.255.248.0=20
> > > > > >
> > > > > >Cheers.
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > >



This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 08:26:12 GMT-3