From: Brian Hescock (bhescock@xxxxxxxxx)
Date: Sat Oct 13 2001 - 17:31:37 GMT-3
First, the forwarding address question: using 0.0.0.0 as the wildcard
bits has absolutely nothing to do with the ospf forwarding address.
You could use non-zero wildcard bits and still have the same problem
with the forwarding address if the outbound interface on the asbr isn't
included within ospf. So as long as ospf is enable on that interface,
through either the use of the interface ip address and 0.0.0.0 wildcard
or the network address and 0.0.0.255 (or whichever wildcard bits you
want), it will work just fine.
Second question about the frame config and having to use 0.0.0.0
wildcard: not true at all, you could use the ip address of the
interface and whatever wildcard bits you want and it would work. Rather
than repeat myself (and others on the list who have posted this a 1,000
times), look up on CCO what the opsf network statement really does
(hint: by itself, the ospf network statement has absolutely nothing to
do with what is advertised by ospf).
And if you can't find it on CCO, use the groupstudy archives, the
information is in there numerous times monthly.
Brian
Ajaz Nawaz wrote:
>>If you use 0.0.0.0, your neighbor statements will not appear and you
>>will never be able to make a neighbor (this is on the hub router of a point
>>to multipoint interface).
>>
>
>If you use the router ospf command 'net 11.11.11.11 0.0.0.0 area 0' this
>associates the router ospf process to 11.11.11.11. In other words you have
>just enabled OSPF on that interface with the IP address of 11.11.11.11.
>
>I have used this configuration (no ip ospf network commands) and
>neighborships DO appear - are we talking about the same issue here ? Please
>can you shed some more light on your explanation i.e. 'If you use 0.0.0.0,
>your neighbor statements will not appear and you will never be able to make
>a neighbor ' ?
>
>regs
>jaz
>
>-----Original Message-----
>From: nobody@groupstudy.com [mailto:nobody@groupstudy.com]On Behalf Of
>John Elias
>Sent: 13 October 2001 15:28
>To: elouie@yahoo.com; xpranax@hotmail.com; bosoro@hotmail.com
>Cc: ccielab@groupstudy.com
>Subject: Re: why not to use 0.0.0.0 in OSPF
>
>
> What I have noticed, if you are using frame relay and you are NOT using
>the 'ip ospf network' statement, you need to specify the exact wildcard
>mask. If you use 0.0.0.0, your neighbor statements will not appear and you
>will never be able to make a neighbor (this is on the hub router of a point
>to multipoint interface).
>
>John E.
>CCIE #8150
>
>
>>From: "routerjocky" <elouie@yahoo.com>
>>Reply-To: "routerjocky" <elouie@yahoo.com>
>>To: "Brian Lodwick" <xpranax@hotmail.com>, <bosoro@hotmail.com>
>>CC: <ccielab@groupstudy.com>
>>Subject: Re: why not to use 0.0.0.0 in OSPF
>>Date: Fri, 12 Oct 2001 19:00:55 -0700
>>
>>Am I just dense, or can someone else explain how specifying the host
>>address
>>and host wildcard mask in the ospf network stmt have anything to do with
>>the
>>OSPF forwarding address?
>>
>>default route is 0.0.0.0 as the network address, rarely if ever used in an
>>ospf net statement as the network (unless the entire router is an internal
>>router, and one doesn't care what networks are installed on it, it'll still
>>be in the same area
>>
>>0.0.0.0 is also a host wildcard mask (as in net 171.68.1.1 0.0.0.0 area 0)
>>
>>hellllppppp me... I've fallen and I can't get up
>>-e-
>>
>>----- Original Message -----
>>From: "Brian Lodwick" <xpranax@hotmail.com>
>>To: <bosoro@hotmail.com>
>>Cc: <ccielab@groupstudy.com>
>>Sent: Friday, October 12, 2001 5:00 PM
>>Subject: why not to use 0.0.0.0 in OSPF
>>
>>
>>>I have finally found one reason for not using the individual host
>>>
>>(0.0.0.0)
>>
>>>for the network command.
>>>I've gotta go home, but check out this issue with the "OSPF Forwarding
>>>Address" if you use the 0.0.0.0 network command on the ASBR external
>>>interface.
>>>http://www.cisco.com/warp/customer/104/10.html
>>>
>>>>>>Brian
>>>>>>
>>>>From: "Bryan Osoro" <bosoro@hotmail.com>
>>>>Reply-To: "Bryan Osoro" <bosoro@hotmail.com>
>>>>To: "'Ben-Shalom, Omer'" <omer.ben-shalom@intel.com>,
>>>><ccielab@groupstudy.com>
>>>>Subject: RE: OSPF interface - a non related question
>>>>Date: Fri, 12 Oct 2001 17:25:49 -0600
>>>>
>>>>My personal habit is to use the individual host, so I can assure that
>>>>
>>no
>>
>>>>other interfaces are grabbed and put into the process. I have yet to
>>>>see actual advantages for any either way that are backed with
>>>>documentation.
>>>>
>>>>Bryan
>>>>
>>>>
>>>>-----Original Message-----
>>>>From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of
>>>>Ben-Shalom, Omer
>>>>Sent: Friday, October 12, 2001 3:52 PM
>>>>To: ccielab@groupstudy.com
>>>>Subject: OSPF interface - a non related question
>>>>
>>>>regarding the below:
>>>>
>>>>>>network 10.10.10.1 0.0.0.0 area 1, network 10.10.10.1 0.0.0.63 area
>>>>>>
>>1,
>>
>>>>or
>>>>
>>>>>>network 10.0.0.0 0.255.255.255 area 1
>>>>>>
>>>>>>
>>>>>>
>>>>>>That should fix you right up
>>>>>>
>>>>>>
>>>>I am pretty certain all of the above will, indeed work fine as the idea
>>>>is
>>>>to get the interface considered in the OSPF domain.
>>>>
>>>>Is one of the forms (the interface address specifically, the subnet it
>>>>belongs to or a supernet/classful net) considered a better form/just
>>>>better
>>>>for any reason or is using any of them just as good.
>>>>
>>>>Personally I find using the subnet the most intuitive as it is a
>>>>'network'
>>>>statement so using a host there seems awkward and the classful one can
>>>>match
>>>>interfaces I don't want included in the same area, but I see not real
>>>>reason
>>>>why this would be better.
>>>>
>>>>Omer.
>>>>
This archive was generated by hypermail 2.1.4 : Thu Jun 20 2002 - 22:33:18 GMT-3