From: Peter Van Oene (pvo@xxxxxxxxxxxx)
Date: Thu Aug 16 2001 - 01:01:12 GMT-3
I've heard of these bugs though I haven't encountered them myself. Most ISP's
I've worked with use full 0's masks simply for clarity. However, redistributio
n of external info into IGP's is usually minimized in service provider networks
which might indicate why I've not seen the bug.
*********** REPLY SEPARATOR ***********
On 8/15/2001 at 2:28 PM Monty.Majszak@Level3.com wrote:
>Just out of curiosity, Yves, you said you know of problems w/ using the
>0.0.0.0 wildcard mask. Peter, you stated, "I personally use the direct mask
>of 0.0.0.0 for all network statements." Can you guys clearify this? How
>much/how long have you been using the 0.0.0.0 Peter?
>
>P.S. in the real world, when would you want to set your network type to
>"non-broadcast" when doing the whole OSPF over Frame-relay fully-meshed
>thing?
>
>
>-Monty
>
>
>
>
>
>-----Original Message-----
>From: Yves Fauser [mailto:Yves@Fauser.de]
>Sent: Wednesday, August 15, 2001 12:57 PM
>To: Jeffrey Levine
>Cc: ccielab@groupstudy.com
>Subject: Re: router configs/OSPF - interfaces or entire networks?
>
>
>Hi Jeffrey,
>
>If you want to take an Interface into ospf it doesn't matter which of the
>statement you use.
>But be aware of the consequences that the 0.0.0.0 wildcard mask has, it
>changes the forwarding address in the database.
>To be short if you use the 0.0.0.0 Mask the forwarding address will be
>0.0.0.0
>in the LSA. If you take 0.0.0.255 the forwarding address will be the
>address
>of the next hop. You can have various Problems with 0.0.0.0 or next hop
>depending on the topology. Another thing is that there is a Bug in all
>versions when redistributing ospf into RIP.
>
>I'll send some links later.
>
>Yves
>
>
>Jeffrey Levine wrote:
>
>> I have noticed in some lab scenarios that when the instructions say that
>a
>> router interface should be advertised in a routing protocol, the labs
>> sometimes advertise only the interface, othertimes an entire subnet.
>>
>> For example, let's say it's an ethernet interface with address
>> 192.168.1.1/24 being advertised under OSPF. I've seen the following:
>>
>> router ospf 64
>> net 192.168.1.1 0.0.0.0 area 0
>>
>> or
>>
>> router ospf 64
>> net 192.168.1.1 0.0.0.255 area 0
>>
>> I wouldn't be concerned if there were some consistency. I've seen the
>lab
>> instructions state "interface" and then seen the entire network
>advertised
>> and vice-versa.
>>
>> Any thought?
>>
>> Jeffrey S. Levine
>>
This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 10:31:51 GMT-3