[c-nsp] Network Address Response

Geoffrey Pendery geoff at pendery.net
Fri Jun 26 14:17:19 EDT 2009


Fascinating.  Thanks, that answers another question I've always had -
why do lots of systems require you to manually enter the broadcast
address?  I always figured they should be able to determine that from
the address and mask you've already entered, and assumed they were
being lazy or obtuse.  But it makes more sense now - they were asking
if you'd prefer to use the 0 address as the broadcast, rather than the
usual "all ones".

Thanks to Ray for the question, and Lee for the answer.


-Geoff


On Fri, Jun 26, 2009 at 12:52 PM, Lee<ler762 at gmail.com> wrote:
> On 6/25/09, Ray Burkholder <ray at oneunified.net> wrote:
>
>> I was wondering the reasoning for routers/switches to respond for the
>> network portion of an ip-address range.
>>
>> For example, a router interface A with 10.0.0.1/30 and interface B with
>> 10.0.0.5/30.
>>
>> Generate a ping from a device several hops away on the A side to the B side
>> network address of 10.0.0.4.  The router will respond with an echo reply
>> with an address of 10.0.0.1.
>>
>> Is this expected behaviour?  And the reason?
>
> Standards compliance?  From RFC-1122
>   3.3.6  Broadcasts
>
>         There is a class of hosts* that use non-standard broadcast
>         address forms, substituting 0 for -1.  All hosts SHOULD
>         recognize and accept any of these non-standard broadcast
>         addresses as the destination address of an incoming datagram.
>         A host MAY optionally have a configuration option to choose the
>         0 or the -1 form of broadcast address, for each physical
>         interface, but this option SHOULD default to the standard (-1)
>         form.
>
> So {network, 0} is a broadcast address, same as {network, -1}.
> Combine that with
>
>   3.2.1.3  Addressing: RFC-791 Section 3.2
>
>             ...   An incoming datagram is destined
>            for the host if the datagram's destination address field is:
>
>            (1)  (one of) the host's IP address(es); or
>
>            (2)  an IP broadcast address valid for the connected
>                 network; or
>
> and you get the router answering a ping to 10.0.0.4
>
> Off the top of my head, I don't know if it's a ciscoism or 'standard'
> behavior to use the IP address of the outgoing interface as the source
> address of the ping reply, but that's why you get the response from
> 10.0.0.1 instead of 10.0.0.5
>
> Regards,
> Lee
> _______________________________________________
> cisco-nsp mailing list  cisco-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/
>


More information about the cisco-nsp mailing list