Subject: Re: dhcpd(8) _cannot_ be completely disabled on an interface
To: Christopher W. Richardson <cwr@nexthop.com>
From: Andrew Brown <atatat@atatdot.net>
List: tech-net
Date: 01/11/2002 18:44:13
>> I don't think so, not if numbers are assigned sequentually at
>> boot and never repeated.  In practice today, a 32-bit serial
>> number will never be repeated; a 64-bit probably will never be
>> repeated for the foreseeable lifetime of NetBSD.
>
>While this might be true,
>
>struct ifnet {                          /* and the entries */
>[...]
>        u_short if_index;               /* numeric abbreviation for this if */
>
>in practice today 16-bit serial numbers are repeated.
>
>Not that this says anything about names and addresses being
>repeated less frequently than if_index, just that 32 or 64 bit
>if_index would be nice. :)

while i'll agree that in theory a 16 bit interface number might roll
over (and what if it did?), in practice i think it won't happen very
much at all.  i can cause the next assigned interface number to
increase by repeatedly popping out and pushing back in my pcmcia
network card, or i can repeatedly create and destroy, for example,
tun0 (that might roll it over in about ten minutes), but short of
doing something wrong or trying to roll it over intentionally, i can't
imagine it would happen naturally.

-- 
|-----< "CODE WARRIOR" >-----|
codewarrior@daemon.org             * "ah!  i see you have the internet
twofsonet@graffiti.com (Andrew Brown)                that goes *ping*!"
andrew@crossbar.com       * "information is power -- share the wealth."