[DNSOP] BIND use of compiled defaults

bmanning@vacation.karoshi.com Tue, 08 June 2010 04:12 UTC

Date: Tue, 08 Jun 2010 04:11:58 +0000
From: bmanning@vacation.karoshi.com
To: dnsop@ietf.org
Message-ID: <20100608041158.GA30130@vacation.karoshi.com.>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Disposition: inline
User-Agent: Mutt/1.4.1i
Cc: bmanning@vacation.karoshi.com
Subject: [DNSOP] BIND use of compiled defaults
Precedence: list

So ISC has allowed BIND to build with some default zones being created.  I think this
is - to coin a phrase - suboptimal and yet more code I have to rip out of the BIND distro...
but that is not the point of this missive... :)

I will use two of the automatically created zones to illistrate a potential point and then
ask a question.  Mark has "bracketed" the IPv4 space with the following two zone stanzas:


0.in-addr.arpa.

and

255.255.255.255.in-addr.arpa.


clearly the first incalulates the entire 0/8 netblock...  while the latter only incalcualtes
an IPv4 /32 or a host entry.


historically, one would define the local network with preceeding zeros, e.g.  


	0.0.0.152  with a netmask of 255.255.255.0  is the host .152 on the local network

and only the "all-zeros" /32 or 0.0.0.0/32 was special - reserved for broadcast.

and yet we see the ISC code reserving the entire /8 as an automatic zone.


If there was any consistancy here, ISC should have created the zone

255.in-addr.arpa.   or the 255/8 netblock 

but they did not.  They created a zone cut for a /32 - which (other than zome of my own
older configurations) seems to be unique.


So the question -  how common do we expect /32 delegations to become in future?


--bill

[DNSOP] BIND use of compiled defaults bmanning
Re: [DNSOP] BIND use of compiled defaults Mark Andrews
Re: [DNSOP] BIND use of compiled defaults bmanning
Re: [DNSOP] BIND use of compiled defaults Mark Andrews
Re: [DNSOP] BIND use of compiled defaults Tony Finch