Unable to query the nameserver

[Kevin Oberman]

> Date: Wed, 6 Oct 2010 14:03:56 -0400
> From: "Lightner, Jeff" <jlightner at water.com>
> Sender: bind-users-bounces+oberman=es.net at lists.isc.org
> 
> Of course some versions of nslookup arent' "standard" even for nslookup.
> The one on HP-UX actually interrogates local /etc/hosts file if
> nsswitch.conf says to use files first.   I got so used to doing that for
> years that when I tried to use nslookup on Linux back in 2005 I was
> miffed because it was "broken" and only looked up from name servers.
> (Someone even had the gall to point out that "ns"lookup was "name
> server" lookup).  :-)
> 
> -----Original Message-----
> From: bind-users-bounces+jlightner=water.com at lists.isc.org
> [mailto:bind-users-bounces+jlightner=water.com at lists.isc.org] On Behalf
> Of Ben McGinnes
> Sent: Wednesday, October 06, 2010 1:52 PM
> To: Kevin Darcy
> Cc: bind-users at lists.isc.org
> Subject: Re: Unable to query the nameserver
> 
> On 7/10/10 4:42 AM, Kevin Darcy wrote:
> >    
> > ISC has tried to kill it, but the beast is resilient and won't die.
> 
> Maybe we should call it a wombat then ...
> 
> > Invocations of nslookup are embedded in thousands of legacy scripts
> and
> > some folks are unable or unwilling to change them.
> 
> Nothing quite like coding/sysadmin laziness is there.  Still, I probably
> can't talk on that front.

Invocations of nslookup are embedded in thousands of BROKEN legacy
scripts. nslookup is broken. It gives answers that are, from any sane
point of view, wrong (though right from some other points of view). Most
of the users of those legacy script are completely unaware of this until
it bites them and they either kludge around the case they hit or fix the
scripts to use host (or, very rarely, dig).

Could we maybe replace nslookup(1) with a script which does a host(1) and
and re-formats the output to look like nslookup(1) output. I don;t know
that this would be easy, but it LOOKS like it would be easy.

Yes, I am sure that some script somewhere depends on some "wrong"
response from nslookup, but I can't see keeping nslookup(1) alive as is
for that amazingly unlikely case.
-- 
R. Kevin Oberman, Network Engineer
Energy Sciences Network (ESnet)
Ernest O. Lawrence Berkeley National Laboratory (Berkeley Lab)
E-mail: oberman at es.net			Phone: +1 510 486-8634
Key fingerprint:059B 2DDF 031C 9BA3 14A4  EADA 927D EBB3 987B 3751