override ttl=0

[Adam Tkac]

On Thu, Jan 03, 2008 at 02:07:48PM +0100, Stephane Bortzmeyer wrote:
> On Wed, Jan 02, 2008 at 08:50:55PM -0500,
>  Kevin Darcy <kcd at chrysler.com> wrote 
>  a message of 159 lines which said:
> 
> > Well, it *shouldn't* be included in BIND, in my opinion. If a DNS
> > record comes from an authoritative source with TTL=0 then it is to
> > be treated as highly dynamic,
> 
> On the other hand, many of these TTL=0 records are set up that way
> because the administrator does not understand the DNS and irrationally
> believes that he has no control of his data if the TTL is non-null (a
> similar effect can be seen on the Web: many Web servers attempt to
> disable caching because they do not understand it).
> 
> Can we ignore the fact that many high-traffic DNS recursors, such as
> those of big ISPs, already incorporate such a patch, in order to
> reduce the humongous amount of DNS traffic that comes from
> ignorant-managed DNS zones?
> 
> If the "bumping" of the TTL is not done by default, is such a patch
> really harmful?
> 
> 

If you include that patch to main source broken servers won't be fixed
and uneducated admins will stay uneducated. I think We should report
that problem to correct place (bad administrator or bad software) and
stop include such hacks into BIND. This is similar problem like EDNS -
many elements doesn't honor RFCs and aren't configured correctly. Try
bypass those problems in BIND is really bad idea.

Adam

-- 
Adam Tkac, Red Hat, Inc.