Software or PHP/PERL scripts for simple network management?

William Allen Simpson william.allen.simpson at
Wed Jun 20 17:39:23 UTC 2007

alex at wrote:
> I agree, DNS should *reflect* reality, but I think it is very much 
> misguided to say that DNS should be the place to have canonical 
> information (i.e. source of all data). Canonical data is in 
> routing/forwarding tables on routers/switches. That's the operational 
> reality.
Others have mentioned this, but that's just wrong.  For 20 years, there's a
reason we've been using policy-based routing, routing arbiters, etc.

> The amount of data that you need to track IP allocations just doesn't fit
> well into DNS - there's no place to store customer id/service id, the
> length of allocation (is this IP part of a /28? /29?), etc. So you'll have
> to have "canonical data" somewhere else anyway.
Others have mentioned this, but of course all that should be stored as
comments in the file.  I never found any automated tool that stored all
the information properly.  Text records with comments are flexible.

And the allocation size is extremely important, as you need pointer records
to the customers' .arpa NS records!  Surely, you don't handle everything on
8-bit boundaries in this day and age....

>> And when the routing table doesn't match, withdraw the route, and fire
>> the miscreant that failed to properly maintain the allocation data!
> Unfortunately, I'll have to say again that this doesn't scale. :)
There's a saying where I grew up:
   Ford is in the business of making cars.
   GM is in the business of making money.

The notion is that GM doesn't really care about the quality of its cars,
as long as it makes money.  Branding the local congresscritter "the
representative from GM" is not a compliment.  (Not so coincidentally, his
considerably younger trophy wife is a GM heiress.)

The 'net is what I've spent most of my adult life making.  'nuff said.

More information about the NANOG mailing list