Dynamic IP & cache DNS

Kevin Darcy kcd at daimlerchrysler.com
Mon Sep 10 21:12:09 UTC 2001


Cricket Liu wrote:

> > Some websites offer services to give a dynamic IP address a domain name.
> > I Tried this, it works very well. But usualy, DNS servers use a cache,
> > and then, I'd like to know how the update can be such fast (some
> > seconds). Is there an insctruction to avoid DNS entries to be cached? Or
> > what ?
>
> Each resource record in DNS carries a time-to-live field
> with it, which tells name servers how long they may cache
> that record.
>
> For example, in this record:
>
> www    60    IN    A    10.0.0.1
>
> the TTL field instructs name server to cache the record
> for 60 seconds.

Unfortunately, gratuitously-low TTLs are detrimental to the public
DNS infrastructure. If a node is going to have a particular IP address
continuously for, say, 24 hours, why force other nameservers to re-lookup the
name as often as every 60 seconds? In a more perfect world, of course, a
dynamic client would set the TTL to *exactly* as long as it was going to have
the IP address, no more, no less. This is, of course assuming that it knows
exactly how long it is going to have that address, that nothing breaks, etc.
etc. In short, it's unrealistic.

I recently proposed a protocol extension that would implement a "callback" to
nameservers when a particular piece of data changes (using a generalized form
of NOTIFY). This would, I think, provide the best of both worlds -- being
able to set a reasonably-long TTL in the general case but at the same time
being able to make sudden changes visible rapidly. Unfortunately, the
proposal was shot down as requiring the maintenance of too much state
information in nameservers. But, I haven't given up hope yet. I'm still
trying to come up with a refined version of the proposal that would be more
nameserver-resource-friendly...


- Kevin





More information about the bind-users mailing list