dns based loadbalancing/failover

Paul Vixie vixie at vix.com
Sun Oct 7 03:40:47 UTC 2001


> There obviously is a need for an 'official' method to do global load
> balancing using DNS.

Ouch!  No, there isn't.  Not "obvious" to me, that is.

> Let's face it, people are doing it now on a not so large scale but that
> is rapidly changing because of the introduction of both hardware and
> software solutions that (mis)use DNS to overcome it's current limitation.

DNS has no current limitation that is relaxed by making it less coherent.
People abuse DNS due to limitations in other parts of the TCP/IP stack, but
DNS coherency introduces no problems of this kind on its own behalf.

> I'm not very interested in the discussion why this behaviour would be
> broken. It's for more interesting to talk about improving DNS so that
> there will be room for things like load balancing or dynamic DNS. In
> such a way that people will not start screaming when they see TTLs of
> 30 seconds or non-linear behaviour of load balancers.

If your goal is to arrange for global content mirroring, and binding of
content clients to whichever content server will give them the best 
measured performance for any given transaction, then using DNS qualifies
for a "you're digging in the wrong place" award.  (You won't find what
you're looking for but you will make a hell of a mess everyplace else.)

Note that if you'd like to debate fine points of DNS, there's a mailing
list (namedroppers at ops.ietf.org) for it, and that such traffic would be
off-topic for this (nanog@) list.



More information about the NANOG mailing list