[sclug] Sticky DNS Glue.

[James Fidell]

Quoting Roland Turner (raz.fpyht.bet.hx at raz.cx):

> > Performance-wise, most people will have cached the NS data after the
> > first query anyway, so whilst there's a small performance gain from
> > having glue records, it's not that significant.
> 
> This argument doesn't fly; the same resolvers that cache the NS
> record will cache whatever other information they obtain during
> the same resolution from the same server(s), and do so for the
> same period of time (oddball TTL configurations in the domains
> notwithstanding). If the NS record is in cache, then it's pretty
> likely that whatever was being looked up in the domain is also in
> cache. Across the total number of non-cached DNS resolutions that
> are performed, the vast majority request just one item from a
> given domain ("what are your MX's?", or "what is the address of
> www?", etc.). N.B. This is not the same as the total number of
> DNS resoltuions Internet-wide; it is the number of non-cached
> reolutions which determines performance.
> 
> Consequently, the absence of glue records will typically double
> (at least) the amount of DNS traffic required to resolve lookups
> in the delegated domain.

The point of caching servers though is that they will provide data
to large numbers of clients.  When a cache is well-utilised, the
difference in performance between using one query to populate the
cache or two isn't the significant issue.

Actually, I realise there may be a disparity in the "perforamnce"
that we're talking about here.  "Performance" as viewed by the
average end-user and by the person running the nameserver aren't
always the same thing and may even be inversely related.

> > How about someone bigger?  Say, Nominet?  They do hold
> > glue records for all my .uk domains, but only because my nameservers
> > appear inside a .uk domain anyway and therefore the glue records are
> > required to exist.  They don't carry glue records where the nameservers
> > are, say, in .com or .net.
> 
> Arguments of the form "those big[1] guys think X is true,
> therefore X must be true" hold about as much sway with me as
> arguments of the form "most people think X is true, therefore X
> must be true"; which is to say "none at all".

I wasn't attempting to prove anything either way, merely provide
a counter-example to your original assertion.  Since you had a .cx
domain and I'm a Nominet tag-holder, they were the first two
possibilities that occurred to me to check.

There's more to it than a counter-example though.  Nominet is, from
a technical point of view fairly well run and in the eight years or
thereabouts it has been running has always had people I know to be
fairly clueful involved at a high level.  So, why do they choose not to
carry glue records for nameservers unless they're strictly necessary?
It's not like they don't expect you to jump through all kinds of other
hoops to register a domain, nor that they wouldn't change policy at some
point if they decided it was a good idea (they've done so a number of
times with other things).  As it happens, one of my clients is a company
run by one of the members of Nominet's Policy Advisory Board, so next
time I talk to him, I'll try to remember to ask.  Even if we've bored
everyone else on the list to tears by now, I'm sure you and I will find
the answer of interest.

James