Hallam-Baker, Phillip wrote:
You are wrong.
I often get the impression that the laws of physics do not apply on the
planet where you live...
Take SMTP email. Imagine that a company has three mail servers at
geographically dispersed locations with independent network connections.
With MX records a local failure of IP connectivity does not need to mean
total failure of IP connectivity.
You want to extrapolate to the whole Internet and all of its
applications from a single, somewhat hypothetical, data point - one that
you haven't even completely analyzed?
In fact provided that core DNS is there, and BGP is not borked it is
possible to contingency plan for pretty much any outage that does not
result in a virtually complete loss of connectivity at the client end.
There's a big difference between what is possible and what is usually
true in practice.
There's also a big difference between the perspective of someone who
operates networks (and who can presumably fix the network if it's
broken), and application writers and users (who are forced to deal with
whatever brain-damage the network - including DNS in their view -
presents to them).
Not a week goes by when I'm not asked to figure out "why people can't
get to a web server" or "why email isn't working". In about 70% of the
web server cases and 30% of the email cases, the answer turns out to be
DNS related. IP failures, by contrast, are quite rare. And the ones
that I see mostly consist of failures of the user's local wireless
Ietf mailing list