The state of resolv.conf

[Tom Lane]

Dan Williams <dcbw at redhat.com> writes:
> Nobody is rolling their own code; they all use gethostbyname.  The
> problem is that glibc's resolver is too simplistic.  I asked the glibc
> guys about this about 3 years ago and they said they wanted to keep
> glibc's resolver simple and that this should be handled either by lwresd
> or a caching nameserver.  The core issue was that doing something like
> polling /etc/resolv.conf for updates would be unecessary on many
> platforms.

Hm, these are the same people who insist on driving date/time
performance into the ground by stat'ing /etc/localtime on every call
to any function that has any interest in timezone-dependent info.
And surely that file is *less* likely to change than /etc/resolv.conf;
not to mention that the ensuing work is orders of magnitude cheaper
than sending an inquiry to your friendly local DNS server.
Something does not compute here.

			regards, tom lane