Tore Anderson <[email protected]> writes:
> Hm, from a very quick glance at the getaddrinfo() calls in the ntpd
> directory of the source tree it seems none of them are passing in the
> AI_ADDRCONFIG hint. That's the one that prevents getaddrinfo() from
> returning "unusable" addresses (i.e., no IPv6 addresses if the node is
> IPv4-only and vice versa).

AI_ADDRCONFIG is useless in all but a tiny minority of cases.  Most *nix
systems today are on IPv4-only networks but have autoconfigured
link-local IPv6 addresses on all interfaces.  The converse (a system on
a pure IPv6 network having useless IPv4 addresses) is probably less
common but not unthinkable.  For instance, a laptop or desktop computer
on an IPv6-only network might correctly discover its IPv6 address
through SLAAC but still attempt DHCP for IPv4 and end up with a 169.254
address.  At the University of Oslo, we have a large cluster that uses
routable IPv6 addresses but also has RFC1918 addresses for internal
communication with software that does not understand IPv6 (including
software from a major player in the Linux arena that crashes if it sees
an IPv6 address in /etc/resolv.conf, which the vendor told us they
*might* fix in the next version...  three years ago).

What it boils down to is that, in the overwhelming majority of cases,
*the only way to find out what will work is to try it*.

/This message was brought to you by "google python lbyl eafp".  Thank
you for reading, and have a nice day./

DES
-- 
Dag-Erling Smørgrav - [email protected]
_______________________________________________
pool mailing list
[email protected]
http://lists.ntp.org/listinfo/pool

Reply via email to