Re: bindv6only again

[Michael Poole <mdpoole@troilus.org>]

Tollef Fog Heen writes:

> ]] Juliusz Chroboczek 
>
> | >> What if it is just installed from the tarball?
> | 
> | > Then that person is still using buggy, non-free software.
> | 
> | Proprietary, granted, but why buggy? 
>
> Because it does not handle non-default values.  This is just like an
> application that didn't handle IFS or PATH being different from its
> default value would be buggy.  If it absolutely needs a given value, it
> should tell the system that.
>
> | bindv6only=0 is assumed by both POSIX and RFC 3493.
>
> As the default value, yes.  Not as the only possible value.

More precisely, RFC 3493 (I haven't checked POSIX) specifies[1] that the
default behavior is that when net.ipv6.bindv6only=0.  RFC 3493 does not
specify that an operating system provide anything like the
"net.ipv6.bindv6only" sysctl option; setting bindv6only to a non-zero
value makes the OS behave in a non-standard manner.  This is quite
different from the IFS or PATH example.

[1]- RFC 3493, section 5.3, IPV6_V6ONLY option for AF_INET6 Sockets: "By
default this option is turned off."

Michael Poole