Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 18 Jun 2019 08:22:53 +0000
From:      Lorenzo Salvadore <phascolarctos@protonmail.ch>
To:        FreeBSD Questions <freebsd-questions@freebsd.org>
Subject:   Re: Eliminating IPv6 (?)
Message-ID:  <osfidavETPv_k8mIOOmWi54rn6Rd5D06sX4Bk7vbwFIb27vsmyjzH9V1iPKwkF0BDDPIdXYinOB4MG8KiajCKxRv7_tKjFfgSUq1MCYaBgg=@protonmail.ch>
In-Reply-To: <18748.1560843874@segfault.tristatelogic.com>
References:  <18748.1560843874@segfault.tristatelogic.com>

next in thread | previous in thread | raw e-mail | index | archive | help
=E2=80=90=E2=80=90=E2=80=90=E2=80=90=E2=80=90=E2=80=90=E2=80=90 Original Me=
ssage =E2=80=90=E2=80=90=E2=80=90=E2=80=90=E2=80=90=E2=80=90=E2=80=90
On Tuesday 18 June 2019 09:44, Ronald F. Guilmette <rfg@tristatelogic.com> =
wrote:

> In message d6a5d6b8-1630-3095-dd0b-22b49213176e@grosbein.net,
> Eugene Grosbein eugen@grosbein.net wrote:
>
> > 18.06.2019 10:10, Ronald F. Guilmette wrote:
> >
> > > How can I turn off IPv6 entirely without rebuilding the kernel?
> >
> > You cannot. GENERIC kernel specifically enables IPv6 support and you ne=
ed to
> > disable it at compile time.
> > And if you do, you better rebuild the world too using WITHOUT_INET6=3Dy=
es in the
> > /etc/src.conf
> > or else some utilities compiled with INET6 by default will query kernel
> > for IPv6-specific data (like routing entries) and complain that your ke=
rnel does
> > not know about it.
> > World built WITHOUT_INET6 has no such rough edges.
>
> OK, so I obviously expressed myself badly. Let me try again.
>
> IPv6 support is enabled in a the stock kernel. OK. Fine. But just because
> that feature is present in the kernel, that does not imply that anything =
in
> userland -has- to actually make any use of it at all.
>
> Something is doing ifconfig on my loopback (lo0) interface. What is that
> thing and how can I get it to stop doing that?
>
> As I have already learned, the /etc/rc.firewall script also assumes both =
the
> presence of, and the desirability of IPv6 support. And unless one edits t=
hat
> file manually... which I have been effectively forced to do... there is n=
o way
> to get it to simply NOT create and install multiple IPv6-related ipfw rul=
es,
> EVEN THOUGH in my particular situation... which is still the most common =
case...
> those extra and entirely superfluous IPv6 ipfw filtering rules are servin=
g
> no earthly purpose whatsoever and are only cluttering up my ipfw rule set=
,
> thus pointlessly making it harder for me to grok and maintain them all.
>
> Clearly, if doesn't have to be this way. Some maintainers just decided th=
at
> I and all other IPv4-only users should get stuck dealing with a lot of us=
eless,
> unnecessary and distracting IPv6 stuff, whether I like it or not, and pre=
sumably
> for our own good.
>
> I really wish that maintainers would allow me a bit more freedom, and sho=
w
> me the courtesy and respect to allow me to decide for myself what is and =
what
> isn't "for my own good".
>
> I can and will most certainly get down and grovel around in the various
> /etc/rc.d/ scripts and will comment out those parts that do things like
> ifconfig'ing my loopback interface for IPv6, whether I like it or not.
> But there ought to be some single /etc/rc.conf variable via which one cou=
ld
> simply select the "No, I don't want to have to deal with IPv6 at all righ=
t
> now" option.
>
> Is that really an unreasonable hope, expectation, and request?
>
> I understand that the kernel will still -offer- the IPv6 support. But if =
no
> -other- software on my system actually takes the kernel up on that offer,
> then the kernel's IPv6 support becomes like the tree that falls in the
> forrest when there is nobody around to hear it. It might as well be said
> that it makes no sound, and no difference to anything at all.
>
> It is clearly not necessary for me or anyone else to have to rebuild the
> kernel... and world... just in order to get rid of what are, for the
> majority of users here in 2019, still a bunch of utterly superfluous IPv6
> "features" that (a) do not help us one iota and that (b) are all just a
> big and pointless distraction that muddles everything and unnecessarily
> complicates and complexifies ordinary system maintenance tasks.
>
> IPv6 is great and I'm sure I'll be using it someday. But today is not tha=
t
> day... not for me, and also not for one hell of a lot of other users. The
> fact that I and others are effectively being forced to even think about i=
t,
> due to an absence of reasonable and easily accessible userland options, i=
s
> actually a big turn-off, and leaves a bad taste in the mouth which will
> be remembered, in future, at every mention of IPv6. I hope that all of th=
e
> IPv6 evanglists will take a moment to stop and think about that, and that
> they'll stop effectively forcing those of us who don't need it to both us=
e
> IPv6 and to think about it, whether we like it or not, and before we are =
ready,
> willing, and able to do so.
>
> Regards,
> rfg
>
> P.S. In case I have again failed to be clear, I am proposing a new /etc/r=
c.conf
> option. Something simple and intutive like:
>
> ipv6=3D"NO"
>
> That in turn should be checked -and- respected by all relevant /etc/rc,d/
> scripts.
>
> I ask again, is this really such an unreasonable thing to hope for?

You can just block ipv6 once and for all with your firewall.
I wanted to disable ipv6 on a machine and the only thing I did was to add
"block quick inet6" on top of my pf rules. I guess ipfw has a similar rule.

This does not solve your issue with ifconfig, but as you understood it will=
 not
remove ipv6 support from your kernel either. It will just drop any inet6 pa=
cket
as soon as it arrives on your system.
If you want to disable ipv6, then firewall it. If you want to remove any ip=
v6
support then, as already stated, you must rebuild from sources (both kernel
and world).

Lorenzo Salvadore.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?osfidavETPv_k8mIOOmWi54rn6Rd5D06sX4Bk7vbwFIb27vsmyjzH9V1iPKwkF0BDDPIdXYinOB4MG8KiajCKxRv7_tKjFfgSUq1MCYaBgg=>