netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Brian Haley <brian.haley@hp.com>
To: greg@enjellic.com
Cc: Neil Horman <nhorman@tuxdriver.com>,
	netdev@vger.kernel.org, davem@davemloft.net
Subject: Re: Global IPV6 auto-configuration does not work as expected.
Date: Tue, 19 Jan 2010 13:21:04 -0500	[thread overview]
Message-ID: <4B55F810.4060406@hp.com> (raw)
In-Reply-To: <201001182108.o0IL8nBB029112@wind.enjellic.com>

greg@enjellic.com wrote:
> On Jan 18,  3:17pm, Neil Horman wrote:
> } Subject: Re: Global IPV6 auto-configuration does not work as expected.
> 
> Hi Neil, thanks for the note.
> 
>>> The only way we have been able to disable the auto-configuration has
>>> been to explicitly disable it at the individual interface level with
>>> net.ipv6.conf.ethN.autoconf=0 directives.

If done after module load, then that's currently the only way to do it.

> The 'sysctl -p' command is issued by the rc.sysinit script very early
> in the boot process.  Long before the interfaces are actually
> 'upped'.  I'm assuming from your description that the:
> 
> net.ipv6.conf.default.autoconf 
> 
> If set to 0 before the network interfaces are configured should
> prevent auto-configuration from occuring.  I will re-test but I don't
> think that is happening.

I can confirm that this works correctly, for example if you set this to
zero then rmmod/modprobe one of your network drivers, you'll see it get
set correctly.

To fix your reported problem, I added a module parameter to control
the setting of autoconf at load time in June 2009
(see Documentation/networking/ipv6.txt).  Basically, just add this to
/etc/modprobe.conf:

	options ipv6 autoconf=0

Your distro might require it somewhere else.

> Also from your description I'm assuming the following:
> 
> net.ipv6.conf.all.autoconf=0
> 
> If set should globally turn off auto-configuration.

Actually, I don't think that's ever done that, the "all" settings only
affects some things, like forwarding, proxy_ndp, and disable_ipv6.

Even if the "all" setting did have control over this, if its setting was 0
and the device setting was 1, shouldn't the device setting override it?

Hope this helps,

-Brian

  parent reply	other threads:[~2010-01-19 18:21 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-01-18 21:08 Global IPV6 auto-configuration does not work as expected greg
2010-01-18 22:30 ` Neil Horman
2010-01-19 18:21 ` Brian Haley [this message]
  -- strict thread matches above, loose matches on Subject: below --
2010-01-22 19:01 greg
2010-01-19 20:34 greg
2010-01-19 20:35 ` David Miller
2010-01-18 19:06 greg
2010-01-18 20:17 ` Neil Horman

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4B55F810.4060406@hp.com \
    --to=brian.haley@hp.com \
    --cc=davem@davemloft.net \
    --cc=greg@enjellic.com \
    --cc=netdev@vger.kernel.org \
    --cc=nhorman@tuxdriver.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).