All of lore.kernel.org
 help / color / mirror / Atom feed
From: Dan Williams <dcbw@redhat.com>
To: Lorenzo Colitti <lorenzo@google.com>
Cc: Sylvain Munaut <s.munaut@whatever-company.com>,
	"netdev@vger.kernel.org" <netdev@vger.kernel.org>
Subject: Re: Who/What is supposed to remove IPv6 address from interface when moving from one network to another ?
Date: Mon, 18 Mar 2013 12:03:11 -0500	[thread overview]
Message-ID: <1363626191.6437.2.camel@dcbw.foobar.com> (raw)
In-Reply-To: <CAKD1Yr1=1kfcSmRYgnrYQBhNPPc=nquB2c1jLbopbvqz848cxQ@mail.gmail.com>

On Mon, 2013-03-18 at 09:51 -0700, Lorenzo Colitti wrote:
> On Thu, Mar 14, 2013 at 12:48 PM, Dan Williams <dcbw@redhat.com> wrote:
> > The kernel does not (and shouldn't) trigger anything on carrier change as that's a
> > site-specific/user-specific policy.
> 
> Actually, it *does* trigger events on carrier change: it creates the
> addresses when you connect. It just doesn't delete them when you
> disconnect. So you can get addresses without a userspace daemon, but
> you can never delete them without a userspace daemon.
> 
> I tried to argue that that's incorrect, but, well, the archives show
> how far I got.

It does handle them when you connect, but only if you've set accept_ra
to something > 0.  And something has to set that :)  But in reality,
it's not a problem to listen for new addresses.  But *deleting*
addresses is way out of the kernel's responsibility, because a carrier
event doesn't tell the kernel anything about whether it's reconnecting
to the same network or a different one and thus it doesn't know whether
it should delete the old address or keep it around.  And that's where
the userspace stuff and policy comes in, and the kernel doesn't do
policy.

Dan

  reply	other threads:[~2013-03-18 17:02 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <CAF6-1L4=xACZdqvJ6AZL4XkY17N0pC9PnPfNfK9Tk7QB47c1tw@mail.gmail.com>
2013-03-14 19:33 ` Who/What is supposed to remove IPv6 address from interface when moving from one network to another ? Sylvain Munaut
2013-03-14 19:48   ` Dan Williams
2013-03-18 16:51     ` Lorenzo Colitti
2013-03-18 17:03       ` Dan Williams [this message]
2013-03-18 18:11         ` Sylvain Munaut
2013-03-18 18:12           ` Lorenzo Colitti
2013-03-18 18:28             ` Sylvain Munaut
2013-03-18 20:42               ` Dan Williams
2013-03-18 20:49                 ` Mikael Abrahamsson
2013-03-18 18:34       ` Hannes Frederic Sowa
2013-03-18 18:39         ` Lorenzo Colitti
2013-03-18 18:58           ` Hannes Frederic Sowa

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=1363626191.6437.2.camel@dcbw.foobar.com \
    --to=dcbw@redhat.com \
    --cc=lorenzo@google.com \
    --cc=netdev@vger.kernel.org \
    --cc=s.munaut@whatever-company.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.