Index: [Article Count Order] [Thread]

Date: Wed, 9 Jan 2008 16:48:10 +0100
From: Karsten Keil <kkeil@suse.de>
Subject: [users:00461] Re: Selftest 1.5.0b2
To: users@tahi.org
Message-Id: <20080109154810.GA17994@pingi.kke.suse.de>
In-Reply-To: <20080109085649.ee7390e8.akisada@tahi.org>
References: <20080108181112.GA16126@pingi.kke.suse.de> <20080109085649.ee7390e8.akisada@tahi.org>
X-Mail-Count: 00461

On Wed, Jan 09, 2008 at 08:56:49AM +0900, Yukiyo Akisada wrote:
> Hi, Karsten.
> 
> Please check RFC 4862 (IPv6 Stateless Address Autoconfiguration).
> 
> Section 5.4.5 says,
> 
>     909    If the address is a link-local address formed from an interface
>     910    identifier based on the hardware address, which is supposed to be
>     911    uniquely assigned (e.g., EUI-64 for an Ethernet interface), IP
>     912    operation on the interface SHOULD be disabled.  By disabling IP
>     913    operation, the node will then:
>     914
>     915    -  not send any IP packets from the interface,
>     916
>     917    -  silently drop any IP packets received on the interface, and
>     918
>     919    -  not forward any IP packets to the interface (when acting as a
>     920       router or processing a packet with a Routing header).
> 

Yes, but it states SHOULD be disabled, not MUST be disabled.
I think it should only check, that the address is not assigned, to be conform to
RFC 4862.

I agree that disabling the interface makes sense and is the better approach
to handle the situation, but not disabling the interface is still a valid
option I think.

-- 
Karsten Keil
SuSE Labs
ISDN and VOIP development
SUSE LINUX Products GmbH, Maxfeldstr.5 90409 Nuernberg, GF: Markus Rex, HRB 16746 (AG Nuernberg)