Re: Forwarding Packets With Link Local Destination Addresses

神明達哉 <> Fri, 08 January 2021 00:56 UTC

Return-Path: <>
Received: from localhost (localhost []) by (Postfix) with ESMTP id F06823A0F47 for <>; Thu, 7 Jan 2021 16:56:59 -0800 (PST)
X-Virus-Scanned: amavisd-new at
X-Spam-Flag: NO
X-Spam-Score: -1.402
X-Spam-Status: No, score=-1.402 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, FREEMAIL_FORGED_FROMDOMAIN=0.249, FREEMAIL_FROM=0.001, HEADER_FROM_DIFFERENT_DOMAINS=0.248, RCVD_IN_MSPIKE_H2=-0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=no autolearn_force=no
Received: from ([]) by localhost ( []) (amavisd-new, port 10024) with ESMTP id vyXX36YUAh_a for <>; Thu, 7 Jan 2021 16:56:58 -0800 (PST)
Received: from ( []) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by (Postfix) with ESMTPS id D8C543A0F48 for <>; Thu, 7 Jan 2021 16:56:57 -0800 (PST)
Received: by with SMTP id a31so2898658uae.11 for <>; Thu, 07 Jan 2021 16:56:57 -0800 (PST)
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;; s=20161025; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc:content-transfer-encoding; bh=hzVirpCvNs27IF0R9/iPWRj4Jpp54oWly7KLMuy2I24=; b=D1oyGOXPilr6P9hWYBSB4PMWWvWSOHANx5zpKHahf4lfSNsbTPvMFMp1NSQ80q5LtU koAsbfzEjy1QP2E0g6gLfcf3Npv5krOB86B8PYeUwlW7S7Og5pfDDN0x4IB2F8jGvfZN 98aAATwYe0jkHkzZKPxkiQ+FzB1zq5zKaCcTprGjp2Lk5ZA334C4vA8muq7tcEURiW2S fDPLGF9GhuF8kNoOCa2B/P3ySsLnULMdLwGVzgpjT/sT+WETSEYfONFnbtnx+K3Fw08Y YruHRQ8ShOQD3Hoh3YK27aw98LRxw/qBy4hyBqU4RKgYL0Qxu78wJKz0uJAajFT8VNy0 4aCA==
X-Gm-Message-State: AOAM532i02Jlo/s0sl7Uv1UPstuXCgAqt+lYklJKRxYVi1cBLMJYYqd7 Ki1lJFftPg/AP6C7Sa0oEjHGL5f8qPqngGjp4ykSamLlbBlKVQ==
X-Google-Smtp-Source: ABdhPJzc+NBpXj2238vCUSZZmoSggKPPhwKzQyFVyza9jc14SGfAkv6jhdsLD6IwgAnY6NR73MN2jgdvQmd1NsOzUmw=
X-Received: by 2002:ab0:6206:: with SMTP id m6mr1107692uao.123.1610067416860; Thu, 07 Jan 2021 16:56:56 -0800 (PST)
MIME-Version: 1.0
References: <> <>
In-Reply-To: <>
From: =?UTF-8?B?56We5piO6YGU5ZOJ?= <>
Date: Thu, 7 Jan 2021 16:56:45 -0800
Message-ID: <>
Subject: Re: Forwarding Packets With Link Local Destination Addresses
To: Fernando Gont <>
Cc: Ron Bonica <>, "" <>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Archived-At: <>
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: "IPv6 Maintenance Working Group \(6man\)" <>
List-Unsubscribe: <>, <>
List-Archive: <>
List-Post: <>
List-Help: <>
List-Subscribe: <>, <>
X-List-Received-Date: Fri, 08 Jan 2021 00:57:00 -0000

At Thu, 7 Jan 2021 18:40:18 -0300,
Fernando Gont <> wrote:

> > According to RFC 4291, “routers must not forward any packets with
> > Link-Local source or destination addresses to other links”.
> >
> > I interpret this statement to include packets that contain routing
> > headers. For example, it forbids an SRv6 packet whose final segment has
> > a locator that begins with FE80.
> >
> > Does everyone share this interpretation? If so, do RFC 4291 or RFC 8200
> > make this sufficiently clear?
> Let me ask a different question:
> Why should this be any different for a routing header? The specs still
> apply.

The RFC 4291 spec still applies, but if it were only with RFC 4291, we
could allow the following:

- A node sends a packet with a routing header that would route the
  packet to a global address (D1) and then to a link-local (final)
  destination (D2).
- The packet leaves the source link and is delivered to the interface
  corresponding to the global address D1.
- The receiving node now sets D2 as the destination address of the
  IPv6 header and forwards the packet to the link where the packet is
  delivered to D2 (so it does not forward the packet to "other link").

RFC 4007 makes it clearer that this scenario is forbidden.

> I'd say that the case of link-locals is probably even "worse", because
> it would mean a link-local address was included in a routing header, to
> be processed by some router elsewhere in the networks. Something that
> doesn't seem to make any sense to me.

Actually, I can imagine a source routing usage where all intermediate
and final destination addresses are link-local, and the packet is
forwarded through nodes within the same link.  It may still be an
imaginary scenario, but it still makes some sense to me (I don't
intend to champion that usage though).

JINMEI, Tatuya