Re: [v6ops] How do you solve 3GPP issue if neither operator nor handset supports PD?

Philip Homburg <> Fri, 27 November 2020 13:15 UTC

Return-Path: <>
Received: from localhost (localhost []) by (Postfix) with ESMTP id 285533A0B89 for <>; Fri, 27 Nov 2020 05:15:57 -0800 (PST)
X-Virus-Scanned: amavisd-new at
X-Spam-Flag: NO
X-Spam-Score: -1.918
X-Spam-Status: No, score=-1.918 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_HELO_NONE=0.001, SPF_NONE=0.001] autolearn=ham autolearn_force=no
Received: from ([]) by localhost ( []) (amavisd-new, port 10024) with ESMTP id u5Ghzhg1VUrR for <>; Fri, 27 Nov 2020 05:15:55 -0800 (PST)
Received: from ( []) (using TLSv1.2 with cipher ECDHE-RSA-CHACHA20-POLY1305 (256/256 bits)) (No client certificate requested) by (Postfix) with ESMTPS id 6AD843A0B87 for <>; Fri, 27 Nov 2020 05:15:54 -0800 (PST)
Received: from (localhost [::ffff:]) by with esmtp (TLS version=TLSv1.2 cipher=ECDHE-RSA-CHACHA20-POLY1305) (Smail #157) id m1kidWT-0000FlC; Fri, 27 Nov 2020 14:10:49 +0100
Message-Id: <>
Subject: Re: [v6ops] How do you solve 3GPP issue if neither operator nor handset supports PD?
From: Philip Homburg <>
References: <> <> <> <>
In-reply-to: Your message of "Fri, 27 Nov 2020 10:16:40 +0000 ." <>
Date: Fri, 27 Nov 2020 14:10:48 +0100
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, 27 Nov 2020 13:15:57 -0000

>You can't run a flexible address assignment protocol without a 
>provisioning database. ND is typically implemented in o/s kernels, so 
>interfacing this with user-mode radius is architecturally troublesome.

It is my experience that RAs are sent by a user space daemon. Maybe
I'm wrong. Which kernels send RAs directly from the kernel. How does the 
RDNSS option get passed to the kernel?

>As a separate issue, adding this level of complexity also goes against 
>many of the design principals that ND was intended to fulfil.  It could 
>be argued that these principals are already being infringed on, but a PD 
>extension would take this several steps further.

Can you explain? RA provides hosts with a prefix to configure addresses from.
How is providing a device with a bigger prefix that can be be used to number
downstream networks are significant departure?

Beyond that we have a flash renumbering problem. In my opinion we should 
try to fix that in ND instead of letting hosts guess whether something has
changed or not.