From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Jason@zx2c4.com Received: from krantz.zx2c4.com (localhost [127.0.0.1]) by krantz.zx2c4.com (ZX2C4 Mail Server) with ESMTP id eb6b3d0b for ; Sat, 31 Dec 2016 02:20:22 +0000 (UTC) Received: from frisell.zx2c4.com (frisell.zx2c4.com [192.95.5.64]) by krantz.zx2c4.com (ZX2C4 Mail Server) with ESMTP id 649b4f7d for ; Sat, 31 Dec 2016 02:20:22 +0000 (UTC) Received: by frisell.zx2c4.com (ZX2C4 Mail Server) with ESMTP id 1c92a784 for ; Sat, 31 Dec 2016 02:20:22 +0000 (UTC) Received: by frisell.zx2c4.com (ZX2C4 Mail Server) with ESMTPSA id ad1726fc (TLSv1.2:ECDHE-RSA-AES128-GCM-SHA256:128:NO) for ; Sat, 31 Dec 2016 02:20:22 +0000 (UTC) Received: by mail-oi0-f47.google.com with SMTP id 3so255058228oih.1 for ; Fri, 30 Dec 2016 18:28:46 -0800 (PST) MIME-Version: 1.0 In-Reply-To: References: <44DAF4D4-00A8-4903-8003-EB0215635B61@danrl.com> <7973130b-159b-a7c9-c2d8-24ca7afa8914@mmoya.org> From: "Jason A. Donenfeld" Date: Sat, 31 Dec 2016 03:28:45 +0100 Message-ID: Subject: Re: [WireGuard] Dual stack? To: =?UTF-8?Q?J=C3=B6rg_Thalheim?= Content-Type: text/plain; charset=UTF-8 Cc: WireGuard mailing list List-Id: Development discussion of WireGuard List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On Thu, Dec 29, 2016 at 10:22 AM, J=C3=B6rg Thalheim = wrote: > On the other hand switching between dual-stack/ipv4 only networks/ipv6 on= ly networks > is problematic at the moment with the tools we have for roaming clients, > because wireguard only supports one endpoint of one address family at the= time. > This might be partially fixable in future by observing the availability o= f default routes > in userspace (switch address family if it become unavailable). However th= e optimal > solution would be something like the happy eyeballs protocol (https://too= ls.ietf.org/html/rfc6555), > which is implemented in modern browser - > only because somebody got a v6/v4 default route does not mean it is also = route able. > I don't know how the latter one would fit into the stateless concept of w= ireguard. > I currently help myself by using an dedicated routing protocoll. Any suggestions on what the right behavior would be to support changes in a dual stack environment?