From owner-freebsd-net@FreeBSD.ORG Mon Aug 2 06:54:39 2010 Return-Path: Delivered-To: freebsd-net@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 24044106566B; Mon, 2 Aug 2010 06:54:39 +0000 (UTC) (envelope-from fli@shapeshifter.se) Received: from mx1.h3q.net (mx1.h3q.net [IPv6:2001:16d8:ffe5:1::f1]) by mx1.freebsd.org (Postfix) with ESMTP id D80928FC1C; Mon, 2 Aug 2010 06:54:38 +0000 (UTC) Received: from smtp-auth.h3q.net (smtp-auth.h3q.net [127.0.0.1]) (Authenticated sender: hidden) by mx1.h3q.net (Postfix) with ESMTPSA id 6381833D11 ; Mon, 2 Aug 2010 08:54:37 +0200 (CEST) Message-ID: <4C566B94.5090405@shapeshifter.se> Date: Mon, 02 Aug 2010 08:54:12 +0200 From: Fredrik Lindberg User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.1.11) Gecko/20100713 Thunderbird/3.0.6 MIME-Version: 1.0 To: perryh@pluto.rain.com References: <201007311310.o6VDA3S6082887@freefall.freebsd.org> <4c54ddf3.dbUwkfoxXwnHLLQz%perryh@pluto.rain.com> In-Reply-To: <4c54ddf3.dbUwkfoxXwnHLLQz%perryh@pluto.rain.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: freebsd-usb@freebsd.org, freebsd-net@freebsd.org, pilzableiter@web.de, bug-followup@freebsd.org Subject: Re: usb/149039: [uhso] Binding problem with uhso X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 02 Aug 2010 06:54:39 -0000 On 08/01/2010 04:37 AM, perryh@pluto.rain.com wrote: >> The following reply was made to PR usb/149039; it has been noted >> by GNATS. >> >> From: Fredrik Lindberg >> To: bug-followup@FreeBSD.org, pilzableiter@web.de >> Cc: Hans Petter Selasky >> Subject: Re: usb/149039: [uhso] Binding problem with uhso >> Date: Sat, 31 Jul 2010 15:00:07 +0200 >> >> I apparently missed some interface flags (that really doesn't make >> sense for this device, it's configured with a /32 mask so broadcast >> etc can only be to itself) that the network stack wants to work >> properly. > > Is a /32 mask even legal? Unless there's a special case involved, > it ought to mean that there are no interfaces on the subnet other > than this one, thus this interface has no peer to communicate with > and might as well not exist. > > Adding net@ in hopes someone there knows what should happen. > Yes, technically a /32 mask defines only one single address, but it's the only mask that really makes sense for this device. /32 masks are "legal" and commonly used for the loopback address of routers. But this is is indeed a very special case. The device has a USB interface that accepts raw IP-packets (with no other encapsulation). Once you have told the device to connect, it will tell you what IP-address you have and what DNS-servers to use, but that's it. My best guess is that the devices does PPP internally in firmware and abstracts the point-to-point link with a IP-packet interface. But since none of these details are available the only (as far as I know) viable thing is to set a /32 mask and set 0.0.0.0 (default route) to be directly reachable through the interface (route add -interface). Fredrik Lindberg