From owner-freebsd-questions@FreeBSD.ORG Wed Jul 9 07:14:42 2003 Return-Path: Delivered-To: freebsd-questions@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 6665037B401 for ; Wed, 9 Jul 2003 07:14:42 -0700 (PDT) Received: from mail1.bluewin.ch (mail1.bluewin.ch [195.186.1.74]) by mx1.FreeBSD.org (Postfix) with ESMTP id 7ABFB43FA3 for ; Wed, 9 Jul 2003 07:14:41 -0700 (PDT) (envelope-from raphael@computer-rental.ch) Received: from computer-rental.ch (62.203.89.117) by mail1.bluewin.ch (Bluewin AG 7.0.016) id 3EFC3F760014B4EB for freebsd-questions@freebsd.org; Wed, 9 Jul 2003 14:14:39 +0000 Date: Wed, 9 Jul 2003 16:13:27 +0200 Mime-Version: 1.0 (Apple Message framework v552) Content-Type: text/plain; charset=US-ASCII; format=flowed From: =?ISO-8859-1?Q?Rapha=EBl_Marmier?= To: freebsd-questions@freebsd.org Content-Transfer-Encoding: 7bit Message-Id: <82D6C134-B217-11D7-A282-000393D67E4A@computer-rental.ch> X-Mailer: Apple Mail (2.552) Subject: pptp connection unreliable with MacOSX client, help! X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 09 Jul 2003 14:14:42 -0000 I've setup a vpn server with poptop 1.1.3 from the ports (FreeBSD 4.8 release). It work fine, with mppe 128 enabled, from Windows 2000 and MacOSX 10.2 (using the built-in vpn client). I use it to mount a volume through nfs and samba, so I can use cvs in local mode (needed). However, on MacOSX, it is unreliable and the mountpoint freezes, until I manage to disconnect and reconnect the vpn. The following appears in /var/log/messages: Jul 9 02:10:56 napalm pptpd[412]: CTRL: EOF or bad error reading ctrl packet length. Jul 9 02:10:56 napalm pptpd[412]: CTRL: couldn't read packet header (exit) Jul 9 02:10:56 napalm pptpd[412]: CTRL: CTRL read failed Jul 9 02:11:58 napalm pptpd[421]: CTRL: EOF or bad error reading ctrl packet length. Jul 9 02:11:58 napalm pptpd[421]: CTRL: couldn't read packet header (exit) Jul 9 02:11:58 napalm pptpd[421]: CTRL: CTRL read failed Jul 9 02:17:13 napalm pptpd[434]: GRE: read(fd=5,buffer=804d580,len=8196) from PTY failed: status = 0 error = No error Jul 9 02:17:13 napalm pptpd[434]: CTRL: PTY read or GRE write failed (pty,gre)=(5,4) Jul 9 02:18:35 napalm pptpd[441]: CTRL: EOF or bad error reading ctrl packet length. Jul 9 02:18:35 napalm pptpd[441]: CTRL: couldn't read packet header (exit) Jul 9 02:18:35 napalm pptpd[441]: CTRL: CTRL read failed I've gone as far as setting nfs to tcp only, which helps go through disconnection/reconnection cycles. But it seems that once the vpn connection is wedged, it doesn't recover by itself. I spent hours pinpointing this to pptp. Curiously, this problem manifests itself only on slow llinks, such as dialups. I can reproduce it by reducing the bandwidth with altq. help... Raph