From owner-freebsd-questions Sun Dec 7 04:15:47 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id EAA06251 for questions-outgoing; Sun, 7 Dec 1997 04:15:47 -0800 (PST) (envelope-from owner-freebsd-questions) Received: from www.webfront.net.au (www.webfront.net.au [203.23.200.10]) by hub.freebsd.org (8.8.7/8.8.7) with SMTP id EAA06233 for ; Sun, 7 Dec 1997 04:15:39 -0800 (PST) (envelope-from livio@mse.com.au) Received: from syrah (yoogali.webfront.net.au [203.23.200.128]) by www.webfront.net.au (8.6.12/8.6.12) with SMTP id XAA07176 for ; Sun, 7 Dec 1997 23:14:08 +1100 Message-Id: <1.5.4.32.19971207120534.00933c20@mse.com.au> X-Sender: livio@mse.com.au X-Mailer: Windows Eudora Light Version 1.5.4 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Sun, 07 Dec 1997 23:05:34 +1100 To: questions@freebsd.org From: Livio Mazzon Subject: User PPP wont release the device Sender: owner-freebsd-questions@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I have just upgraded an existing system from FreeBSD 2.1.6 to 2.2.5. It is an ISP server that uses user-ppp to accept dial-ins from attached modems. Before the upgrade the ppp dial-ins worked fine! After the upgrade (finally got the machine going)....I now have a problem where users can dial in, login and connect, and they properly get a ppp session running. All routing etc is fine I use a script on connection to run exec /usr/sbin/ppp -direct. However, when the user hangs up, the modem disconnects, DCD drops, but the PPP process stays attached to the port. The user stays logged onto the system even though the physical connection has dropped. When the next person logs in, the modem answers, but they never see the login prompt (thats because the getty programme never gets the device, ppp is still holding on to it!). This happens EVERY logon that uses PPP. I've checked all the config files, they are the same as before... when they were working. And no, I dont think its a cable problem, or modem config problem because if the user logs in via a shell, the device is released properly to getty. This is now becoming critical for us. Any help would be greatly, greatly appreciated. Thanks Livio