From owner-freebsd-questions Wed Dec 10 02:48:05 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id CAA04845 for questions-outgoing; Wed, 10 Dec 1997 02:48:05 -0800 (PST) (envelope-from owner-freebsd-questions) Received: from mother.sneaker.net.au (akm@mother.sneaker.net.au [203.30.3.2]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id CAA04777 for ; Wed, 10 Dec 1997 02:46:47 -0800 (PST) (envelope-from akm@mother.sneaker.net.au) Received: (from akm@localhost) by mother.sneaker.net.au (8.8.5/8.8.5) id VAA02207; Wed, 10 Dec 1997 21:55:21 +1100 (EST) From: Andrew Kenneth Milton Message-Id: <199712101055.VAA02207@mother.sneaker.net.au> Subject: Re: iijppp problems with proxy arp To: markus@nrgnet.com (Mark Plummer) Date: Wed, 10 Dec 1997 21:55:20 +1100 (EST) Cc: freebsd-questions@FreeBSD.ORG In-Reply-To: <199712100415.XAA00276@zaphod.nrgnet.com> from "Mark Plummer" at Dec 9, 97 11:15:43 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-questions@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk +-----[ Mark Plummer ]------------------------------ | | hello, | | i'm using iijppp on a 2.1.0-RELEASE freebsd machine. occasionally, | the ppp dialup becomes unusable; it connects and authenticates fine, but | when packets are to be passed, nothing works. i've noticed that the | routing table believes the slave side ip address is to be reached through | the ethernet card (there's a specific entry for it), and the arp table | contains an entry for it as incomplete. neither entry goes away. this | seems to happen when another machine pings the ip address when the slave | side isn't connected (i suspect any traffic would do it). the pinging | machine (or atleast the one i tested) got an icmp redirect when it pinged. | what i'd like to know is: (a) would using pppd solve this problem, (b) | would upgrading to a later rev of free bsd solve this problem, or (c) is | there something i am doing wrong? my ppp.conf follows. please direct any | questions or replies direct to me as well as the list. thanks in advance. I had this problem using pppd as well. Routes would hang around, and arp-entries would hang around. I made ip-up remove the route and the arp entry and re-add it. That works for us because we had the auto-pap getty, which runs as root. It's not exactly an elegant solution, but, it worked so I left it. I haven't seen the same problems since upgrading to 2.2.2, although now, I have a different set of problems to contend with. ------------ #!/bin/sh # # A program or script which is executed when the link # is available for sending and receiving IP packets # (that is, IPCP has come up). It is executed with # the parameters interface-name tty-device speed # local-IP-address remote-IP-address. Interface=$1 Tty=$2 Speed=$3 Local=$4 Remote=$5 (/usr/sbin/arp -d $Remote; /sbin/route delete $Remote; /sbin/ifconfig $Interface $Local $Remote; /usr/sbin/arp -s $Remote 0:0:1:18:25:86 pub proxy) 2>> /tmp/ifconfig.$Interface; -- ,-_|\ SneakerNet | Andrew Milton | GSM: +61(41)6 022 411 / \ P.O. Box 154 | akm@sneaker.net.au | Fax: +61(2) 9746 8233 \_,-._/ N Strathfield +--+----------------------+---+ Ph: +61(2) 9746 8233 v NSW 2137 | Low cost Internet Solutions |