From owner-freebsd-questions Wed Mar 31 17:30: 2 1999 Delivered-To: freebsd-questions@freebsd.org Received: from osa.qcislands.net (osa.qcislands.net [209.205.50.22]) by hub.freebsd.org (Postfix) with SMTP id 2A15D14C1B for ; Wed, 31 Mar 1999 17:29:56 -0800 (PST) (envelope-from paz@ccstores.com) Received: (4128 bytes) by osa.qcislands.net via sendmail with P:stdio/R:inet_hosts/T:smtp (sender: ) id for freebsd-questions@freebsd.org; Thu, 1 Apr 1999 01:29:35 +0000 (GMT) (Smail-3.2.0.101 1997-Dec-17 #3 built 1998-Feb-7) From: Jim Pazarena To: freebsd-questions@freebsd.org Subject: problem with dual NIC's in computer X-Mailer: SCO Shell Date: Wed, 31 Mar 1999 17:18:33 -0800 (PST) Message-ID: <9903311718.aa14955@dick.ccstores.com> Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG I have two machines connected to a third ns 209.53.238.2 -- ciu |-- 209.53.238.1 209.53.238.3 -- ns2 the third has two nic's, connecting one network to another ciu 209.53.238.1 209.205.50.8 uw7 This machine in turn is connected to the outside world via my 209.205.50/26 IPs. In my examples below I refer to machine "r-qci" which is 209.205.50.1 and is hanging on the other side of the dual NICs machine. 209.53.238/24 <------> 209.53.238.1 209.205.50.8 <------> 209.205.50/26 "outside world" ciu/uw7 PROBLEM: While logged into ciu/uw7 (the dual NIC's) I get unsual ping and traceroutes to (ns) but perfectly normal one for the other machine in the network (ns2). Here is some output: ----- # traceroute ns2 traceroute to ns2.qcislands.net (209.53.238.3), 30 hops max, 80 byte packets 1 ns2.qcislands.net (209.53.238.3) 10 ms 0 ms 0 ms # traceroute ns traceroute to ns.qcislands.net (209.53.238.2), 30 hops max, 80 byte packets 1 ciu (209.53.238.1) 0 ms 0 ms 0 ms 2 ciu (209.53.238.1) 0 ms 0 ms 0 ms 3 ciu (209.53.238.1) 0 ms 0 ms 0 ms 4 ciu (209.53.238.1) 0 ms 10 ms 0 ms 5 ciu (209.53.238.1) 0 ms 0 ms ns.qcislands.net (209.53.238.2) 0 ms ----- A route get shows nothing unusual: # route get ns route to: ns.qcislands.net destination: ns.qcislands.net interface: net1 flags: rpipe spipe ssthresh rtt,msec rttvar hopcount mtu expire tos 0 0 0 0 0 0 1500 978 0 # route get ns2 route to: ns2.qcislands.net destination: ns2.qcislands.net interface: net1 flags: rpipe spipe ssthresh rtt,msec rttvar hopcount mtu expire tos 0 0 0 0 0 0 1500 946 0 ----- A ping from the r-qci to ns2 looks normal. A ping from r-qci to ns: 64 bytes from ns.qcislands.net (209.53.238.2): icmp_seq=4 ttl=254 time=0 ms 64 bytes from ns.qcislands.net (209.53.238.2): icmp_seq=4 ttl=254 time=0 ms (DUP!) 64 bytes from ns.qcislands.net (209.53.238.2): icmp_seq=4 ttl=254 time=0 ms (DUP!) 64 bytes from ns.qcislands.net (209.53.238.2): icmp_seq=4 ttl=254 time=0 ms (DUP!) 64 bytes from ns.qcislands.net (209.53.238.2): icmp_seq=4 ttl=254 time=0 ms (DUP!) ----- Can ANYONE suggest what might be wrong with "ns" when "ns2" is fine? I suspect the dual NICs on CIU/UW7 are screwing up.. but I can't determine exactly what is going wrong. ns & ns2 are both FreeBSD3.1 machines with SMC 8432BT NICs. uw7 is a UnixWare 7 machine. BTW, I'm in transition from one upstream provider to another and I'm trying to make the move seamless for my customers. Hence 209.53.238/24 is "mine" but in-accessible from the outside world as yet. -- Jim Pazarena mailto:paz@ccstores.com http://www.qcislands.net/paz To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message