Date: Mon, 15 Sep 1997 09:53:51 +0930 From: Greg Lehey <grog@lemis.com> To: Gary Schrock <root@eyelab.psy.msu.edu> Cc: Doug White <dwhite@resnet.uoregon.edu>, freebsd-questions@FreeBSD.ORG Subject: Re: arplookup messages Message-ID: <19970915095351.00157@lemis.com> In-Reply-To: <3.0.3.32.19970914115337.007e7440@eyelab.msu.edu>; from Gary Schrock on Sun, Sep 14, 1997 at 11:53:37AM -0400 References: <3.0.3.32.19970913192927.007e4210@eyelab.msu.edu> <Pine.BSF.3.96.970913204208.918A-100000@localhost> <3.0.3.32.19970914115337.007e7440@eyelab.msu.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, Sep 14, 1997 at 11:53:37AM -0400, Gary Schrock wrote: > At 08:43 PM 9/13/1997 -0700, you wrote: >>> Ok, we've been getting messages like this one lately: >>> Sep 13 16:23:42 anguish /kernel: arplookup 205.138.224.173 failed: host is >>> not on local network > >> I'd be interested in seeing the output of `ifconfig de0' and `netstat >> -rn'. You should also double-check with your network admin that your >> netmask is set properly, and see if you have a single gateway (router) >> that you should be pointed at. > > Ok, here we are: > anguish:~$ netstat -rn > Routing tables > > Internet: > Destination Gateway Flags Refs Use Netif Expire > default 208.25.49.1 UGSc 95 104881 de0 > 127.0.0.1 127.0.0.1 UH 1 26841 lo0 > 208.25.49 link#1 UC 0 0 > 208.25.49.1 0:e0:34:48:c0:0 UHLW 96 0 de0 1115 > 208.25.49.5 127.0.0.1 UGHS 1 2700 lo0 > 224/4 link#1 UCS 0 0 > anguish:~$ ifconfig de0 > de0: flags=8863<UP,BROADCAST,NOTRAILERS,RUNNING,SIMPLEX,MULTICAST> mtu 1500 > inet 208.25.49.5 netmask 0xffffff00 broadcast 208.25.49.255 Well, the message is correct. Your local network has the address range 208.25.49.0 to 208.25.49.255. Since it's arp that's complaining, it would appear that you have non-local hosts on the network. Now you might come and say "but Microsoft does it". Yes, that's right. Microsoft is broken. This Must Not Work. > Once again, my appologies for any formatting problems that has (I'm sure it > ends up wrapping funny). No, it looks OK to me, and I'm known for complaining about formats. > To the best of my knowledge the gateway and netmask are set properly > (the isp hosting the machine actually set those numbers themselves, > but they do look right to me). Quite honestly though, I don't know > exactly what the physical layout of the network is (this particular > machine is located on the opposite side of the country from me). Funny. Then I don't know why arp's complaining. But that's not what my traceroute tells me: traceroute to 208.25.49.1 (208.25.49.1), 30 hops max, 40 byte packets ... 9 144.228.40.17 (144.228.40.17) 358.376 ms 357.313 ms 338.419 ms 10 144.228.147.22 (144.228.147.22) 348.382 ms 355.980 ms * traceroute to 208.25.49.5 (208.25.49.5), 30 hops max, 40 byte packets ... 9 144.228.40.17 (144.228.40.17) 348.040 ms 345.633 ms * 10 144.228.147.22 (144.228.147.22) 348.942 ms 365.984 ms 348.413 ms 11 anguish.inreach.net (208.25.49.5) 358.312 ms 443.735 ms 348.347 ms traceroute to 205.138.224.173 (205.138.224.173), 30 hops max, 40 byte packets ... 9 144.228.40.17 (144.228.40.17) 398.139 ms 345.544 ms 340.080 ms 10 144.228.147.22 (144.228.147.22) 428.206 ms 346.212 ms 358.527 ms 11 205.138.224.173 (205.138.224.173) 378.206 ms 355.866 ms 348.344 ms In other words, both 208.25.49.5 and 205.138.224.173 are behind 144.228.147.22, which is probably another interface for gatekeeper.inreach.com (208.25.49.1), which is on your local net. I'd guess that 144.228.147.22 is doing proxy arp, and it's incorrectly configured. Can you ping the host? Can you traceroute? If this is the only other machine on the net, as the netstat -nr output suggests, then just disable proxy arp. Greg
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19970915095351.00157>