Date: Wed, 14 Jun 2000 08:34:55 -0400 (EDT) From: Kenneth W Cochran <kwc@world.std.com> To: freebsd-questions@freebsd.org Subject: One-way cable-modem, Ethernet packet losses Message-ID: <200006141234.IAA06468@world.std.com>
next in thread | raw e-mail | index | archive | help
Previously, >Date: Tue, 13 Jun 2000 16:18:19 -0400 (EDT) >To: freebsd-stable@freebsd.org >Subject: One-way cable-modem, Ethernet packet losses Hi, I'm having strange network packet losses on the Ethernet "side" with FreeBSD connected to a one-way (analog voice-return) cable modem (& this is difficult to explain...). Per advice I may need to take this to -net. (?) Several times per day, traffic across the Ethernet interface just "vanishes," only to "return" in, say, 5-20 minutes. I'm not quite sure, but I don't *think* this is happening with Linux (Slackware 7, kernel 2.2.16). I can ping the default-route/gateway (analog/serial/pppd side), but can't "get past" it because of the (lack of) Ethernet "return." Also, I've never been able to ping "myself" at my cable-modem ip-address with FreeBSD. My ip-address is ping-able just fine from elsewhere, just not this machine. Under Linux, I *can* ping my cable-modem ip-address. OS is 4.0-STABLE, cvsup'ed 2000/06/10, but this was happening with 3.x-stable too. NIC is Intel Pro100+ using fxp driver (fxp0). This is the "downstream" side. It is ifconfig'ed at address 10.0.0.1 netmask 255.255.255.0, as per cablemodem company's instructions. I've noticed that it also "works" at ip address 192.168.0.1, btw. Cable modem is a Com21 CP-2000. "Upstream" side is via kernel ppp (pppd 2.3.5, which, btw, appears to be very old nowadays). Upon connecting, the machine gets its "real" ip-address (+ gateway/defaultroute) from the remote-end, static in this case, apparently based on the "user-id." Both FreeBSD and Linux are using the same pppd setup/config files (with the obvious exception of the different modem-device :). Here is output of netstat -r on FreeBSD: Internet: Destination Gateway Flags Refs Use Netif Expire default (remote ip, pppd) UGSc 9 206 ppp0 10/24 link#1 UC 0 0 fxp0 => myname 0:90:27:c7:bc:3d UHLW 1 615 lo0 10.0.0.14 0:a0:73:15:8a:e9 UHLW 0 1 fxp0 1177 (remote ip, pppd) (local ip, pppd) UH 10 0 ppp0 localhost localhost UH 0 13 lo0 "myname" can be either 10.0.0.1 or 192.168.0.1, depending on how I'm "testing." 10.0.0.14 is the MAC address of the cable-modem, but where did that ip-address come from? It is not "configured" anywhere I can find... Also, sometimes it "disappears" & then later "reappears." IIRC, that .14 address is "around" if I have fxp0 on the 192-block. Netstat -r: (from Slackware Linux 7, kernel 2.2.16) Kernel IP routing table Destination Gateway Genmask Flags MSS Window irtt Iface (remote ip, pppd) * 255.255.255.255 UH 0 0 0 ppp0 10.0.0.0 * 255.255.255.0 U 0 0 0 eth0 loopback * 255.0.0.0 U 0 0 0 lo default (remote ip, pppd) 0.0.0.0 UG 0 0 0 ppp0 Netstat -rn (also from Slackware) Kernel IP routing table Destination Gateway Genmask Flags MSS Window irtt Iface (remote ip) 0.0.0.0 255.255.255.255 UH 0 0 0 ppp0 10.0.0.0 0.0.0.0 255.255.255.0 U 0 0 0 eth0 127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 lo 0.0.0.0 (remote ip) 0.0.0.0 UG 0 0 0 ppp0 Can somebody shed some light on this? Am I looking at a bug or a Feature (or some config thing :)? Naturally, book/doc/faq pointers are most welcome. Speaking of which, where can I find Good Description(s) of these routing table displays? (tcpdump output too) Perhaps one of the W.Stevens books? Btw, fwiw, even though my cablemodem company Officially Doesn't Support anything outside Winxx (or perhaps Mac), the support guys themselves seem to have a nice, umm, "collection," of *BSD & Linux boxes... :) (& we're all stumped, so far...) Many thanks, -kc To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200006141234.IAA06468>