What NIC driver are you using? I had the same behavior when I moved from
Dachstein to Bering. The DS driver appeared to work, but would randomly
drop the connection. Probably not recompiled for the Bering's
kernel/compiler.
Harry Kitt
lbilyeu wrote:
> swfla.rr.com == aka == timewarner/roadrunner cable
> I'm using the default setup on the Bering_1.0rc3 floppy1680 image
> ---except I went ahead and removed norfc1918 from
> /etc/shorewall/interfaces' eth0
>
> Initially the users connected behind the firewall are able to use
> services, then after a random amount of time the internet dissapears.
> The internal network can still reach weblet on the firewall, but all
> requests NAT to the internet fail.
>
> From the firewall/Bering box itself, pump -s says I still have valid
> lease, but it sure doesn't act like it.
>
> If I issue shorewall stop, svi networking stop, power cycle the modem,
> svi networking start, shorewall start, svi networking restart; the
> connection to the internet at large is restored. It doesn't appear to
> be a physical ISP failure, because I can do this immediately after the
> disconnect. I can also reboot bering & powercycle the modem and get
> immediate connection. The disconnect appears after a random amount of
> time, sometimes a few minutes, sometimes after more that 12 hours.
>
> eth0 is outside - connected to the rr cablemodem
> eth1 is inside - connected to hub
> eth2 is unused
>
> what follows are via the firewall/bering box console.
>
> uname=
> Linux firewall 2.4.18 #4 Sun Jun 9 09:46:15 CEST 2002 i486 unknown
>
> ip addr show ===
> 1: lo: mtu 16436 qdisc noqueue
> link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
> inet 127.0.0.1/8 brd 127.255.255.255 scope host lo
> 2: dummy0: mtu 1500 qdisc noop
> link/ether 00:00:00:00:00:00 brd ff:ff:ff:ff:ff:ff
> 3: eth0: mtu 1500 qdisc pfifo_fast qlen 100
> link/ether 00:00:c5:04:db:e8 brd ff:ff:ff:ff:ff:ff
> inet 65.34.117.132/23 brd 255.255.255.255 scope global eth0
> 4: eth1: mtu 1500 qdisc pfifo_fast qlen 100
> link/ether 00:40:f6:f4:e5:d4 brd ff:ff:ff:ff:ff:ff
> inet 192.168.1.254/24 brd 192.168.1.255 scope global eth1
> 5: eth2: mtu 1500 qdisc noop qlen 100
> link/ether 00:40:33:90:fc:3a brd ff:ff:ff:ff:ff:ff
>
> ip route show ==
> 192.168.1.0/24 dev eth1 proto kernel scope link src 192.168.1.254
> 65.34.116.0/23 dev eth0 proto kernel scope link src 65.34.117.132
> default via 65.34.116.1 dev eth0
>
> pump -s=
> Device eth0
> IP: 65.34.117.132
> Netmask: 255.255.254.0
> Broadcast: 255.255.255.255
> Network: 65.34.116.0
> Boot server 65.32.2.175
> Next server 0.0.0.0
> Gateway: 65.34.116.1
> Hostname: firewall
> Domain: swfla.rr.com
> Nameservers: 65.32.1.70 65.32.2.130
> Renewal time: Sat Aug 10 05:29:08 2002
> Expiration time: Sat Aug 10 06:59:08 2002
>
> netstat -nr-
> netstat: not found
>
> traceroute: not found
>
>
> ping a FQN that will normally respond
> never responds, have to kill process.
> I am not getting info back from the nameserver.
>
>
> All pings that follow are to numerical addresses x.x.x.x
> see pump -s section above for appropriate address numbers
> These ping(s) will succeed when firewall is first booted.
>
> ping one, then the other nameserver-
> never responds, have to kill process. 100% packet loss
>
> ping gateway-
> never responds, have to kill process. 100% packet loss
>
> ping bootserver/DHCPserver--
> never responds, have to kill process. 100% packet loss
>
> ping address outside ISP that normally responds-
> never responds, have to kill process. 100% packet loss
>
> iptables -nvL===
> Chain INPUT (policy DROP 2 packets, 144 bytes)
> pkts bytes target prot opt in out source
> destination
> 560 52190 ACCEPT ah -- lo * 0.0.0.0/0
> 0.0.0.0/0
> 565 160K eth0_inah -- eth0 * 0.0.0.0/0
> 0.0.0.0/0
> 373 24430 eth1_inah -- eth1 * 0.0.0.0/0
> 0.0.0.0/0
> 0 0 common ah -- * * 0.0.0.0/0
> 0.0