pppoe 3.5.1 keeps disconnecting

[lexton]

Hello all,

I'm using RH 9.0 and roaring penguin's pppoe 3.5.1. I removed the original pppoe software that came with redhat and downloaded and installed pppoe 3.5.1 directly from the roaring penguin site. I found this version to be more reliable and easier to setup then RH's version.


I'm able to connect to the internet but after about 1 - 2 hours the connection gets dropped and I have to reconnect. Here is what I find in the /var/log/messages when the connection gets dropped:

Sep 11 01:00:14 localhost pppoe[7554]: recv (receivePacket): Network is down
Sep 11 01:00:14 localhost pppoe[7554]: recv (receivePacket): Network is down
Sep 11 01:00:14 localhost network: Shutting down interface eth0: succeeded
Sep 11 01:00:14 localhost dhcpd: receive_packet failed on eth1: Network is down
Sep 11 01:00:14 localhost network: Shutting down interface eth1: succeeded
Sep 11 01:00:15 localhost network: Shutting down loopback interface: succeeded
Sep 11 01:00:15 localhost sysctl: net.ipv4.ip_forward = 0
Sep 11 01:00:15 localhost network: Disabling IPv4 packet forwarding: succeeded
Sep 11 01:00:16 localhost pppoe[7554]: send (sendPacket): Network is down
Sep 11 01:00:16 localhost pppd[7553]: Modem hangup
Sep 11 01:00:16 localhost pppd[7553]: Connection terminated.
Sep 11 01:00:16 localhost pppd[7553]: Connect time 62.0 minutes.
Sep 11 01:00:16 localhost pppd[7553]: Sent 211011 bytes, received 9121463 bytes.
Sep 11 01:00:16 localhost pppd[7553]: Exit.
Sep 11 01:00:16 localhost adsl-connect: ADSL connection lost; attempting re-connection.
Sep 11 01:00:16 localhost /etc/hotplug/net.agent: NET unregister event not supported
Sep 11 01:00:16 localhost apmd[1633]: System Standby
Sep 11 01:00:21 localhost pppd[8739]: pppd 2.4.1 started by root, uid 0
Sep 11 01:00:21 localhost pppd[8739]: Using interface ppp0
Sep 11 01:00:21 localhost pppd[8739]: Connect: ppp0 <--> /dev/pts/0
Sep 11 01:00:21 localhost pppoe[8740]: send (sendPacket): Network is down
Sep 11 01:00:36 localhost last message repeated 2 times
Sep 11 01:00:52 localhost pppd[8739]: LCP: timeout sending Config-Requests
Sep 11 01:00:52 localhost pppd[8739]: Connection terminated.
Sep 11 01:00:53 localhost /etc/hotplug/net.agent: NET unregister event not supported
Sep 11 01:00:56 localhost pppoe[8740]: Timeout waiting for PADO packets
Sep 11 01:00:56 localhost pppd[8739]: Exit.
Sep 11 01:00:57 localhost adsl-connect: ADSL connection lost; attempting re-connection.
Sep 11 01:01:02 localhost pppd[8760]: pppd 2.4.1 started by root, uid 0
Sep 11 01:01:02 localhost pppd[8760]: Using interface ppp0
Sep 11 01:01:02 localhost pppd[8760]: Connect: ppp0 <--> /dev/pts/0
Sep 11 01:01:02 localhost pppoe[8761]: send (sendPacket): Network is down
Sep 11 01:01:17 localhost last message repeated 2 times
Sep 11 01:01:33 localhost pppd[8760]: LCP: timeout sending Config-Requests
Sep 11 01:01:33 localhost pppd[8760]: Connection terminated.
Sep 11 01:01:34 localhost /etc/hotplug/net.agent: NET unregister event not supported
Sep 11 01:01:37 localhost pppoe[8761]: Timeout waiting for PADO packets
Sep 11 01:01:37 localhost pppd[8760]: Exit.


I have tried tweaking the LCP_interval and PPP_timeout settings. I made sure that PPP_timeout is 4x as much as the LCP_interval.

example:

LCP_INTERVAL = 20
PPP_TIMEOUT = 80.

but this did not help. What else can I try to keep this connection from dropping? Thanks in advance.....

[Dolda2000]

Doesn't it say anything before the start of those messages? Considering that it works for 1-2 hours, I would guess that it has with DHCP to do, ie. something along the lines of that the lease is about to expire, and then when the DHCP client tries to refresh it, it cannot, and therefore it shuts down the interface instead.

[lexton]

/var/log/messages/ does not print anything thats related to the network connection before the code that I provided in my last post. In addition I tried just connecting with pppoe and disable my dhcpd server and still the connection was dropped.

Also there are times when I do have the dhcpd server running and when the connection goes down (it's always the same errors that I posted in the last post) my eth1 card also goes down.

are there things I can try to get this working without timing out all the time?

[Dolda2000]

Are you running this on a laptop?
I spotted a line in the logs where apmd says that the system is going into standby mode; maybe you should check your power management settings.

[blue]

I just came across this post while searching for a similar error. Are you sure you are using rp-pppoe v3.5? Judging by the entry at Spe 11, 01:00:21, it appears you are using 2.4.1.

Assuming you still have the 3.5.1rpm still around type the following

#/sbin/adsl-stop
#rpm -e rp-pppoe
#rm /etc/ppp/ppp.conf
#rpm -Uvh rp-pppoe-3.5-1.i386.rpm
#cd /usr/sbin
#./adsl-setup
---- follow instructions here ----
#./adsl-start

I found that the version 2.4.1 (installed with RH9) left adsl-* scripts in the /sbin directory where the 3.5-1 scripts are located in /usr/sbin/. That screwed me over 'cause when I typed adsl-setup, it re-setup 2.4.1 and then adsl-start started 2.4.1.... a full upgrade for nuthin. Anyway... Following the steps above will clear out that junk and you are now on your way again.

Good luck.
J.