IPCOP using IBURST Restarting problem

[Anton-AHA]

Hi there,
I am using IPCOP PPPoE ethernet connection.
It works fine but frequently drops out and continuously attempts to restart
the connection, with no joy.
If I restart the IPCOP it immediately reconnects...
Could this be an MTU problem? or and IBURST signal problem?
If anyone can offer any advice, it'll be much appreciated ;-)
cheers,
Anton

You can never discover new oceans or lands until you are prepared to lose sight of your own shores...

here is my /var/log/messages log:

Feb 7 20:00:40 ipcop snort: pcap_loop: recvfrom: Network is down
Feb 7 20:00:40 ipcop snort: Final Flow Statistics
Feb 7 20:00:40 ipcop snort: Snort exiting
Feb 7 20:00:40 ipcop pppd[2027]: Couldn't increase MTU to 1500
Feb 7 20:00:40 ipcop pppd[2027]: Couldn't increase MRU to 1500
Feb 7 20:00:41 ipcop ipcop: PPP has gone down on ppp0
Feb 7 20:00:43 ipcop dnsmasq[19626]: started, version 2.17 cachesize 150
Feb 7 20:00:43 ipcop dnsmasq[19626]: read /etc/hosts - 3 addresses
Feb 7 20:00:43 ipcop dnsmasq[19626]: reading /var/state/dhcp/dhcpd.leases
Feb 7 20:00:43 ipcop dnsmasq[19626]: reading /var/ipcop/red/resolv.conf
Feb 7 20:00:43 ipcop dnsmasq[19626]: using nameserver 196.7.142.132#53
Feb 7 20:00:43 ipcop dnsmasq[19626]: using nameserver 196.7.0.138#53
Feb 7 20:00:44 ipcop ipcop: already rc.connectioncheck reconnect running
Feb 7 20:00:46 ipcop pppd[2027]: Connection terminated.
Feb 7 20:00:46 ipcop pppd[2027]: Connect time 147.3 minutes.
Feb 7 20:00:46 ipcop pppd[2027]: Sent 5478071 bytes, received 94571421 bytes.
Feb 7 20:00:46 ipcop pppd[2027]: Exit.
Feb 7 20:00:47 ipcop red: Connectioncheck: will connect again
Feb 7 20:00:49 ipcop ipcop: Starting RED device eth1.
Feb 7 20:00:49 ipcop ipcop: Dialling IBURST.
Feb 7 20:00:49 ipcop pppd[19673]: Plugin rp-pppoe.so loaded.
Feb 7 20:00:49 ipcop pppd[19673]: RP-PPPoE plugin version 3.3 compiled against pppd 2.4.2
Feb 7 20:00:49 ipcop pppd[19674]: pppd 2.4.2 started by root, uid 0
Feb 7 20:00:49 ipcop ipcop: Dynamic DNS ip-update for aha.gotdns.com: failure
Feb 7 20:01:00 ipcop fcron[19689]: Job [ -f "/etc/FLASH" ] && /etc/rc.d/rc.flash.down started for user root (pid 19691)
Feb 7 20:01:00 ipcop fcron[19690]: Job /usr/sbin/logrotate /etc/logrotate.conf started for user root (pid 19692)
Feb 7 20:01:00 ipcop fcron[19689]: Job [ -f "/etc/FLASH" ] && /etc/rc.d/rc.flash.down terminated (exit status: 1)
Feb 7 20:01:24 ipcop pppd[19674]: Timeout waiting for PADO packets
Feb 7 20:01:24 ipcop pppd[19674]: Unable to complete PPPoE Discovery
Feb 7 20:01:24 ipcop pppd[19674]: Exit.
Feb 7 20:01:26 ipcop red: Connectioncheck: RED fail to connect
Feb 7 20:01:38 ipcop red: Connectioncheck: Restarting 1/10
Feb 7 20:01:40 ipcop ipcop: Starting RED device eth1.
Feb 7 20:01:40 ipcop ipcop: Dialling IBURST.
Feb 7 20:01:40 ipcop pppd[19744]: Plugin rp-pppoe.so loaded.
Feb 7 20:01:40 ipcop pppd[19744]: RP-PPPoE plugin version 3.3 compiled against pppd 2.4.2
Feb 7 20:01:40 ipcop pppd[19745]: pppd 2.4.2 started by root, uid 0
Feb 7 20:02:15 ipcop pppd[19745]: Timeout waiting for PADO packets
Feb 7 20:02:15 ipcop pppd[19745]: Unable to complete PPPoE Discovery
Feb 7 20:02:15 ipcop pppd[19745]: Exit.
Feb 7 20:02:17 ipcop red: Connectioncheck: RED fail to connect
Feb 7 20:02:29 ipcop red: Connectioncheck: Restarting 2/10
Feb 7 20:02:31 ipcop ipcop: Starting RED device eth1.
Feb 7 20:02:31 ipcop ipcop: Dialling IBURST.
Feb 7 20:02:31 ipcop pppd[19810]: Plugin rp-pppoe.so loaded.
Feb 7 20:02:31 ipcop pppd[19810]: RP-PPPoE plugin version 3.3 compiled against pppd 2.4.2
Feb 7 20:02:31 ipcop pppd[19812]: pppd 2.4.2 started by root, uid 0
Feb 7 20:03:06 ipcop pppd[19812]: Timeout waiting for PADO packets
Feb 7 20:03:06 ipcop pppd[19812]: Unable to complete PPPoE Discovery
Feb 7 20:03:06 ipcop pppd[19812]: Exit.
Feb 7 20:03:08 ipcop red: Connectioncheck: RED fail to connect
Feb 7 20:03:20 ipcop red: Connectioncheck: Restarting 3/10
Feb 7 20:03:22 ipcop ipcop: Starting RED device eth1.
Feb 7 20:03:22 ipcop ipcop: Dialling IBURST.
Feb 7 20:03:22 ipcop pppd[19876]: Plugin rp-pppoe.so loaded.
Feb 7 20:03:22 ipcop pppd[19876]: RP-PPPoE plugin version 3.3 compiled against pppd 2.4.2
Feb 7 20:03:22 ipcop pppd[19877]: pppd 2.4.2 started by root, uid 0

Feb 7 20:03:57 ipcop pppd[19877]: Timeout waiting for PADO packets
Feb 7 20:03:57 ipcop pppd[19877]: Unable to complete PPPoE Discovery
Feb 7 20:03:57 ipcop pppd[19877]: Exit.
Feb 7 20:03:59 ipcop red: Connectioncheck: RED fail to connect
Feb 7 20:04:11 ipcop red: Connectioncheck: Restarting 4/10
Feb 7 20:04:13 ipcop ipcop: Starting RED device eth1.
Feb 7 20:04:13 ipcop ipcop: Dialling IBURST.
Feb 7 20:04:13 ipcop pppd[19942]: Plugin rp-pppoe.so loaded.
Feb 7 20:04:13 ipcop pppd[19942]: RP-PPPoE plugin version 3.3 compiled against pppd 2.4.2
Feb 7 20:04:13 ipcop pppd[19944]: pppd 2.4.2 started by root, uid 0
Feb 7 20:04:48 ipcop pppd[19944]: Timeout waiting for PADO packets
Feb 7 20:04:48 ipcop pppd[19944]: Unable to complete PPPoE Discovery
Feb 7 20:04:48 ipcop pppd[19944]: Exit.
Feb 7 20:04:50 ipcop red: Connectioncheck: RED fail to connect

 

[DFantom]

Crash had the same problems and it turned out to be a bug in the PPPoE stuff. He was using Mandrake 10 and updated to the latest PPPoE patch for Mandrake 10 and that fixed it right up

 

[Anton-AHA]

Thanks DFantom,
I have the latest Patch for IPCOP and have been "googling" for anyone else having had similar problems... very strange indeed...! I still have a feeling it's got todo with MTU's...
At the moment it's set to 1500. I have been told by IBURST that it should be set to 1352.

any more feedback welcome ;-)

 

[KikzAzz]

I have been using Windows XP and funny enough it was plug and play. No issues!

 

[Anton-AHA]

yeah, IBURST seems to be fine on my windows XP machine too, but I have several networks behind, and want a secure linux firewall upfront, ie. IPCOP
There must be a way to stop the drop-outs and constant restarts...
Just trying to establish where the error originates...
strange that a IPCOP reboot always establishes immediate connection thereafter...
almost sounds like a buffering issue... flushed on reboot!
Cos it usually takes from 5 minutes to one hour, and then suddenly drops and cannot re-establish a connection..?

 

[Gatecrasher]

I still have a feeling it's got todo with MTU's...
At the moment it's set to 1500. I have been told by IBURST that it should be set to 1352.

any more feedback welcome ;-)

I can confirm MTU of 1352. But I reckon it should only improve speed/efficiency, not cure disconnects. Much more likely there is a signal problem.

 

[bb_matt]

I have been using Windows XP and funny enough it was plug and play. No issues!

And what's between you and the rest of the world, zonealarm ?

 

[Anton-AHA]

IPCOP is a firewall ;-) bb_matt
Thanks Gatecraher.. I tend to agree... but then why does it always reconnect of system reboot?

 

[DFantom]

IPCOP <> PPPoE
You need to make sure you update your PPPoE side of your linux box as well as your firewall.

 

[Gatecrasher]

IPCOP is a firewall ;-) bb_matt
Thanks Gatecraher.. I tend to agree... but then why does it always reconnect of system reboot?

I don't know how good your signal is, but before reloacting my modem, my signal was variable, and sometimes the only way to reconnect after a disconnect was to switch the modem power off and on. I don't recall having to reboot my pc though.

After the next disconnect you could try leaving your system as-is and just reboot the modem. If that doesn't fix it then more likely you have a software conflict, rather than any problem with your signal/modem.

 

[Anton-AHA]

Thanks Gatecrasher!~ I am going to try and relocate my IBURST modem to a better high gain vantage point, hopefully that'll help resolve signal strength issues if that's the problem. Do you know how to hardcode the eth interface to a specific MTU value? ie. on bootup.
Anton

 

[Anton-AHA]

It's a linux machine that I am refering to in the previous post ;-)

 

[ic]

I think this thread needs to be moved to the iBurst forum since it is more about disconnects & signal & mtu & iBurst.

The thing is iBurst is still unstable, and it doesn't surprise me that your Linux box is getting confused. Your starting point is to get your MTU set correctly. I have just checked, and despite what WBS Helplessnessdesk are telling everyone, MTU via the base-station & routers I'm connecting to is still 1392, [post=162875]click here to see my test results[/post].