Re: Lots of Unsupported protocol errors and high CPU uilization.

G'day Tobias,

To me, after reading your description and logs, what appears to have
happened is:

1.  the tunnel is up and running fine from 04:16 to 07:20, packets are
being processed by the kernel without any significant involvement of the
pppd process,

2.  at 07:20, incorrectly formatted packets begin to be received from
the peer, these are rejected by the kernel ppp module and passed up to
the pppd process for handling, where they are logged,

3.  at 08:09, the peer pppd terminates the connection.

Your observations regarding bandwidth and CPU usage appear consistent
with #2 above.

I see no evidence that the user triggered the disconnection; an LCP
termination request can be sent by the peer for other reasons.

What can cause this?  There's not enough information here to isolate to
a specific cause.  You would have to include within the scope the
hardware and software of both the server, the client, and the routers
between them.

I've have certainly seen older problems from many years ago where a peer
handles an MPPE rekeying differently, or spontaneously enables MPPC
without negotiating it.

-- 
James Cameron
http://quozl.linux.org.au/

------------------------------------------------------------------------------
Nokia and AT&T present the 2010 Calling All Innovators-North America contest
Create new apps & games for the Nokia N8 for consumers in  U.S. and Canada
$10 million total in prizes - $4M cash, 500 devices, nearly $6M in marketing
Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store 
http://p.sf.net/sfu/nokia-dev2dev