[ previous ] [ next ] [ threads ]
 
 From:  Rick Preston <rickjpreston at gmail dot com>
 To:  m0n0wall at lists dot m0n0 dot ch
 Subject:  mpd terminating randomly
 Date:  Wed, 9 Mar 2005 13:14:16 -0500
Hi eh,

I am running m0n0wall 1.1 on a adsl connection with a static IP
address.  We have enabled the internal PPTP server on a separate
subnet, 192.168.7.96/28, from our LAN subnet, 192.168.10.0/24.  The
PPTP server address is 192.168.7.95.  Added a couple of users and all
was\is working fine.

One of the users, using MS W2k sp4's native VPN client called and said
he received the error message:
Error 781: The encryption attempt failed because no valid certificate was found.

Shortly after everyone was complaining they could not connect.  This
statement is a little misleading.  We don't use PPTP very often.  We
use IPSec gateway tunnels to the branch offices, but when someone is
out of the office they can use PPTP.

I checked the status page under processes and mpd was running. 
Luckily I am running a syslog server, so I started going through all
the logs.  Sure enough mpd had been terminated but I could not find a
valid reason(weird that it said it was still running under processes).
 Here is the output from the log: (I added the entry just before and
just after for reference.)

13:21:32 racoon: INFO: isakmp.c:1574:isakmp_ph1delete(): ISAKMP-SA
deleted <removed>
13:56:25 mpd: mpd: caught fatal signal term
13:56:25 mpd: [pt0] IPCP: Down event
13:56:25 mpd: [pt0] IFACE: Close event
...
13:56:25 mpd: [pt15] IPCP: Down event
13:56:25 mpd: [pt15] IFACE: Close event
13:56:27 mpd: mpd: pid 98040, version 3.18 (root at nb dot neon1 dot net 13:08  6-Aug-2004)
13:56:27 mpd: mpd: already running as process 19900
13:56:27 mpd: mpd: process 19900 terminated
13:56:32 racoon: INFO: isakmp.c:1368:isakmp_open(): 127.0.0.1[500]
used as isakmp port (fd=6)

(Does it look like racoon restarted? sorry off topic)

I went into the PPTP settings and turned off PPTP, apply, then enabled
the service again and everything started working fine.  This is the
second time this has happened in a 68 day uptime.

Does any know why this might happen?  Is there more information I can
share that would be helpful in solving this?  Has anyone else seen
this problem?  I did google the mail archive and there is lots of
stuff on mpd and PPTP but I didn't see any with this issue.

Any pointers would be greatly appreciated.
Cheers,
Rick