[pptp-server] Dropped connection using NT4

[Gareth Marlow]

I have had a couple of connections dropped after approx 15 mins of idle
time. This is an Australia-UK connection. It looks from the log as if
it's timed out due to network problems. Does anyone know how I can
increase the timeout? Relevant log extracts are as follows:

Feb  2 11:47:48 h8a pptpd[8444]: CTRL: Session timed out, ending call
Feb  2 11:47:48 h8a pppd[8445]: Modem hangup
Feb  2 11:47:48 h8a pppd[8445]: Script /etc/ppp/ip-down started (pid 8487)
Feb  2 11:47:48 h8a pppd[8445]: Connection terminated.
Feb  2 11:47:48 h8a pppd[8445]: Connect time 42.4 minutes.
Feb  2 11:47:48 h8a pppd[8445]: Sent 88767 bytes, received 262829 bytes.
Feb  2 11:47:48 h8a pppd[8445]: Waiting for 1 child processes...
Feb  2 11:47:48 h8a pppd[8445]:   script /etc/ppp/ip-down, pid 8487
Feb  2 11:47:48 h8a pptpd[8444]: CTRL: Client 210.84.66.39 control
connection finished
Feb  2 11:47:48 h8a pppd[8445]: Script /etc/ppp/ip-down finished (pid
8487), status = 0x0
Feb  2 11:47:48 h8a pppd[8445]: Exit.
Feb  2 11:47:50 h8a pptpd[8444]: CTRL: Unexpected control message 6 in
disconnect sequence
Feb  2 11:47:51 h8a pptpd[8444]: CTRL: Asked to free call when no call
open, not handled well
Feb  2 11:47:51 h8a pptpd[8444]: CTRL: Could not free Call ID [call
clear]!
Feb  2 11:47:51 h8a pptpd[8444]: CTRL: Got call clear request after call
manually shutdown - buggy client

When the person attempts to reconnect, they are generally unable to do so.
The log shows the following:

Feb  2 11:51:11 h8a pppd[8499]: sent [CHAP Success id=0x1 "Welcome to
h8a."]
Feb  2 11:51:11 h8a pptpd[8498]: CTRL: Ignored a SET LINK INFO packet with
real ACCMs!
Feb  2 11:51:11 h8a pppd[8499]: rcvd [CHAP Response id=0x1
<b9e92c667613a6b482f6f5b6320fddce0000000000000000c2fbe771bbad22245581fb4da68e7b4c19f9e80a231fd8da00>,
name = "richardk"]
Feb  2 11:51:11 h8a pppd[8499]: sent [CHAP Success id=0x1 "Welcome to
h8a."]
Feb  2 11:51:11 h8a pppd[8499]: rcvd [CHAP Response id=0x1
<d442e40c57d7ecafc4eae92509f757800000000000000000845594235b798d515eb2a83566b59269a06d287a7042775a00>,
name = "richardk"]
Feb  2 11:51:11 h8a pppd[8499]: sent [CHAP Success id=0x1 "Welcome to
h8a."]
Feb  2 11:51:12 h8a pppd[8499]: rcvd [LCP TermReq id=0x4 00 00 02 ce]
Feb  2 11:51:12 h8a pppd[8499]: LCP terminated by peer (^@^@^BM-N)
Feb  2 11:51:12 h8a pppd[8499]: sent [LCP TermAck id=0x4]
Feb  2 11:51:13 h8a pppd[8499]: rcvd [LCP TermReq id=0x5 00 00 02 ce]
Feb  2 11:51:13 h8a pppd[8499]: sent [LCP TermAck id=0x5]
Feb  2 11:51:15 h8a pppd[8499]: Connection terminated.
Feb  2 11:51:15 h8a pppd[8499]: Exit.
Feb  2 11:51:15 h8a pptpd[8498]: GRE: read(fd=4,buffer=804d7c0,len=8196)
from PTY failed: status = -1 error = Input/output error
Feb  2 11:51:15 h8a pptpd[8498]: CTRL: PTY read or GRE write failed
(pty,gre)=(4,5)
Feb  2 11:51:15 h8a pptpd[8498]: CTRL: Client 210.84.66.39 control
connection finished

If they leave it for a few minutes, they can usually reconnect without a
problem.

Please can anyone help?

Thanks,
Gareth