[pptp-server] Administration of dead PPTP tunnels

Gord Belsey gord at amador.ca
Thu Oct 12 09:15:15 CDT 2000


Pascal:

I fixed this problem by setting the following in the /etc/ppp/options file:

lcp-echo-failure 10
lcp-echo-timeout 1

This is sort of the ppp/pptp version of a ping.  In this case, if the client
doesn't respond in 1 second (lcp-echo-time-out) I consider it a failed echo,
and 10 failed echos (lcp-echo-failure) in a row means the link has been
dropped.  PoPToP will then let the connection "die".

You can play around with both these settings to your liking.

Hope this helps

Gord Belsey


----- Original Message -----
From: Pascal Fremaux <pascal.fremaux at sxb.bsf.alcatel.fr>
To: <pptp-server at lists.schulte.org>
Sent: Thursday, October 12, 2000 1:39AM
Subject: [pptp-server] Administration of dead PPTP tunnels


> Sometimes in our tests, when clients are bad configured or somethings
> like that, a tunnel goes down for the client side, but stay up on the
> server.
> So there are more and more 'zombi' tunnels which 'eat' all the range of
> IP addresses.
> How could you detect automatically those zombies, and clean them ?
> How do you manage that ?
>
> --
>  Pascal Fremaux, SSII Alten
>  Study Engineer at Alcatel Telecom
>  R&D, Illkirch, France
>
>
>
> _______________________________________________
> pptp-server maillist  -  pptp-server at lists.schulte.org
> http://lists.schulte.org/mailman/listinfo/pptp-server
> List services provided by www.schulteconsulting.com!
>




More information about the pptp-server mailing list