'Idle-Timeout' and the 'Acct-Session-Time' argument

[YvesDM]

On Sun, Feb 24, 2008 at 2:59 PM, Ivan Kalik <tnt at kalik.net> wrote:

> >'Idle-Timeout' timeout is an argument in Radius. Chillispot use this
> >radius argument.
>
> Yes.
>
> >I think freeradius configuration can update automatically the
> >Acct-Session-Time argument if it see a user is IDLE.
> >What do you think about this?
>
> No. Chillispot does the accounting and sends data to freeradius. Radius
> server has no idea about whether the user is connected or not. It is up
> to the NAS to send that information to it. If Chillispot disconnects the
> user and doesn't send accounting Stop packet to freeradius the session
> in radius database will remain open. As far as I can see that's how you
> set up the Chillispot - disconnect the session only on clicking the
> logout link. Set up Chillispot properly.
>
> Ivan Kalik
> Kalik Informatika ISP
>

Yes, and a correct functioning chillispot definitely sends an accounting
stop packet when idle-timeout is reached and user gets logged out.
I got idle-timeout in radgroupreply too,  and as soon as chillispot reaches
the idle-timeout for a user the user gets disconnected (by chillispot) and
the accounting session is closed in radacct. In acctTerminateCause you
should find idle-timeout as termination cause in this scenario. This is
standard behaviour for chillispot in a correct setup. So I don't understand
why this isn't working for the OP.
Sniff your network and see if an acctstop packet is sent by your chillispot
when idle timeout is reached.

Kind regards

Y.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freeradius.org/pipermail/freeradius-users/attachments/20080224/a47d16ed/attachment.html>