Re: Limiting dial-in users to one login at a time

Mark Damrose (med@npcts.edu)
Tue, 5 Sep 1995 16:59:26 -0600

>We do in fact offer unlimited usage for all our customers, and at the
>current cost per telephone line ($17.50), we cannot possibly justify a
>dedicated or pseudo dedicated line per customer since we only charge
>$19.95 per month. Thats the reason that we asked the initial
>question concerning limiting users to one login at a time. We also
>offer additional mailboxes at $5.00 per month. And we have some
>customers that have dialed-in from two machines simultaneously to
>check mail.
>
>The limiting of hours has been requested by a school, to inhibit
>students from dialing in when they get home. If they didn't give out
>the password this would not have been a problem, but the kids would
>have eventually found out anyhow.

If the school only want kids dialing in at school, why not set up a dial
back for them?

>
>Since I use RADIUS, I am planning on writing a perl scipt to peruse
>the detail file to determine if a user is connected multiple times,
>and if so, we have decided to bill them the daily amount as if they
>had multiple accounts rather than preventing the multiple logins.
>
>--Larry
>
>
>On 5 Sep 95 at 9:44, Charles Scott shaped the electrons to say:
>
>> Date: Tue, 5 Sep 1995 09:44:55 -0400 (EDT)
>> From: Charles Scott <cscott@freeway.net>
>> Subject: Re: Limiting dial-in users to one login at a time
>> To: Brett Hays <bhays@hsonline.net>
>> Cc: tech@arakis.sugar-river.net, portmaster-users@livingston.com
>> Reply-to: Charles Scott <cscott@freeway.net>
>
>>
>> Brett:
>> Yes. The easiest way to limit users is to not provide unlimited
>> accounts. If you limit a standard user to some defined amount of on-line
>> time, then why should you care if he shares it with his friends. This is
>> particularly effective if you charge for use beyond the daily or monthly
>> limit. A user who sees their time gobbled up by associates and friends is
>> going to change their password in short order. Besides, there are
>> reasons that a user might at times want to have more that one connection
>> or login session at a time.
>> The flipside of this is that if you do sell unlimited accounts, that
>> you need to prevent them from multiple logins. The best way to handle
>> that is to restrict them to a single phone line and therefore serial port,
>> dedicated to them alone. You would of course get enough for the unlimited
>> account to cover the cost of a line/modem/PortMaster port, wouldn't you?
>>
>> Chuck Scott, Pres.
>> Freeway, Inc.
>> cscott@freeway.net
>>
>>
>> On Mon, 4 Sep 1995, Brett Hays wrote:
>>
>> > I have seen a number of questions asked on this subject, but no
>> > answers and I am eager to know if there is indeed a way to limit
>> > callers to one login at a time on the PM. Anyone have any ideas?
>> >
>> > Brett Hays
>> > Hoosier On-Line Systems
>> >
>> > > Date: Sun, 3 Sep 1995 03:44:06 -0400 (EDT)
>> > > From: System Administrator <tech@arakis.sugar-river.net>
>> > > To: larryt@goldrush.com
>> > > Cc: portmaster-users@livingston.com
>> > > Subject: Re: Limiting dial-in users to one login at a time
>> > > Reply-to: System Administrator <tech@arakis.sugar-river.net>
>> >
>> > >
>> > >
>> > > On Sat, 2 Sep 1995 larryt@goldrush.com wrote:
>> > >
>> > > > I am using RADIUS on my BSDI v2.0 system. I have a PM2ER10, and I am
>>using
>> > > > dynamic (assigned) IP address for dial-in PPP lines. I would like to
>>limit
>> > > > users to only one login at a time, and optionally, I might like to
>>limit users
>> > > > to certain hours of the day. Does anyone have any ideas on how to do
>>this? I
>> > > > looked at the acct.c and radiusd.c code, and I think that I could
>>modify them
>> > > > to achieve the results, but I was hoping that someone might already
>>have it
>> > > > working.
>> > > >
>> > > I would be interested in any repliy in this area, too.....
>> > >
>> > > Also, does anyone have a COMPILED radiusd for BSDI that I can FTP? I
>> > > tried to compile the 1.16 from Livingston, but got a stack of errors,
>> > > relating to mixed pointer types........ I did not see anything at
>> > > Livingston that refered to this (possible) portability problem. Did I
>> > > miss something? This error came from 2 specific functions in the source
>> > > code, I can pinpoint the problem areas if this would be of use. But,
>>then
>> > > again, if I can simply FTP the finished product, so much the better....
>> > > I wonder why there isn't a compiled one availible directly from
>> > > Livingston?
>> > > Oh well, thanks to anyone that can shed some light on this subject;
>> > > T. Parssinen (odin@arakis.sugar-river.net)
>> > >
>> > >
>> >
>>
>>
>____________________________________________
>A+ Professionals, Inc. & Goldrush World Access
>1163 South Main Street, Angels Camp, CA 95222
>+1 209 736-0111 Voice +1 209 736-9252 FAX
>larryt@goldrush.com http://www.goldrush.com
>____________________________________________

Mark Damrose med@npcts.edu (312)244-5541
Telecommunications Technician
North Park College & Theological Seminary
3225 West Foster Ave
Chicago, IL 60625