The only thing I can think of is to write a wrapper that makes/checks a
timestamp and allows/disallows access to pop.

Honestly for what its worth though this is a headache to implement, a waste
of disk/cpu resources, and is pointless as long as you can ask them to stop
and deny access if they continue.  In my experience management will support
you if you just explain it an abuse of resources.  Technology cannot solve
problems where problems only exist in limited minds of individuals.

 --tim

-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
Sent: Thursday, January 04, 2001 2:23 PM
To: Edward Chaves
Cc: [EMAIL PROTECTED]
Subject: Re: Restricting frequency of POP3 connects



dear edward,
i understand what u mean. however, pls note that its not always about
taking action and blocking access. at times, its better to restrict users
from doing something that we wud rather avoid. mine is such a case where
the users are employees and i cant simply block one of them coz he checks
emails frequently.

in any case, the purpose and the justification of the same aside, i m
interested in a technical input on my requirement. i m sure its going to be
a hard thing to express how and why i need to limit the frequency.

any technical suggestions will be greatly appreciated.

regards

pati
Edward Chaves writes:

> On Thu, Jan 04, 2001 at 09:57:19AM -0600, James Beam wrote:
> >
> > Think of it this way - you have a sever with 1000 users, if every one of
> > them was POPing in every 1 minute or so, imagine the strain on your POP3
> > daemon (not to mention wasted bandwidth - it may only be 1 or 2kb, but
that
> > adds up for those of us who co-locate our servers).
> >
>
> do you allow thousands of connections to the pop3d?  what about
tcpserver's
> -c option? that's your throttle.
>
> if someone is needlessly filling up pop3 connections, and blocking other
> users, isn't that more a terms of service issue? in some cases, that's
> network abuse, and grounds for termination.
>
> you can always block the offender with a tcpserver rule, right?
>
> --
> Edward




Reply via email to