Valentin Chopov
valentin@valcho.net
Thu Apr 3 13:51:37 UTC 2003
Vernon,
Maybe the database can contain {IP address, rejections/[hdw]}
In this case we will keep avarage rejections per hour/day/week from the IP
address. If this amount of rejections/[hdw] are less then
"limit-rej-thold", "hard-rej-thold" will be in action, if these
rejections reach "limit-rej-thold", "soft-rej-thold" will be action.
Thanks,
Val
On Wed, 2 Apr 2003, Vernon Schryver wrote:
> > From: Tim Wicinski <tim@meer.net>
> > To: Valentin Chopov <valentin@valcho.net>
>
> > > "rej-thold".
> > > "hard-rej-thold" - this is the same as thw current "rej-thold"
> > > "soft-rej-thold" and "limit-rej-thold"
> > > I'll try to explain my idea with an example:
> > > hard-rej-thold=1000
> > > soft-rej-thold=100
> > > limit-rej-thold=10
> > >
> > > If we rejected at least 10 messages with "hard-rej-thold=1000" from
> > > the IP address A.B.C.D, after that to start rejecting messages with
> > > "soft-rej-thold=100" from the same IP address.
>
> That's interesting, but I see implementation hassles in maintaining
> the database {IP address, recent rejections}.
>
> Or do I understand the idea?
> It seems that every time you rejected a message, you'd increment a
> counter corresponding to the IP address of the SMTP client. Separately,
> the rejection threshold would be affected by the previous total number
> of rejected messages. The rejection counter for each IP address would
> decrease, perhaps by exponential decay, as in halving it once per day
> or week.
>
> Is that right?
>
>
> Vernon Schryver vjs@rhyolite.com
> _______________________________________________
> DCC mailing list DCC@rhyolite.com
> http://www.rhyolite.com/mailman/listinfo/dcc
>
>
More information about the DCC
mailing list