On Sun, 10 Nov 1996, Brock Rozen wrote:
> I suggested something of the sort, just not on a per user level, rather
> per list level. Regardless, Jason said that the overhead would be too
> great and that people (who?) are already complaining about that.
The overhead is rather large when you're hit by 43000 help messages from
the same guy over a 3 day period... And that isn't an isolated
occurance. Since yesterday when I started ignoring mail from
postmaster/mailer-daemon, I have stopped 117 mail loops originating from
34 unique e-mail addresses. That's pretty good for a simple pattern
match check.
> The message-id one probably wouldn't add too much overhead and might solve
> somebody's problems. I also can't imagine anybody objecting to it(except
> for overhead).
I can't even see an overhead issue. Implement it as a dbm file and the
access time will be very fast. Have it configurable, and then people who
don't have access to the PERL db*() functions can just turn it off.
If you implement any of these things using dbm's the access overhead will
be minimal. A message ID, a "banned users" list and a number of mails
from a particular user restriction would solve just about every possible
problem situation.
Evan
--
Evan Champion * Director, Network Operations
mailto:evanc@synapse.net * Directeur, Exploitation du reseau
http://www.synapse.net/ * Synapse Internet
References:
|
|
