[ietf-nntp] unsubscribing users too aggressively

[Russ Allbery]

Rob Siemborski <rjs3 at andrew.cmu.edu> writes:
> On Mon, 21 Jun 2004, Ruud H.G. van Tol wrote:

>> Please compare the list archive to what you actually received. In
>> that way you'll find out whether the problem is at your end.

> The problem does indeed appear to be the viruses (the archive doesn't
> have Message-ids, but the missing messagers are the ones with "Subject:
> Re: Hi" and "Subject: Re: Yahoo!" mostly from rra at stanford.edu.

Yeah, my e-mail address gets forged on viruses pretty routinely since I've
used the same e-mail address for many years.

> Is there anything that can be done to either implement a virus scanner
> on the list or make it much less sensitive to bounces? [e.g. if the
> "you've been kicked off the list" message gets through, reenable
> delivery?]

Stan, I can't get into the Mailman configuration; could you crank the
bounce ratio up to something like 10.0 for this group (so that it takes
ten bounces without a successful delivery before someone is unsubscribed)?
There may also be some way of filtering out bad content-types, although I
don't know if that would cause problems for some things people might want
to send to the list.  (I think most of the viruses are attached as simple
application/octet-stream.)

BTW, so that people are aware, Stan added me to the list of moderators for
the list so that I could see what the list server was rejecting in case it
caught anyone's posts by accident, but I'm afraid that turned out to be a
lost cause.  The sheer amount of spam that this list gets that's turned
away by SpamAssassin and non-member filters runs in the hundreds of
messages a day.  The most I've been able to do is just keep the most
recent ones in a folder so that I can grep through if anyone runs into
problems.  Also, because of the sheer quantity of spam, I can't get into
the moderation interface to Mailman to approve or reject posts; it looks
like the number of pending posts is so large that it's overwhelmed the
Mailman database.

If we weren't so close to being done, I'd advocate abandoning this
particular address in favor of another.  When we recharter, I think it
would be a good idea to do that.

-- 
Russ Allbery (rra at stanford.edu)             <http://www.eyrie.org/~eagle/>