Purposely bouncing selected mail? (SUMMARY)
Jay Lawlor
jlol at REMUS.EE.BYU.EDU
Fri Sep 28 01:13:29 AEST 1990
I promised a summary and have been getting some requests for one so
here it is.
A couple people mentioned that this might be kind of an anti-social
thing to do. I hadn't really thought of it that way before. But
considering that so many people are happy to share their time and
knowledge on the net and through email, I guess some of us could stand
to be a little more tolerant of mail we consider junk.
Actually, I made the request on behalf of a friend. I'm pretty happy
with the mailing lists that I'm on currently.
But here goes... I haven't tried these yet. Some of them look rather
fun.
------------------------------------------------------------------
>>>>> On Fri, 21 Sep 90 21:27:42 -0400, David Lesher <wb8foz at mthvax.cs.miami.edu> said:
David> I use filter from elm. Here's how. (I'm chewing on Phil, a vampire
David> type I know who has mailboxes on a dozen machines, but never answers
David> his mail.)
David> ~/.elm/filter-rules:
David> If (from "ncoast!smith") then execute "elm -s garlictoast Count"
David> If (from "smith at skybridge") then execute "elm -s garlictoast Count"
David> If (from "smith at pyrite") then execute "elm -s garlictoast Count"
David> If (from "hal9000") then execute "elm -s garlictoast Count"
David> If (from "val9000") then execute "elm -s garlictoast Count"
David> If (from "Phil Smyth") then execute "elm -s garlictoast Count"
David> If (from "smith at ncoast") then execute "elm -s garlictoast Count"
David> If (from "sidrat") then execute "elm -s garlictoast Count"
David> where Count is a alias within elm as follows:
David> ~/.elm/aliases.text:
David> c1 = Phil Smith = ncoast!hal9000!smith at usenet.ins.cwru.edu
David> c2 = Phil Smith = ncoast!sidrat!smith at usenet.ins.cwru.edu
David> c3 = Phil Smith = ncoast!hal9000!val9000!smith at usenet.ins.cwru.edu
David> c5 = Phil Smith = ncoast!smith at usenet.ins.cwru.edu
David> c6 = Phil Smith = ncoast!hal9000!root at usenet.ins.cwru.edu
David> c7 = Phil Smith = ncoast!hal9000!val9000!root at usenet.ins.cwru.edu
David> c8 = Phil Smith = ncoast!sidrat!root at usenet.ins.cwru.edu
David> c9 = Phil Smith = smith at pyrite.som.cwru.edu
David> Count = Phil Smith = c1, c2, c3, c5, c6, c7, c8, c9
David> with c1-9 needed because elm can only map its own alili on a
David> multiple destination line.
David> You could, of course, be less anti-social, and just use
David> filter to delete the mail. It will look at subject, from, and
David> a bunch of other things...
David> --
David> A host is a host from coast to coast.....wb8foz at mthvax.cs.miami.edu
David> & no one will talk to a host that's close............(305) 255-RTFM
David> Unless the host (that isn't close)......................pob 570-335
David> is busy, hung or dead....................................33257-0335
---------------------------------------------------------------------------
>>>>> On Fri, 21 Sep 90 21:35:08 PDT, raymond at math.berkeley.edu (Raymond Chen) said:
Raymond> How about "sendmail -t" with a spoofed return address?
Raymond> In general, though, this is a Very Antisocial thing to do...
-------------------------------------------------------------------------
>>>>> On Sat, 22 Sep 90 13:33:08 EDT, Rodney Peck II <rodney at ipl.rpi.edu> said:
Rodney> that's really not such a good idea -- if the admin doesn't answer and
Rodney> take you off, he's probably not watching at all and all you will do is
Rodney> clog up the net with messages.
Rodney> Something better would be to use a filter like mh or something you write
Rodney> yourself to catch your incoming mail, check it to see if its from the list,
Rodney> and if it is, compose a letter to the list-request asking to be removed,
Rodney> and tossing the received message in /dev/null.
Rodney> this would solve your problem since you would never see the message again.
--------------------------------------------------------------------------
>>>>> On Mon, 24 Sep 1990 15:55:02 EDT, mskucherawy at watmath.waterloo.edu (Murray S. Kucherawy) said:
Murray> Return-Receipt-To: mskucherawy at watmath.waterloo.edu
Murray> Mush can do it (it's not BSD or Unix V, but a mail package that can be told
Murray> to do what you're asking for). If you want more info, I cn provide it.
Murray> --
Murray> ============================== Murray S. Kucherawy ==========================
-----------------------------------------------------------------------------
>>>>> On Mon, 24 Sep 90 13:29 EDT, tct!chip at uunet.UU.NET (Chip Salzenberg) said:
Chip> I suggest you get my Deliver program. It can do just about anything
Chip> with arriving mail, including throw it away. You control its actions
Chip> by writing shell scripts, called "delivery files". Here's mine:
Chip> ================================================================
Chip> # Delivery for Chip
Chip> PATH=/bin:/usr/bin:/u/local/bin ; export PATH
Chip> user="$1"
Chip> case "$SENDER" in
Chip> *!NOVELL)
Chip> echo DROP
Chip> ( sed -e 's/!uunet![A-Za-z]*\.PRINCETON\.EDU!/!/g' $HEADER
Chip> cat $BODY ) |
Chip> /u/local/bin/postmail mail.novell >&2
Chip> exit
Chip> ;;
Chip> esac
Chip> echo $user
Chip> for tty in `who | awk '$1 == "'"$user"'" { print $2 }'`
Chip> do
Chip> case "$tty" in
Chip> ttyp*) continue ;;
Chip> esac
Chip> f="`header -f from $HEADER`"
Chip> [ "" != "$f" ] || f="$SENDER"
Chip> m="Hey! You have new mail from $f."
Chip> echo "\r\n>>> $m <<<\r\n" >>/dev/$tty 2>/dev/null
Chip> done
Chip> ================================================================
Chip> You see the part where my Novell mailing list is gatewayed to a local
Chip> newsgroup, based on the $SENDER value (an environment variable set by
Chip> Deliver before it runs the script). To throw away mail, just replace
Chip> the "postmail" code with "echo DROP; exit".
Chip> Deliver works with System V, BSD, Xenix and V7 (!). When used with
Chip> Sendmail or Smail 3, you put "|/usr/bin/deliver" in your .forward.
Chip> With Smail 2.5, you configure Smail to run /usr/bin/deliver for
Chip> delivery of local (and/or remote) mail. Otherwise, you have to hack
Chip> it in manually.
Chip> Note that current version is 2.0 PL12. Once you have 2.0 PL11, send
Chip> me mail and I'll send you patch #12.
Chip> **FTP: Get Deliver 2.0 PL11 by anonymous FTP from tut.cis.ohio-state.edu.
Chip> For those of you without DNS, that's [128.146.8.60]. Here's the relevant
Chip> part of the Ohio State archive info:
Chip> Deliver 2.0
Chip> -----------
Chip> Source is Chip Salzenberg <ateng!tct!chip at uunet.uu.net>.
Chip> Root is ~/deliver/deliver-2.0-part-[1234].Z [4 parts], varying sizes
Chip> less than 25Kbytes each.
Chip> Note that "~" is the anonymous FTP "pub" directory. So, after making an
Chip> FTP connection as user "anonymous," type the following commands to get
Chip> Deliver into the current directory:
Chip> binary
Chip> cd pub/deliver
Chip> get deliver-2.0-part-1.Z deliver.01.Z
Chip> get deliver-2.0-part-2.Z deliver.02.Z
Chip> get deliver-2.0-part-3.Z deliver.03.Z
Chip> get deliver-2.0-part-4.Z deliver.04.Z
Chip> **UUCP: Get Deliver 2.0 PL11 by anonymous UUCP from alfred:
Chip> alfred Any ACU 2400 14072909799 "" \d in:--in: anonuucp word: unknown
Chip> alfred Any ACU 1200 14072909799 "" \d\r\d in:--in: anonuucp word: unknown
Chip> Alfred's complete file listing is alfred!~/ANON/ls-lR.
Chip> Deliver is stored on alfred as four compressed sharchives in the
Chip> ~/ANON/deliver directory, with a total size of less than 100K.
Chip> To get Deliver 2.0 from alfred into the current directory, type:
Chip> uucp alfred!~/ANON/deliver/deliver.01.Z .
Chip> uucp alfred!~/ANON/deliver/deliver.02.Z .
Chip> uucp alfred!~/ANON/deliver/deliver.03.Z .
Chip> uucp alfred!~/ANON/deliver/deliver.04.Z .
Chip> Note that wildcards ("deliver.??.Z") will *not* work.
Chip> --
Chip> Chip Salzenberg at Teltronics/TCT <chip at tct.uucp>, <uunet!pdn!tct!chip>
---------------------------------------------------------------------------
>>>>> On Wed, 26 Sep 90 20:20:00 PDT, andy at spl28.spl.fac.com (Andrew Purshottam) said:
Andrew> Ask shipley at berkeley.edu for "autobounce", you'll love it.
Andrew> I reccomend against installing it in you .forward as it has
Andrew> a few bugs still. You can use it against particular pieces
Andrew> of mail or have it screen all mail and bounce with selected
Andrew> probabilaties though uses in your ".s*itlist"!
Andrew> Cheers,
Andrew> Andy
----------------------------------------------------------------------------
>>>>> On 23 Sep 90 16:55:54 GMT, karl_kleinpaste at charcoal.com said:
karl> The following suggestion is entirely untested, but the principle
karl> should be reasonably obvious.
karl> In .forward, place this:
karl> \jlol, "|/home/dirs/jlol/bin/bouncer"
karl> Then the bouncer script could look like...
karl> ----------------
karl> #!/bin/csh -f
karl> #
karl> # Bouncer: a script to scan incoming mail to see if it comes
karl> # from the "wrong" place, and fakes a bounce-o-gram back if so.
karl> #
karl> set t1=/tmp/bounce.a$$
karl> set t2=/tmp/bounce.b$$
karl> #
karl> # Save the mail; then pull out the From_ address.
cat > $t1
karl> set origin="`head -1 $t1 | awk '{ print $2 }'`"
karl> #
karl> # If it comes from the wrong place (hopefully, a *-request@* form)...
karl> if ("$origin" == bad at addr.ess) then
karl> #
karl> # Bad juju. Preformat a pseudobounce-o-gram.
karl> cat << EoF > $t2
karl> To: $origin
karl> Subject: Returned mail: User unknown
karl> From: mailer-daemon at remus.ee.byu.edu
karl> ----- Transcript of session follows -----
karl> 554 jlol...User unknown
karl> EoF
karl> #
karl> # Stuff it directly into sendmail.
karl> (cat $t2 ; tail +2 $t1) | /usr/lib/sendmail -t
karl> endif
karl> #
karl> # Clean up.
karl> rm -f $t1 $t2
karl> exit 0
karl> ----------------
karl> Embellish to taste.
karl> --karl
More information about the Comp.unix.admin
mailing list