1993-03-03 - REMAIL: Usage stats.

Header Data

From: Hal <74076.1041@CompuServe.COM>
Message Hash: 46e6435f99a8ff5af82a0511ca68ec4e115c34c979ccbdd53e61297221d00a7d
Message ID: <93030306510974076.1041_DHJ45-1@CompuServe.COM>
Reply To: _N/A

UTC Datetime: 1993-03-03 07:01:06 UTC
Raw Date: Tue, 2 Mar 93 23:01:06 PST

Raw message

From: Hal <74076.1041@CompuServe.COM>
Date: Tue, 2 Mar 93 23:01:06 PST
Subject: REMAIL: Usage stats.
Message-ID: <930303065109_74076.1041_DHJ45-1@CompuServe.COM>
MIME-Version: 1.0
Content-Type: text/plain


I got mail from a remailer operator asking how he could keep some
information about how much his remailer was being used without keeping
logs of the messages.  This seemed like a good feature so I have added
it to my maildelivery file.  This is how it looks now:

# field                 pattern action/ string 
#                               result  (quote included spaces)
Request-Remailing-To    ""      pipe R  "date >> LOG.REMAIL"
Request-Remailing-To    ""      pipe A  remail.pl
Anon-To		        ""      pipe R  "date >> LOG.REMAIL"
Anon-To			""	pipe A  remail.pl
Encrypted               PGP     pipe R  "date >> LOG.ENCRYPTED"
Encrypted               PGP     pipe A  pgpmail.pl
*                       ""      pipe ?  recurse.pl

The lines which say "date >> LOG.XXX" will append one line to the
LOG.* file each time a remailing or decryption is done.  That line
will tell when it is done.  No other information is kept to allow
reconstruction of who sent the message or what it was.  You can use
"wc" to count how many lines are in the files, archive and/or delete
them periodically, or do whatever you like.

A couple of notes: When you have more than one line that matches a
particular field/pattern (defined in the first two columns), use the
letter R for the result code for the first ones, and A for the last.

Also, be SURE you put quotes around the command string if it is more
than one word.  Otherwise ALL of your mail will DISAPPEAR!  (I need to
do a little work on the slocal.pl script to make it more robust about
this. :-)

If you're using slocal rather than slocal.pl this should still work.

Hal Finney

Version: 2.1