-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hi,
I have found many people claiming the current OpenSMTP (SMTP-
after-POP) is useless for them because some MTAs (like
Netscape? Outlook? I don't really know) try to do SMTP first and
the retrieve mail.
A revolutionary approach would be POP-after-SMTP. Here is my
outline and I hope you can tell me if it's too unsafe or what are the
weak point (hell, you might even write the scripts!):
1. SMTP relay looks kind of open - and relay attempts are
accepted (not sent out, just accepted). qmail-smtpd is invoked
through a script which does something like
if (not set RELAYCLIENT)
set RELAYCLIENT="@relay-attempt-"+TCPREMOTEIP;
qmail-smtpd
(what's the right syntax for it?)
2. There is a virtual domain for relay attempts which stores all the
messages in some directory, easily identified by IP address (like
relay/aaa.bbb.cc.dd) - kind of ppp outgoing directory.
3. POP authentication triggers a script which - through serialmail
and with RELAYCLIENT="" - stuffs the contents of
relay/aaa.bbb.cc.dd directory to qmail at localhost (it means that
for localhost there is a RELAYCLIENT="" line in tcp.smtp.cdb)
4. There is a cron job which periodically bounces or deletes mails
in relay/whatever (which were not sent out after successful POP3
authentication).
What do you think, is it a DoS in disguise, or is there a decent
chance to get it working?
(Any volunteers to code it? I don't currently need it - therefore I
can't spend any time on that.)
-----BEGIN PGP SIGNATURE-----
Version: PGP 6.0.2 -- QDPGP 2.60
Comment: http://community.wow.net/grt/qdpgp.html
iQA/AwUBNymJxlMwP8g7qbw/EQLfPgCeMSjWZQrRfgbQ6IMlae02L517tr8AoN4l
WcvkkL2X93xeXrH4JfCJf6Wc
=5gK9
-----END PGP SIGNATURE-----
--
Petr Novotny, ANTEK CS, antek.cz node administrator
[EMAIL PROTECTED]
PGP key ID: 0x3BA9BC3F
-- Don't you know there ain't no devil there's just God when he's drunk.
[Tom Waits]
