ietf-smtp
[Top] [All Lists]

Re: per user post-data rejects, Processing after the end of DATA

2010-08-14 09:14:05

Hello,

On 11.08.2010 21:36, David MacQuigg wrote:
Дилян Палаузов wrote:
 >
 > On 10.08.2010 20:18, David MacQuigg wrote:
 >> My system is too small to draw any conclusions, but my impression is
 >> that we are overly-concerned about a very small fraction of the
mailflow
 >> - those few messages that are legit, but get a false reject from a
 >> statistical filter. We really don't care about the proper SMTP response
 >> if it is spam. The only advantage I see in running the spam filter
 >> before ACCEPT is a few less messages at the bottom of my spam bucket. I
 >> wish I had some numbers from a larger system.
 >
 > A user might have installed redirect for her incoming mails to
another mail provider. The other provider might SMTP reject incoming
spam. If there is spam for your user, your server does accept it, then
sends it to the other mail provider, and that second server SMTP rejects
it, then your server will have to send DSN for that SPAM (provided that
you do not want to discard emails). With rejection-preferences similar
on both servers, the amount of such DSNs will decrease, when both
servers reject after DATA. You have no influence on what other servers
do with the spam (smtp rejecting or inserting in spam folders).

We have quite a bit of influence. We insist that our recipients turn off
all spam filtering, chain-forwarding, SPF checks, whatever might
conflict with our acting as Receiver for their box67.com email address.
Typically, they will have a private mailbox known only to our
Receiver/Forwarder. Failures resulting from SMTP rejects, DSNs, or any
other problem at the recipient's mailbox are treated the same as a
mailbox being suddenly offline. Incoming mail to that address is
rejected until we can resolve the problem with the recipient.

I do not think it is good to insist on such things. One might want to communicate to a sender, that her mailbox does not exist anymore and SMTP reject the mails from that sender. Your service will then conclude, that destination mailbox is suddenly offline, but it is not. [I used to receive a newsletter from my mobile operator, that contained no instructions how to unsubscribe and writing replies has not unsubscribed me. The fastest solution (for me) was to instruct my mailbox to SMTP reject mails with that newsletter).

 >> Would be nice if SMTP had a 'conditional ACCEPT': "Sorry, your
 >> transmitter ID 'yahoo.com' cannot be verified, Your message has been
 >> routed to the recipient's quarantine." Imagine the effect that would
 >> have on all the lame excuses for an invalid or hard-to-verify HELO
name.
 >
 > What do you want to communicate to the sender with this conditional
ACCEPT: "Dear Sender, your suspicious email has arrived. It might be
read soon, it might be read late, or it might not be read at all."?

No, the message I suggested had a specific reason for the failure. It
should also have a link to a webpage with instructions on how to fix the
failure, like what we have now on REJECT messages:
http://open-mail.org/3strikes/REJECT.html

The link above shall be send with SMTP permanent reject (not conditional accept).

My rewording does not change the meaning of David`s message -- the sender wants to know if the message is delivered, or not; and the message "Sorry, your transmitter ID 'yahoo.com' cannot be verified, Your message has been routed to the recipient's quarantine." does not clarify this.

 > The problem with the spam folders are the false positives. Many users
do check the spam folders/quarantines not that often as their Inbox and
with less attention. This might lead to reading a message too late, or
even overseeing it. I prefer to immediately notify the sender that her
message was not delivered, including in the SMTP response alternative
ways to contact the recipient (and spamassassin-evaluation of the mail),
rather than let the sender hope that her email was properly delivered,
while there is a risk that the recipient oversees that message.

False rejects are indeed the main problem with modern email systems, but
this problem isn't solved by rejecting a small fraction of messages with
the highest spam scores. There are still plenty of messages, almost all
spam, that get a score low enough to be kept in quarantine, and there is
always a risk that the recipient will overlook a legitimate message. The
only good solution to the false reject problem is to offer senders a way
to avoid the statistical filters entirely. That is what a reputation
system does. When a message comes from yahoo.com (or any other A-rated
domain), it goes straight to our recipient's inbox.

Informing the sender, that her message was evaluated as spam and is therefore returned, including information how to alternatively contact the recipient, is also good solution. Or rather: giving the mailbox owners the one [smtp spam reject] or the other [spam folders] option is a good solution.

By the way, in the Russian newspapers you can read job offers for marketing agents. Their job is to send mails using free mail providers to a provided long lists of recipients by copy&paste the list of addresses [= to send spam]. Thus, if the message comes from an A-rated domain, it does not mean the message is ham.

In article <01NQK8EKPYCU000CVY(_at_)mauve(_dot_)mrochek(_dot_)com> you write:
>>> Hello,
>>>
>>> if rejecting after CRLF.CRLF is better than bouncing, what about
>>> discussing draft-hall-prdr-00 and doing it more normative? ...
>>
>>I never understood why the interest level in this was so low.
> It seems much more likely in that scenario that rather than rejecting, a system would drop it in the second recipient's spam folder, or if they're reasonably confident in their filters, just not deliver it to the second recipient.

The sender shall be informed, that only the second recipient has not received the message. Just not deliver to the second recipient is not good enough. I hate when mails are discarded:

This year I got a mail from a user, who has not received the mails forwarded from my server. I replied him to check in his spam folder. In two weeks the same happened. In two more weeks, when he wrote me again, I called him and figured out, that my answers were discarded and have not reached him (they were not in the spam folder, they were discarded).

A propos spam folders: What happens when a spam is sent to a mailing list, where everybody can post (so the decision can be made after DATA). Mailing lists do not have spam folders. You either SMTP reject the message, or accept it, and later wrap it and send it to the list moderators [causing them manual work], or send it back to the indicated sender [not good]. If this message is sent in the same envelope to a user with settings, which permit delivering her the message, then there is no choice, but to accept the message and bounce it [or accept for the first recipient, and TEMP fail for all the rest; when the sender replies later, do the same and so on, until no recipients are left -- accept only one recipient per delivery try].

> So if they've decided it's spam, what's the problem with throwing it away?

False positive.

Със здраве
  Дилян

<Prev in Thread] Current Thread [Next in Thread>