reporting known/unknown recipients

Vernon Schryver vjs@calcite.rhyolite.com
Thu Apr 15 13:10:32 UTC 2010


> From: Matus UHLAR - fantomas <uhlar@fantomas.sk>

> I would like to ask your opoinion on reporting received and apparently even
> rejected mail to DCC. Do you find it OK to
>
> - report rejected mail, as if it was acepted for all recipients?
>
> - count the number of provided recipients, not only those accepted?
>   (e.g. for 3 known and 7 unknown recipients report 10)

I think the count for each DCC checksum is the number of addressees or
target mailboxes, and not the number successfully hit advertising
targets.  So dccm (with recent versions of sendmail) and dccifd in proxy
mode (e.g. as a postfix before-queue filter) count SMTP Rcpt_To commands
regadless of their validity.

> - double or multiple the number of non-existent recipients in addition to
>   number of those existing?
>   (e.g. for 3 known and 7 unknown recipients report 17, 24 or even more)

If you think that an attempt to send mail to a particular mailbox
is proof that the message is spam, why not report it with a count
of "many"?  Some people run spam traps.  See
http://www.dcc-servers.net/dcc/FAQ.html#spamtrap

Many SpamAssassin+DCC installations use DCC thresholds of "many."  The
only spam they detect with DCC are copies of messages sent to spam traps
or otherwise reported with counts of "many"


Vernon Schryver    vjs@rhyolite.com



More information about the DCC mailing list

Contact vjs@rhyolite.com by mail or use the form.