Charlie Brady wrote:
> Would you care to estimate the number of Internet connected unrepaired
> Lotus Notes systems which say EHLO with aol.com or hotmail.com or my IP
> address?
SWAG, about 0. ;-)
But that's _also_ not my point. In the specific case of spamhelo, we can pretty
much assume that no further communiction from the remote server is possible and
so Instant Death(TM) is justified. On the other hand, there are a number of
other hooks which would be well served by handling RFC abuse in the core, rather
than on a case by case basis.
If we have issued a 5xx or 4xx message, the only acceptable (to us) next
commands from the remote MTA are QUIT or RSET (and possibly the latter only in
the case of 4xx messages). I know for a fact that there is additional chatter
than occurs before zombie spawn get the picture (or hit something that does
disconnect them). So why don't we codify that in the core.
In pseudocode:
IF return == DENY or return == DENYSOFT
transaction->denied(1)
In the preamble for each SMTP command except QUIT and RSET:
if transaction->denied() == 1
DISCONNECT
John
