> Why not check during the SMTPD to see if the recipient is valid

--this is already the case--

> does  the  mail  box  have  room and is it under the allowed message
> size?

My take has been that SMTPD tries to do as little as possible in order
to  conserve  sockets  and  shift processing load to less easily DoSed
resources.  On  the other hand, one by-product is that net disk I/O is
often      increased      when     certain     checks     that     are
part-envelope/part-content into are shifted to the delivery processing
phase,  and then are failed. But while some of these checks would be a
cinch  to  perform  at  the  close of the envelope if there are only a
couple  of  recipients,  you  have to consider scaleability: with many
recipients  which  each  have  their own thresholds, the disk might be
traversed  for  seconds, or even minutes, at the system's most exposed
point.  I'd  venture  that  by  comparison,  always  so  shifting  the
anti-spam blacklist in actions in IMail 8--which are all-envelope!--is
more glaringly poor design.

> Has  this  behavior changed in 8.0x? (The server will be upgraded to
> 8.03 pending a resolution of a ODBC issue.)

Nope.

What ODBC issue?

-Sandy


------------------------------------
Sanford Whiteman, Chief Technologist
Broadleaf Systems, a division of
Cypress Integrated Systems, Inc.
e-mail: [EMAIL PROTECTED]
------------------------------------


To Unsubscribe: http://www.ipswitch.com/support/mailing-lists.html
List Archive: http://www.mail-archive.com/imail_forum%40list.ipswitch.com/
Knowledge Base/FAQ: http://www.ipswitch.com/support/IMail/

Reply via email to