-----BEGIN PGP SIGNED MESSAGE-----
>>>>> "PB" == Peter Bowyer <pbowyer@verity.com> writes:
PB> and the characteristics of the new MTA are that it handles mail more
PB> effectively if it gets one message with all 2000 list recipients
PB> specified, rather than lots of mails with a subset of the
PB> recipients.
On general principle, I find this difficult to believe. At the very
least the memory requirements skyrocket (need to keep 2000 mailboxes
times however many instances of the MTA are running concurrently in
available memory to work on them), which can (and probably will) cause
increased VM swapping, which will slow down the machine, thus negating
any benefits of the supposedly better delivery mechanism.
Lemme guess... you have bought into the hype behind qmail, right? :)
qmail is full of such "improvements" that ignore the environment it runs
in.
PB> Now, the sendmail config I'm using on the MD box splits each mail up
PB> into chunks of about 20 recipients, which uses a lot of LAN traffic
PB> to get long mails to the smarthost, and as described above, is
PB> sub-optimal for the effeciency of the smarthost.
Stop using bulk_mailer and let sendmail transfer the queue over to the
smart_host just once. The problem is that there is no way to "split"
the queue on the smart_host without hacking up the MTA.
The only way to really win is to move your lists over to the mail hub.
-----BEGIN PGP SIGNATURE-----
Version: 2.6.3a
Charset: noconv
iQCVAwUBMskkY56VRH7BJMxHAQEFaQQAwXJRHmuKFaiWydh7mIDsd+xCq4/l2V4E
bKEhc81/ybgipUFpMQkiHY+Y7I+hgvG4ZBMtI816dclWqbhYOfGQYsd6V1aIIIem
1U2kSBahIDVAFTio6RKFWsrqAwuI76T5BZA4Sd1o+B8B6ovZYaVDAv7ZAqbntOdJ
o+pViZMYY2o=
=8jVD
-----END PGP SIGNATURE-----
--
Rich Pieri <rich.pieri@prescienttech.com> | Caution: Happy Fun Ball may
Prescient Technologies, Inc. | suddenly accelerate to dangerous
A Stone & Webster Company | speeds.
I speak for myself, not PTI or SWEC |
References:
|
|
