-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,

it seems that my users (at a site in Italy I administer) would not 
obey the rules and would keep on trying to send out 
simultaneously large e-mails (about 5MB) - it congests the line, 
and the connections time out, and many retries are neccessary.

I therefore consider imposing some kind of policy on outgoing mail.

The idea is to have _two_ outgoing queues: One is without 
limitations, with concurrencyremote at 20 or 40, and is used for 
small mails (where the latency - roundtrip time - plays a large role 
in the speed of delivery). The other is for outgoing files larger than a 
certain size (like 512kB or 1MB) and concurrencyremote is set to 1.

I believe it would solve most of the problems I am seeing. (The line 
upgrade is being discussed for the last two years, with no avail 
whatsoever.)

Now how do I implement this policy? (The users inject the mail by 
SMTP, never by qmail-inject.)

What I could think of has to do with three qmail installations: One 
accepting SMTP connections, and having a catch-all virtual 
domain. The .qmail-catchall-default would then check the size of 
the message, and invoke qmail2/forward or qmail3/forward 
accordingly. (What should the forward line look like BTW?)

I am not sure if it can't be done more efficiently (with less qmail 
installations and/or with more effective switch).


Could you please comment? Thanks

-----BEGIN PGP SIGNATURE-----
Version: PGP 6.0.2 -- QDPGP 2.60 
Comment: http://community.wow.net/grt/qdpgp.html

iQA/AwUBOIWryVMwP8g7qbw/EQLjawCgrMGcRBk9XvjUkTF2BzQtxCvIRxoAmwUU
3vI/VcWJsx0tRpgIlvxgBCyG
=u1Om
-----END PGP SIGNATURE-----
--
Petr Novotny, ANTEK CS
[EMAIL PROTECTED]
http://www.antek.cz
PGP key ID: 0x3BA9BC3F
-- Don't you know there ain't no devil there's just God when he's drunk.
                                                             [Tom Waits]

Reply via email to