On Wed, Jan 19, 2000 at 12:19:22PM -0000, Petr Novotny wrote:
> 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 can think of another way: Have 2 qmail installations, one for small
mails and the other for large mails. However, make the decision to
inject the message into one of the 2 queues at the SMTP level, by making
qmail-smtpd invoke one of the 2 qmail-queues. There's a patch somewhere
where you can make qmail-smtpd invoke a program other than qmail-queue,
which can do some checks, and _then_ invoke a qmail-queue of your
choice. Or you could patch your existing qmail-smtpd.c to invoke
different qmail-queues based on size.
--
See complete headers for more info
