Rodrigo Severo writes:
Let's admit the following three .courier files:
1.
use...@example.com,use...@example.com
2.
use...@example.com
use...@example.com
3.
use...@example.com
| /some_delivery_slave
use...@example.com
Is there any difference in the way courier treats these files? I am
particulary interested in the example.com deliveries.
The only difference is that the third example involves running an external
program.
I believe that from the first .courier file Courier starts one single
delivery connection to mx.example.com for both users. My question is,
what would Courier do from the last two .courier files?
In both cases the message is forwarded to the same recipient list. A
message with multiple listed recipients in the same domain will deliver all
recipients with one delivery attempt, regardless of the message's origin.
The only major difference is that if the external program terminates with a
non-zero exit code the message is deferred or bounced, but also forwarded to
the first recipient. If the message keeps getting deferred, the fist
recipient will get a slow mailbomb. That's why programs invoked from a
.courier file must be vetted for their error recovery.
2 messages in net.sourceforge.lists.courier-users[courier-users] Re: .courier entries ...
