Yes.
>If so, that mail is subject to the "maximal_queue_lifetime", so that after
>that
>period, it gets returned to the sender. Correct?
Yes.
>While in the deferred queue, it is bypassed by normal queue processing.
>Correct?
Not quite. The qmgr checks to see if it's time to attempt delivery again,
or if maximal_queue_lifetime is exceeded. If the transport is deferred, no
delivery attempt is made. When maximal_queue_lifetime is exceeded, the
mail is returned to sender.
>Delivery attempts will only occur if the queue is flushed (sendmail -q or
>portqueue -f). Correct?
Yes, or if the mail is requeued with postsuper -r QUEUEID
>If I use HOLD form an access map, the mail goes into the "hold" queue.
>Correct?
>
>It stays there until "postsuper -H" is issued. Correct?
>
>It is immune to "postqueue -f". Correct?
Yes, yes, yes. Mail is also released from the hold queue when requeued
with postsuper -r QUEUEID or ALL, but will end up on HOLD again unless you
change the rule that caused the original HOLD.
>What happens if the mail has been in the "hold" queue for longer than the
>maximal_queue_lifetime when it is "released" (by postsuper -H)?
When mail is released from the hold queue with postsuper -H, delivery is
attempted once regardless of the age of the mail. If delivery fails and
the mail is deferred, the mail is then checked to see if
maximal_queue_lifetime has been exceeded.
>Colin
>--
>Colin Campbell
>Unix Support/Postmaster/Hostmaster
>Citec
>+61 7 3227 6334
--
Noel Jones
-
To unsubscribe from the postfix-users list, click the link below:
<mailto:majo...@postfix.org?body=unsubscribe%20postfix-users>
The first question was about a message in the deferred queue. Those
messages are processed with "postqueue -f" or "postfix flush". Some folks
define a transport named "hold" and add that to the "defer_transports"
list. This is not the same as the HOLD queue function, but the similar
name can cause confusion.
The second question was about mail on the "hold" queue, which gets there
with either the HOLD access table function, or "postsuper -h
QUEUEID". Messages in the hold queue are not processed by "postqueue -f"
or "postfix flush".