>There's a crucial detail or three missing from this description. ...
Sorry I was not clear. We use listproc and it handles the AOL mailbox full
messages just fine. I was describing a situation and asking if there was
something 'better' that could be done.
>The point is, there are already well-established protocols to prevent
>undesirable behavior from bounced mail. Rather than try to invent new
>ways to solve this problem, we should use the ones that are
Our lists have
and the To: pointed to the list.
What else could I do?
>>Build a list of addresses that do not want notice that someone's mailbox is
>>full. It should be easy to do and it will make life a little easier for some
>>of us list managers.
>The question is, is it really a good idea to suppress bounces when the
>mailbox is full? The whole idea is to reduce the usage of netbandwidth.
>If people have a full mailbox, they're evidently not going to benefit from
>further messages from your mailinglist. I.e. they might as well be
>If the problem really is transient, these bounce messages could be suppressed
>(if coming from a mailinglist); the problem is determining what is a transient
>problem and what not.
Good point. If it's a transient problem, don't use up bandwidth sending mail
back to a mailing list. If the problem (full mailbox in my example) does not
go away, send a message so the list owner can remove the address from the
If I unsubscribe someone as soon as I get a message that they have a full
mailbox, that saves bandwidth. It also removes someone that may have had a
VERY transient problem from the list.
Is the proper thing to do to supress sending warning messages back if the
mail has 'Precedence: bulk' and send an error message to the 'Errors-To:' if
an account is no longer active?