Oh it was nothing at all really....Embarrassingly simple as a matter
of fact. The POP server only allowed 1 concurrent SMTP session from
the same host. Postfix was configured to try as many as 10. So for
every message sent, a whole slew of them were deferred because the
POP server rejected the connection attempts.

EIMS only opens one connection for each host in it's outgoing queue,
and spits out each message for that host in sequence through the same
connection. It also wouldn't have deferred individual messages due
to a failed connection attempt....it would have attempted to open
it's single connection and if it failed it would have deferred every
message for that host.

Since EIMS is the only mail server I've used before, I foolishly
expected similar behavior.

>What were the problem+solution? Posting a description could help
>someone else who experiences the same problem.
>
> Wietse
>
>Adam Moffett:
>> >
>> >This is an impossible scenario. It would happen only when the mail
>> >delivery request contains zero recipients. How reproducible is
>> >this, and what does Postfix actually log?
>> >
>> > Wietse
>>
>> I'm sorry...you're right. I musta drank too much coffee yesterday
>> because the RSET command was from a different TCP stream than the one
>> where a message was actually being sent.
>>
>> I found the problem in about 5 minutes this morning. I'll try to
>> refrain from jumping to conclusions in the future.
>>
>> -
>> To unsubscribe, send mail to majordomopostfix.org with content
>> (not subject): unsubscribe postfix-users
>>
>>
>
>-
>To unsubscribe, send mail to majordomopostfix.org with content
>(not subject): unsubscribe postfix-users

-
To unsubscribe, send mail to majordomopostfix.org with content
(not subject): unsubscribe postfix-users

• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]