On 2012-06-07 at 11:10 +0200, Totoro wrote:
> Actually one of our client can't send mail to one domaine name.
>
> Our client mail all the time go in quarantine, the remote server return a
> Bad Header - Bad Form error.
That does not sound like an Exim error.
On 2012-06-07 at 11:44 +0200, Totoro wrote:
> Jun 7 10:55:28 mail postgrey[828]: action=pass, reason=triplet found,
> delay=2005, client_name=mail.domaine.problem,
> client_address=XXX.XXX.XXX.XXX, sender=sender.domaine.problem,
> recipient=recipient.domaine
> Jun 7 10:55:28 mail postfix/cleanup[11083]: CDDE6D5002C:
> message-id=<201206071022.00579.sender.domaine.problem>
> Jun 7 10:55:28 mail postfix/qmgr[965]: CDDE6D5002C:
> from=<sender.domaine.problem>, size=2428, nrcpt=1 (queue active)
> Jun 7 10:55:29 mail postfix/cleanup[11083]: 52111D50076:
> message-id=<201206071022.00579.sender.domaine.problem>
> Jun 7 10:55:29 mail amavis[10715]: (10715-02) Passed BAD-HEADER,
> [XXX.XXX.XXX.XXX] [XXX.XXX.XXX.XXX] <sender.domaine.problem> ->
> <recipient.domaine>, quarantine: n/badh-naJNEpVK0y0e, Message-ID:
> <201206071022.00579.sender.domaine.problem>, mail_id: naJNEpVK0y0e, Hits:
> 0.492, size: 2428, queued_as: 52111D50076, 416 ms
> Jun 7 10:55:29 mail postfix/smtp[11084]: CDDE6D5002C:
> to=<recipient.domaine>, relay=127.0.0.1[127.0.0.1]:10024, delay=0.52,
> delays=0.09/0.01/0/0.42, dsn=2.0.0, status=sent (250 2.0.0 Ok, id=10715-02,
> from MTA([127.0.0.1]:10025): 250 2.0.0 Ok: queued as 52111D50076)
> Jun 7 10:55:29 mail postfix/qmgr[965]: 52111D50076:
> from=<sender.domaine.problem>, size=1714, nrcpt=1 (queue active)
> Jun 7 10:55:29 mail dovecot: deliver(recipient.domaine): sieve:
> msgid=<201206071022.00579.sender.domaine.problem>: stored mail into mailbox
> 'INBOX'
> Jun 7 10:55:29 mail postfix/pipe[11088]: 52111D50076:
> to=<recipient.domaine>, orig_to=<recipient.domaine>, relay=dovecot,
> delay=0.08, delays=0.03/0.01/0/0.04, dsn=2.0.0, status=sent (delivered via
> dovecot service)
I do not know Postfix well enough to pick apart the logs there, but I
*think* that this is Postfix processing message CDDE6D5002C, rejecting
it, and generating 52111D50076 as an error response and then
successfully delivering the error response.
If true, then this is the Postfix quarantine rules rejecting your mail,
not Exim.
> Exim Server side log:
>
> 2012-06-07 10:22:01 1ScXyj-00039i-BA <= sender.domaine.problem
> H=(sender.domaine.problem.localnet) [XXX.XXX.XXX.XXX] P=esmtp S=2061
> id=201206071022.00579.sender.domaine.problem T="test"
> 2012-06-07 10:22:02 1ScXyj-00039i-BA == recipient.domaine R=dnslookup
> T=remote_smtp defer (-44): SMTP error from remote mail server after RCPT
> TO:<recipient.domaine>: host mail.domaine [XXX.XXX.XXX.XXX]: 450 4.2.0
> <recipient.domaine>: Recipient address rejected: Greylisted, see
> http://postgrey.schweikert.ch/help/domaine.html
> 2012-06-07 10:55:27 1ScXyj-00039i-BA => recipient.domaine R=dnslookup
> T=remote_smtp H=mail.domaine [XXX.XXX.XXX.XXX]
> 2012-06-07 10:55:27 1ScXyj-00039i-BA Completed
That is Exim receiving the mail, trying to deliver, reporting that the
first attempt failed because the remote server is using "greylisting" to
delay mail from new senders; then 30+ minutes later, Exim tries again,
the mail is accepted and delivery is complete.
So Exim is not the source of the problems here. It delivered the mail
just fine.
This would be slightly easier to check the connections between the two
if you have Exim log the SMTP confirmation from the remote side, because
that will typically include the receiving system's message-id.
If you check your Exim configuration file for "log_selector =" and add
to it "+smtp_confirmation", or if log_selector is not found, then add to
the *main* section of the configuration file this:
log_selector = +smtp_confirmation
then the "=>" line above will give you that extra piece of information.
-Phil
