1

I have a strange phenomenon that I have the following message in my rejectlog:

2012-04-13 14:19:17 1SIfTB-0007y2-5S H=(mail.myhost.com) [0.0.0.0] F=<[email protected]> temporarily rejected after DATA
Envelope-from: <[email protected]>
Envelope-to: <[email protected]>
P Received: from [0.0.0.0] (helo=mail.myhost.com)
    by mail.myhost.com with esmtp (Exim 4.34)
    id 1SIfTB-0007y2-5S
    for [email protected]; Fri, 13 Apr 2012 14:19:17 +0200
* Return-Path: <[email protected]>
I Message-ID: <[email protected]>
  Date: Fri, 13 Apr 2012 14:19:17 +0200
  Subject: My Subject
F From: [email protected]
T To: [email protected]
  MIME-Version: 1.0
  Content-Type: multipart/alternative;
 boundary="_=_swift_v4_13343195574f8819c529888_=_"

(Please note that myhost.com, otherhost.com, mywebhost.com and otherhost2.com are fictitious in order to maintain privacy)

The client (SwiftMailer) spits out the following error message:

Expected response code 250 but got code "451", with message "451 Temporary local problem - please try later"

The weird thing is, that exim only reports the error message above in its rejectlog - there is no sign of the mail in mainlog, paniclog, mail.log, syslog or any other logfile.

Any clues?

It's an Exim 4.34 and works fine besides these strange rejects...

4
  • Well, more information is required for this. Did it only happen once? If no, does it always happen with ONE specific message or ONE specific host? For now it looks like a temporary problem, like dns resolving, crashed database-server, crashed spamd, network-problems. Apr 17, 2012 at 23:57
  • It happens randomly when sending out newsletters with SwiftMailer (at least this is where I recognized the error). Recipient addresses doesn't seem to have any pattern, it also happened with local recipients. When sending out 500 mails, there are ~5-10 who don't get sent. Randomly, not 10 in a row. When trying to re-send minutes later, it works fine, so yes it's a temporary problem (like the error message suggests), but nevertheless I'd like to find the cause of this.. :/
    – Nuramon
    Apr 18, 2012 at 9:06
  • Check your smtp_accept_max_per_connection setting, if Swiftmailer is sending out all messages in a single connection, you might hit some system default(100?). Also, set log_selector = +all maybe you've suppressed some log messages. It might also be a problem with your acl_smtp_data which might temp-reject mails under certain conditions. Apr 18, 2012 at 9:25
  • Swiftmailer is set to send out 10 per connection (AntiFloodPlugin) and throttles to 100 Mails per Minute (ThrottlerPlugin). In Exim smtp_accept_max is set to 200, smtp_accept_max_per_connection is 1000. Thanks for the tip with log_selector, I set it to +all and will check the log next time a newsletter goes out...hopefully it'll reveal more.
    – Nuramon
    Apr 19, 2012 at 7:33

2 Answers 2

0

It's probably greylisting or similar by the other server: it's decided to give you a 451 temporary failure to see if your server is actually trying to deliver mail (in which case the retry later will be allowed through). Plenty of spam systems will not retry these failures.

1
  • Can't be...1) if it's greylisting, there should be also an entry in mainlog, containing the corresponding server who greylisted the message and 2) we also had the problem with recipients lying on the very same webserver (which has no greylisting)...it has to be something before the message goes out...there doesn't seems to be any communication with outside servers, the message just gets rejected
    – Nuramon
    Apr 16, 2012 at 9:41
0

yes because the mail did't get sent at the first place,

try doing >telnet your_server_ip 25

the expected response should be :

250 ************
welcome to xxx smtp mail server 

the response you are getting is 451 not 250 and that is due to some restriction on the client or properly miss-configuration for exim mail server

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .