Koozali.org: home of the SME Server
Obsolete Releases => SME Server 7.x => Topic started by: drlizau on December 19, 2011, 07:36:18 AM
-
My mail inwards is being collected by qmail but not being passed onwards by dovecot.
I am unaware of what has been the cause of the stoppage - It worked last on Friday it seems and I didn't do any updates on Sunday (my usual update day).
Logging in and trying to start the dovecot service I get
# /etc/init.d/dovecot start
Starting Dovecot Imap: Error: Missing auth section for server default
Fatal: Invalid configuration in /etc/dovecot.conf
/etc/dovecot.conf contains
#------------------------------------------------------------
# !!DO NOT MODIFY THIS FILE!!
#
# Manual changes will be lost when this file is regenerated.
#
# Please read the developer's guide, which is available
# at http://www.contribs.org/development/
#
# Copyright (C) 1999-2006 Mitel Networks Corporation
#------------------------------------------------------------
maildir_copy_with_hardlinks = no
Obviously this is a mission critical failure - but where are the correct template fragments so that I can check them and find out what has broken?
Any other suggestions gratefully received.
-
Your dovecot configuration file is the default on stock SME Servers. There is nothing wrong with that.
What is the output of:
service status imap
My mail inwards is being collected by qmail but not being passed onwards by dovecot.
IIRC that is normal behavior of SME Server.
-
Your dovecot configuration file is the default on stock SME Servers. There is nothing wrong with that.
What is the output of:
service status imapIIRC that is normal behavior of SME Server.
thanks for the reply.
# service imap status
gives
run: /service/imap: (pid 4505) 29315s, normally down; run: log: (pid 3417) 29391s
So what other things do I need to check for a failure in mail delivery?
-
I suggest you look through the /var/log/qmail and /var/log/qpsmtpd logfiles.
-
I suggest you look through the /var/log/qmail and /var/log/qpsmtpd logfiles.
All I have managed to confirm is that no mail has been sent for local delivery since 13:19 local time on Friday.
Messages for remote delivery have been accepted and sent (and received).
I can't see any real clues in the logs - no mention of incoming mail for any local addresses at all.
back to my first item - should dovecot be running? because it is not running, and won't start for me
-
back to my first item - should dovecot be running? because it is not running, and won't start for me
Dovecot is running as you showed by:
# service imap status
gives
run: /service/imap: (pid 4505) 29315s, normally down; run: log: (pid 3417) 29391s
-
Dovecot is running as you showed by:
thankyou for your time, but i still seem to be going in circles.
Following the last incoming message, the next event on /var/log/messages is a ppp dropout and restart.
Once this has messed up my static IP and hence my incoming mail, but I have restarted the server, and can log in remotely to it.
I have sent test mails, and they don't result in delivery failure notices but they aren't turning up in the qmail or qpsmtpd logs.
I don't understand where they are lost in the system.
-
how are you sending emails?
-
the system won't let me reply
try 3
but from external mailservers
-
My mail inwards is being collected by qmail but not being passed onwards by dovecot.
You are jumping to lots of conclusions; many of them wrong. Start by reporting your observations - what you see and what you don't see. Others will better be able to help you that way.
What do you see which makes you say that "mail inwards is being collected by qmail"? What makes you say "not being passed onwards by dovecot"?
Logging in and trying to start the dovecot service I get
# /etc/init.d/dovecot start
Starting Dovecot Imap: Error: Missing auth section for server default
Fatal: Invalid configuration in /etc/dovecot.conf
SME server does not use the "/etc/init.d/dovecot" script, and does not use dovecot for imap authentication. You can ignore that message, and fortunately your attempt to start dovecot via the /etc/init.d script won't have done anything bad.
Obviously this is a mission critical failure - but where are the correct template fragments so that I can check them and find out what has broken?
Looking at template fragments won't help you to work out what is broken. Start by describing what your problem symptoms are, then look for evidence of problems in log files. Are you sure that your problem is not at your ISP (e.g. filtering port 25)?
-
Thanks for a helpful suggestion.
Because I had not received a "undeliverable message" notice I thought that the message had reached the mail server on the SME box. Now I have learnt that Exim4 doesn't send these routinely, nor does gmail.
The final result was that the ppp dropout on the Friday was significant
(A) upstream provider to ISP loaded new router rules
(B) new router rules didn't put my ISPs IP address range in the 'allow' section
(C) somehow I was the first to find that I had no emails