postfix lost connection with 127.0.0.1

[Kolab Users]

Hi,

Each day we get a few of these messages pile up in the postfix mail queue:

# mailq
-Queue ID- --Size-- ----Arrival Time---- -Sender/Recipient-------
3AA2983A65     4682 Fri Mar 22 05:25:13  postmaster at host1.domain.com
(lost connection with 127.0.0.1[127.0.0.1] while sending end of data -- 
message may be sent more than once)
                                          first.last at domain.com.au

E40C283F8F     4095 Fri Mar 22 10:39:42  postmaster at host2.domain.com
(lost connection with 127.0.0.1[127.0.0.1] while sending end of data -- 
message may be sent more than once)
                                          first.last at domain.com.au

and then:

E40C283F8F     4095 Fri Mar 22 10:39:42  postmaster at host1.domain.com
(delivery temporarily suspended: lost connection with 
127.0.0.1[127.0.0.1] while sending end of data -- message may be sent 
more than once)
                                          first.last at domain.com.au


The above messages that are being sent are simply daily "logwatch" 
emails sent from our internal physical hosts. The hosts are email 
filtering hosts, so report on various bad/spam URL's etc within their 
content.

All other hosts with logwatch reports (that aren't email filtering 
hosts) have no trouble sending emails to the "first.last at domain.com.au" 
Kolab3 account, which is expected as they don't contain any "bad URL" 
content.

After extensive internet searches, the "lost connection" and "delivery 
temporarily suspended" errors from postfix are very generic errors with 
many and diverse root causes. In our case, we believe the root cause is 
specifically related to amavisd not being able to scan the emails 
because of the spam URL's contained in them.

We don't need amavisd on the Kolab3 server at all (all our scanning is 
done on other dedicated machines), and have tried disabling amavisd in 
/etc/postfix/main.cf by commenting out:

content_filter = smtp-amavis:[127.0.0.1]:10024

then restarting postfix, running "postfix flush" but to no avail, the 
"lost connection" messages happen immediately after attempting the flush.

The only thing we can do is delete them from the postfix queue, but the 
new emails get re-queued the very next day.

We've also tried whitelist our domain in amavisd by adding the line:

@whitelist_sender_maps = (['.domain.com']);

adjusting the line:

$mynetworks

to define our internal network, and restarting amavisd, then attempting 
a postfix flush. To no avail.

Policy bank:

$policy_bank{'MYNETS'} = {   # mail originating from @mynetworks
   originating => 1,  # is true in MYNETS by default, but let's make it 
explicit
   os_fingerprint_method => undef,  # don't query p0f for internal clients
};

is also set.

We really don't want amavisd on the Kolab server, it's unnecessary in 
our setup.

How can we:

 1. effectively disable amavsid from the Kolab3 server?

Thanks.

Michael.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.kolab.org/pipermail/users/attachments/20130322/b7484776/attachment.html>