[rt-users] RT stop sending emails

[Swart, Tom]

I am able to send an email by using msmtp from the command line and I
can telnet into my smtp relay on port 25. I think it is a problem
between RT and msmtp. Here are some details of my configuration.

The full path for the msmtp.conf file is
/etc/request-tracker3.6/msmtp.conf and which msmtp returned
/usr/bin/msmtp.

Here is the contents of the msmtp script.

#!/bin/bash
/usr/bin/msmtp -t -C /etc/request-tracker3.6/msmtp.conf
/usr/bin/logger -t RTmailer -p syslog.info -- CALL /usr/bin/msmtp -nt
"$@" RETURNED $?

Here is the SendmailPath line in the SiteConfig file:
Set ($SendmailPath, "/etc/request-tracker3.6/msmtp");

Apache has been restart after editing SiteConfig.

-----Original Message-----
From: Joop [mailto:JoopvandeWege at mococo.nl] 
Sent: Friday, March 27, 2009 3:29 AM
To: Swart, Tom
Cc: rt-users at lists.bestpractical.com
Subject: Re: [rt-users] RT stop sending emails

Swart, Tom wrote:
> I have gone back to using msmtp and have an external SMTP relay
defined.
> In the /var/log/mail.log file the only current entries are for the
> results of my auto fetchmail cronjob. There is results from yesterday
> when it was working but nothing from today.
>   
What you could try is see if msmtp has a -v (verbose) flag and manually 
send an email to see what the mailservers are telling each other. If it 
hasn't this then you could resort to:
telnet mailserver 25
And by hand do a conversation with it.
I've been having problems lately with Centos4.7 and its sendmail and 
each time it turned out that the other end didn't like my
user at host.domain.
This being caused by the fact that host.domain didn't have a DNS entry 
in the wide world.
So using a couple of option within Sendmail (masqerading) fixed this.

You could have a look and see it you're not hit by this either.

Regards,

Joop