Not receiving emails into Webmin

14 posts / 0 new
Last post
#1 Fri, 06/17/2011 - 23:26
rjbathgate

Not receiving emails into Webmin

Hi,

I am really lost.

I have a few virtual servers set up, and working fine, expect incoming mail.

I can send mail from the Webmin > Servers > Read User Mail > Compose -- it is sent fine.

However nothing is ever received, if I reply to the sent email, so back to the sender, and then back into Read User Mail, nothing at all, for any of my Virtual Servers.

I also cannot connect to imap/pop3, but surely I should at least be receiving these emails into Webmin > Read User Mail?

Thanks

Fri, 06/17/2011 - 23:30
andreychek

Howdy,

You may want to try restarting dovecot, by running "/etc/init.d/dovecot restart" on the command line as root.

If it still doesn't work after that -- send yourself an email, then take a peek in the email logs, /var/log/mail.log or /var/log/maillog (depending on your distro). That would contain some info about what's going on, and should help you track down the issue.

-Eric

Sat, 06/18/2011 - 00:44
rjbathgate

Thanks for the reply.

Restarted dovecot, no change.

So, sent myself an email internally (sent it FROM the Read User Mail >> Compose function) straight back to the same address. It is received back into the Read User Mail.

However, sending to the address from EXTERNALLY does not work. But I can send from INTERNAL out and that works fine.

The log file includes success messages for the internal send, but nothing at all for the external sending in.

Help!!

Thanks

Sat, 06/18/2011 - 00:47
rjbathgate

Note,

this is the same server as one I messaged about seperatly in this same forum, which is a result of incorrect IP addresses on the DNS, which I am trying to (not very successfully) fix.

The IP address for the MX record is set as the IP address for the SHARED IP of the VPS. However this domain sits on its own PRIVATE IP

So, would the fact the MX IP is NOT the same as the Private IP of the account be an issue? Even though the MX IP is all within the VPS?

Thanks

Re the IP/DNS - I requested a DNS zone update with my provider but it's not happened (and being the weekend cannot get support). I did have BIND running on VPS, which I have since disabled thinking that was holding onto the DNS, thus not letter my provider change, but DNS is still unupdated....

Sat, 06/18/2011 - 08:47
andreychek

So, would the fact the MX IP is NOT the same as the Private IP of the account be an issue? Even though the MX IP is all within the VPS?

Well, the only thing that should matter is that the IP address of the MX record needs to be publicly accessible, and you'll want to make sure that port 25 is visible from the outside world (many ISP's block port 25).

One thing you can try is to go to intodns.com, and have it run a DNS report on your domain... that could help you figure out what's going on.

-Eric

Sat, 06/18/2011 - 14:27
rjbathgate

Hey

Thanks for the reply.

http://www.intodns.com/newzealand-motorhomes.com

Doesn't seem to return any errors.

Re port 25 -- the VPS is physically hosted not by me, but a hosting company, so I presume the issue of ISP blocking port 25 is with regard to their end, and nothing to do with my personal ISP.

And since they are a hosting provider, one would presume there is no port 25 issue? I'm running free (GPL) version by the way if that makes any difference.

Thanks

Sat, 06/18/2011 - 15:04
andreychek

Howdy,

Well, when I try to telnet to port 25 on your server, I get a connection refused message.

That suggests to me that either there's a Postfix error that's preventing it from starting properly, or that your provider is blocking port 25.

Let's start with the Postfix error option -- my suggestion there would be to log into your server as root, and restart Postfix by running this command:

/etc/init.d/postfix restart

After doing that -- take a look in the mail logs, do you see any errors or warnings?

-Eric

Sat, 06/18/2011 - 15:14
rjbathgate

Hi,

After restart, the log says this:

Jun 19 08:12:19 vps dovecot: pop3-login: Aborted login (tried to use disabled plaintext auth): rip=208.115.212.119, lip=119.47.122.53 Jun 19 08:12:21 vps last message repeated 3 times Jun 19 08:12:21 vps dovecot: pop3-login: Aborted login (tried to use disabled plaintext auth): rip=208.115.212.119, lip=119.47.122.72 Jun 19 08:12:22 vps dovecot: pop3-login: Aborted login (tried to use disabled plaintext auth): rip=208.115.212.119, lip=119.47.122.53 Jun 19 08:12:22 vps dovecot: pop3-login: Aborted login (tried to use disabled plaintext auth): rip=208.115.212.119, lip=119.47.122.72 Jun 19 08:12:25 vps last message repeated 4 times Jun 19 08:12:26 vps dovecot: pop3-login: Aborted login (tried to use disabled plaintext auth): rip=208.115.212.119, lip=119.47.122.53 Jun 19 08:12:26 vps dovecot: pop3-login: Aborted login (tried to use disabled plaintext auth): rip=208.115.212.119, lip=119.47.122.72 Jun 19 08:12:26 vps dovecot: pop3-login: Aborted login (tried to use disabled plaintext auth): rip=208.115.212.119, lip=119.47.122.53

Plus lots of very very similar lines....

Humz

Thanks for your continuing assistance by the way :)

Sat, 06/18/2011 - 15:20
andreychek

Hmm, well, I only see Dovecot messages in that output, nothing about Postfix. When restarting Postfix, you should see something like this:

Jun 19 00:18:29 test postfix/master[22409]: terminating on signal 15
Jun 19 00:18:30 test postfix/master[22493]: daemon started -- version 2.5.5, configuration /etc/postfix

Also, what output do you receive if you run this command:

grep disable_plaintext_auth /etc/dovecot/dovecot.conf

Sat, 06/18/2011 - 15:24
rjbathgate

Hi,

Output for that command:

disable_plaintext_auth = yes for authentication checks). disable_plaintext_auth is also ignored for

# NOTE: See also disable_plaintext_auth setting.

And log returns more of the same message as before: Jun 19 08:22:29 vps dovecot: pop3-login: Aborted login (tried to use disabled plaintext auth): rip=208.115.212.119, lip=119.47.122.53 Jun 19 08:22:39 vps last message repeated 14 times Jun 19 08:22:39 vps dovecot: pop3-login: Aborted login (tried to use disabled plaintext auth): rip=208.115.212.119, lip=119.47.122.72 Jun 19 08:22:39 vps dovecot: pop3-login: Aborted login (tried to use disabled plaintext auth): rip=208.115.212.119, lip=119.47.122.53 Jun 19 08:22:40 vps dovecot: pop3-login: Aborted login (tried to use disabled plaintext auth): rip=208.115.212.119, lip=119.47.122.72 Jun 19 08:22:40 vps dovecot: pop3-login: Aborted login (tried to use disabled plaintext auth): rip=208.115.212.119, lip=119.47.122.53 Jun 19 08:22:41 vps dovecot: pop3-login: Aborted login (tried to use disabled plaintext auth): rip=208.115.212.119, lip=119.47.122.72

Thanks

Sat, 06/18/2011 - 16:13
rjbathgate

Hi,

After 24 hours I have a delivery failure report from external mail service (gmail) to the domain in question:

Google tried to deliver your message, but it was rejected by the recipient domain. We recommend contacting the other email provider for further information about the cause of this error. The error that the other server returned was: 553 553 5.3.0 xxxxxx@yyyyy.com... User unknown (state 17).

Unknown user.... humz, the user is indeed the main root account for this domain...

(ps i have tried other user accounts too, but no errors reports yet (not been 24 hours on those attempts).

Sat, 06/18/2011 - 18:29
andreychek

Hmm, what distro/version are you using? And how did you perform your installation, did you by chance use the install.sh?

On a typical installation, it would explicitly set "disable_plaintext_auth" to "no". That's a setting I'd recommend adding to your dovecot.conf, the line should look like this:

disable_plaintext_auth = yes

As for your issue receiving email --

Google tried to deliver your message, but it was rejected by the recipient domain.

Well, I've tried from 3 different ISP's, and each time, I get a "Connection Refused" message.

So, I suspect Gmail is running into the same issue. Something is preventing outside systems from accessing port 25 on your server.

Let's try one more thing... what do you get if you run this command:

netstat -an | grep :25

That should show where Postfix is listening -- we want to verify that it's listening on a public IP address. If it is, then something along the way is blocking the connection.

-Eric

Sun, 06/19/2011 - 21:47
rjbathgate

Hi,

Thanks for your continuing help.

Following that command, I get:

tcp        0      0 127.0.0.1:25            0.0.0.0:*               LISTEN

Doesn't look very public to me...

I have also had a few other email bounce backs since, with 'connection refused' errors too.

Versions: Virtualmin version 3.82.gpl GPL Webmin version 1.530

I have since changed the DNS for this particular domain so I can get it working externally, but another (same) domain in the same situation (if required for testing) is speak-up.co.nz

Thanks

Sun, 06/19/2011 - 21:53
andreychek

So did your VPS provider perform your OS installation for you? It's possible that their OS image has external SMTP access disabled by default.

You may need to edit your /etc/postfix/main.cf file, and check your "inet_interfaces" parameter.

A typical setup would have it set like this:

inet_interfaces = all

After changing that, restart Postfix, and then see if it begins working for you after that.

-Eric

Topic locked