Exim rejects recipient address on my domain

Question

I have a dedicated server (debian) on which I have installed Exim and Dovecot. Everything worked fine until around a month ago. I tried to reinstall and reconfigure exim but I keep having all the incoming emails rejected.


Outlook says:

A message that you sent could not be delivered to one or more of its recipients. This is a permanent error. The following address(es) failed:

[email protected] SMTP error from remote mail server after RCPT TO:: host mail.mydomain.com [94.76.##.##]: 550 relay not permitted

GMAIL:

Delivery to the following recipient failed permanently:

[email protected]

Technical details of permanent failure: 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: 550 550 relay not permitted (state 14).

On the server side, my rejectlog file shows:

2011-01-04 17:09:21 H=mail-qw0-f53.google.com [209.85.216.53] F=<####@gmail.com> rejected RCPT : relay not permitted

... and the mainlog file:

2011-01-04 17:00:01 1PaAEr-0007vN-DX <= root@ETC_MAILNAME U=root P=local S=869
2011-01-04 17:00:01 1PaAEr-0007vN-DX ** root@etc_mailname: Unrouteable address
2011-01-04 17:00:01 1PaAEr-0007vY-Kn Error while reading message with no usable sender address (R=1PaAEr-0007vN-DX): at least one malformed recipient address: root@ETC_MAILNAME - malformed address: _MAILNAME may not follow root@ETC
2011-01-04 17:00:01 1PaAEr-0007vN-DX Process failed (1) when writing error message to root@ETC_MAILNAME (frozen)
2011-01-04 17:09:21 no IP address found for host MAIN_RELAY_NETS (during SMTP connection from mail-qw0-f53.google.com [209.85.216.53])
2011-01-04 17:09:21 H=mail-qw0-f53.google.com [209.85.216.53] F=<####@gmail.com> rejected RCPT : relay not permitted

then after the message becomes frozen:

2011-01-04 17:28:44 1PaAEr-0007vN-DX Message is frozen

Thank you for your help, any idea/comment is welcomed as I am really running out of idea to fix this issue,
Nicolas.

Oh and the PHP mail() function does not do anything as well, would it be linked to? I think mail() uses sendmail from my php.ini.

Answer 1

Dunno excatly yet, but I have the same problem here since about 22hrs ago...

Maybe you should have a look at this http://forums.debian.net/viewtopic.php?f=11&t=58312 and that http://www.reddit.com/r/netsec/comments/en650/details_of_the_root_kit_that_got_installed_on_my/

Maybe you got rooted?

Answer 2

A very handy trick to see what exim is doing with an email address is to run:

exim -bt [email protected]

If it immediately uses the remote_smtp transport, then you need to tell exim what domains it's responsible for.

Answer 3

Looks as though you're using /etc/exim4/exim4.conf.template as your /etc/exim4/exim4.conf file, verbatim, which is broken.

You should look at /usr/share/doc/exim4-base/README.Debian.gz as this is all Debian-specific. Stock Exim from upstream does not have any macro pre-processing or split configurations or the like. Section 2.1.6.1 talks about how you use a single config file, which is what I think you've tried to do.

Try first mv /etc/exim4/exim4.conf /etc/exim4/exim4.conf-old and then running: update-exim4.conf. (You could also first check that update-exim4.conf.conf is up-to-date.).

Ultimately, you'll want exim -bV to show that the configuration file in use is /var/lib/exim4/config.autogenerated, which means that the mass of tangled definitions all comes into play.

Answer 4

Had the same issue. Here's what I did to troubleshoot.

See where it fails by running (change [email protected] to your destination)

exim -bt [email protected]

Check if your hosts are getting resolved properly by running these commands on the terminal.

# the following should show the short 'hostname'
hostname

# following should show the full domain name (without www or subdomains)
hostname --fqdn

# following should show the external IP (not 127.0.0.1)
hostname -i

If all is good, and if the block is still unroutable, force the exim configuration to route it. So edit /etc/exim/exim4.conf.template file, and find the block within .ifdef DCconfig_internet. Add this as the first rule. (Change example.com to your domain)

my_domain_routing:
  debug_print = "R: dnslookup for $local_part@$domain"
  driver = dnslookup
  domains = example.com
  transport = remote_smtp
  # ignore private rfc1918 and APIPA addresses
  ignore_target_hosts = 0.0.0.0 : 127.0.0.0/8 : 192.168.0.0/16 :\
                        172.16.0.0/12 : 10.0.0.0/8 : 169.254.0.0/16 :\
                        255.255.255.255
  no_more

Run the routing test again to see if it passes now.

Note about exim4.conf.template file. This can get regenerated if you run the config utility again. To avoid this, duplicate this change in the /etc/exim/conf.d/routing/ folder as well.

Reference: SMTP unroutable address Understanding exim configuration files https://stackoverflow.com/questions/17351634/unrouteable-address-with-exim4

Answer 5

I wonder if you have somehow mixed up debian-specific Exim configuration (they use a packager-friendly config that is not part of the standard Exim distribution) with non-debian-specific? It appears that your outbound mail is being automatically tagged with the sender domain ETC_MAILNAME. I'm guessing that's a macro which is supposed to be replaced with the contents of /etc/mailname, but somehow the macro-definition is missing, so you just get the literal string.

Does /etc/mailname exist on your server? If not, you might try creating it and putting your server's FQDN (or whatever you want your default domain to be) in it. Beyond that I don't know anything about debian's packaging so you might seek help on the debian-exim4 help list.