(private) HKS wrote:
I recently implemented "smtpd_sender_restrictions =
check_sender_mx_access cidr:/etc/postfix/sender_mx_access" on Postfix
2.3.3 running on Ubuntu 6.10.

$ cat /etc/postfix/sender_mx_access
0.0.0.0/8 REJECT MX in IANA reserved network
127.0.0.0/8 REJECT MX in loopback network
10.0.0.0/8 REJECT MX in non-routable network
169.254.0.0/16 REJECT MX in non-routable network
172.16.0.0/12 REJECT MX in non-routable network
192.168.0.0/16 REJECT MX in non-routable network
224.0.0.0/4 REJECT MX in multicast network
240.0.0.0/4 REJECT MX in IANA reserved network


Turning this on, however, led to some unexpected failures when email
was sent from my own domain:

Jul 30 11:43:34 mail.example.com postfix/smtpd[28463]: NOQUEUE:
reject: RCPT from server1.example.com[10.1.0.1]: 554 5.7.1 <
[EMAIL PROTECTED]>: Sender address rejected: MX in
loopback network; from=<[EMAIL PROTECTED]>
to=<[EMAIL PROTECTED]> proto=ESMTP helo=<server1.example.local>


A dig shows that the MX record for sub1.example.com is
mail.example.com (a public IP address).

Not here.

dig mx mail.example.com

; <<>> DiG 9.4.2 <<>> mx mail.example.com
;; global options:  printcmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NXDOMAIN, id: 36605
;; flags: qr rd ra; QUERY: 1, ANSWER: 0, AUTHORITY: 1, ADDITIONAL: 0

;; QUESTION SECTION:
;mail.example.com.              IN      MX

;; AUTHORITY SECTION:
example.com. 10800 IN SOA dns1.icann.org. hostmaster.icann.org. 2007051703 7200 3600 1209600 86400

;; Query time: 184 msec
;; SERVER: 127.0.0.1#53(127.0.0.1)
;; WHEN: Wed Jul 30 19:22:02 2008
;; MSG SIZE  rcvd: 95

but you may be living in parallel internet. if you manage to get back to our internet, we may be able to help you. In the meantime, say helo to the martians.


 How does postfix decide that
it's running on the loopback network, then?


postfix does not decide. you decide. you tell it when to reject mail, and it will. postfix does not invent DNS records.

Reply via email to