I am seeing this SPF softfail for emails sent from my server, and I notice that 2a01:4f8:121:2062::2 is an IPV6 address. When I check 2a01:4f8:121:2062::2 it is translated as 0.0.0.2 ipv4, so that ipv6 address seems bogus.
I have a Centos server that does not have ipv6 configured, and I left everything at default.
How can I solve this? Should I disable ipv6 on my server? Is postfix responsible for this? Why postfix sends mail with ipv6 when I never configured it?
As requested (I redacted my valid ipv4 address with XX)
[root@host ~]# ifconfig -a
eth0 Link encap:Ethernet HWaddr XX:XX:XX:2B:97:82
inet addr:XX.XXX.XXX.XX Bcast:XX.XXX.XXX.XX Mask:255.255.255.255
inet6 addr: 2a01:4f8:121:2062::2/64 Scope:Global
inet6 addr: fe80::4261:86ff:fe2b:9782/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:5906959065 errors:0 dropped:0 overruns:0 frame:0
TX packets:3933424961 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:6766726597523 (6.1 TiB) TX bytes:1315274953503 (1.1 TiB)
lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
inet6 addr: ::1/128 Scope:Host
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:34502551 errors:0 dropped:0 overruns:0 frame:0
TX packets:34502551 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:37797695084 (35.2 GiB) TX bytes:37797695084 (35.2 GiB)`
We've established that your server is, in fact, configured for ipv6, and has a perfectly globally-valid address. I think that disposes of the original question.
If you didn't know you were running v6, you should probably check your
ip6tables
rules, since it might be participating in all sorts of conversations you neither know about nor want! In case further proof was wanted (I have v6 to my desktop):If you'd prefer not to be running v6, you can disable it with
in
/etc/sysconfig/network-scripts/ifcfg-eth0
, though personally I think v6 is the coming wave and you've got a great opportunity to play with it!