Re: [Exim] Sudden change in exim relay behavior

On Wednesday, February 5, 2003, at 09:56 AM, Giuliano Gavazzi wrote:

> At 18:23 +0530 2003/02/05, Suresh Ramasubramanian wrote:
>> On Wednesday, February 05, 2003 5:06 PM [GMT+0530=IST],
>> Mark Edwards <mark@???> wrote:
>>
>>> I updated to Exim 4.12 from Exim 4.10, and from Mailman 4.something
>>> to
>>> 4.1 today on my FreeBSD box.
>>>
>>> hostlist relay_from_hosts = 127.0.0.1 : localhost
>>>
>>> fixed the problem. Does anyone have an idea why all of a sudden
>>> Exim
>>> stopped recognizing 127.0.0.1 as localhost in this case? Yes, the
>>> two
>>
>> localhost also maps to ::1
>>
>
> do you mean that Mailman uses IPv6 to connect? Mmmm, even if it did I
> guess Mark is not listening on ::1 anyway and probably did not even
> compile IPv6 support in.
> So, in the above is true, wouldn't 127.0.0.1 be enough?

Okay, I'm starting to understand what's going on here. The FreeBSD
port for Exim started defaulting to include IPv6 support since I last
built it. And, unlike other processes that communicate with Exim on my
machine, Mailman apparently identifies itself as ::1 (IPv6) instead of
127.0.0.1 (IPv4). Thus, the setup broke.

Now, my only question would be what's the best way to cope with this?
Which is better:

hostlist relay_from_hosts = 127.0.0.1 : ::::1

or

hostlist relay_from_hosts = localhost

I'm guessing the former, as the latter would be easier to spoof?

Are there other things related to having inadvertantly enabled IPv6 in
Exim, that might bite me?

The only other place in my configure file that has an IP address is

ignore_target_hosts = 0.0.0.0 : 127.0.0.0/8

which is in the dnslookup router taken from the default configure file.
Should I alter that as well?

Thanks for the help!

--
Mark Edwards
San Francisco, CA