I like this idea. Please start a new thread. Once it's refined a bit we
can start the new wiki with it on github.
Will someone please start a new thread for this?
Thanks.
--
-Eric 'shubes'
On 03/12/2014 07:21 AM, Scot Needy wrote:
Lets make one!
Here are tid bit’s I have picked up over the past few weeks.
*Implementation*
* Host MUST be on a NAT or outside address where it can see the real
IP address of the last hop.
* Internet facing Reverse DNS on your hostname must work to prevet
getting blocked by remote rDNS filtering rules.
* (I removed the previous 2 because they don't really apply. QMT
> on COS6 with spamdyke are pretty much ready to go)
*Configuration*
* NOTE: you cannot have a trailing : on /var/qmail/control/simcontrol
"attach=.exe:.pif:.src:" = All mail even NO attachments
On Mar 12, 2014, at 9:32 AM, Jim Shupert <[email protected]
<mailto:[email protected]>> wrote:
On 3/11/2014 3:09 PM, Eric Shubert wrote:
On 03/11/2014 11:53 AM, Scot Needy wrote:
Yea,
I guess that was my original question when I asked about NAT.
Forgive me if I ask a dumb question but..
I thought that the source TCP address wasn’t an issue because much
of the spam prevention will look at mail headers not just the TCP
source IP of the last relay before mail got to qmail.
Depends on what you consider to be "much of the spam prevention".
There's not really much in the headers that spamdyke or spamassassin
relies on. Source IP address is probably the biggest single factor
that spamdyke uses.
You really need to get QMT on the perimeter (or behind a simple NAT)
in order to have decent anti-spam effectiveness.
I do not wish to hiJack a thread ... but would like to say that a
comprehensive "best practice" suggestion list would be wonderful.
I am presently trying to 'knuckel down' on the spam problem.
thanks
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
<mailto:[email protected]>
For additional commands, e-mail: [email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
