On 05-29-2021 1:16 pm, Shawn Heisey wrote:
Adding to the reply from Wietse, which I have to agree with:
On 29.05.21 16:40, [email protected] wrote:
I was just assuming that a connection happens first before postfix
could know if the PTR resolves or not.
Otherwise how does postfix know a client hostname needs to be checked
if it has never connected?
maybe you could try using postscreen:
http://www.postfix.org/POSTSCREEN_README.html
which will start processing incoming connection before resolving DNS record,
it's great for stopping spam/bots too.
So following that logic, connection happens, log connection, resolve
PTR, log the warning.
I did not know that postfix can first know if someone has a valid PTR
before they connect. (And still don't understand how)
Now for my purposes in parsing logs, it would have been more
convenient for my mouse trap to rely on a new event ALWAYS starting
with a connect message since those always happen where as a warning is
only sometime.
more than one mail can be delivered on single connection.
Since it can't be convenient, i have to look for a
warning or connect for a new event then treat the connect message
different based if there was a previous warning or not.
I think guys here mentioned a script that parsed and reorders matching log
lines,
unfortunately I forgot its name...
--
Matus UHLAR - fantomas, [email protected] ; http://www.fantomas.sk/
Warning: I wish NOT to receive e-mail advertising to this address.
Varovanie: na tuto adresu chcem NEDOSTAVAT akukolvek reklamnu postu.
Save the whales. Collect the whole set.
