On Fri, Jan 23, 2015, at 05:56 AM, Wietse Venema wrote: > rogt3...@proinbox.com: > > > > On Fri, Jan 23, 2015, at 03:41 AM, Wietse Venema wrote: > > > rogt3...@proinbox.com: > > > > Hi, > > > > > > > > I'm setting up a postscreen instance in a multi_instance setup. > > > > > > > > I've read the docs on the Deep Protocol Tests stage. The tests > > > > seem helpful but for course there's the fact that on a PASS of > > > > good-to-this-point mail, it "waits for the client to disconnect. > > > > The next time the client connects it will be allowed to talk to a > > > > Postfix SMTP server process to deliver its mail." > > > > > > If you are worried about delays, then do not turn on the "deep > > > protocol" tests. > > > > As documented, if Deep Protocol tests are used there'll be a delay. > > > > That's not the question I asked. > > You have no control over how often remote senders retry deliveries, > nor do you control whether they will retry from the same IP address. > Some providers will retry from a random IP address in a large pool, > and those will never get past postscreen's deep protocol tests.
Not what I asked > When you ask "what is the best way to shoot myself into the foot", > then my answer is "don't do that", even if you did not ask for that > answer. I see. So one lesson I learn is that when the docs say "Wietse enables "deep protocol tests" on his own internet-facing mail server. " I'll jjust ignore what you do or suggest as self-puffery. Less arrogant attitude that way too.
