Wietse Venema wrote in
 <[email protected]>:
 |natan:
 |> Hi
 |> I have one specific question
 |> 
 |> in main.cf i have:
 |> ...
 |> smtpd_hard_error_limit = 5
 |> smtpd_soft_error_limit = 2
 |> ...
 |> 
 |> It is possible to change number *_error_limit for one IP ?
 |
 |This is not per client configurable. I suggest that you use the Postfix \
 |defaults.
 |
 |$ postconf -d smtpd_hard_error_limit smtpd_soft_error_limit
 |smtpd_hard_error_limit = ${stress?{1}:{20}}
 |smtpd_soft_error_limit = 10

It would be interesting to know how these numbers "come".
For my very restricted private use case (no relay but for local
users, except occasional relay of mailing-list posts) i am driving
very well with

  smtpd_hard_error_limit = 2
  smtpd_soft_error_limit = 1

I do not use any policy etc. servers except for my own s-postgray
that i announced on this list (yet -- i like DKIM), and ever since
that offers [--]focus-sender mode only one error can be generated
by this policy service per message.

I think what i want to say is that i was totally surprised to see
that a single message can create numerous errors.  A single
connection, yes, but more than one per MAIL FROM:<> i would not
have expected.  Surely my problem!

(By the way, i use reject_unauth_pipelining in client_restrictions
as well as in data_restrictions, which i think is still a bug in
my configuration?)

So except for the fact that policy servers may create multiple
errors, my question would be how any number other than one for an
error limit can spring into existence?  I mean -- these are client
_errors_, which to me either means telnet(1) mistyping or
something malicious?

--steffen
|
|Der Kragenbaer,                The moon bear,
|der holt sich munter           he cheerfully and one by one
|einen nach dem anderen runter  wa.ks himself off
|(By Robert Gernhardt)

Reply via email to