Postfix legacy releases 3.3.3, 3.2.8, 3.1.11, 3.0.15

2019-02-26 Thread Wietse Venema
[An on-line version of this announcement will be available at http://www.postfix.org/announcements/postfix-3.3.3.html] This announcement concerns fixes for problems that were introduced with Postfix 3.0 and later. This is the final update for Postfix 3.0. Fixed in Postfix 3.3 and later: * When

Re: randmap() lookup with percentage

2019-02-26 Thread Wietse Venema
If randmap could nest, then one could say randmap:{smtp1,randmap:{smtp1,randmap:{smtp1,randmap:{smtp1,smtp2 and approximate the desired ratio with 15/16. But it does not nest, because 'randmap' is like 'static' combined with multiple choice. That's enough humor for now. Wietse

Re: Ignore smtpd_recipient_restrictions list for one sender

2019-02-26 Thread Viktor Dukhovni
On Tue, Feb 26, 2019 at 02:45:52PM -0300, Emanuel wrote: > Hello, within the postfix configuration I have created a list of blocked > recipients. My question, it is possible to ignore this list for a one > sender.? > > smtpd_recipient_restrictions = > check_recipient_access > mysql:/etc/

Ignore smtpd_recipient_restrictions list for one sender

2019-02-26 Thread Emanuel
Hello, within the postfix configuration I have created a list of blocked recipients. My question, it is possible to ignore this list for a one sender.? smtpd_recipient_restrictions = check_recipient_access mysql:/etc/postfix/mysql-virtual-recipient-access.cf regexp:/etc/postfix/regex_destinat

Re: randmap() lookup with percentage

2019-02-26 Thread Viktor Dukhovni
On Tue, Feb 26, 2019 at 04:01:15PM +0100, Tom Sommer wrote: > > # Picks smtp1 90% of the time, smtp2 10% of the time. > > randmap:{smtp1,smtp1,smtp1,smtp1,smtp1,smtp1,smtp1,smtp1,smtp1,smtp2} > > True, that is a way - although a litte bit of a workaround; setting 93% > is a little bit of a mess :

Re: randmap() lookup with percentage

2019-02-26 Thread Wietse Venema
Wietse: > # Picks smtp1 90% of the time, smtp2 10% of the time. > randmap:{smtp1,smtp1,smtp1,smtp1,smtp1,smtp1,smtp1,smtp1,smtp1,smtp2} Tom Sommer: > True, that is a way - although a litte bit of a workaround; setting 93% > is a little bit of a mess :) I didn't think that such a high level of pr

Re: randmap() lookup with percentage

2019-02-26 Thread Tom Sommer
On 2019-02-26 15:51, Wietse Venema wrote: Tom Sommer: I have a feature request for a lookup table like randmap(), but with the option to control a percentage or a count of mails sent to the transport. Even if the percentage was just a "chance" of selection, it would be nice. So you could do :

Re: randmap() lookup with percentage

2019-02-26 Thread Wietse Venema
Tom Sommer: > I have a feature request for a lookup table like randmap(), but with the > option to control a percentage or a count of mails sent to the > transport. > > Even if the percentage was just a "chance" of selection, it would be > nice. > > So you could do : > > sender_dependent_defa

Re: how to use (open)dmarc when already doing before-queue content filtering?

2019-02-26 Thread Matus UHLAR - fantomas
On 24/02/19 08:47, Patrick Proniewski wrote: I'm also using postscreen, but it doesn't provide the same filtering as a regular greylist. And milter-greylist is nice enough to synchronise between multiple MX servers in a load balanced environment, which is what I'm using at work. On 26.02.19 08:

Re: how to use (open)dmarc when already doing before-queue content filtering?

2019-02-26 Thread Matus UHLAR - fantomas
On Feb 25, 2019, at 2:46 PM, Matus UHLAR - fantomas wrote: Milter is BQCF. It processes the mail during scanning, and afaik its functionality can replace both smtp_proxy and check_policy_service. And, you can do multiple milters. It just can't modify mail body, only headers, but you usually don

randmap() lookup with percentage

2019-02-26 Thread Tom Sommer
I have a feature request for a lookup table like randmap(), but with the option to control a percentage or a count of mails sent to the transport. Even if the percentage was just a "chance" of selection, it would be nice. So you could do : sender_dependent_default_transport_maps = randmap:{