dovecot mailing list (this mailing list), DKIM, SPF and DMARC
justina colmena ~biz
justina at colmena.biz
Fri Oct 21 22:01:11 UTC 2022
Trojitá, a fast Qt IMAP e-mail client
http://www.trojita.flaska.net/
I also use
http://opendkim.org/
http://www.trusteddomain.org/opendmarc/
as milters on Postfix
Active development, I'm sure they could all use some help, or forks for
alternatives, I don't know, I'm not involved in development per se, just a
user, and I have to get off the property of any of these places with my
code before anything happens. All that Finnish osalliyhdistys and by the
time a Swede gets online all hell breaks loose./
On Friday, October 21, 2022 1:50:43 PM AKDT, hi at zakaria.website wrote:
> On 2022-10-11 14:05, Benny Pedersen wrote:
>> hi at zakaria.website skrev den 2022-10-11 13:42: ...
>
> Indeed, it's because you set the following headers in dkim signing headers:-
>
> from : subject :
> date : to : message-id
>
> Although not sure why you've added some space, as per standards
> I think only colon separated list its the compliant format like
> the following:-
>
> from:subject:date:to:message-id
>
> Anyhow this is my final update, the previous headers set which
> I included wasnt perfect as cc header was causing a trouble,
> given it can fail at some point e.g. when replying more than one
> time to the same recipient through a mailing list, and mind me
> OX and iRedMail, I had to check your signing headers set,
> hopefully you are ok for me to present it here as the optimal
> one to avoid DKIM failures:-
>
> OX:-
> Date:From:To:In-Reply-To:References:Subject:From
>
> IRM:-
> x-mailer:message-id:in-reply-to:to:references:date:subject
> :mime-version:content-transfer-encoding:content-type:from
>
> iRedMail seems to be the best headers set given it includes
> X-Mailer header, which enhances signature validity, when client
> uses specific mail client app, although it can be faked yet one
> must know which client app the sender would use and if was able
> to have information to this length I guess signature validity
> would be an easy task to break it further.
>
> Also, I was advised by a friend to duplicate the signing
> headers in order to disallow spoofing signature further, while I
> couldnt see how nor populate a proof of concept, I removed it
> but if someone understand it, I would appreciate their
> elaboration, surely with thanks :)
>
> Good luck.
>
> Zakaria.
>
>
More information about the dovecot
mailing list