Re: opensmtpd-filter-dnsbl

[Martijn van Duren]

On Sun, 2022-02-20 at 14:33 +, Pete Long wrote:
> Hi all,
> 
> I hope this is ok but I just wanted to say thanks to the developer of 
> 'opensmtpd-filter-dnsbl' whom I believe is Martijn van Duren. This filter is 
> superb and just what I was looking for in order to use the great resource 
> that is Spamhaus.
> 
> I can't seem to find a personal email address on the following site so that's 
> why I'm saying thanks on the list: 
> https://openports.pl/path/mail/opensmtpd-filters/dnsbl
> 
> Here's how I've used it in my smtpd.conf on OpenBSD 7.0:
> 
> filter dnsbl proc-exec "filter-dnsbl -mv zen.spamhaus.org"
> 
> listen on em0 tls pki mydomain.tld \
> hostname mydomain.tld filter { check_rdns, dnsbl }
> 
> I will shortly remove the -m flag which just adds headers rather than 
> rejecting the mail.
> 
> 
> Pete.

You're welcome.
And there's nothing wrong with using -m if that suits your needs, it's
designed with maildir's junk flag in mind.

martijn@

Re: dkimsign uses unexpected domain in signature on message from Ubuntu/Postfix relay client

[Martijn van Duren]

On Fri, 2022-01-28 at 00:09 +0100, Martijn van Duren wrote:
> On Mon, 2022-01-24 at 16:20 +0100, Tim van der Molen wrote:
> > Hi,
> > 
> > Martijn van Duren (2022-01-23 20:13 +0100):
> > > >  From: r...@relayclient.example.com (Cron Daemon)
> > > 
> > > According to RFC5322 section 3.4[0] this is not a valid e-mail format.
> > 
> > Just to point out this actually is valid. Text in parentheses is a
> > comment. See RFC 5322 section 3.2.2. Also this quote from section 3.4
> > which describes the format used above:
> > 
> >   Note: Some legacy implementations used the simple form where the
> >   addr-spec appears without the angle brackets, but included the
> >   name of the recipient in parentheses as a comment following the
> >   addr-spec.  Since the meaning of the information in a comment is
> >   unspecified, implementations SHOULD use the full name-addr form of
> >   the mailbox, instead of the legacy form, to specify the display
> >   name associated with a mailbox.  Also, because some legacy
> >   implementations interpret the comment, comments generally SHOULD
> >   NOT be used in address fields to avoid confusing such
> >   implementations.
> > 
> > Best,
> > Tim
> 
> Thanks for pointing this one out, it made me take a closer look at the
> spec. So when I originally responded I was only thinking in terms of
> what characters are used in a domain name, but the " (Cron Daemon)" part
> is not to be interpreted as domain characters, but a CFWS (or comment
> folding whitespace). This means that it should not be returned by
> osmtpd_mheader_from_domain() as part of the domain. So it is valid
> syntax, but I shouldn't have returned it when comparing against the
> known domain list. Similarly there were also a couple of FWS that I
> could ignore.
> 
> As for Paul's remark "SHOULD NOT" when it comes to this syntax: It's
> part of the current syntax (e.g. not obsolete), so I don't see any
> reason not to. Sorry for the confusion.
> 
> I have the following changes lined up in my repo[0]:
> - Fix a couple of memory leaks in error paths (pointed out by
>   Peter)
> - Add support for -D file, where file contains one domain per line. All
>   other rules from -d apply. (requested by Mischa and Renaud)
> - Fix FWS and CFWS issues when parsing a domain (pointed out by
>   Paul/Tim)
> 
> If people could help me test the latest code (or even check the diffs
> of revision 75-HEAD) that would help prepare for a new release.
> 
> martijn@
> 
> [0] http://imperialat.at/dev/filter-dkimsign/

Did anyone test this yet? I would like to have other people test this
before creating a new release.