Bill, On 28/04/2019 20:37, Bill Cole via mailop wrote: > On 28 Apr 2019, at 13:05, Grant Taylor via mailop wrote: > >> On 4/27/19 11:43 PM, Bill Cole wrote: >>> I can't say "should" because that's a site-specific/sender-specific >>> choice. >> >> As is the choice to (over)sign headers, even non-existent headers; >> List-*, Sender, etc. > > Qualitatively different choices. > > Signing non-existent Sender, List-*, and Resent-* headers has no > positive effects outside of absurdly contrived cases and causes tangible > problems. Fixing that is zero-maintenance with no negative side-effects.
The question here is how to distinguish headers which absence should be treated by ignorance (e.g. `Sender`) from headers which absence should be oversigned, e.g. `Subject` or `Reply-To`. In fact, I was asked about somehow similar issue in Rspamd that also performs DKIM signing. So far, I'm not quite sure if there is a simple solution for this problem. > <skipped> _______________________________________________ mailop mailing list mailop@mailop.org https://chilli.nosignal.org/cgi-bin/mailman/listinfo/mailop
