On Sat, Jul 18, 2020 at 8:58 PM John C Klensin <john-ietf(_at_)jck(_dot_)com>
--On Saturday, July 18, 2020 18:56 -0400 E Sam
I think the max size of the header name does need to be
addressed in RFC 5322. Seeing how email and Usenet can
sometimes be the "wild west" I have seen really long headers
and developing mail software the question of header max limits
have also come into my mind.
Out of curiosity and just to be precise, are you talking about
the header field name or the field value?
I was originally talking about the header field name, but I think the
field value could use some addressing. I once saw a spam email in my
inbox which 95 percent of the actual email was an extremely long
All the big email providers like Gmail and new email providers with
considerable leverage like hey.com and Protonmail keep their header
value sizes to reasonable amounts so I don't think its as important as
the header field.
The problem with header value limits is the length of some headers are
deterministic (like the Path header in Usenet, X-Beenthere in mailing
lists, and to an extent dkim signatures).
I think a header value limit for important headers stored in overview
systems (From:, To:, Newsgroups:, Bcc:, etc...) could be a good
idea... this could be difficult but it is something that we could
consider - I don't know if there is already something out there that
already sets the rules for this and I don't know about it. SMTP
software developers - anyone from EXIM, Sendmail, OpenSMTPD, or
Postfix on this list can elaborate if they want in my reply.
Time zones, etc., permitting, please try to participate. Much
of the BOF is precisely about deciding what is in or out of
I'll try my best. Would I see the meeting link here or will I see it
after I purchase my registration to IETF 108?
ietf-smtp mailing list