Problems making local postings to inn2

[Julien ÉLIE]

Hi Jeffery,

>  2: The Received: headers are converted to X-Rec##: because I wanted
>     inn2 to retain them.  Unfortunately, it still deletes them when
>     posting.  (Is there a way to control which headers inn2 will retain?)

INN retains X-Rec##: header fields.
Both inews and nnrpd keep them.

POST
340 Ok, recommended message-ID <oh6v1s$de$1 at news.trigofacile.com>
From: =?UTF-8?Q?Julien_=c3=89LIE?= <iulius at nom-de-mon-site.com.invalid>
Newsgroups: trigofacile.test
Subject: Header testing
Date: Tue, 6 Jun 2017 21:00:00 +0200
X-Rec1: from dymaxion.cjsa2.com (localhost [127.0.0.1])
  by dymaxion.cjsa2.com (8.15.2/8.15.2/Debian-8ubuntu1) with ESMTP id 
v54N92ck031389
  for <jeff at localhost>; Sun, 4 Jun 2017 16:09:02 -0700

Test!
.
240 Article received <oh6v1s$de$1 at news.trigofacile.com>
HEAD <oh6v1s$de$1 at news.trigofacile.com>
221 0 <oh6v1s$de$1 at news.trigofacile.com> head
Path: news.trigofacile.com!.POSTED.localhost.localdomain!not-for-mail
From: =?UTF-8?Q?Julien_=c3=89LIE?= <iulius at nom-de-mon-site.com.invalid>
Newsgroups: trigofacile.test
Subject: Header testing
Date: Tue, 6 Jun 2017 21:00:00 +0200
Message-ID: <oh6v1s$de$1 at news.trigofacile.com>
Injection-Date: Tue, 6 Jun 2017 19:18:22 -0000 (UTC)
Injection-Info: news.trigofacile.com; posting-account="<localhost>"; 
posting-host="localhost.localdomain:127.0.0.1";
	logging-data="430"; mail-complaints-to="abuse at trigofacile.com"
X-Rec1: from dymaxion.cjsa2.com (localhost [127.0.0.1])
  by dymaxion.cjsa2.com (8.15.2/8.15.2/Debian-8ubuntu1) with ESMTP id 
v54N92ck031389
  for <jeff at localhost>; Sun, 4 Jun 2017 16:09:02 -0700
Xref: news.trigofacile.com trigofacile.test:576
.

As you see, the X-Rec1: header field is still present.

I don't know what's happening in your case.



>  3: Note that the DKIM-Signature: header was deleted as it was too long
>     for inn2 and long headers were causing posting rejection.

The DKIM-Signature: header field is valid; inews should not reject it.
I've filled a bug report:
   https://inn.eyrie.org/trac/ticket/144



>  4: Header continuation lines are coming in with spaces at the beginning
>     of the lines.  All leading whitespace is converted to a <tab>.

Spaces are OK.



>  5: If missing, an Approved: and Lines: header are added to the end of
>     the header block.

Note that the Lines: header field is obsolete.  You could stop 
generating it.



>  6: Unless I'm blind, there are no empty headers or unusual characters
>     anywhere.

An empty header field is a header field whose value consists of only 
whitespace characters.  In your quoted article, the following header 
field is considered as empty:
X-Source:

-- 
Julien ÉLIE

« Avez-vous remarqué qu'à table les mets que l'on vous sert vous
   mettent les mots à la bouche ? » (Raymond Devos)