Problems making local postings to inn2

[Jeffery Small]

Russ Allbery <eagle at eyrie.org> writes:

Russ, thanks for the reply:

>Are you perhaps sending just CR and not CRLF at the end of lines?  That's
>the first thing that comes to mind here.

Well, this is a Linux system and I am filtering the email messages through
a perl script where I chomp the end of the lines and then print them all
out with just a newline.  However, I'm doing this for every message received
and most of them post without problem, so I don't see how this could be
the problem.

>Is there any chance you could include the full headers of one of these
>rejected messages?

I'll monitor for one and post it soon.  It's never clear which email relates
to which error message unless I'm monitoring closely.

Is there possibly some way to make inn2 less restrictive and go ahead and
post even if certain errors are detected?  The code snipet below doesn't
show any method for this, but maybe elsewhere?  (Unlikely, I know!)

Regards,
--
Jeff


>Jeffery Small <jeff at cjsa.com> writes:

>> These messages get left in my inbox and I have investigated the headers of
>> these failed posts in great detail.  I can find no instances of:

>>     * A header that does not have a "Name:<space>" format
>>     * A header with an empty field after the :<space>
>>     * Any odd characters: e.g., no non-ascii-7 characters

>> Everything appears normal, so I am at a loss to see why these fail.
>> Any suggestions as to where to look for the problem?

>Are you perhaps sending just CR and not CRLF at the end of lines?  That's
>the first thing that comes to mind here.

>Is there any chance you could include the full headers of one of these
>rejected messages?

>> Also, is there a way to get inn2 to generate better diagnostic output.
>> Despite all these (and other) problems, the log file /var/log/news/errlog
>> is always empty and the other news logs have little useful information
>> to help isolate a problem.

>Well, this isn't an error from the INN perspective -- rejecting malformed
>articles is just normal.  I think the rejection may be logged to
>news.notice by nnrpd, though.  But it's not going to contain any more
>information than the 411 error you already included, because that's all
>the information there is.

>        if (hp->Value) {
>            hp->Len = TrimSpaces(hp->Value);
>            /* If the header is empty, we just remove it.  We do not reject
>             * the article, contrary to what an injecting agent is supposed
>             * to do per Section 3.5 of RFC 5537.  (A revision to RFC 5537
>             * may someday allow again that existing and useful feature.) */
>            if (hp->Len == 0) {
>                hp->Value = hp->Body = NULL;
>            } else if (!IsValidHeaderBody(hp->Value)) {
>                snprintf(Error, sizeof(Error),
>                         "Invalid syntax encountered in %s: header field "
>                         "body (unexpected byte or empty content line)",
>                         hp->Name);
>                return Error;
>            }
>        }

>The check is really pretty simple and very low-level: there has to be at
>least one non-whitespace character in the field body, CR can't appear by
>itself, LF or CRLF has to be followed by whitespace or another header, and
>every character has to satisfy the C function isgraph() ("any printable
>character except space") or be a space or tab.

>-- 
>Russ Allbery (eagle at eyrie.org)              <http://www.eyrie.org/~eagle/>

>    Please send questions to the list rather than mailing me directly.
>     <http://www.eyrie.org/~eagle/faqs/questions.html> explains why.
>_______________________________________________
>inn-workers mailing list
>inn-workers at lists.isc.org
>https://lists.isc.org/mailman/listinfo/inn-workers