Historical functional question - rejecting messages that have blankheaders
julien at trigofacile.com
Tue Feb 2 18:32:39 UTC 2010
> OLD/news.notice.2.gz:Jan 31 22:42:08 jik3 rnews: rejected 437 Body of header
> is all blanks in "Keywords" header
> Am I correct that INN 2.5.0 no longer rejects messages with blank headers,
> and that this was an intentional change?
Yes, you're correct.
That check is no longer done by innd, as a relaying and serving agent.
The change was intentional for INN 2.5.0.
> Personally, I think this is a correct change, because while blank headers
> may be rude, I'm pretty sure the NNTP RFCs don't prohibit them, which means
> that INN shouldn't be rejecting messages containing them.
See Section 2.2 of RFC 5536:
o Every line of a header field body (including the first and any
that are subsequently folded) MUST contain at least one non-
NOTE: This means that no header field body defined by or
referenced by this document can be empty.
What should be done is reject such messages posted via nnrpd (injecting
It means that rnews (normally using innd) will be able to post these
messages but inews (normally using nnrpd) will not.
« -- Il t'arrive une tuile ?
-- Oui, je ne peux pas payer mon ardoise. »
More information about the inn-workers