Newsgroups: news.software.nntp,news.software.b
Subject: ANNOUNCE: INN 1.4unoff3 release
Summary: 
Followup-To: news.software.b
Distribution: 
Organization: Mathematics Department, The Pennsylvania State University
Keywords: 
Cc: 

This is an unofficial release of INN, called "1.4unoff3".  This has
been released without the permission of the author, Rich $alz.
Rich is busily working on INN 1.5 along with his "real life"
work.  This release is just my way of saying "thanks" to Rich
for his effort, and to make his job easier without having to
answer more questions like "I noticed this bug in INN 1.4"
and "how can I apply this patch...".

This release is only a bugfix release over unoff2.  No new features
have been added.  Here's a list of changes:

	- fix to innxmit to stop core dumps on blank lines or corrpted
	  lines in batchfile. [From Joe Ramey <ramey@jello.csc.ti.com>]
	- fix to art.c to stop core dumps on null messageid's
	  [ From Joel Gallun <joel@mci.net> ] and to cast some int's
	  as unsigned char to avoid core dumps 
	  [ From KIZU takashi <kizu@ics.es.osaka-u.ac.jp>].
	- fix to post.c to fix nnrpd core dump with tracing on.
	  [ "Ronald F. Guilmette" <rfg@segfault.monkeys.com> ]
	- newnews.patch, Fix NEWNEWS command's handling of distributions.
	  [ dmtech@magi.com (Andrew Mileski) ]
	- install permission changes.  rnews is now 550 instead of 2555.
	  rnews isn't meant to be used by general users.  inndstart
	  installed setuid root, since rc.news expects it.
	- patch to nc.c and chan.c form Ian Dickinson (via INN FAQ)
	  to work around poll() bugs in Solaris 2.4.


I _still_ haven't integrated XBATCH, nor the ident stuff.

Again, this release is my fault, and thus you shouldn't bug
real author of INN with bugs you find here.

I have been running this release for several days.  I don't expect
there to be many problems, since the changes have been fairly minor.
The only known bugs are still the "DEBUG ERROR SITEspool: trashed"
messages.  This is apparently due to the innd.spool patch.  As
far as I can tell the message is "mostly harmless".  I have tracked
it down as far as WCHANflush() getting called with the handle
of a channel (which is a socket), except as the comment to the
function says it's only supposed to be used on file channels.