[Date Prev] [Date Index] [Date Next] [Thread Prev] [Thread Index] [Thread Next]

conserver-8.0.6 is available

Bryan Stansell bryan@conserver.com
Sun, 16 Nov 2003 11:55:30 -0800 (PST)


Anyone running 8.0.0 through 8.0.5 should really consider upgrading to
8.0.6.  The reason?  NULL characters in console data will cause data
loss.  Unfortunate, but true.  The good thing is it doesn't happen very
often.

The really cool thing in this release is the calculation of port
numbers.  This way, you can specify the formula a terminal server or
serial card uses to access the various ports and then use those physical
port numberings (1 through n) in console definitions...and conserver
will do the math of what socket to access or what device to open.  The
'portbase', 'portinc', 'devicesubst' and 'execsubst' are the various
items that play a role.  I've given some examples (for terminal servers)
in conserver.cf/conserver.cf.

There are many other things added and improved.  The manpages will give
a little bit more detail on the specifics.  Here's the complete list of
changes:

version 8.0.6 (Nov 16, 2003):
	- code was missing regarding the 'setproctitle' option - patch
	  by Dmitry Morozovsky <marck@rinet.ru>
	- fixed the order of access list checking when
	  --with-trustrevdns is used
	- fixed various spelling errors - patches by Matt Selsky
	  <selsky@columbia.edu>
	- added console 'idletimeout' and 'idlestring' options to
	  trigger the sending of 'idlestring' after a lack of activity
	  for 'idletimeout' seconds - suggested by Ian Potts
	  <ian.potts@db.com>
	- added console 'portbase' and 'portinc' options to allow
	  specifying a formula for referencing ports - inspired by Todd
	  Stansell <todd@stansell.org>
	- fixed problem where console 'port' option didn't take port
	  names (manpage said it did)
	- added server -U option and configuration option 'unifiedlog'
	  to allow a copy of all console activity to go to a single file
	- removed all sprintf() calls
	- added 'devicesubst' and 'execsubst' console options for doing
	  replacements with calculated port numbers on the 'device' and
	  'exec' values
	- NULL characters in console data caused data loss because of
	  strlen() usage - reported by Toby Gerhart <toby.gerhart@eds.com>
	- fixed a couple errors in contrib/redhat-rpm/conserver.spec -
	  reported by Martin Evans <m.d.t.evans@qmul.ac.uk>
	- added capability to use '!' in user access lists to prevent a
	  user from having access - suggested by Matt Selsky
	  <selsky@columbia.edu>

Bryan Stansell