>Number:         22169
>Category:       misc
>Synopsis:       Some variables in bsd.*.mk inconsistent in usage
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    misc-bug-people
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Thu Jul 17 22:34:00 UTC 2003
>Closed-Date:
>Last-Modified:
>Originator:     The Grey Wolf
>Release:        NetBSD 1.6U
>Organization:
	Star Wolf Innovations
>Environment:
System: NetBSD rivendell.starwolf.com 1.6U NetBSD 1.6U (RIVENDELL) #9: Wed Jul 2 21:32:34 PDT 2003 greywolf@rivendell.starwolf.com:/sys/arch/i386/compile/RIVENDELL i386
Architecture: i386
Machine: i386
>Description:
	Some variables in the various bsd.*.mk files are inconsistent in
	their usage; of note in this report is MANZ vs. MKMAN; if MANZ
	is defined at all, it is considered positive, where MKMAN may
	be specified with "yes" or "no".  MANZ is also not documented
	in BUILDING, though this could arguably be considered a separate
	issue.  MANZ should be specifiable with "yes" or "no"; since it
	goes by definition/undefinition, its default should probably be
	"no".

	There are doubtless other variables which fall into this category,
	some of which may be desirable to access by the more expert users.
	MANZ doesn't seem to require a degree in rocket science to
	understand...

>How-To-Repeat:
	* define MANZ=no in /etc/mk.conf
	* build a system
	* Note that a release is unbuildable because all the man pages are
	  compressed, and the compressed forms are not listed in the master
	  flist generated from the mtree.
>Fix:
	* Make MANZ (and other user-likely candiates a) yes/no variable(s).
	* Permit the mtree to be regenerated from a desired system, or
	  permit an alternate mtree file to be used to generate the flist.
	* update BUILDING to reflect some of these other variables, placing
	  them in a section heavily guarded by warnings if need be.
>Release-Note:
>Audit-Trail:
>Unformatted: