Date: Wed, 18 Sep 2013 12:48:34 +0200 From: =?utf-8?Q?Dag-Erling_Sm=C3=B8rgrav?= <des@des.no> To: Bryan Drewery <bdrewery@FreeBSD.org> Cc: svn-src-head@freebsd.org, svn-src-all@freebsd.org, src-committers@freebsd.org Subject: Re: svn commit: r255634 - head Message-ID: <86a9jabe2l.fsf@nine.des.no> In-Reply-To: <5238548D.9080503@FreeBSD.org> (Bryan Drewery's message of "Tue, 17 Sep 2013 08:09:33 -0500") References: <201309171259.r8HCxb3o032697@svn.freebsd.org> <5238548D.9080503@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Bryan Drewery <bdrewery@FreeBSD.org> writes: > Hm, I know you noticed this, but by making this conditional, it will not > check if building WITHOUT_UNBOUND, but the passwd and mtree files are > still using it and will blow up. Partly true: - 'make installworld' will succeed (except, I think, in the NO_ROOT case), because it doesn't run mtree. - 'make distributeworld' will fail, but that's a bug: it should use DB_FROM_SRC. - mergemaster will also fail, but in an exceedingly strange manner. The fact that neither 'make installworld' nor mergemaster create missing directories is probably a bug. We should, at the very least, document the need to run 'make hierarchy' before 'make installworld'. > All of these conditions should probably be removed until a better > mechanism for passwd/mtree can be thought up. Yep, but for now I decided to follow precedent. DES --=20 Dag-Erling Sm=C3=B8rgrav - des@des.no
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?86a9jabe2l.fsf>