Date: Wed, 15 Aug 2007 20:12:21 +0200 From: Nikola Lecic <nlecic@EUnet.yu> To: Christopher Key <cjk32@cam.ac.uk> Cc: Rakhesh Sasidharan <rakhesh@rakhesh.com>, freebsd-questions@freebsd.org Subject: Re: portupgrade question Message-ID: <200708151824.l7FIORUE022485@smtpclu-5.eunet.yu> In-Reply-To: <46C2EF03.4040102@cam.ac.uk> References: <46C20CB8.3010706@cam.ac.uk> <200708142245.l7EMjQ8o027148@smtpclu-2.EUnet.yu> <20070815083210.M54184@obelix.home.rakhesh.com> <200708150810.l7F8AJEv032092@smtpclu-2.EUnet.yu> <46C2EF03.4040102@cam.ac.uk>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, 15 Aug 2007 13:18:11 +0100 Christopher Key <cjk32@cam.ac.uk> wrote: [...] > I've gone for a portconf based solution for now, although, when I get=20 > the chance, I'll try to test how portupgrade behaves wrt > dependencies. Please don't forget to try switching to portupgrade-devel in that case. According to /usr/ports/UPDATING recommendations (20070519): # portupgrade -f -o ports-mgmt/portupgrade-devel portupgrade # ... (etc., do the rest in order to rebuild /var/db/ bits and to build or fetch the new INDEX.) If dependencies bug ever really existed (I can't remember that anyone confirmed that offering an example), it was related to non-devel version. > I would prefer to use pkgtools.conf for several reasons: [...] > 2) MAKE_ARGS get echoed when things are being built, whereas > arguments in make.conf don't seem to Agree here, this is sometimes the only 100% sure method to know how the binary in question was really built. Nikola Le=C4=8Di=C4=87
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200708151824.l7FIORUE022485>