Date: Tue, 26 Jun 2012 09:21:41 -0700 From: "Simon J. Gerraty" <sjg@juniper.net> To: Eric van Gyzen <eric@vangyzen.net> Cc: Tim Kientzle <tim@kientzle.com>, freebsd-arch@freebsd.org, sjg@juniper.net, =?ISO-8859-1?Q?Dag-Erling_Sm=F8rgrav?= <des@des.no> Subject: Re: Allow user install Message-ID: <20120626162141.8CE5058081@chaos.jnpr.net> In-Reply-To: <4FE9D84E.7080402@vangyzen.net> References: <20120626063017.D05DA58081@chaos.jnpr.net> <86wr2uwdgf.fsf@ds4.des.no> <C31B93F4-674C-4183-9F3F-5F7C48980204@kientzle.com> <4FE9D84E.7080402@vangyzen.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, 26 Jun 2012 10:42:06 -0500, Eric van Gyzen writes: >Tim's idea sounds great, and would cover several use-cases. >Specifically, it leaves the build artifacts in the usual places so >other, later builds can build against them, whereas writing the Yes, this is almost exactly what we are planning to do. >artifacts directly to a tar file does not. Building a manifest would >also be very handy, and even necessary for correct packaging of the >artifacts from an unprivileged build, where the on-disk meta data are >not "correct". Exactly. >I'm already doing something like this at $WORK, but not using >buildworld/installworld (to my dismay). I manually maintain an mtree FWIW we don't plan to use buildworld either ;-) But until we can offer a useful alternative, incremental improvements can be helpful. >file which gets fed to makefs to build an mfsroot. Although I like the >fascist control of this method, it's more work to maintain. Automation >is good. We've been using essentially this method for many years (10+), but using a hacked version of mkisofs - for freebsd we are re-doing it for makefs. The maintenance isn't that big a deal - and yes automation wins. --sjg
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20120626162141.8CE5058081>