DragonFly BSD
DragonFly users List (threaded) for 2005-05
[Date Prev][Date Next]  [Thread Prev][Thread Next]  [Date Index][Thread Index]

ramfs, ports, and cvsupd

From: "George Georgalis" <george@xxxxxxxxx>
Date: Mon, 16 May 2005 14:14:19 -0400


BSDCan was great, got lots of great ideas and direction...
but without a laptop, a lot of it was only committed to memory.

The first item I want to tackle is a good means to get all the cvsup
repositories and host my own cvsupd. My plan is to specify a ram disk
for the builds vs in the tree(s), so I can cvsup and install world and
run cvsupd from the same directory points.

In the past, I've nfs mounted distfiles, to save those downloads, but
getting ports from scratch from a mirror is still problematic. When I
tried copying ports from a local box, it would work but I generally had
to make clean before I tried to make; which is not optimal.

My thinking is to define a build directory in a ramfs, cvsup, and run
cvsupd (per the example config once I find it). That way I can install
packages or world locally, while hosting a cvsupd of the tree, for local
checkouts, at the same time, and still nfs the distfiles directory.

* how do I make the ram filesystem? mount_null(8) seems the doc, but I'm
not sure, I have 1GB ram and 2GB swap.
* How do I specify a directory separate from the sources for the builds
(world, ports, et al), to keep the source tree pristine for cvsupd?
* Where are the cvsupd config examples (ports/dfly src,doc,site...)?
* Is there a defined pattern I can exclude from the sources, to
get started with ports (eg rsync a local copy)?

Most importantly, am I going about this the right way? Reinventing the
wheel or missing important considerations?

// George

George Georgalis, systems architect, administrator Linux BSD IXOYE
http://galis.org/george/ cell:646-331-2027 mailto:george@xxxxxxxxx

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