Developing and maintaining a rapidly deployable image of an installed system

[Polytropon]

On Thu, 07 May 2009 12:10:42 +0800, Warren Guy <warren.guy at calorieking.com> wrote:
> I'm just wondering if there is an established best practice for
> developing and maintaining a rapidly deployable image of an installed
> FreeBSD system?

One of the common (at least I think so) methods is using
the system's standard tools dump and restore. You create
the installation as you need it on a separate machine or
in a jailed environment, and you end up with partitions
as you want them to have on the machines for deployment.
Then you dump these partitions into files. On the (fresh)
machines, you boot into a minimal FreeBSD system that
allows you to read the dump files, from optical media,
tape, or across the network. First you slice, partition
and newfs the partitions (can easily be scripted if you
know what you want), then you restore the partitions (as
on the machine used for preparing them) from the dump
files.

That's for deploying. For maintaining... it's possible
to use a similar method where you only need to dump and
restore partitions where you did major changes. For minor
ones it should be sufficient to alter files "the usual
way" (can be scripted, too).



> If anyone can point me towards documentation or any other resources that
> might be of use I would greatly appreciate it.

There has been a discussion thread some days ago on this
list which covers a bit of this topic (deploying).



-- 
Polytropon
>From Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...