How do security updates work, can they be automated / automatic ?

From: Dave Arbok (rocker1_at_shegolfs.com)
Date: 08/21/04


Date: 20 Aug 2004 22:15:14 -0700

We are thinking about using a Red Hat Enterprise rebuild (White Box or
Tao Linux) on a server. The main reason we want to do this is the 5
year security update lifecycle. I hear that some people get very long
system uptimes because Linux can apply most updates without rebooting.

Anyway, my question is, can the machine be set to get all security
updates automatically, is there a special utility for this or is it a
setting, how do you do it? Will a reboot ever be necessary? If a
reboot is necessary, what is the easiest and recommended way to get
all services running again, is it easy to set them all to run on
startup?

Further info 1: (might be needed to understand question)
The goal is for the machine to sit in a locked room for 5 years or
until its hard drive dies (maybe until both its RAID mirrored drives
die, depending on costs), whichever comes first, and maintain security
without being touched - no administration via the web / webmin either
once it is set up. For the sake of argument, imagine that the
administrator is retiring with no replacement.

Further info 2: (probably NOT needed to answer my question)
The DSL will come into either our Linksys DSL "router" or perhaps a
similar "firewall" (<$100), and then the server will have 2 network
cards and will be the only point of contact to the other 15 computers.
It will be used as a firewall, an internet sharing host, and as a sort
of mail temporary storage server (not one that gets unsolicited
connections from the outside world- it will poll our isp, virus scan,
and hold it until the local clients want it, then trash or encrypt it
after 2 weeks if it has been gotten by the local user).

Regards, D