Brian,
I don't want to sound contrarian, but there are a lot of organizations where
bouncing every member server and workstation monthly is not practical. Some
companies have strict change control policies that mandate patches of that
nature only be applied once a quarter, or at other intervals of scheduled
maintenance. The intervals in password change policies do not always line up
with these schedules.
On the other side of that, some people cannot simply tell their auditors,
"I'm sure the password must have gotten reset sometime in the last week
because we patched all our computers."
At some point they may need to show a log, or proof. What if some
workstations or servers were turned off for an extended period of time and
were not patched? Sometimes the patching process fails and computers are
not properly patched or rebooted. These hosts would not update their
passwords if you were relying on that scripted process... and the only way
to determine that would be to look at patch logs or some other type of log.
When you have a dependancy like this, if your patch process develops a
problem and does not patch each and every host, you may have a "finding" in
your audit for failure to patch. But, if you rely on patching to force
other processes such as this one, then you may multiply how many "findings"
you have in an audit. Again, this is not a strong argument that would apply
to everyone, but for some people it is a very important consideration.
There are some people out there that need to have a more precise change
mechanism, or more importantly, a more precise logging mechanism for
accountability.
Anyway, those are some of the considerations or "down sides" to using the
GPO/startup script method... in addition to what Joe mentioned. They
certainly are not an issue for everyone, but there are still a lot of people
that do need workarounds for these concerns.
--
Ken Aldrich
DSRAZOR for Windows
Visual Click Software, Inc.
www.visualclick.com