Hi,
I am not sure where you see the problem? Is the problem that you have to
reboot the server?
If this is the case, you can deploy clusters where you reboot one server
while the other takes the load and takes care of any user requests. Then you
patch and reboot the other node.
Personally I don't have any problem with rebooting server once a month (this
is how often Microsoft will in general release patches).
My practice is to automatically install and reboot client computers, but to
only download patches on server. After the update has been evaluated that it
will not cause any problem it is installed on servers...
There are also quite a few tools that will allow you to remotely deploy and
control installation of patches on server. One tool that comes to mind is
Microsoft SMS (Microsoft System Management Server) or WSUS (Windows Server
Update Services) that will soon be released.
When was last time you updated your active network equipment (routers,
switches etc)? E.g. CISCO, IBM, Juniper, Symantec, 3COM, etc... all have
same problem that is described in MS05-19 for Microsoft. It is a critical
bug that could allow DoS against your network. Let me know how it went with
rebooting routers and switches.
http://news.com.com/2102-1002_3-5669...=st.util.print
--
Mike
Microsoft MVP - Windows Security
"Duse" <> wrote in message
news:...
> The recent rend for critical patches for win 2003 to require rebooting the
> server is USELESS!!!
>
> Win 2003 is a SEVER, it cannot be continually rebooted and offer any
> Server level to its users.
>
> Additionally the fact that only a partial installation occures ( which
> leaves the server in a unstable state) is also USELESS...
>
> FYI: server are NOT workstations and do not have people sitting at them to
> monitor and react to your auto updates and installs..
>
> Is this truely what you think improving security is....
>
> Unhappy..
>
>
>