As a follow-up to January’s post http://weblogs.uni.edu/conklinc/2009/01/wakeonlan_and_windows_suspends.html
A couple of new things I learned today about WOL and Windows desktops:
- The “Only Allow management stations to wake the computer” GUI checkbox means that only Magic Packets from any machine will wake the computer in question.
- Windows XP and Vista machines with the “Allow this device to wake the computer” box checked (but NOT the “only allow management stations…” checked) will resume from sleep when any TCP traffic is directed at them. This makes it very convenient for a desktop machine to go to sleep and be awakened by a Remote Desktop Protocol TCP packet sent on 3389… there is a pause as the machine awkens, but not long enough to timeout the RDP client application.
- New machines with Intel AMT/vPro throw out all the old rules for magic packets and WOL. From what I can tell, with my Dell Optiplex 760, you have to use Intel (or SCCM, Altris, or similar) management software to wakeup or shutdown machines remotely using AMT’s encrypted methods. I haven’t found a way to “dumb down” my Opti760 to respond to magic packets, nor have I dove into the “new” management methods.