I am having a problem with Windows 2003 Server reactivating itself although there have been no significant hardware changes. Here is some background on our problem.
We are using a VME-based Single Board Computer with the following setup. There are two Ethernet ports on the board. We are using port 0 as a static IP for an internal network, and port 1 for a network with access to the internet for product activation, updates, etc. When we add software to the hard disks we use a SCSI CD-Rom which is not present in the system during normal operation.
Recently, I have seen more than a few of these systems requiring Windows reactivation and this made me wonder about the licensing process. I know that there is a hardware hash created by 10 hardware characteristics and that it is a tolerance based on a voting mechanism and all this makes me wonder if the nature of my development environment has created a system that is vulnerable to the windows reactivation process.
If I understand the process correctly, there needs to be 7 or more matching points to keep the activation current. In my case since the CD-rom is not present in the system that is in the field, that is one change in the hardware hash. To avoid confusion, I disable Ethernet port 1 (the one with access to the internet used to activate windows) which, if I understand the process, counts for 3 in the voting scheme used to determine activation status (making the total of my hardware changed 4 and putting me over the threshold for activation, thus requiring reactivation).
My question is two-fold.
1. Is my understanding of the activation process accurate?
2. Assuming my thought process is correct, instead of disabling Ethernet 1, if I keep it active will this prevent Windows from requiring reactivation?
Any information that could help in figuring out why this is happening would be great.