G
Guest
We are preparing to perform in-place upgrade from windows 2000 to 2003 AD.
We have single domain, one central location and several remote office
locations (each remote office has Domain Controller).
We have VoIP integrated into our Active Directory (Cisco call manager).
There are high-level concerns as to how to Rollback if the upgrade is not
successful.
A suggestion has been made to shut down all regional office domain
controllers, and then perform the forestprep, domainprep and the Operating
System upgrade on the schema master and the infrastructure master at our
central location (all role holders are at the central location).
Once satisfied that this has worked properly, the regional office domain
controllers will be turned back on.
The basis for this suggestion, is to limit potential "damage" to the remote
offices, in the event that the upgrade fails at some point.
FYI, I am concerned about this approach, as I have reviewed some of the
microsoft migration-related documentation, which states that the Forestprep
should be completed before the domain prep is executed. I am wondering if by
turning off the remote servers, we are contributing to a failed upgrade, in
our efforts to PREVENT a failed upgrade.....
We are planning to test this theory in a test lab environment......
Please feedback your comments or recomendations regarding this proposed
strategy.
Thank you.
We have single domain, one central location and several remote office
locations (each remote office has Domain Controller).
We have VoIP integrated into our Active Directory (Cisco call manager).
There are high-level concerns as to how to Rollback if the upgrade is not
successful.
A suggestion has been made to shut down all regional office domain
controllers, and then perform the forestprep, domainprep and the Operating
System upgrade on the schema master and the infrastructure master at our
central location (all role holders are at the central location).
Once satisfied that this has worked properly, the regional office domain
controllers will be turned back on.
The basis for this suggestion, is to limit potential "damage" to the remote
offices, in the event that the upgrade fails at some point.
FYI, I am concerned about this approach, as I have reviewed some of the
microsoft migration-related documentation, which states that the Forestprep
should be completed before the domain prep is executed. I am wondering if by
turning off the remote servers, we are contributing to a failed upgrade, in
our efforts to PREVENT a failed upgrade.....
We are planning to test this theory in a test lab environment......
Please feedback your comments or recomendations regarding this proposed
strategy.
Thank you.