Eric Quackenbush
asked on
What is the best way to apply targeted group policies?
How should we deal with Active Directory Group Policy Objects that have limited scope? We have a newly-created, global Active Directory forest that has been built with scalability and ease of management in mind. We started with a global IP addressing change so that all our offices and datacenters can co-exist on the same WAN and have a logical layout so that we can predict what a particular ip address would be simply by knowing the physical location and purpose of the device to which the ip address is assigned. We then built our Active Directory Sites and subnets around this global ip infrastructure. Now it is time to expand our Group Policies a bit and we want to use the same care in planning them as we did in planning our Active Directory environment.
We have decided that at a global level we should create as few policies as possible. ('Global Policy - Users', 'Global Policy - Computers', 'Global Policy - Servers', etc.)
Some settings are not appropriate to be applied globally but work well for sites, such as setting dns search domain names or mapping printers, so we have some Site GPOs as well.
For managing servers, it is reasonable to control certain things via group policy that are not universally applicable for all servers. The instance that has arisen lately is controlling Remote Desktop Services connection policies for our terminal servers. There are several ways to apply certain policies to only the Terminal Servers. Here are the ones that I have considered:
I know that there are probably as many opinions as there are A/D engineers, but I'm looking for just that -- your opinion, the logic behind it and examples where you have already implemented your suggested solution.
We have decided that at a global level we should create as few policies as possible. ('Global Policy - Users', 'Global Policy - Computers', 'Global Policy - Servers', etc.)
Some settings are not appropriate to be applied globally but work well for sites, such as setting dns search domain names or mapping printers, so we have some Site GPOs as well.
For managing servers, it is reasonable to control certain things via group policy that are not universally applicable for all servers. The instance that has arisen lately is controlling Remote Desktop Services connection policies for our terminal servers. There are several ways to apply certain policies to only the Terminal Servers. Here are the ones that I have considered:
Create sub-OUs, move the Terminal Server computer objects into them and apply GPOs to the sub-OU(s). It seems at first glance to be an effective solution but there are two problems:
1. It adds complexity to the OU structure.
2. Some servers would potentially need to belong to multiple OUs to get all applicable policies but this is only possible if the OUs are nested.
Apply the GPOs universally but use Security Filtering to target specific servers by group membership.
1. This is an elegant solution but it can be difficult to keep track of which policies apply to which server objects.
Modify the local Group Policy on every computer that needs the specific policy.
1. This is the preferred method for testing but it isn't scalable nor centrally managed.
I know that there are probably as many opinions as there are A/D engineers, but I'm looking for just that -- your opinion, the logic behind it and examples where you have already implemented your suggested solution.
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.