One of the ideals we hold in high regard for ConfigMgr 2012 is the reliability of the system and, specifically, achieving very high success percentages for management operations and deployments. Successful management can be a challenge, with laptops coming and going from the network and environmental issues causing problems for user desktops.
There are a couple of high level principles in our approach to client health. First, an administrator should have a clear view of the ‘reachable’ clients for management operations for any group of machines. Second, the ConfigMgr clients should as much as possible ‘heal’ if problems are encountered.
The integrated Client Status feature of System Center 2012 ConfigMgr adheres to these principles. The system monitors when clients are connecting to the server infrastructure (Management Point) and, if communication is normal in the last 7 days (configurable), then the client is deemed ‘Active’. In addition to this, a series of checks are run on the machine to determine if the components required by the ConfigMgr client are working and configured. An example, well known to ConfigMgr administrators, is the correct functioning of WMI. Many other checks monitor Windows components and services required by the client. If a problem is found, then certain checks have an associated remediation action, such as starting a service or repairing a component. The client is reported as passing or failing these client checks.
So how does this get represented in the console? The display is geared to present actionable data. The Client Status node homepage summary line presents the percentage of Active Clients that are currently passing client check. If this figure is below, say 95%, then it’s something worth looking into.
A useful feature is the ability to define an alert against this value for a collection. When these alerts are triggered, they appear in the lower portion of the homepage. These can be configured on the collection properties page shown below:
The Client Status homepage also shows a view of the top client check errors, which can indicate common issues on machines that need to be investigated. You also have the ability to show the separation of Client Activity and Check results with individual pie charts and trends. The pie charts represent all devices, including those without ConfigMgr clients, to give a full view of the manageability of a collection.
At the feature level, application and update deployment failures may be a result of the functional action or the ConfigMgr client itself not working. In the monitoring view, it’s possible to drill through to a list of problem clients and Client Activity and Check state. Enhanced information for Client Activity and Check is available per client as well. Endpoint protection in particular has integrated Client Status data into the dashboard view.
In conclusion, Client Health in ConfigMgr is an integrated feature that can be leveraged for individual features or viewed as a whole in the Client Status dashboard. Overall the system is designed to be hands off, with clients monitoring and repairing themselves.
For more information, see the Monitoring the Status of Client Computers in Configuration Manager section in the TechNet topic, Introduction to Client Deployment in Configuration Manager.
I hope you like what you see in System Center 2012 Configuration Manager.
Lead Program Manager
System Center Configuration Manager