Weekly quickTIP

Server Confidence Enhancement

Once you've got Systems Center Essentials running, what's next? It's time to act on what SCE is telling you is working -- or not working.

• By Greg Shields
• 02/11/2008

In your small IT environment today, do you know when systems go down? Does your pager or mobile device alert you when a system stops responding to pings? Or, are you forced to wait for users to call in complaints? Lacking a centralized monitoring solution, it is painfully difficult to proactively solve problems before users find out. Included with SCE are components of Systems Center Operations Manager that collect Event Log and PerfMon information into a centralized database. When problems occur, you can configure an alert to hit your pager before your phone rings.

With an installed SCE instance connected to clients, click on the Monitoring node. There, you'll see a link for Active Alerts. Herein lies the beauty with SCE and its SCOM foundation. SCE incorporates a set of filters that looks through the mountains of log data submitted by the systems in the environment. It brings forward to the screen only that data it feels is necessary to be acted upon. Unlike a lot of log centralization systems that force the user to create their own filters, SCE and its Management Packs starts you right out of the gate by showing you only the alerts that are important.

Even more useful under Monitoring is the Computers node. Here, each managed computer is lined up and given a rating. If no problems are present with a system, that system is marked as Healthy. If something may be wrong, that system can be Critical. Right-clicking an unhealthy computer and choosing to open the Health Explorer provides a drill-down view for that computer that discusses what's wrong, when it went wrong and, in many cases, how to fix it.

As an example, upon installing SCE into my network, I immediately discovered that one of my systems was showing a high Percentage of Committed Memory in Use. I knew that system was running slowly, but I had no deep knowledge of why. With SCE monitoring the Event Log and PerfMon for counters out of its expected thresholds, I immediately knew that I needed to add RAM to the machine.

To be notified when problems occur, SCE allows for the creation of Notification Channels, Subscriptions and Recipients. Notification Channels identify how the notification will arrive to the admin: via e-mail, instant message, SMS message or a custom command. Subscriptions provide a way of identifying which messages are of interest. Recipients are the individuals who want to receive subscribed alerts. These three combine to enable a high level of granularity to which alerts individuals want to receive. As an example, if Bob the mail admin only wants to receive high-priority Exchange error alerts, and only between the hours of 5 p.m. and 8 a.m. when he's not in the office, that level of granularity can be assigned to him.

As you can see here -- and, hopefully, throughout this entire series -- System Center Essentials mates the software deployment and inventory capabilities of System Center Configuration Manager (formerly SMS) with the monitoring and alerting capabilities of System Center Operations Manager (formerly MOM). It then skins it with an interface that's easy to use for admins and designed specifically with the smallest of networks in mind. If you're tired of firefighting in your network like me, you may find SCE to be the tool that gets you back out onto the slopes for a few runs.