Viewing the Health of Your Instances

Hi, folks. Here's a quick guide on how to view the health of your instances, using public Foglight for Databases. We're going to begin with alarms. Foglight comes with a suite of pre-built alarms that you can configure and set actions to. For more information on this, check out our alarm management video. There are three severities of alarms, warning, critical, and fatal. The number within each tile indicates the number of alarms of that severity on that instance. For more information, click on the number. This will show you a list of alarms for the severity chosen. You can see exactly when the alarm was triggered, if it's been acknowledged or cleared, and more information, such as the host an instance. You can get more information by clicking on the alarm. First of all, on the alarm summary we can see the parent an object. In this example, the parent is the host server, and the object is the sequence server service, which appears to be down. In the details, we get more information about the alarm. On their status, we can see exactly when the alarm was triggered. And under things affected, we can see other objects that had been affected by this alarm, such as the underlying instance. Under background you get some extra information about the alarm, and suggestions we give you recommendations on how to resolve it. Under notes and history tab, you can see notes that other users may have entered against this alarm. Next we're going to look at performance baselines. To access the baselines, click on the Quick View button, next to the instance name. First we're going to focus on the baseline activity of the underlying host, focusing on CPU, memory, and disk activity. You'll see on the top of each one of these counters is a little bar, which indicates your baseline activity range for that particular counter. That is the normal level of activity for this host, at this time of day, of this day of the week. You can see that a disk activity is well outside the baseline range, whereas memory and CPU are falling within the range. Similarly on workload. The light blue area indicates the baseline range for this instance, whereas the blue line indicates the actual level of workload for the instance. This allows us to easily see if there is anything out of the ordinary happening on our instance. There is also a range of alarms and advisories on baseline deviation. Next we're going to look at the instance overview screen. To access this, click on the instance name. The instance overview screen uses tiles to represent different areas of interest of the instance. Each tile has a number of counters, which are color coded to make it easy to identify potential issues. If the counters are colored green, then there is no issue. But as issues arise, the counters will change from green, to yellow, to orange, to red, depending on the severity. To get more information on a particular counter, simply click on it to see the recent history, and there will be a link to the relevant drill down dashboard to get more information, and to get to the root cause of the issue. Finally, we're going to take a more long term view of your instance health, and look at advisories. Since Foglight for databases is monitoring your instance 24/7, 365, and gathering performance metrics, it's in a position to analyze that data and make recommendations and advisories. To access those, click advisories at the top of the screen. Foglight for databases gives advice around areas, such as performance baseline deviation, resource bottlenecks, reducing contention, and optimizing workloads. In this example, I have two advisories on how to deal with excess of IO Wait, and excessive memory pressure. To get more information and advisory, click on the name. First of all, it will detail why it has created this advisory. Then it will list the worst offending sequence statements contributing towards this issue. Then make recommendations, such as missing indexes. Other recommendations will also be listed. Then you will give additional reading advice on how to better understand the issue, and next steps in getting to the root cause and alleviating the issue. So that's how you view the health of your instances using Foglight for databases. For more videos and content, go to quest.com/foglight-for-databases. Thanks for tuning in.