Monitor MongoDB Databases

On this page:

Watch the video:

AppDynamics Pro has interceptors that track calls to MongoDB and an AppSphere MongoDB monitoring extension that works with the AppDynamics Pro machine agent to capture statistics from the MongoDB server and display them in the AppDynamics Metric Browser. Now from AppDynamics Pro you can drill down through an SQL statement in a Call graph on AppDynamics Pro to AppDynamics for Databases where you can dig deeper into MongoDB performance issues. You can also monitor the MongoDB directly from AppDynamics for Databases.

The following are sample screenshots from an AppDynamics for Databases MongoDB collector. They illustrate the type of information collected for this database platform.

MongoDB Current Window

% CPU statistics are available on the Current window when the Host Collector is enabled for the database collector. You can choose to monitor the host system when you setup the database collector. By default, the system presents statistics for the entire MongoDB cluster. For the specified time period, the chart shows the percentage of available CPU resources consumed by users and the system, and the percentage of time the CPU was idle during I/O wait states. High User CPU may indicate the more CPU resources are required. High System CPU may indicate that your system is running at its maximum level and for example, may need to be upgraded with a more powerful processor. High I/O Wait states may indicate, for example that more disks or faster disks are required.

MongoDB Activity Window

For MongoDB, AppDynamics for Databases supports multi-shard detection. When you configure the collector, you use the IP address and port number of a single shard in the cluster and AppDynamics for Databases detects all the shards and replica sets in the cluster. The default Activity view shows a consolidated view of the MongoDB database.

The Multi-Instance View - In Mongo Over Time chart shows statistics for the Mongo Cluster as a whole and for each shard. Here you can see:

Executions: Throughput of the instance, the total number of SQL statements that were executed on the cluster during that time period. Executions are represented by a red line in the chart
In Mongo: The total number of seconds spent running SQL statements on each shard during that time period. Each shard is represented by a different colored bar in the chart as shown in the legend. If the MongoDB collector detects that the server it is configured to monitor is a shard, it automatically detects the other shards in the cluster. The shard names are detected when the MongoDB collector starts.

MongoDB Statistics Window

The Statistics window presents standard MongoDB statistics to help you manage your environment. Statistics are available for the MongoDB cluster as a whole and individually for each shard in the cluster.