Monitor MongoDB Using vRealize Operations

The latest addition to the VMware Solution Exchange—the Blue Medora VMware vRealize Operations Management Pack for MongoDB—was just released today. And it’s about to make monitoring MongoDB on VMware a whole lot easier.

So… why monitor MongoDB inside vROps?

For infrastructures that depend on MongoDB to run their business-critical applications, monitoring database performance—especially with insight into virtual layer it’s running on—is crucial to minimizing bottlenecks and slow-downs.

The Management Pack for MongoDB stands out from existing solutions by harnessing the power of several key vROps features, including:

• Complete visibility and virtualization-aware context from MongoDB to VMware VMs
• Monitoring dashboards for both “big picture” views of your MongoDB environment as well as detailed “drill-downs” for troubleshooting
• Key metrics, including counters for top operations and connections, network traffic, cache usage, memory, etc.
• Detailed alerts and recommendations based on performance metrics pulled from MongoDB
• Capacity analysis and reporting for mongos and mongod resources

8 comprehensive monitoring dashboards

The Management Pack’s out-of-the-box dashboards include four overview dashboards that provide a high-level view of the health, performance, and availability of your MongoDB environment:

• MongoDB Overview – View at-a-glance heatmaps depicting the overall health of your MongoDB resources (mongos instances, mongods, shards, databases, etc.).

• MongoDB Environment – View your entire MongoDB environment from your top-level resources, down through the parent-child hierarchy. Selecting a resource from the Environment Selector will populate KPIs such as page faults, network traffic, current connections, asserts, and opcounters, as well as health and properties information for that particular resource.

• MongoDB Health Investigation – Investigate any current active alerts in your MongoDB environment. When an alert is selected, health information and key performance indicators (KPIs) related to the affected resource are shown, helping you identify the source of the problem immediately.

• MongoDB IOPS – View top IOPS metrics for the selected mongos/mongod resource alongside IOPS metrics for an associated virtual machine or datastore to gain insight into the virtual layer’s impact on your MongoDB environment.

In addition, there are four detail-level dashboards that enable drill-downs into the most critical MongoDB resources:

• MongoDB Mongos Instances – Analyze opcounters, asserts, memory, connections, network, and page faults information for the selected mongos instance.

• MongoDB Mongods – Analyze opcounters, memory, topology (relationships), cache, connections, and network information for the selected mongod resource.

• MongoDB Databases – Analyze health, KPIs, topology (relationships), and top alerts for the selected database.

• MongoDB Replica Sets – Analyze opcounters, network, memory, topology (relationships), cache, and connections information for the selected replica set.

Metrics

The Management Pack collects and displays more than 450 metrics regarding MongoDB resources. Some of the most useful key metrics provided by the MongoDB management include:

• Indexes (Size, Total Counts, etc.)
• Extent Counts
• Operation Counters (Read/Write/Insert/Query, etc.)
• Replication Operation Counters
• Assert Counts
• Connections (Total/Current/Available, etc.)
• Page Fault Counts
• Lock Counts
• Documents (Moved/Inserted/Deleted, etc.)
• Network (Traffic In/Out)

Alerts

The Management Pack for MongoDB creates alerts from key metrics from MongoDB resources and displays them in vRealize Operations. These alerts contain detailed recommendations to help remedy the issue at hand. Examples of alerts include:

• Mongos Total Connections Reaching Maximum
• Mongos Ratio of Page-Faults to Operations Greater than or Equal to 1
• Replica Set has less than 3 data bearing nodes
• Mongod Has a Delay Behind the Primary
• Mongod Journaling is disabled

MongoDB Capacity Analysis & Reporting

Harnessing the predictive analytics capabilities of the vROps Analysis Badges using capacity definitions, the management pack displays Connections Capacity Remaining for a selected MongoDB mongos instance or mongod.

In addition, a MongoDB Capacity Report—which can be scheduled, exported, and shared with key stakeholders—makes it easier to demonstrate and communicate the need to provision more resources to (or reclaim over-provisioned resources within) a MongoDB environment.

To learn more about the Management Pack for MongoDB, the data-tier of the True Visibility Suite for vRealize Operations, or to download a free trial, please visit the True Visibility Suite for VMware vRealize Operations page on Blue Medora’s website.

This post appeared first on the VMware Cloud Management blog.