MongoDB Atlas monitoring

Monitor your MongoDB Atlas projects, clusters, and nodes from a single console with Site24x7. Track connections, operations, replication lag, cache, CPU, memory, and disk metrics, so you can detect and resolve database issues before they affect your applications.

Use cases

Site24x7 MongoDB Atlas monitoring helps you identify and troubleshoot database performance issues through the following use cases:

• Track cluster state, availability, and node health across all your Atlas projects, and get alerted when a node is down or in a critical state.
• Identify slow queries and inefficient query patterns by using slow query logs, suggested indexes, and query targeting ratios.
• Detect replication lag and oplog growth early to prevent data inconsistency between replica set members.
• Plan capacity by monitoring storage, disk usage, cache fill ratio, and memory trends.
• Audit your Atlas environment by reviewing API keys, roles, backup status, encryption at rest, and termination protection of each cluster.

Benefits

Monitoring MongoDB Atlas with Site24x7 provides the following benefits:

• Monitor projects, clusters, and nodes without installing an agent.
• Discover clusters automatically, or add them manually.
• View database-, operation-, and host-level metrics for every node in one place.
• Reduce mean time to resolution (MTTR) with threshold-based alerts on key database metrics.

Supported components

Site24x7 creates the following monitors for MongoDB Atlas.

Prerequisites

Create a service account in MongoDB Atlas with the Organization Read Only permission:

1. In MongoDB Atlas, navigate to Identity & Access > Applications > Service Accounts.
2. Click Add new > Service Account.
3. On the Create Service Account page, provide the following details:
   • Name: Enter a name for the service account.
   • Description: Enter a description for the service account.
   • Client Secret Expiration: Select the expiration period for the client secret.
   • Organization Permissions: Select Organization Read Only.
4. Click Create.
5. Copy the generated Client ID and Client Secret for configuring MongoDB Atlas monitoring in Site24x7.

Additional prerequisites for slow queries

To view data in the Slow Queries tab of the MongoDB Atlas Node monitor, the service account must have the Project Data Access Read Only role for each project you want to monitor.

The Organization Read Only permission alone does not provide access to Atlas Performance Advisor data.

To assign the required project-level permission:

1. Log in to MongoDB Atlas and open the project you want to monitor.
2. Navigate to Security > Project Identity & Access > Users.
3. Click Invite to Project.
4. In the search box, enter the Client ID of the service account created for Site24x7 monitoring.
5. Select the service account from the search results.
6. Under Give your members access permissions below, click Select Roles.
7. Select Project Data Access Read Only (this means the user can access a cluster’s data and indexes).
8. Click Invite to Project to assign the permission.
9. To verify the permission, navigate to Project Identity & Access > Applications > Service Accounts. The Project Permissions column for the service account should display Project Data Access Read Only.

Adding your MongoDB Atlas organization as a monitor in Site24x7

1. Log in to Site24x7.
2. Go to Apps > Cloud Application > MongoDB Atlas.
3. Click the MongoDB Atlas icon. The Add MongoDB Atlas Monitor form opens.
   
   
4. Configure the fields in the Monitor details, Configuration Profiles, and Alert Settings sections below.
5. Click Save. A Successfully Added message appears and the monitor shows up in the monitor list.

Monitor details

Configuration Profiles

Alert Settings

Add an OAuth provider

Click the + icon next to Provider Name. The Add OAuth Provider dialog opens.

Click Check Configuration to test the connection. A successful check returns the HTTP response headers and the response body, which contains the access token. Then click Authorize to complete the setup. Select the new provider in the Provider Name list and click Save.

Discover and add clusters manually

If you set Auto Create Cluster Monitor to No, add clusters yourself:

1. Open the MongoDB Atlas monitor and go to the Clusters tab.
2. Click Discover Cluster. Site24x7 discovers the clusters in your organization. This may take a few moments.
3. In the Discover Cluster dialog, select the clusters to monitor. Each row shows the cluster Name and its Project Name. The dialog also shows the Last Discovered Time and a Search Filter.
4. Click Add. To remove a cluster from monitoring, select it and click Delete.

Monitors created in Site24x7

After you add the organization, Site24x7 lists the following under Apps > Cloud Application > MongoDB Atlas. The status circles (Down, Critical, Trouble, Up) summarize them.

• The organization monitor (type MongoDB Atlas), for example, Atlas.
• One cluster monitor per cluster (type MongoDB Atlas Cluster), for example, Cluster0.
• One node monitor per node (type MongoDB Atlas Node), named after the node hostname, for example, cluster0-shard-00-00.xxxx.mongodb.net.

The list shows each monitor's performance count and when it was last polled. Click a monitor name to open it.

View your MongoDB Atlas monitor

1. Go to Apps > Cloud Application > MongoDB Atlas.
2. Click the monitor name to open it.

The monitor has the following tabs:

Summary

The Summary tab provides an overview of the MongoDB Atlas organization’s availability, downtime, and projects.

API Keys

The API Keys tab provides details about the API keys configured in the MongoDB Atlas organization.

Clusters

The Clusters tab displays a graph showing the total number of clusters over time and a table listing the clusters added to Site24x7, along with their configuration details and availability status. Click Discover Cluster to discover and add clusters manually. Click a cluster name to view its associated cluster monitor.

View your MongoDB Atlas cluster monitor

Click a cluster in the Clusters tab to open its cluster monitor.

Summary

The Summary tab provides an overview of the cluster’s availability, downtime, and performance.

Nodes

This tab lists the node hostnames mapped to the cluster monitor, along with their availability status. Click a node to open its node monitor.

View your MongoDB Atlas node monitor

The node monitor collects metrics at one-minute granularity, except database-level metrics. The MongoDB Atlas API doesn't return per-minute data for database-level metrics, so Site24x7 collects them once per poll cycle (every five minutes by default).

Performance metrics

This section describes the metrics available under each tab of the MongoDB Atlas node monitor, helping you analyze database operations, query performance, resource utilization, replication, and overall node health.

Summary

This tab provides an overview of the node’s availability, database operations, connections, memory, CPU, cache, and network performance.

Databases

This tab displays database-level metrics, including data size, storage usage, collections, indexes, and object counts.

Operations

This tab provides insights into database operations, including queries, inserts, updates, deletes, commands, and document activity.

Performance

This tab displays query execution times, query efficiency, concurrency, throttling, and other performance metrics.

Replication

This tab provides insights into replication performance, including oplog activity and replication lag.

Cache & Storage

This tab displays cache utilization, database storage, index sizes, and cursor activity.

Connections & Asserts

This tab provides insights into active connections, network traffic, and database assertions.

CPU

This tab displays CPU utilization metrics for the MongoDB process and host system, including normalized and peak CPU usage.

Each metric below has a normalized (per core) variant, named with Normalized, and a peak variant, named with the Max prefix, for example, Max Process Normalized CPU User.

Memory & Swap

This tab provides insights into memory consumption, available memory, swap usage, and swap I/O activity.

Each host memory and swap metric has a peak variant with the Max prefix, for example, Max System Memory Used.

Network

This tab displays network traffic and request metrics for the MongoDB process and host system.

Full-Text Search

This tab provides insights into Atlas Search resource utilization, including CPU, memory, and disk usage.

Disk

This tab displays disk performance metrics, including IOPS, latency, throughput, disk space, and queue depth.

Slow Queries

This tab displays slow query logs to help identify queries that may affect database performance.

Suggested Indexes

This tab displays index recommendations from the Atlas Performance Advisor to help optimize query performance.

Other tabs

The MongoDB Atlas Organization, Cluster, and Node monitors also include the following common tabs:

• Outages: Displays the outage history of the cluster monitor, including downtime details and duration.
• Notes: Allows you to add and view notes related to the cluster monitor.
• Log Report: Displays the historical monitoring data collected for the cluster.
• Alert Logs: Displays the alerts generated for the cluster monitor, including alert status and notification details.