Use Managed Service for Prometheus to monitor ECS instances - Managed Service for Prometheus

Workflow

To monitor ECS instances, complete these steps:

Enable Host Monitor: Select a VPC to auto-install open source exporters. The managed Prometheus agent collects data automatically.
(Optional) Modify Host Monitor: Change settings such as the service port to fix errors or meet new requirements.
(Optional) Discard metrics: Drop unneeded metrics to reduce collection volume and costs.
(Optional) Configure alerting: Set up alert rules to detect metric changes and troubleshoot issues promptly.

Prerequisites

A VPC with one or more ECS instances is created. For more information, seeCreate and manage an ECS instance in the console (express version).
Alibaba Cloud Resource Center is activated. For more information, see Activate Resource Center.

Note
Resource Center must be activated before monitoring. Managed Service for Prometheus relies on VPC and ECS data of the current account from Resource Center for service discovery.

1. Enable Host Monitor

When you enable Host Monitor for a VPC, Node Exporter and Process Exporter are installed on each ECS instance by default. The managed Prometheus agent then collects, stores, and visualizes data with unified alerting. About 1,000 metrics are collected per ECS instance per scrape.

1.1 Enable the component

In the left-side navigation pane of the ARMS console, click Integration Center. In the Infrastructure section, click Host Monitor.
Select the VPC and configure parameters as described in Monitor ECS instances.
Click OK and wait 1 to 2 minutes.

1.2 View the dashboards

In the left-side navigation pane of the ARMS console, click Integration Management. On the Integrated Environments tab, click ECS Instance. Click the VPC ID to open the environment details page.
On the Component Management tab, click Dashboards in the Addon Type section to view the built-in Grafana dashboards.

Note
If the dashboards have no data, check the security group settings. For more information, see Why do the dashboards have no data?

2. (Optional) Modify Host Monitor

Modify Host Monitor settings such as service discovery, service port, or data collection interval.

2.1 Procedure

In the left-side navigation pane of the ARMS console, click Integration Management. On the Integrated Environments tab, click ECS Instance. Click the VPC ID to open the environment details page.
Find the target exporter and clickSettings.
Adjust the settings and click OK. Parameter descriptions are in Monitor ECS instances.

2.2 Verification

Refresh the page and click Settings to verify the changes.
Check the dashboards to verify data accuracy. For more information, see the 1.2 View the dashboards section.

3. (Optional) Discard metrics

Discard unneeded metrics to simplify data analysis and reduce costs.

3.1 Procedure

In the left-side navigation pane of the ARMS console, click Integration Management. On the Integrated Environments tab, click ECS Instance. Click the VPC ID to open the environment details page.
In the Discard Metrics section of the Metric Scraping tab, select the metrics to discard and click Update. For information about metrics, see Metrics.

Note
You cannot discard the basic metrics about ACK clusters.

3.2 Verification

Click Update and refresh the page to verify the changes.
Check the dashboards to verify data accuracy. For more information, see the 1.2 View the dashboards section.

4. (Optional) Configure alerting

Configure alert rules to get notified when metric values change, enabling faster maintenance and troubleshooting.

Managed Service for Prometheus supports built-in and custom alert rules. Configure either type to match your requirements.

4.1 Configure built-in alert rules

Built-in alert rules generate events by default. Configure notifications manually.

In the left-side navigation pane of the ARMS console, click Integration Management. On the Integrated Environments tab, click ECS Instance. Click the VPC ID to open the environment details page.
In the Addon Type section of the Component Management tab, click Alert Rule. Click View Alert Event or Edit to view or modify an alert rule.
Adjust the alert rule and click OK. Configuration details are in Prometheus alert rules.

4.2 Configure custom alert rules

Create custom alert rules if the built-in rules do not meet your requirements.

In the left-side navigation pane of the ARMS console, click Integration Management. On the Integrated Environments tab, click ECS Instance. Click the VPC ID to open the environment details page.
On the Component Management tab, click the VPC next to Default Metric Storage in the Basic Information section.
On the Alert rules page, create, modify, or view custom alert rules. See Prometheus alert rules.

4.3 Verification

Refresh the page to verify the changes.
Create a low-risk test alert rule, trigger it, and verify that it works as expected.

Note
Notification delivery depends on the alert rule configuration.

FAQ

Why do the dashboards have no data?

The ECS instance security groups may not allow the required access. Ensure the following:

Each ECS instance's security group must allow inbound access from 100.64.0.0/10 and 192.168.0.0/18 to the Node Exporter port (default: 9100) and Process Exporter port (default: 9256). Use your custom ports if modified. For information about how to view the security group rules of an ECS instance, see Search for security groups.

Why was Node Exporter not automatically installed on the ECS instance?

Check the following:

Verify that the ECS instance is running.
Check whether an open source Node Exporter is already installed and using port 9100. If so, go to the Component Management tab, find the Alibaba Cloud Node Exporter, and click Settings to change the port.

How do I check whether Node Exporter has been installed?

Visit http://<ECS-IP>:<PORT>/metrics. If metric data is returned, Node Exporter is installed and running.

How do I manually configure security group rules?

Log on to the ECS console. Add an inbound rule to each ECS instance's security group to allow access from the VPC CIDR block to the Node Exporter port (default: 9100) and Process Exporter port (default: 9256). Use your custom ports if modified.

How do I troubleshoot network issues when integrating ECS instances?

Ensure mutual network access between ECS instances and the Prometheus agent within the VPC. Verify the VPC route table for correct traffic direction, then check that firewall and security group rules allow the traffic.

Why is the metric data inaccurate or missing?

If Node Exporter and Process Exporter are enabled, verify that they are running. Use cURL to query the exporter metrics endpoints and confirm that data is returned. If errors occur, check the exporter logs.

How do I enable process status data collection?

Process Exporter collects process status data on port 9256 by default. Allow port 9256 in the security group of each ECS instance. Process data collection consumes minimal system resources and generally does not affect performance. However, if system resources are insufficient, proceed with caution.