Help Center/ MapReduce Service/ User Guide/ MRS Cluster O&M/ Checking MRS Cluster Health/ Performing a Health Check for an MRS Cluster
Updated on 2026-10-09 GMT+08:00
View PDF
Share

Performing a Health Check for an MRS Cluster

Big data clusters often contain dozens or hundreds of nodes and multiple components. Administrators must regularly check if each component works properly and identify potential risks. However, manual checks are slow, labor-intensive, and can miss hidden issues, delaying fault detection. How can you check cluster health quickly and thoroughly? MRS provides a cluster health check feature that automatically monitors the health status, alarms, and custom metrics for MRS Manager, services, and hosts. This helps administrators detect issues early and keep the cluster running stably.

The cluster health check includes three key items: object health, related alarms, and custom metrics. The check result may not match the health status shown on the UI. A health check includes MRS Manager, service-level, and host-level health checks:

  • MRS Manager health checks focus on whether the unified management platform can provide management functions properly.
  • Service-level health checks focus on whether components can provide services properly.
  • Host-level health checks focus on whether host indicators are normal.

To ensure that cluster parameters, configurations, and monitoring are correct and that the cluster can run stably for a long time, you can perform a health check during routine maintenance.

Video Tutorial

This video shows how to perform routine health checks on MRS clusters and nodes.

The UI may vary depending on the version. This tutorial is for reference only.

Prerequisites

  • The IAM users have been synchronized in advance. You can do this by clicking Synchronize next to IAM User Sync on the Dashboard page of the cluster details.
  • You have logged in to MRS Manager. For how to log in, see Accessing MRS Manager.

Performing a Health Check on MRS Manager

This section describes how to perform a health check on MRS Manager. For MRS 3.x or later, the operation path differs from that for MRS 2.x or earlier. The actual path may vary.

  1. Log in to FusionInsight Manager of the MRS cluster.

    For details about how to log in to FusionInsight Manager, see Accessing MRS Manager.

  2. Go to the health check page to view the existing check report. Choose O&M > Health Check.

    By default, all saved check reports appear in a list. For details about the report parameters, see Table 1.

  3. Start a manual health check or configure a periodic check based on your O&M requirements.

    • Start a manual health check.

      Click Start Check. In the displayed dialog box, click OK to start a health check.

    • Configure a periodic health check.

      Click Configuration, select Enable, set the check period to Daily, Weekly, or Monthly based on your needs, and click OK to save the configuration.

    Verifying the check result:

    After the health check completes, find the latest report in the check report list. Check the result column. If any item shows abnormal status, handle the fault promptly. For details about viewing and exporting reports, see Viewing and Exporting a Health Check Report.

Health Check on the MRS Management Console

Only MRS 1.9.2 clusters support health check management on the MRS console.

  1. Log in to the MRS console.
  2. On the Active Clusters page, select a running target cluster and click its name to go to the cluster details page.
  3. Start a manual health check or configure a periodic check based on your O&M requirements.

    • Perform the health check for all services.

      Choose Management Operations > Start Cluster Health Check.

    • Perform the health check for a service.

      Click Components. In the service list, click the target service name and choose More > Start Service Health Check.

    • Perform the health check for a host.

      Click Nodes, expand the node group information, select the check box of the target hosts, and choose Node > Start Host Health Check.

    Verifying the check result:

    After the health check completes, find the latest report in the check report list. Check the result column. If any item shows abnormal status, handle the fault promptly. For details about viewing and exporting reports, see Viewing and Exporting a Health Check Report.

FAQs

Q: What should I do if a health check fails?

  1. Check the report to identify the specific abnormal item and determine which component or host is affected.
  2. On MRS Manager, choose O&M > Alarms and locate the alarm details using the associated alarm ID.
  3. Clear the fault by following the handling suggestions in the alarm description.
  4. After fixing the fault, run the health check again to confirm that the issue is resolved.

Q: What should I do if a scheduled health check does not run as expected?

  1. Ensure that the periodic health check feature is enabled. Choose O&M > Health Check > Configuration to verify the status.
  2. Verify that the check period (daily, weekly, or monthly) is configured correctly.
  3. Check whether the MRS Manager service is running properly. If MRS Manager is abnormal, scheduled tasks may not trigger.
  4. Check the MRS Manager background logs to see if there are any execution errors.

Q: What should I do if the health check result differs from the status on the web UI?

A: The health check result and the real-time health status on the web UI come from two independent data sources, so a short time delay may occur between them. The health check result is a snapshot at a specific point in time, while the web UI health status is a real-time value from continuous monitoring. If the status difference persists, check the health check report and look for relevant alarms.

Helpful Links

Parent Topic: Checking MRS Cluster Health