Intelligent Maintenance is a smart operations and maintenance system of Alibaba Cloud Elasticsearch. Intelligent Maintenance is designed to diagnose the health state of Alibaba Cloud Elasticsearch instances, detect potential risks, and provide suggestions. This topic describes how to use Intelligent Maintenance to resolve issues found on your Elasticsearch instance. Using Intelligent Maintenance can help you save operations and maintenance costs and guarantee that your instance is secure, healthy, and easy to use.

Supported regions

Intelligent Maintenance is supported in the following regions:

  • China (Hangzhou)
  • China (Shanghai)
  • China (Qingdao)
  • China (Beijing)
  • China (Shenzhen)

Cluster overview

Notice
  • If this is the first time that you have activated the Cluster Overview feature, you must grant Intelligent Maintenance the permissions to use the basic information and log data of the cluster. Intelligent Maintenance does not need to use other data.
  • After you activate Cluster Overview, the system automatically diagnoses the cluster at a specific time every morning and generates a diagnostic report.
  1. Go to the instance management page of your Elasticsearch instance.
  2. In the left-side navigation pane, choose Intelligent Maintenance > Cluster Overview.
  3. On the Cluster Overview page, you can view the health state of your cluster.

    The Cluster Overview page shows the health states of the cluster in the last seven days to help you diagnose the health state of the cluster.

    Intelligent Maintenance uses red, yellow, and green to indicate the health state of a cluster. For more information, see Diagnosis results.

Cluster diagnosis

In addition to the daily diagnosis, you can click Diagnose Now to manually diagnose your cluster. By default, you can use this feature to diagnose a cluster for up to five times per day. Each diagnostic process takes about three minutes.

  1. Go to the instance management page of your Elasticsearch instance.
  2. In the left-side navigation pane, choose Intelligent Maintenance > Cluster Diagnosis.
  3. On the Cluster Diagnosis page, specify the Indexes and Diagnosis Items.
    Notice By default, all the Indexes and Diagnosis Items are selected. You can specify up to 10 indexes for each diagnostic request.
  4. Click Diagnose Now.
    After the system diagnoses the cluster, it generates a diagnostic report containing descriptions about the diagnostic items, diagnostic results, and suggestions.
    Note After the system generates a diagnostic report, you can click submit your feedback or suggestions to inform us what we can do to optimize the performance of Intelligent Maintenance.

Previous reports

  1. Go to the instance management page of your Elasticsearch instance.
  2. In the left-side navigation pane, choose Intelligent Maintenance > Previous Reports.
  3. The Previous Reports page lists the 20 latest diagnostic reports.

Diagnosis results

Intelligent Maintenance uses red, yellow, and green to indicate the health state of a cluster.

  • Red: indicates a severe issue or potential risk that may adversely affect the usage of the cluster and requires immediate attention. If you do not resolve this issue, data loss or cluster crash may occur.
  • Yellow: indicates a moderate issue that may adversely affect the usage of the cluster and requires prompt attention.
  • Green: indicates that the cluster is healthy.