All Products
Search
Document Center

Elasticsearch:Upgrade versions

Last Updated:Aug 13, 2026

You can upgrade the version of an Alibaba Cloud Elasticsearch instance or apply a new kernel patch to ensure a seamless transition for your services. This topic describes the version limitations, restart modes, and the upgrade procedure.

Version limitations

The instance that you want to upgrade must meet the following version requirements.

Current version

Target version

Procedure

5.5.3

5.6.16

Upgrade earlier versions and kernel patches

5.6.16

6.3.2

6.3.2

6.7.0

7.x

Direct upgrades to other versions, including 8.x, are not supported.

To upgrade to an 8.x version, you must purchase a new 8.x instance and migrate your data.

8.x

The latest version

Upgrade an 8.x instance to the latest version

Important

If your current instance version does not meet the requirements, you cannot directly upgrade it. Purchase a new instance that runs the required version.

Restart modes

An upgrade triggers a cluster restart. The restart typically takes about one hour but can vary depending on the data volume, data structure, and specifications of your cluster. Alibaba Cloud Elasticsearch supports two restart modes. The system automatically selects the appropriate mode based on your upgrade plan.

Restart mode

Mechanism

Risks and recommendations

Applicable versions

Blue-green deployment

The system adds higher-version nodes to the cluster, migrates data from the lower-version nodes to the new ones, and then removes the lower-version nodes.

  • The upgrade cannot be canceled after it starts. During the upgrade, you can continue to read data from and write data to the cluster, but you cannot perform other configuration changes. Perform the upgrade during off-peak hours.

  • The IP addresses of the nodes change. If you specified IP addresses in your cluster configuration or client-side code, you must update them after the upgrade is complete.

  • 5.5.3 to 5.6.16

  • 6.3.2 to 6.7.0

  • 8.x to the latest version

  • Kernel patch upgrade

In-place change

All nodes are shut down and then restarted.

The upgrade process requires you to install a TLS certificate for the cluster, and causes service downtime. This process does not cause data loss. The upgrade duration depends on the data volume and cluster specifications. Plan the upgrade in advance.

Note

After an in-place restart, the IP addresses of the cluster nodes do not change. You do not need to modify the instance configuration.

5.6.16 to 6.3.2

Prerequisites

  1. Be aware of potential breaking changes, minor changes, and deprecations between your current and target versions. For more information, see the Migration guide.

    Note

    Versions 8.9 and 8.15 include breaking changes. Other 8.x versions involve only minor changes. For more information, see Migrating to 8.15 and Migrating to 8.9.

  2. If you use Elasticsearch plugins, ensure that each custom plugin version is compatible with the target version. Otherwise, the plugins might not work correctly. System plugins, such as aliyun-optimization and kmonitor, are maintained by Alibaba Cloud. You cannot uninstall or upgrade them.

  3. If you also use other Elastic Stack components such as Logstash, APM, Beats, Fleet/Elastic Agent, Elastic Security, or Enterprise Search, ensure that they are compatible with the target version.

  4. Before you start the upgrade, take a snapshot of the latest data. If the upgrade fails, you must restore the data from the snapshot.

  5. Before you upgrade a production cluster, validate the upgrade process in a test environment or on a test instance.

  6. Do not create new indexes during the upgrade process. Perform upgrades on production clusters during off-peak hours.

  7. The version upgrade also upgrades the Kibana version. During the upgrade, Kibana is unavailable.

Procedure

Upgrade earlier versions and kernel patches

  1. Log on to the Alibaba Cloud Elasticsearch console.

  2. In the left navigation menu, choose Elasticsearch Clusters.

  3. Navigate to the target cluster.

    1. In the top navigation bar, select the resource group to which the cluster belongs and the region where the cluster resides.

    2. On the Elasticsearch Clusters page, find the cluster and click its ID.

  4. On the Basic Information page, click Update and Upgrade in the upper-right corner.

  5. In the Update and Upgrade dialog box, select an operation type and a target version.

    Two operation types are supported:

    • Kernel patch update: This operation upgrades the kernel version without changing the version of the ES instance. Only kernel-enhanced instances support kernel version upgrades. For more information about the features of each kernel version, see Kernel Version Release Notes.

      Note

      When the system detects that a new kernel patch is available, the A new kernel patch is available. message appears next to the Version field on the Basic Information page. You can click this message to upgrade the kernel patch.

    • Elasticsearch version upgrade: Upgrades only the version of the ES instance and does not simultaneously upgrade the kernel patch.

  6. Click Upgrade Check. The system checks for configuration compatibility, cluster health, snapshot backups, and underlying resources.

    After the checks are complete, resolve any reported issues and click Re-check. For more information, see Upgrade check.

    The following items are checked:

    • Configuration compatibility check

      Checks for incompatible configurations between the current and target versions. This is primarily for upgrades across major versions, such as an upgrade from 5.x to 6.x. If this check fails, the upgrade stops. View the check results and recommended solutions to resolve the issues and then restart the upgrade process. For more information, see Configuration compatibility check and Adjust incompatible configurations.

    • Cluster health check

      Checks whether the cluster health is Green and whether the cluster load meets the requirements. If this check fails, refer to Cluster health check to verify that the cluster load is at a normal level.

    • Cluster snapshot backup

      Checks whether a snapshot has been successfully created within the last hour. If the upgrade fails, you can restore data from the snapshot.

      Note
      • If a snapshot has not been created in the last hour, you can click Manual Snapshot in the Update and Upgrade dialog box to manually trigger a snapshot. For more information about snapshot backups, see Data backup.

      • The time required to create a snapshot depends on the data volume of the cluster. If automatic snapshot backup is not enabled and the data volume is large, creating the first snapshot might take a long time.

  7. After all checks pass, click Upgrade.

    During the upgrade, you can view the progress on the Tasks page. After the upgrade is complete, you can view the new instance version or kernel patch on the instance's Basic Information page.

Upgrade an 8.x instance

  1. Log on to the Alibaba Cloud Elasticsearch console.

  2. In the left navigation menu, choose Elasticsearch Clusters.

  3. Navigate to the target cluster.

    1. In the top navigation bar, select the resource group to which the cluster belongs and the region where the cluster resides.

    2. On the Elasticsearch Clusters page, find the cluster and click its ID.

  4. In the upper-right corner of the Basic Information section, click Update and Upgrade.

  5. On the Version Selection wizard page of the Update and Upgrade control panel, select the target version.

    1. For Elasticsearch Version Upgrade, select Elasticsearch Version Upgrade.

    2. You can only select the latest version of Alibaba Cloud Elasticsearch.

    3. Read the risk warnings and select the checkbox to acknowledge the risks.

    4. Click Next step.

  6. In the Upgrade Check step, resolve any reported issues.

    1. If a cluster snapshot has not been created in the last hour, click Manual Backup to back up the cluster.

    2. If the cluster health or basic resource checks fail, see Upgrade check to perform a manual check.

    3. Click Check Again until all check items pass, and then click Next step.

  7. In the Plug-in Configuration step, configure plugins based on the on-screen instructions.

    The plugin configuration is complete when Check Passed is displayed in the Plug-in Configuration section.

  8. Click Upgrade, and then click OK in the dialog box that appears.

    During the upgrade, service jitters may occur due to node changes. Perform this operation during off-peak hours. The version upgrade consists of two stages:

    1. Phased change: After the phased change is complete, check the cluster stability and monitor your services. If any exceptions occur, submit a ticket to Alibaba Cloud technical support for a rollback.

    2. Full change: After the phased change is complete and you have confirmed that your services are stable, click Full Change in the console to initiate the full change.

      After the full change is initiated, rollbacks are not supported.

FAQ