Alibaba Cloud Elasticsearch allows you to upgrade the version or kernel of your Alibaba Cloud Elasticsearch cluster. This ensures that your business is up-to-date. This topic describes how to upgrade the version of an Elasticsearch cluster and provides the related precautions.

Prerequisites

  • The cluster whose version you want to upgrade meets the following requirements.
    Type Limit
    V5.5.3 to V5.6.16 None
    V5.6.16 to V6.3.2 None
    V6.3.2 to V6.7.0 None
    Note If the version of your Elasticsearch cluster does not meet the preceding requirements, you cannot directly upgrade the version. We recommend that you cancel the subscription of or release the cluster and purchase a cluster of the desired version.
  • A precheck is performed for the version upgrade.

    For more information about the related check items, see Check items for cluster status and Check for and modify incompatible configurations before you perform an upgrade from V5.6 to V6.3.

  • Kernel upgrades: A new kernel version is available.
    You can go to the Basic Information page of your cluster to check whether a new kernel version is available for the cluster. Upgrade a kernel
  • Upgrades from V5.6.16 to V6.3.2: The configurations of your cluster are modified to prevent configuration incompatibility. V6.X may be incompatible with some configurations in V5.X. If these configurations are not modified, the cluster may not provide services as expected after the upgrade. For more information about the incompatible configurations and how to modify these configurations, see Check for and modify incompatible configurations before you perform an upgrade from V5.6 to V6.3. For more information about major changes in Elasticsearch V6.X, see Breaking changes in 6.0.
    Notice If your client is connected to the cluster whose version you want to upgrade, you must update the client version before the upgrade. This ensures that the client is compatible with the cluster version after the upgrade. For more information about the version compatibility between clients and clusters, see Compatibility.

Precautions

When you upgrade the version or kernel of your cluster, the system restarts the cluster. The following table describes the restart modes that are provided by Alibaba Cloud Elasticsearch.
Restart mode Principle Risk and recommendation Scenario
Blue-green restart The system adds nodes of a later version to the cluster, migrates data stored on original nodes to the added nodes, and removes the original nodes from the cluster.
  • An ongoing upgrade cannot be canceled. During the upgrade, you can only read data from or write data to your cluster. Therefore, we recommend that you perform an upgrade during off-peak hours.
  • After the upgrade, the IP addresses of the nodes in the cluster are changed. If you have specified the IP addresses in the configuration file of the cluster or in the code that enables clients to access the cluster, update the configuration file or code after the upgrade.
  • V5.5.3 to V5.6.16
  • V6.3.2 to V6.7.0
  • Kernel upgrades
Full restart The system disables all nodes in the cluster and restarts the cluster. During an upgrade, the system installs a TLS certificate on your cluster. This may interrupt your services but does not cause data loss. Upgrade duration depends on the specifications of the cluster and the volume of data stored on the cluster. We recommend that you plan the upgrade in advance.
Note A full restart does not change the IP addresses of the nodes in the cluster. Therefore, you do not need to update the IP addresses.
V5.6.16 to V6.3.2

Check items for cluster status

Before you upgrade the version or kernel of your cluster, check whether the cluster is in a normal state and whether its load is normal. You can upgrade the version or kernel of your cluster only when the cluster is in a normal state and its load is normal.
Check item Normal state
Cluster status The cluster is in the Active state (indicated by the color green).
JVM heap memory usage The JVM heap memory usage of the cluster is less than 75%.
Disk usage The disk usage of nodes is less than the value of cluster.routing.allocation.disk.watermark.low.
Replica shards All indexes are configured with replica shards.
Snapshots The cluster created snapshots during the last hour.
Custom plug-ins The cluster does not have custom plug-ins installed.
Elastic Compute Service (ECS) instances in the zone where the cluster resides The zone where the cluster resides has sufficient ECS instances.
Note When you upgrade the version or kernel of a cluster, the system adds nodes of a later version to the cluster. Then, the system migrates data stored on original nodes to the added nodes and removes the original nodes from the cluster. Therefore, before the upgrade, make sure that the zone where the cluster resides has sufficient ECS instances.
YML configuration file The cluster of a later version is compatible with the YML configuration file in an earlier version.

Procedure

  1. Log on to the Elasticsearch console.
  2. In the left-side navigation pane, click Elasticsearch Clusters.
  3. Navigate to the desired cluster.
    1. In the top navigation bar, select a resource group and a region.
    2. In the left-side navigation pane, click Elasticsearch Clusters. On the Elasticsearch Clusters page, find the desired cluster and click its ID.
  4. In the upper-right corner of the Basic Information page, click Update and Upgrade.
  5. In the Upgrade dialog box, select the desired version.
    Notice A kernel upgrade does not change the version of the cluster. The entry point of a kernel upgrade is displayed only after the system detects a new kernel version. After the kernel is upgraded, the entry point is no longer displayed. For more information about the new features that are provided by the upgraded kernel, see AliES release notes.
  6. Click Precheck.
    Then, the system checks the configuration compatibility, status, snapshots, and basic resources of the cluster. Upgrade precheck
    After the check is complete, handle exceptions as prompted. The following description provides specific instructions on the check items:
    • Configuration compatibility check

      Check whether the later version is compatible with all configurations in the earlier version, especially for major version upgrades such as upgrades from V5.X to V6.X. If the cluster fails the check, the upgrade is terminated. In this case, modify the incompatible configurations based on the instructions provided in Check for and modify incompatible configurations before you perform an upgrade from V5.6 to V6.3 and perform the upgrade again.

    • Cluster status check

      Check whether the cluster is in the Active state (indicated by the color green) and whether its load is normal. Before the check or if the cluster fails the check, you can check whether the cluster load is normal based on the instructions provided in Check items for cluster status.

    • Snapshot check
      Check whether the cluster has created snapshots during the last hour before the upgrade. If the cluster has not created snapshots during the last hour, you can click Create Snapshots in the Upgrade dialog box to trigger snapshot creation.
      Notice If the upgrade fails, you can restore data from the snapshots. The time that is required to create snapshots depends on the volume of data stored on the cluster. If the Auto Snapshot feature is disabled for the cluster and the data volume is large, a longer time is required to create the first snapshot.
  7. After the cluster passes the precheck, click Upgrade.
    During the upgrade, you can view the upgrade progress in the Tasks dialog box.
    After the upgrade, you can view the cluster or kernel version on the Basic Information page.

FAQ