Login
      Contents x
      Search Engine Service FAQ
        • Print
        • PDF
        Contents

        Search Engine Service FAQ

          • Print
          • PDF
          Article Summary

          Available in Classic and VPC

          The Search Engine Service FAQ provides answers to frequently asked questions. If you haven't found the answer to your question in the FAQ below, then search the user guide for what you would like to know.

          Q. What is the required specification for using clusters?

          A. Check and determine the data throughput by cluster specifications (refer to Data throughput by cluster specifications). Be sure to use it in a real environment after conducting a test.

          Q. Can I change the specifications of the cluster I'm using?

          A. It does not provide a feature to change the specifications of the cluster in use.

          Q. How can I check the status of a cluster?

          A. In the Search Engine Service, the cluster status is displayed in 4 ways: running, warning, error, unknown. For a description of each status, see Cluster status (VPC) or Cluster status (Classic).

          Q. How many shards are appropriate?

          A. Determine by considering the size of the data node memory, the index data volume, and the recommended shard size.

          • It is recommended that the shard size not exceed the JVM heap size (half of the data node memory size).
          • <example> If the data node memory size is 32 GB and the index data volume is 150 GB
            • the JVM heap size is 16 GB. If you consider the recommended shard size to be about 15 GB, 10 shards would be appropriate.
          Note

          You can set the number of shards for a specific index in the Console in Kibana or OpenSearch Dashboards.

          • This applies only to newly created indexes, not to existing indexes.
          • <example> Create a template that sets the number of shards to 10 for an index starting with logstash-
            PUT _template/logstash-template
{
  "index_patterns": ["logstash-*"],
  "settings": {
    "number_of_shards": 10
  }
}

          Q. When adding a data node, is there any effect on the currently running cluster?

          A. It may have the following effects on clusters:

          • When a data node is added, change the settings file of the running cluster and restart the cluster. Temporary service disruption may occur during this process.
          • The shards are reallocated to the added data nodes, and resource usage may increase more than usual during this process.

          Q. Can I install plugins I want to use separately?

          A. After accessing the server as sshuser, you can save and install the plugin file.

          • For Elasticsearch clusters, you can install plugins with the following command:
            sudo /usr/share/elasticsearch/bin/elasticsearch-plugin install "Name of saved plugin file"
          • For OpenSearch clusters, you can install plugins with the following command:
            sudo /usr/share/opensearch/bin/opensearch-plugin install "Name of saved plugin file"
          Note

          sshuser can use the sudo command without a password.

          Caution

          If the cluster does not operate normally due to the user access to and running of the search engine service cluster, NAVER Cloud Platform is not responsible for such operational quality.

          Q. Is it possible to change from OSS license of Elasticsearch to the basic license?

          A. Only OSS licenses are available for the Search Engine Service, and basic or higher licenses cannot be used.

          Q. What caused the snapshot creation to fail?

          A. If snapshot creation fails, check the following:

          • The snapshot creation is not available for the following cases:

            • If you are creating or deleting another snapshot
            • In case of setting API authentication key of the Sub Account which has an insufficient level of permission related to Object Storage

          • Snapshots may not be created successfully in the following cases:

            • If the cluster usage is high (if there is high data inflow or search queries)
              • Try taking snapshots during periods when the cluster usage is low.
            • If the cluster is being restored, or if a shard in the initializing status exists
            • In case of timeout due to too much data in the Snapshot repository
            • If 2 or more clusters create a snapshot in the same Snapshot repository (the same base_path as the bucket name)
            • When disabling the API authentication key set during snapshot creation
            • When deleting a bucket during snapshot creation or setting access control to the bucket
            • When deleting or changing the Object Storage Plugin (s3-repository) or the API authentication key (s3.client.default.access_key, s3.client.default.secret_key)
          Caution
          • In the Snapshot repository, you can perform only one job (snapshot creation, deletion, restoration) at the same time, and NAVER Cloud Platform is not responsible for any problems with the snapshot file by performing multiple jobs at the same time.
          • If you arbitrarily edit or delete snapshot-related files in the bucket, snapshot-related jobs may not be performed normally, and NAVER Cloud Platform is not responsible for this.
            • Delete the folder or create a new bucket to create a snapshot.

          Q. Why do I see the snapshot names that I didn't create?

          A. If the status of the created snapshot is PARTIAL, the Search Engine Service creates a snapshot with the name, name of existing snapshot name+-retry.

          Q. How can I see the progress of a snapshot?

          A. You can check the progress of a snapshot in the following ways:

          • You can check the approximate progress in the Snapshot (VPC) menu (refer to Snapshot).
          • In Kibana or the Console in OpenSearch Dashboards, you can find how many indexes are running or completed for your current snapshot by using the following command:
            GET _snapshot/_status

          Q. How do I collect logs for a cluster?

          A. Clusters created before January 13, 2022 don't collect logs. If you want to collect logs, Contact us at NAVER Cloud Platform support > Support center.

          Q. Why can't I see the Visualization of TSVB Type in Kibana?

          A. The use of Visualization of TSVB-Type is restricted due to the CVE-2020-7015 vulnerability in Kibana OSS version 7.7.0 provided by the search engine service.

          • If it is unavoidable to use Visualization of TSVB Type, edit the kibana.yml file and restart Kibana as follows:
            1. Access the manager node with the sshuser account (refer to SSH connection to the manager node (VPC) or SSH connection to the manager node(Classic)).
            2. Run the sudo vi /etc/kibana/kibana.yml command to open the kibana.yml file.
            3. Delete the metrics.enabled: false contents of the bottom line of the file.
            4. Run the sudo systemctl restart command to restart Kibana.
          Caution
          • If you want to edit the kibana.yml file and use it, it may be difficult to respond to the CVE-2020-7015 vulnerability.
          • For more information about the CVE-2020-7015 vulnerability, see CVE-2020-7015.
          Was this article helpful?
          What's Next
          Change password!
          Changing your password will log you out immediately. Use the new password to log back in.
          Change profile
          Success!
          First name must have atleast 2 characters. Numbers and special characters are not allowed.
          Last name must have atleast 1 characters. Numbers and special characters are not allowed.
          Enter a valid email
          Enter a valid password
          Your profile has been successfully updated.
          Logout