Configuring an indexing database

If you choose not to use the recommended Storage as a Service (STaaS) setup but instead choose self-hosted storage, you also need to set up an OpenSearch or Elasticsearch indexing database in your DMS.

An indexing database is required for many DataMiner features, including:

• Pattern matching

• Time-scoped relations

• DataMiner Object Models (DOM)

• DataMiner Service & Resource Management (SRM)

• DataMiner User-Defined APIs

• Specific GQI data sources

A dedicated clustered storage setup always requires indexing, among others to store active and historical alarms and information events.

Important

Elasticsearch is only supported up to version 6.8. As this version is no longer supported by Elastic, we recommend using OpenSearch instead.

Note

For information on the system requirements for OpenSearch/Elasticsearch, refer to DataMiner Compute Requirements.

Tip

See also:

• Securing the OpenSearch database
• Securing the Elasticsearch database

Dedicated clustered storage setup

In a dedicated clustered storage setup, an indexing database is required. Once you have set up Cassandra or a Cassandra-compatible database for the general system storage, you need to set up one of the following databases as the indexing database:

• OpenSearch is supported from DataMiner 10.3.0/10.3.3 onwards. To install OpenSearch, first install an OpenSearch database (ideally on a separate Linux machine), then configure the OpenSearch settings as detailed under Cassandra database.

• Elasticsearch is only supported up to Elasticsearch 6.8 and is therefore not recommended. To install Elasticsearch, first install Elasticsearch on a separate Linux machine, then configure the Elasticsearch settings as detailed under Cassandra database.

Note

Amazon OpenSearch Service is also supported from DataMiner 10.3.0 [CU0] up to 10.3.0 [CU8] and from DataMiner 10.3.3 up to 10.3.11, but this is now deprecated.

Storage per DMA

While this is no longer recommended, it is still possible to use a self-hosted storage setup with a Cassandra database per DataMiner Agent. In such a setup, you should also install an OpenSearch or Elasticsearch indexing database in order to get access to additional DataMiner features, including User-Defined APIs, DOM, and SRM.

• OpenSearch is supported from DataMiner 10.3.0/10.3.3 onwards. To install OpenSearch, first install an OpenSearch database (ideally on a separate Linux server), and then connect your DMS to the OpenSearch database.

• Elasticsearch is only supported up to Elasticsearch 6.8 and is therefore not recommended. You can install Elasticsearch from within DataMiner Cube, or you can install Elasticsearch on a separate Linux machine like for a dedicated clustered storage setup and then manually connect your DMA to the database.

When the database has been deployed, you can configure the indexing settings in DataMiner Cube.