Seperate Databases for Clustering¶

WSO2 Identity Server uses a database to store information such as user management details and identity data. By default, WSO2 Identity Server is shipped with an embedded H2 database that works for all types of data.

This section guides you through the logical separation of data that you can do when clustering WSO2 Identity Server.

Embedded H2 is NOT RECOMMENDED in production

The embedded H2 database is not recommended for enterprise testing and production environments. It has lower performance, and clustering limitations, and can cause file corruption failures. Therefore, use an industry-standard RDBMS such as Oracle, PostgreSQL, MySQL, or MS SQL instead.

Before you begin

Creating separate databases as shown below is not mandatory. Instead, you can point to all the following data sources according to the default data structure. This will not make a difference in performance.

The following image shows the default database structure and a recommended database structure for the logical separation of data.

Separate-databases-for-clustering

Database Name	Description
`USERSTORE_DB`	Mainly contains users and roles of the system Contains authorization manager configurations, internal permissions, and roles In the default database configuration, user data and registry data are located inside `SHARED_DB`. . For more information, see Configure Userstores.
`IDENTITY_DB`	This contains identity-related data. For example, user sessions, OAuth 2.0, SAML 2.0, User Managed Access (UMA), etc. In the default database configuration, all the UMA, consent, and identity data are located in the `IDENTITY_DB`. In a deployment, both identity and UMA can be configured in a single database while consent data is configured to a separate database `CONSENT_MGT` .
`BPS_DB`	This is used to create, drop, and truncate data pertaining to the workflow feature.
`SHARED_DB`	This is the shared database for config and governance registry mounts in WSO2 Identity Server's nodes. This includes data on tenants and keystores. In the default database configuration, user data and registry data are located inside `SHARED_DB`. During a deployment, user data can be configured to a different database, `USERSTORE_DB` while the registry data remains in the `SHARED_DB` .
`CONSENT_MGT`	Used for data pertaining to user consents In the default database configuration, UMA, consent, and identity data are located `IDENTITY_DB`. In a deployment, both identity and UMA can be configured in a single database while consent data is configured to a separate database.