An external system database is required for any deployment using a load balancer. In addition, best practices recommend the use of an external system database for maintenance and recovery purposes. The following databases have been verified for use as external system database. (Refer to "Hybrid Data Pipeline External Database Compatibility Matrix" in the DataDirect Product Compatibility Guide for version-specific information.)

Note: Hybrid Data Pipeline supports Amazon RDS instances that are compatible with these supported database versions.
  • Microsoft Azure SQL Database
  • Microsoft SQL Server
  • MySQL Community Edition
  • Oracle
  • PostgreSQL

Depending on the database you are using, certain requirements must be met. See the following sections for details.

Microsoft Azure SQL Database and SQL Server

If you plan to store system information in a Microsoft Azure SQL Database or SQL Server database, you must take the following steps when setting up the database.
Note: Hybrid Data Pipeline supports failover when an Azure SQL Database is used as an external system database. See Azure SQL system database failover for more information.
  1. Create a database schema to be used for storing Hybrid Data Pipeline system information.
  2. Create an administrator who can access the newly created schema. The administrator must have the CREATE TABLE privileges.
  3. Create a user who can access the newly created schema. The user must have the CREATE SESSION privileges.

After the database has been setup, you must provide the following information during installation:

  • Hostname (server name or IP address)
  • Port information for the database. The default is 1433.
  • Database Name
  • Schema Name
  • Administrator and user account information
    • An administrator name and password. The administrator must have the CREATE TABLE privileges.
    • A user name and password for a standard account. The user must have the CREATE SESSION privileges.

MySQL Community Edition

If you plan on to use a MySQL Community Edition database as an external system database, you must provide the following.

  • The MySQL Connector/J driver during the deployment process
  • Hostname (server name or IP address)
  • Port information for the database. The default is 3306.
  • Database Name
  • Administrator and user account information:
    • An administrator user name and password. The administrator must have the following privileges:
      • ALTER
      • CREATE
      • DROP
      • DELETE
      • INDEX
      • INSERT
      • REFERENCES
      • SELECT
      • UPDATE
    • A user name and password for a standard account. The standard user must have the following privileges:
      • DELETE
      • INSERT
      • SELECT
      • UPDATE

Oracle

If you plan to store system information in an external Oracle database, you must provide the following information.

  • Hostname (server name or IP address)
  • Port information for the database. The default is 1521.
  • SID or Service Name
  • Administrator and user account information
    • An administrator name and password. The administrator must have the following privileges:
      • CREATE SESSION
      • CREATE TABLE
      • CREATE ANY SYNONYM
      • CREATE SEQUENCE
      • CREATE TRIGGER
    • A user name and password for a standard account. The standard user must have the CREATE SESSION privileges.

PostgreSQL

Important: When installing the Hybrid Data Pipeline Server in FIPS mode and connecting to a PostgreSQL 14 or later database using scram-sha-256 authentication, the administrator and user passwords must be greater than 112 bits in length (equivalent to 14 UTF-8 Latin block characters). If this requirement is not met, the product will time out or return the error org.bouncycastle.crypto.fips.FipsUnapprovedOperationError: password must be at least 112 bits.

If you plan to store system information on an external PostgreSQL database, you must take the following steps when setting up the PostgreSQL database.

  1. Enable the citext PostgreSQL extension.
  2. Create an administrator with privileges to create tables.
  3. Create a user with privileges to select, insert, update, delete, and sequence tables.

By default, Hybrid Data Pipeline tables are created in the PostgreSQL public schema. However, tables may be created in a custom schema by specifying the custom schema with the search_path parameter for each Hybrid Data Pipeline system database user. The following SQL ensures that Hybrid Data Pipeline tables will be created and accessed in the hdp custom schema.

ALTER USER hdp_db_admin SET search_path TO hdp;
ALTER USER hdp_db_user SET search_path TO hdp;

After the PostgreSQL database has been setup, you must provide the following information during installation:

  • Hostname (server name or IP address)
  • Port information for the database. The default is 5432.
  • Database Name
  • Administrator and user account information
    • An administrator name and password. The administrator must have privileges to create tables.
    • A user name and password for a standard account. The user must have privileges to select, insert, update, delete, and sequence tables.