Overview of Creating a DB System

This article provides an introduction to the various settings required to create DB systems.

When you create a DB system using the Console, the API, or the CLI, a system is provisioned to support Oracle AI Database, and an Oracle AI Database is created based on the options you provide and some default options described later in this article.

Required IAM Policy

To use Oracle Cloud Infrastructure, you must be granted security access in a policy by an administrator. This access is required whether you’re using the Console or the REST API with an SDK, CLI, or other tool. If you get a message that you don’t have permission or are unauthorized, verify with your administrator what type of access you have and which compartment to work in.

For administrators: The policy in Let database admins manage Oracle Cloud database systems lets the specified group do everything with databases and related Database resources.

If you’re new to policies, see Getting Started with Policies and Common Policies. If you want to dig deeper into writing policies for databases, see Details for the Database Service.

Prerequisites

You’ll need the following items to create a DB system:

• The public key, in OpenSSH format, from the key pair that you plan to use for connecting to the DB system via SSH. A sample public key, abbreviated for readability, is shown below.

  ssh-rsa AAAAB3NzaC1yc2EAAAABJQAA....lo/gKMLVM2xzc1xJr/Hc26biw3TXWGEakrK1OQ== rsa-key-20160304

• A correctly configured virtual cloud network (VCN) to launch the DB system. Its related networking resources (gateways, route tables, security lists, DNS, and so on) must also be configured as necessary.
• Oracle recommends using a service gateway to enable necessary access, if you plan to back up your DB system or use the managed update feature.
• For a multi-node RAC DB system, ensure that port 22 is open for both ingress and egress on the subnet, and that the security rules you create are stateful (the default). Otherwise, the DB system might fail to provision successfully.

For more information, see:

• Managing Key Pairs on Linux Instances
• VCN and Subnets

Default Options for the Database

To simplify creating a DB system in the Console, and when using the API, the following default options are used for the database.

• Console enabled: False
• Create container database: True
• Create instance only (for standby and migration): False
• Database home ID: Creates a new database home
• Database language: AMERICAN
• Database sizing template: odb2
• Database storage: Automatic Storage Management (ASM). Optionally, for faster provisioning, single-node DB systems can be provisioned using Logical Volume Manager.
• Database territory: AMERICA
• Database unique name: The user-specified database name and a system-generated suffix, for example, dbtst_phx1cs.
• PDB admin name: pdbuser

For more information, see:

• Create a DB System
• Automatic Storage Management

Use a Backup to Create the Database

When creating a new DB system using a backup stored in the Recovery Service or Object Storage as the source of the database, you have the following options:

• Daily automatic backup. Requires that you have automatic backups enabled and an available backup to use. If you are creating a database from an automatic backup, you can choose any level 0 weekly backup, or a level 1 incremental backup created after the most recent level 0 backup.
• On-demand full backup.
• Standalone backup.
• Last archived redo log backup. Requires that you have automatic backups enabled. This backup combines data from the most recent daily automatic backup and data from archived redo logs, and represents the most current backup available. The time of the last archived redo log backup is visible on the Database Details page in the Last backup time field.

• Point-in-time out of place restore. Specify a timestamp to create a new copy of the database that included data up to a specified point in time. The timestamp must be earlier or equal to the Last backup time time displayed on the Database Details page. Note the following limitations when performing a point-in-time out of place restore:

  • The timestamp must be within the recovery window of the database.
  • The timestamp must be available within the database incarnation of the available automatic backups.
  • The timestamp cannot fall within two overlapping database incarnations.
  • The create database operation will fail if the database has undergone structural changes since the specified timestamp. Structural changes include operations such as creating or dropping a tablespace.
  • The create database operation cannot be started if another point-in-time database copy operation is in progress.

For more information, see Back Up a Database Using the Console.

Custom IP Addresses for the DB Systems

When creating a new single-node DB system or cloning an existing DB system, you can optionally define the IP address of the DB system being provisioned. This is useful in development contexts where you create and delete the same DB system over and over, and you need each new iteration of the DB system to use the same IP address.

Note: This facility is not available when creating a multi-node RAC DB system.