Considerations and Support Information

Oracle Cloud Infrastructure Data Integration is integrated with Oracle Cloud Infrastructure Vault to enable you to manage sensitive information using vaults, keys, and secrets.

A vault is a container for keys and secrets. Secrets store credentials such as required passwords for connecting to data sources. You use an encryption key in a vault to encrypt and import secret contents to the vault. Secret contents are based64-encoded. Oracle Cloud Infrastructure Data Integration uses the same key to retrieve and decrypt secrets when creating a data asset and connecting to the data source.

For most data source types, you can use a secret in OCI Vault to store the password for the default connection in a data asset. To create a vault, key, and secret, see Overview of Vault.

When you create a data asset, you provide the OCID of the secret in the connection details, so you don't have to enter the actual password. To copy the secret OCID, see View Secret Details.

For Oracle Database, Oracle Autonomous Data Warehouse, and Oracle Autonomous Transaction Processing sources, you have the option to use secrets for the Oracle wallet and passwords instead of uploading the wallet and entering the wallet password when you create your data asset.

To use an Oracle wallet with secrets in OCI Vault, you must:

1. Provide a wallet password when you download the wallet.
2. Remove the .p12 file from the dowloaded wallet zip.
3. Use any base64 encoder to encode the modified wallet zip to base64.
4. Copy the base64-encoded data to a secret in a vault.
5. Create a secret for the wallet password.
6. Create a secret for the database password.

To use secrets in OCI Vault, ensure that you have the following policy:

allow any-user to read secret-bundles in compartment <compartment-name> where ALL {request.principal.type = 'disworkspace', request.principal.id = '<workspace-ocid>'}

The following policy enables a group of users who are not administrators to use secrets with Oracle Autonomous Data Warehouse and Oracle Autonomous Transaction Processing:

allow group <group-name> to read secret-bundles in compartment <compartment-name>

The following table lists the data sources that you can use with Data Integration.

Data Integration supports hierarchical data types in source and target data entities.

To read and write data with hierarchical data types, currently you can use only generic REST data assets and file storage data assets such as OCI Object Storage, Amazon S3, and HDFS. The JSON file format is supported by generic REST data assets. For file storage data assets, the following file formats are supported:

• JSON and multi-line JSON
• Avro
• Parquet

The supported hierarchical data types are the Array, Struct, and Map complex types. You can perform any type of file to file transformation, such as JSON to Avro or Avro to JSON.

Before using hierarchical data entities and complex types, ensure that you understand the supported capabilities in Data Integration when working with components and performing tasks.

{
   "CUST_ID":1333,
   "NAME":{
      "F_NAME":"Sam",
      "L_NAME":"Smith"
   },
   "EMAIL":[
      {
         "EMAILID":"abc@oracle.com",
         "EMAILtype":"work"
      },
      {
         "EMAILID":"abc@othermail.com",
         "EMAILtype":"personal"
      }
   ],
   "GENDER":"Male"
}

Data Integration supports the Unicode standard, which is a universal character encoding standard for written characters and text in any language. The Unicode standard provides a unique number for every character regardless of the platform, device, or application. For example, 0041 is the Unicode character for the English letter "A".

Data Integration supports Unicode characters (including multibyte characters) in data and metadata.

Unicode support in data means that the attributes and attribute values in your source and target data assets can include Unicode and multibyte characters. You can also enter Unicode and multibyte characters in expressions. For JSON and CSV Object Storage data assets, the default encoding is UTF-8, and you cannot change it.

Unicode support in metadata means that the data entity and schema names of your data assets can include Unicode and multibyte characters. You can also enter Unicode and multibyte characters for names and descriptions when working with objects in Data Integration.

In the Unicode standard, a unique number assigned to a Unicode character is a code point. Currently, Data Integration supports the following Unicode code points and range of code points:

Data types from the source and target systems you use are mapped to and mapped from a core set of generic data types in Oracle Cloud Infrastructure Data Integration.

In the set of generic data types, some of the types have additional length or scale properties that you can use to further constrain the data type.

The Expression operator in Data Integration does not yet support all the generic data types. You can create a new attribute based on a generic data type only if the generic type is supported.

Your network configurations depend on the source and target data assets you're using with the Data Integration service, and where the assets are located.

See the Understanding VCN Configuration for Data Integration blog to identify the options for your needs.

A workspace can have an attached Virtual Cloud Network (VCN). For data sources in a private network, create a VCN with at least one regional subnet. Only regional subnets are supported, and DNS hostnames must be used in the subnets. Depending on the location of your data sources, you might have to create other network objects such as service gateways, network security groups, and Network Address Translation (NAT) gateways.

In general, for data sources that are accessible from the internet:

• If a workspace has an attached VCN: Data Integration can connect directly through a Network Address Translation (NAT) gateway on the VCN of the workspace.
• If a workspace does not have an attached VCN: Data Integration can connect directly using public IP addresses.

Resources, such as workspaces, with private IPs defined in any subnet can access other private resources in different virtual cloud networks and regions through Service gateways or NAT gateways using local or remote peering gateways.

You can also combine gateways when you need to access both Object Storage and Autonomous Data Warehouse. For example, for public connectivity, you would need both a Service gateway for Object Storage and a NAT gateway for Autonomous Data Warehouse.

For data sources that are not accessible from the internet, other options include:

• Create a workspace with a private endpoint enabled, with the private endpoint in the same subnet as the data source.

• Use Oracle Cloud Infrastructure FastConnect.

• Use Oracle Cloud Infrastructure VPN Connect (also referred to as an IPSec VPN).

Note, however, that Oracle Cloud Infrastructure FastConnect and Oracle Cloud Infrastructure VPN Connect must be used when these sources are in private domains:

• Oracle on Amazon RDS
• MySQL on Amazon RDS
• Amazon RDS for SQL Server
• Microsoft Azure SQL Database

Oracle Cloud Infrastructure Networking documentation:

• Networking Overview
• VCNs and Subnets
• Network Security Groups
• Service Gateway
• VCN Peering
• FastConnect
• Network Path Analyzer
• Blog: Using Network Path Analyzer (troubleshoot, verify, and validate)