A newer version of Hazelcast Platform is available.

View latest

CREATE DATA CONNECTION

The CREATE DATA CONNECTION statement creates the metadata required to connect to an external system for later use.

Include data connections in mapping statements to allow the SQL service to connect to, read from, write to, or query data from your external systems. To simplify setup, you can use the same data connection in multiple SQL mappings.

Syntax Summary

This code block is a quick reference with all the parameters that you can use with the CREATE DATA CONNECTION statement.

See some practical examples.

CREATE [OR REPLACE] DATA CONNECTION [IF NOT EXISTS] data_connection_name
TYPE connection_type
[[NOT] SHARED]
OPTIONS ('option_name' = 'option_value' [, ...] );

Parameters

The CREATE DATA CONNECTION statement accepts the following parameters.

You must provide the following:

  • data_connection_name

  • connection_type

Parameter Description

OR REPLACE

Overwrite an existing data connection of the same name, if it exists.

If a data connection of the same name has been configured programmatically or in the configuration files of Hazelcast members, then the data connection will not be overwritten. Configured data connections always take precedence.

IF NOT EXISTS

Create a new data connection only if a connection of the same name does not already exist.

data_connection_name

Unique name for the data connection.

TYPE connection_type

The type of data connection required to connect to the external system. The following types of connection are supported: JDBC, Kafka, Mongo (Beta), HZ. All connection types are case insensitive.

NOT SHARED

Every time you issue a query against a SQL mapping, a new physical connection to the external system is created.

SHARED (default)

A reusable data connection. See the implementation of each data connection type for full details of reusability: HazelcastDataConnection, KafkaDataConnection, MongoDataConnection.

OPTIONS

Configuration options for the chosen connection_type. See examples and Configuring Data Connections to External Systems for valid parameters for specific connections.

Dependency Management

Replacing a data connection will not affect any queries that are already running. Only the following queries or jobs are affected by the changes:

  • New queries on existing mappings

  • Queries or jobs that are restarted after a data connection is updated

Permissions

Enterprise

If security is enabled, you can grant or deny permission for certain clients to use this statement. See SQL Permissions.

Examples

Create a Kafka Data Connection

CREATE DATA CONNECTION IF NOT EXISTS myKafkaSource
TYPE Kafka
SHARED
OPTIONS (
    'bootstrap.servers' = '127.0.0.1:9092');

Create a JDBC Data Connection

CREATE DATA CONNECTION IF NOT EXISTS myDatabase
TYPE JDBC
SHARED
OPTIONS (
  'jdbcUrl'='jdbc:mysql://dummy:3306',
  'user'='xyz',
  'password'='xyz');

For a full list of configuration options, see HikariCP configuration.

Create a MongoDB Data Connection

CREATE DATA CONNECTION myMongoDBSource
TYPE Mongo
SHARED
OPTIONS (
    'connectionString'='mongodb://mongodb0.example.com:27017',
    'database'='myDatabaseName');

Update a Data Connection

If you want to overwrite an existing data connection, use the CREATE OR REPLACE DATA CONNECTION statement.

CREATE OR REPLACE DATA CONNECTION myDatabase
TYPE JDBC
SHARED
OPTIONS (
  'jdbcUrl'='jdbc:mysql://dummy:3306',
  'user'='xyz',
  'password'='xyz');