camel-oracle-database-source-kafka-connector source configuration
Connector Description: Query data from an Oracle Database. In your Pipe file, you must explicitly declare the Oracle Database driver dependency in spec->integration->dependencies. - "mvn:com.oracle.database.jdbc:ojdbc11:<version>"
When using camel-oracle-database-source-kafka-connector as source make sure to use the following Maven dependency to have support for the connector:
<dependency>
<groupId>org.apache.camel.kafkaconnector</groupId>
<artifactId>camel-oracle-database-source-kafka-connector</artifactId>
<version>x.x.x</version>
<!-- use the same version as your Camel Kafka connector version -->
</dependency>
To use this source connector in Kafka connect you’ll need to set the following connector.class
connector.class=org.apache.camel.kafkaconnector.oracledatabasesource.CamelOracledatabasesourceSourceConnector
The camel-oracle-database-source source connector supports 7 options, which are listed below.
Name | Description | Default | Priority |
---|---|---|---|
Required The server name for the data source. Example: localhost. | HIGH | ||
The server port for the data source. | "1521" | MEDIUM | |
Required The username to access a secured Oracle Database. | HIGH | ||
Required The password to access a secured Oracle Database. | HIGH | ||
Required The query to execute against the Oracle Database. Example: INSERT INTO accounts (username,city) VALUES (:#username,:#city). | HIGH | ||
Required The name of the Oracle Database. | HIGH | ||
A query to run on a tuple consumed. Example: DELETE FROM accounts where user_id = :#user_id. | MEDIUM |
The camel-oracle-database-source source connector has no converters out of the box.
The camel-oracle-database-source source connector has no transforms out of the box.
The camel-oracle-database-source source connector has no aggregation strategies out of the box.