IBM MQ Sink Connector for Confluent Platform
The Kafka Connect IBM MQ Sink Connector is used to move messages from Kafka to
an IBM MQ cluster.
Features
At least once delivery
This connector guarantees that records are delivered at least once to the Kafka
topic. If the connector restarts, there may be some duplicate
records in the Kafka topic.
Prerequisites
The following are required to run the Kafka Connect IBM MQ Sink Connector:
- Kafka Broker: Confluent Platform 3.3.0 or above, or Kafka 0.11.0 or above
- Kafka Connect: Confluent Platform 4.1.0 or above, or Kafka 1.1.0 or above (requires header support in Connect)
- IBM MQ 8.0.0
or above, or IBM MQ on Cloud service
- IBM MQ Client Jar (See Client Jars section)
- Java 1.8
Install IBM MQ Sink Connector
You can install this connector by using the Confluent Hub client installation
instructions or by
manually downloading the ZIP file.
Prerequisites
Note
You must install the connector on every machine where Connect will run.
An install of the Confluent Hub Client.
Note
This is installed by default with Confluent Enterprise.
An install of the latest (latest) connector version.
To install the latest connector version, navigate to your Confluent Platform
installation directory and run the following command:
confluent-hub install confluentinc/kafka-connect-ibmmq-sink:latest
You can install a specific version by replacing latest with a version
number as shown in the following example:
confluent-hub install confluentinc/kafka-connect-ibmmq-sink:1.0.0-preview
Important
This connector does not include the IBM MQ client JAR. See the
IBM MQ Client Library section for more details.
IBM MQ Client Library
The Kafka Connect IBM MQ connector does not come with the IBM MQ client
library.
If you are running a multi-node Connect cluster, the IBM MQ connector and IBM
MQ client JAR must be installed on every Connect worker in the cluster.
See below for details.
Installing IBM MQ Client Library
This connector relies on a provided com.ibm.mq.allclient client JAR
distributed by IBM. The connector will not run if you have not installed the
JAR on each Connect worker node.
The installation steps are:
- Follow IBM’s guide on Obtaining the IBM MQ classes for JMS separately
to download the IBM MQ client JAR. #. The installation should have created a
wmq/JavaSE directory. From this directory, copy only the
com.ibm.mq.allclient.jar file into the
share/java/kafka-connect-ibmmq-sink directory of your Confluent Platform installation on
each worker node.
- Restart all of the Connect worker nodes.
Note
The share/java/kafka-connect-ibmmq-sink directory mentioned above is for
Confluent Platform. If you are using a different installation, find the location of the
Confluent IBM MQ sink connector JAR files and place the IBM MQ client JAR
file into the same directory.
Forwarding Kafka Properties to JMS
The connector can be configured to forward various values from the Kafka record
to the JMS Message.
- Enable
jms.forward.kafka.key to convert the record’s key to a String and
forward it as the JMSCorrelationID.
- Enable
jms.forward.kafka.metadata to forward the record’s topic,
partition, and offset on JMS Message properties.
- Kafka topic is applied to the message as a String property named
KAFKA_TOPIC.
- Partition is applied to the message as an Int property named
KAFKA_PARTITION.
- Offset is applied to the message as a Long property named
KAFKA_OFFSET.
- Enable
jms.forward.kafka.headers to add each header from the SinkRecord to
the JMS Message as a String property.
Note
The connector converts the record key and headers to a sensible string
representation that is similar to the JSON representation, with the
exception of simple string values (not in objects or arrays) which are
unquoted. No other conversion is done to the key and headers before
forwarding them on the JMS Message. If another format is needed,
out-of-the-box or custom Single Message
Transformation can be used with the connector
to transform the record keys and/or headers to the desired string
representation before the JMS sink connector processes each record.
Quick Start
This quick start uses the IBM MQ Sink Connector to consume records from Kafka and
send them to an IBM MQ broker running in a Docker container.
Start the IBM MQ broker.
docker run -d \
-p 1414:1414 -p 9443:9443 \
-e LICENSE=accept \
ibmcom/mq:9.1.2.0
Install the connector through the Confluent Hub Client.
# run from your Confluent Platform installation directory
confluent-hub install confluentinc/kafka-connect-ibmmq-sink:latest
Start Confluent Platform.
Tip
The command syntax for the Confluent CLI development commands changed in 5.3.0.
These commands have been moved to confluent local. For example, the syntax for confluent start is now
confluent local services start. For more information, see confluent local.
confluent local services start
Produce
test data to the sink-messages topic in Kafka.
seq 10 | confluent local services kafka produce sink-messages
Create a ibmmq-sink.json file with the following contents:
{
"name": "IbmMqSinkConnector",
"config": {
"connector.class": "io.confluent.connect.jms.IbmMqSinkConnector",
"tasks.max": "1",
"topics": "sink-messages",
"mq.username": "app",
"mq.channel": "DEV.APP.SVRCONN",
"mq.hostname": "localhost",
"mq.port": "1414",
"mq.password": "",
"mq.queue.manager": "mq",
"mq.transport.type": "client",
"jms.destination.type": "queue",
"jms.destination.name": "DEV.QUEUE.1",
"value.converter": "org.apache.kafka.connect.storage.StringConverter",
"key.converter": "org.apache.kafka.connect.storage.StringConverter",
"confluent.topic.replication.factor": "1",
"confluent.topic.bootstrap.servers": "localhost:9091"
}
}
Load the IBM MQ Sink Connector.
Caution
You must include a double dash (--) between the topic name and your flag. For more information,
see this post.
confluent local services connect connector load mq --config ibmmq-sink.json
Important
Don’t use the Confluent CLI in production
environments.
Confirm that the connector is in a RUNNING state.
confluent local services connect connector status mq
Navigate to the IBM MQ Console to
confirm the messages were delivered to the DEV.QUEUE.1 queue.
Tip
The default credentials for the IBM MQ Console are admin/passw0rd