Manage Kafka Connect Connectors

This guide shows how to manage Kafka Connect connectors with Jikkou. For the full resource specification, see the KafkaConnector reference.

Before you begin

• One or more reachable Kafka Connect clusters.
• Each cluster declared under the kafkaConnect.clusters[] setting in your Jikkou configuration — see the Kafka Connect provider configuration.

1. Describe a connector

The kafka.jikkou.io/connect-cluster label selects which configured cluster the connector is created in.

file: kafka-connector-filestream-sink.yaml

---
apiVersion: "kafka.jikkou.io/v1beta1"
kind: "KafkaConnector"
metadata:
  name: "local-file-sink"
  labels:
    kafka.jikkou.io/connect-cluster: "my-connect-cluster"
spec:
  connectorClass: "FileStreamSink"
  tasksMax: 1
  config:
    file: "/tmp/test.sink.txt"
    topics: "connect-test"
  state: "running"

2. Preview and apply

jikkou apply --files ./kafka-connector-filestream-sink.yaml --dry-run   # review
jikkou apply --files ./kafka-connector-filestream-sink.yaml             # apply

3. Control the connector lifecycle

Set spec.state and re-apply to change the runtime state of a connector:

• running — connector and tasks are active (default).
• paused — message processing is paused until resumed.
• stopped — connector and tasks are shut down; the config is kept.

4. Inspect connector status

jikkou get kafkaconnectors --expand-status

The status.connectorStatus block reports the live state of the connector and its tasks as returned by the Kafka Connect REST API.

Related

• KafkaConnector reference
• Restart connectors action