Confluent Certified Developer for Apache Kafka Certification Examination Sample Questions:

1. (You are implementing a Kafka Streams application to process financial transactions.
Each transaction must be processed exactly once to ensure accuracy.
The application reads from an input topic, performs computations, and writes results to an output topic.
During testing, you notice duplicate entries in the output topic, which violates the exactly-once processing requirement.
You need to ensure exactly-once semantics (EOS) for this Kafka Streams application.
Which step should you take?)

A) Set enable.idempotence=true in the internal producer configuration of the Kafka Streams application.
B) Set processing.guarantee=exactly_once_v2 in the Kafka Streams configuration.
C) Set enable.exactly_once=true in the Kafka Streams configuration.
D) Enable compaction on the output topic to handle duplicates.

2. You are developing a Java application using a Kafka consumer.
You need to integrate Kafka's client logs with your own application's logs using log4j2.
Which Java library dependency must you include in your project?

A) Just the log4j2 dependency of the application
B) None, the right dependency will be added by the Kafka client dependency by transitivity.
C) SLF4J implementation for Log4j2 (org.apache.logging.log4j:log4j-slf4j-impl)
D) SLF4J implementation for Log4j 1.2 (org.slf4j:slf4j-log4j12)

3. (Your application consumes from a topic configured with a deserializer.
You want the application to be resilient to badly formatted records (poison pills).
You surround the poll() call with a try/catch block for RecordDeserializationException.
You need to log the bad record, skip it, and continue processing other records.
Which action should you take in the catch block?)

A) Throw a runtime exception to trigger a restart of the application.
B) Log the bad record and call consumer.skip() method.
C) Log the bad record; no other action is needed.
D) Log the bad record and seek the consumer to the offset of the next record.

4. (You have a Kafka Connect cluster with multiple connectors deployed.
One connector is not working as expected.
You need to find logs related to that specific connector to investigate the issue.
How can you find the connector's logs?)

A) Modify the log4j.properties file to add a dedicated log appender for the connector.
B) Make no change; there is no way to isolate connector logs.
C) Modify the log4j.properties file to enable connector context.
D) Change the log level to DEBUG to include connector context information.

5. You have a consumer group with default configuration settings reading messages from your Kafka cluster.
You need to optimize throughput so the consumer group processes more messages in the same amount of time.
Which change should you make?

A) Decrease the session timeout of each consumer.
B) Remove some consumers from the consumer group.
C) Increase the number of bytes the consumers read with each fetch request.
D) Disable auto commit and have the consumers manually commit offsets.

Solutions: