CCAAK Confluent Certified Administrator for Apache Kafka Practice Questions
Prepare for CCAAK with more than an answer.
- Exam fee
- $150 USD
- Level
- Administrator
- Valid for
- 2 years
Domains covered on the exam 7
- Apache Kafka Fundamentals15%
- Apache Kafka Security15%
- Deployment Architecture12%
- Kafka Connect12%
- Apache Kafka Cluster Configuration22%
- Observability10%
- Troubleshooting15%
- 1
True or False: A single Kafka Connect worker running in standalone mode can execute tasks for multiple connectors simultaneously.
Show answer details
Correct answer: A
This is true. A single worker process, even in standalone mode, can be configured to run multiple connectors. Each connector will have its own configuration file specified on the command line, and the worker will instantiate and manage the tasks for all of them within its single process.
- 2
You are trying to produce a message larger than the default 1MB limit to a Kafka topic. You have correctly increased
max.request.sizeon the producer andmessage.max.byteson the topic configuration. However, the producer still receives aRecordTooLargeException. What additional broker configuration must be adjusted to allow the larger message?Show answer details
Correct answer: C
This broker-level configuration defines the maximum size of a message that follower brokers can replicate from the leader. If a producer sends a message larger than this value, the leader broker will accept it (assuming
message.max.bytesis set), but replication to followers will fail, ultimately causing the produce request to fail. This setting must be at least as large asmessage.max.bytes. - 3
An administrator needs to add a new broker to an existing Kafka cluster to increase capacity. After starting the new broker with the correct
broker.idandzookeeper.connectsettings, partitions are not being automatically moved to it. What is the reason for this behavior?Show answer details
Correct answer: A
By default, Kafka does not automatically rebalance existing partitions when a new broker is added to the cluster. The new broker will join the cluster and will be eligible to host replicas for any topics created after it joins, but existing data will not be moved to it automatically.
- 4
A financial data processing application requires strict ordering of transactions for each customer account. All transactions for a single account must be processed in the order they were generated. How should an administrator advise the development team to configure their producers and topics to achieve this?
Show answer details
Correct answer: B
This is the standard Kafka pattern for achieving per-entity ordering. The producer's default partitioner will hash the message key (the account ID) to consistently select the same partition for all messages with that key. Since Kafka guarantees order within a single partition, this ensures all transactions for a specific account are processed in order while allowing the topic to scale across multiple partitions.
- 5
Case Study
Company Background: A large e-commerce platform, ShopSphere, uses Kafka for its real-time inventory management system. During peak holiday sales, the system experiences significant performance degradation. The lead administrator is tasked with identifying and resolving the bottleneck.
Observed Symptoms:
- Producers writing to the
inventory-updatestopic experience high latency and frequentTimeoutExceptionerrors. - The topic has 24 partitions and a replication factor of 3.
min.insync.replicasis set to 2. - Monitoring dashboards show that one of the three brokers hosting replicas for this topic, Broker-103, has consistently high disk I/O wait times (avg. > 80%) and a shrinking In-Sync Replica (ISR) set for partitions on that topic.
- The other two brokers, Broker-101 and Broker-102, appear healthy with normal I/O wait times.
Troubleshooting Diagram:
sequenceDiagram participant P as Producer participant B1 as Broker-101 (Leader) participant B2 as Broker-102 (Follower) participant B3 as Broker-103 (Follower) P->>B1: Produce Request (acks=all) B1->>B2: Replicate Data B1->>B3: Replicate Data B2-->>B1: Ack Note right of B3: High I/O Wait B3--x B1: Ack Delayed / Fails Note left of P: Request Times OutGiven this information, what is the most immediate and effective action the administrator can take to restore producer performance while a long-term fix for Broker-103 is investigated?
Show answer details
Correct answer: C
The root cause is the performance of Broker-103, which is a follower for the topic's partitions. Since producers are using
acks=all, they must wait for acknowledgements from the ISR set, which includes the slow Broker-103. This is causing timeouts. The most effective immediate action is to move the problematic partitions off the unhealthy broker entirely. Usingkafka-reassign-partitions.shto generate and execute a plan will migrate the replicas from Broker-103 to other healthy brokers in the cluster. Once the reassignment is complete, Broker-103 will no longer be part of the ISR for this topic, and producer performance will be restored. Restarting the broker is temporary, and changingackscompromises data durability. - Producers writing to the
- 6
An administrator is investigating a Kafka cluster where producers are experiencing high latency and frequent timeouts. The topic in question has a replication factor of 3 and
min.insync.replicasis set to 2. Monitoring reveals that one of the three brokers hosting partitions for this topic is consistently exhibiting high disk I/O wait times, causing it to frequently fall out of the In-Sync Replica (ISR) set. What is the most effective initial action to mitigate the producer latency while ensuring data durability is not compromised below the configured minimum?Show answer details
Correct answer: C
This is the most effective initial step. By moving the partition leadership to healthy brokers, producers will no longer be dependent on the slow broker for acknowledging writes. This directly addresses the latency issue without changing durability settings, allowing the administrator to investigate the disk I/O problem on the affected broker without impacting production traffic.
- 7
A financial services company is deploying a new Kafka cluster using KRaft mode. The architecture calls for a dedicated controller quorum of 5 nodes for high availability. What is the correct value for the
controller.quorum.votersconfiguration property on each controller node?Show answer details
Correct answer: B
The correct format for
controller.quorum.votersis a comma-separated string where each entry specifies thenode.id, the hostname or IP address, and the controller listener port for each node in the quorum. For example:1@controller1:9093,2@controller2:9093,.... - 8
An administrator needs to enforce client quotas to prevent a single misbehaving application from monopolizing cluster resources. The goal is to limit the network bandwidth consumed by a specific user, 'app-user-1'. Which two properties must be configured to enforce a produce quota for this user? (Select TWO)
Show answer details
Correct answer: B, C
This property sets the byte-rate threshold for producers. When combined with the user principal, it defines the specific bandwidth limit for that user's produce requests.
This property specifies the user principal to which the quota will be applied. To target 'app-user-1', this must be set to that principal name.
- 9
True or False: When
unclean.leader.election.enableis set totruefor a topic, it is possible for messages that were not replicated to all in-sync replicas to be lost during a leader failover.Show answer details
Correct answer: A
This statement is true. The purpose of unclean leader election is to prioritize availability over consistency. If no in-sync replicas are available to become the new leader, this setting allows an out-of-sync replica to be elected. This new leader may not have all the messages the old leader had, leading to data loss.
- 10
A new administrator is tasked with setting up log compaction on a topic named
user-profiles. The goal is to retain only the most recent value for each user ID, which is used as the message key. After enabling compaction, the administrator observes that old records are not being removed, and the topic's log segments are growing indefinitely. Which configuration setting is the most likely cause of this issue?Show answer details
Correct answer: A
Log compaction works by retaining the latest message for each unique key. If messages are produced with null keys, Kafka cannot determine which records to compact. Messages with null keys are not eligible for compaction and will be retained until their segment is deleted by time or size-based retention, if also configured. This is the most common reason for compaction appearing not to work.
