Consuming Messages

Karafka framework has a long-running server process responsible for fetching and consuming messages. Consumers should inherit from the ApplicationConsumer. You need to define a #consume method that will execute your business logic code against a batch of messages. Karafka fetches and consumes messages in batches by default.

Consuming Messages in Batches¶

Data fetched from Kafka is accessible using the #messages method. The returned object is an enumerable containing received data and additional information that can be useful during the processing.

To start the Karafka server process, use the following CLI command:
```
bundle exec karafka server
```

To access the message batch, use the #messages method:


  class EventsConsumer < ApplicationConsumer
  def consume
    # Access the batch via messages method
    batch = messages
  end
end

Select one of two processing approaches based on your use case:
- Process each message one by one
- Process all payloads together to leverage batch database operations provided by many ORMs

Access message payloads.

For individual message iteration, use the #payload method available for each received message:

class EventsConsumer < ApplicationConsumer
  def consume
    # Print all the payloads one after another
    messages.each do |message|
      puts message.payload
    end
  end
end

For bulk operations, use the #payloads method to access all payloads at once:

class EventsConsumer < ApplicationConsumer
  def consume
    # Insert all the events at once with a single query
    Event.insert_all messages.payloads
  end
end

Consuming Messages One At a Time¶

While batch processing is recommended to leverage in-memory computation and batch database operations provided by many ORMs, you may need to process messages individually for certain use cases.

To start the Karafka server process, use the following CLI command:
```
bundle exec karafka server
```

Define a reusable base consumer that handles the single-message iteration pattern:

class SingleMessageBaseConsumer < Karafka::BaseConsumer
  attr_reader :message

  def consume
    messages.each do |message|
      @message = message
      consume_one

      mark_as_consumed(message)
    end
  end
end

class Consumer < SingleMessageBaseConsumer
  def consume_one
    puts "I received following message: #{message.payload}"
  end
end

Result: The #consume_one method will be called for each message in the batch, allowing you to process messages individually while maintaining the benefits of Karafka's batch fetching.

Accessing Topic Details¶

If your logic depends on specific routing details, you can access them from the consumer, using the #topic method.

Use Case

You could use it, for example, when you want to perform a different logic within a single consumer based on the topic from which your messages come.

To access the topic details, call the #topic method within the consume method:

class UsersConsumer < ApplicationConsumer
  def consume
    send(:"topic_#{topic.name}")
  end

  def topic_a
    # do something
  end

  def topic_b
    # do something else if it's a "b" topic
  end
end

To extract all the details that are stored in the topic at once, use the #to_h method:

class UsersConsumer < ApplicationConsumer
  def consume
    puts topic.to_h #=> { name: 'x', ... }
  end
end

Setting Initial Offset Position¶

By default, Karafka starts consuming messages from the earliest available offset. Use this procedure to configure the initial offset position for your consumers.

To configure the initial offset globally:

Open your Karafka application configuration file.

Set the initial_offset value in the setup block.

To start from the earliest offset (default behavior):

# This will start from the earliest (default)
class KarafkaApp < Karafka::App
  setup do |config|
    config.initial_offset = 'earliest'
  end
end

To start from the latest offset:

# This will make Karafka start consuming from the latest message on a given topic
class KarafkaApp < Karafka::App
  setup do |config|
    config.initial_offset = 'latest'
  end
end

Result: All topics will use this offset position as the default.

To configure the initial offset for specific topics:

Open your Karafka routing configuration.

Add the initial_offset setting to individual topic definitions:

class KarafkaApp < Karafka::App
  routes.draw do
    topic :events do
      consumer EventsConsumer
      # Start from earliest for this specific topic
      initial_offset 'earliest'
    end

    topic :notifications do
      consumer NotificationsConsumer
      # Start from latest for this specific topic
      initial_offset 'latest'
    end
  end
end

Result: Each topic will use its configured offset position, overriding the global default.

Note

This setting applies only to the first execution of a Karafka process. All following executions will pick up from the last offset where the process ended previously.

Detecting Revocation Midway¶

When working with a distributed system like Kafka, topic partitions can be distributed among different consumers in a consumer group for processing. However, there are cases where a partition needs to be removed from one consumer and reassigned to another. This process is known as a partition revocation.

Partition revocation can be voluntary, where a consumer willingly gives up the partition after processsing the current batch, or it can be involuntary. Involuntary partition revocation usually happens due to events such as a rebalance triggered by changes in the consumer group or a failure of a consumer that makes it unresponsive. It is important to remember that involuntary revocations can occur during data processing. if you are aware that a partition has been removed, you may not want to continue processing messages. This is where the #revoked? method is beneficial.

By monitoring the status of the #revoked? method, your application can detect that your process no longer owns a partition you are operating on. In such cases, you can choose to stop any ongoing, expensive processing. This can help you save resources and reduce the number of potential reprocessings.

As shown in the following example, you can check for revocation after processing each message:

def consume
  messages.each do |message|
    Message.create!(message)

    mark_as_consumed(message)

    return if revoked?
  end
end

It is worth noting, however, that under normal operating conditions, Karafka will complete all ongoing processing before a rebalance occurs. This includes finishing the processing of all messages already fetched. Karafka has built-in mechanisms to handle voluntary partition revocations and rebalances, ensuring that no messages are lost or unprocessed during such events. Hence, #revoked? is especially useful for involuntary revocations.

In most cases, especially if you do not use Long-Running Jobs, the Karafka default offset management strategy should be more than enough. It ensures that, after batch processing and upon rebalances, all offsets are committed before partition reassignment. In a healthy system with stable deployment procedures and without frequent short-lived consumer generations, the number of re-processings should be close to zero.

Note

The #revoked? method detects partition revocation immediately. You don't need to mark messages as consumed for it to detect revocation.

Note

With Long-Running Jobs, #revoked? result also changes independently from marking messages.

Consumer Persistence¶

Karafka consumer instances are persistent by default. A single consumer instance will "live" as long as a given process consumes a given topic partition. This allows you to:

Maintain database connections across batches
Keep in-memory state and caches
Buffer messages for batch processing
Reuse expensive resources

Karafka recreates the consumer instance only when a partition is lost and reassigned.

Note

When buffering messages in memory, use manual offset management. Without it, you'll lose buffered data, if the process crashes before flushing.

The following example contains a consumer that buffers messages until it reaches 1,000 of them before flushing:

# A consumer that will buffer messages in memory until it reaches 1000 of them. Then it will flush
# and commit the offset.
class EventsConsumer < ApplicationConsumer
  # Flush every 1000 messages
  MAX_BUFFER_SIZE = 1_000

  def initialized
    @buffer = []
  end

  def consume
    # Print all the payloads one after another
    @buffer += messages.payloads

    return if @buffer.size < MAX_BUFFER_SIZE

    flush
  end

  private

  def flush
    Event.insert_all @buffer

    mark_as_consumed @buffer.last

    @buffer.clear!
  end
end

Shutdown and Partition Revocation Handlers¶

Karafka consumer, aside from the #consume method, allows you to define two additional methods to free resources used during specific events:

#revoked - it will be executed when there is a rebalance resulting in the given partition being revoked from the current process.
#shutdown - it will be executed when the Karafka process is being shutdown.

The following code demonstrates all three lifecycle methods:

class LogsConsumer < ApplicationConsumer
  def initialized
    @log = File.open('log.txt', 'a')
  end

  def consume
    messages.each do |message|
      @log << message.raw_payload
    end
  end

  def shutdown
    @log.close
  end

  def revoked
    @log.close
  end
end

Shutdown Edge Case Alert

When you use #shutdown with the filtering API or Delayed Topics, there are scenarios where #shutdown and #revoked may be invoked without prior #consume running and the #messages batch may be empty.

Initial State Setup¶

Karafka consumers provide a dedicated #initialized method called automatically after the consumer instance is fully prepared and initialized.

Use this method to set up any additional state, resources, or connections your consumer may need during its lifecycle. Karafka's consumer instance is not entirely bootstrapped during the #initialize method. This means crucial details, like routing information, topic details, and more, may not yet be available. Using #initialize to set up dependencies might result in incomplete or incorrect configurations. On the other hand, #initialized is executed once the consumer is fully ready and contains all the details it might need. By default, #initialized does nothing. Still, you can override it to include custom setup logic for your consumer.

The following example shows two methods on how to override the #initialized" method in a Karafka consumer to set up resources after the consumer is fully ready.

class EventsConsumer < ApplicationConsumer
  def initialized
    # Any setup logic you want to perform once the consumer is fully ready
    @connection = establish_db_connection
    puts "Consumer is initialized with topic: #{topic.name}"
  end

  def consume
    messages.each do |message|
      # Process messages using the setup done in #initialized
      puts message.payload
    end
  end

  private

  def establish_db_connection
    # Custom logic to establish a database connection
  end
end

Using #initialized provides access to the whole consumer context, as it is called after the consumer has been fully set up. This offers several benefits, such as establishing database connections, setting up loggers, or initializing API clients that require topic-specific information. By deferring resource setup to #initialized, you avoid potential issues that can arise when specific resources or states are unavailable during the construction phase.

Early Message Yielding (`enable.partition.eof`)¶

In typical Karafka consumption scenarios, when a consumer reaches the end of a partition, it might still wait for new messages to arrive. This behavior is governed by settings such as max_wait_time or max_messages, which dictate how long a consumer should wait for new data before the polling operation completes or returns. While this can benefit continuous data streams, it may introduce unnecessary latency in scenarios where real-time data processing and responsiveness are critical.

The enable.partition.eof configuration option changes how Karafka responds when the end of a partition is reached during message consumption. By default, when Karafka reaches the end of a partition, it waits for additional messages until either max_wait_time or max_messages is reached. However, if enable.partition.eof is set for a subscription group to true, Karafka will immediately delegate already accumulated messages (if any) for processing, even if neither max_wait_time nor max_messages has been reached.

Benefits of Early Yield¶

Reduced Latency: Immediate message yielding upon reaching the end of a partition can significantly reduce latency. This is particularly beneficial in environments where data must be processed and acted upon quickly.
Increased Responsiveness: Systems that require high responsiveness will benefit from not having to wait for the timeout conditions (max_wait_time or max_messages) to be met, allowing subsequent processing steps to commence without delay.
Efficient Resource Utilization: By avoiding unnecessary waiting times, system resources can be better utilized for processing rather than idling, potentially leading to cost optimizations and improved throughput.

Downsides of Early Yield¶

Potential for Increased CPU Usage: In highly active systems where new messages are frequently published, constantly checking for the end of partition could lead to increased CPU utilization. This is because the system needs to manage and check state transitions more frequently.
Complexity in Batch Processing: For applications that are optimized for batch processing, this setting might disrupt the batching logic, as messages could be processed in smaller batches, potentially leading to inefficiencies.

Configuring `enable.partition.eof`¶

The enable.partition.eof is one of the kafka-scoped options. Depending on your use case, you can configure enable.partition.eof either globally for all subscription groups in the setup block, or on a per-subscription-group basis in your routing configuration, as shown in the following example:

class KarafkaApp < Karafka::App
  setup do |config|
    config.client_id = "my_application-#{Process.pid}-#{Socket.gethostname}"

    config.kafka = {
      'bootstrap.servers': '127.0.0.1:9092',
      'enable.partition.eof': true
    }
  end

  # You can also do it per topics in a subscription group
  routes.draw do
    subscription_group :fast do
      topic 'events' do
        consumer EventsConsumer
        kafka(
          'bootstrap.servers': '127.0.0.1:9092',
          'enable.partition.eof': true
        )
      end
    end
  end
end

This configuration ensures that as soon as the end of a partition is reached, any accumulated messages are immediately processed, enhancing the system responsiveness and efficiency.

Consuming with the Iterator API¶

Karafka Pro provides the Iterator API that allows you to subscribe to topics and to perform lookups from Rake tasks, custom scripts, Rails console, or any other Ruby processes.

The following example demonstrates searching for messages with a specific header value:

# Note: you still need to configure your settings using `karafka.rb`

# Select all the events of user with id 5 from last 10 000 messages of
# each partition of the topic `users_events`

user_5_events = []

iterator = ::Karafka::Pro::Iterator.new(
  { 'users_events' => -1000 }
)

iterator.each do |message|
  # Cast to integer because headers are always strings or arrays of strings
  next unless message.headers['user-id'].to_i == 5

  user_5_events << message
end

puts "There were #{user_5_events.count} messages"

For more details on this feature, see Iterator API.

Avoiding Accidental Overwriting of Consumer Instance Variables¶

When working with Karafka consumers, it is essential to be mindful of certain instance variables used by the consumer instances. Unintentionally overwriting these variables can lead to critical processing errors and result in issues, such as worker.process.error, which can be seen in the Karafka Web UI. The following are the primary instance variables that you should be careful about:

@id: Consumer instance identifier
@messages: Messages batch for the subscribed topic
@client: Kafka connection client
@coordinator: Message processing coordinator
@producer: Producer instance
@used: Internal flag tracking whether the consumer has actively processed messages

Accidental overwriting of any instance variables can disrupt the normal functioning of the consumer, leading to:

Inability to correctly process or retrieve messages.
Loss of connection to the Kafka server.
Failure in coordinating message processing.
Inability to produce messages properly.

Such disruptions often manifest as "worker.process.error" in the web UI, indicating critical processing failures.

Reaching the End of a Partition (EOF)¶

Karafka includes dedicated handling for end-of-partition (EOF) scenarios, allowing you to execute specific logic when the end of a partition is reached. For this feature to work, you must enable the enable.partition.eof kafka setting in your configuration.

Enabling EOF Handling¶

To use EOF features, ensure that both the enable.partition.eof option and the eofed setting are configured properly:

class KarafkaApp < Karafka::App
  setup do |config|
    config.kafka = {
      'bootstrap.servers': '127.0.0.1:9092',
      'enable.partition.eof': true
    }
  end

  routes.draw do
    topic 'events' do
      consumer EventsConsumer
      # Ensure EOF handling is activated
      eofed true
    end
  end
end

Implementing EOF Handling¶

EOF signaling can happen in two ways:

Via #eofed Method: This method is triggered when no more messages are polled.
Alongside #consume Method: EOF can be signaled together with messages if some messages were polled.

Full Coverage of EOF

To ensure full coverage of EOF scenarios, use both the #eofed method and the #eofed? method. This ensures that EOF is handled whether it occurs with or without new messages.

`#eofed` Method¶

Define the #eofed method in your consumer to handle cases where no more messages are polled alongside the EOF information:

class EventsConsumer < ApplicationConsumer
  def consume
    messages.each do |message|
      # Process each message
      puts message.payload
    end

    # Check if EOF was signaled alongside messages
    if eofed?
      puts "Reached the end of the partition with messages."
      # Implement any additional logic needed when EOF is reached with messages
    end
  end

  def eofed
    puts "Reached the end of the partition with no more messages."
    # Implement any additional logic needed when EOF is reached
  end
end

Handling EOF in `#consume` Method¶

If EOF is signaled together with messages, the #eofed method will not be triggered. In such cases, Karafka offers the #eofed? method, which allows you to detect when EOF has been signaled along with the messages.

The #eofed? method allows you to detect EOF within the #consume method:

class EventsConsumer < ApplicationConsumer
  def consume
    messages.each do |message|
      # Process each message
      puts message.payload
    end

    # Check if EOF was signaled alongside messages
    if eofed?
      puts "Reached the end of the partition with messages."
      # Implement any additional logic needed when EOF is reached with messages
    end
  end
end

Use Cases for EOF Handling¶

Knowing when a partition has reached EOF can be helpful in several scenarios:

Batch Processing Completion: When processing data in batches, knowing when you have processed all available data allows you to finalize batch operations, such as committing transactions or aggregating results.
Data Synchronization: When synchronizing data between different systems, you can use EOF as a signal that all current data has been consumed and it is safe to start a new synchronization cycle.
Resource Cleanup: After reaching the end of a partition, you may want to release or reallocate resources that are no longer needed, optimizing your application performance.
Logging and Monitoring: EOF events help track data consumption and detect when no more messages are available to process, aiding in debugging and performance tuning.
Triggering Downstream Processes: EOF signals when to start processes that require all messages to be consumed first.

Wrapping the Execution Flow¶

The #wrap method allows you to execute custom logic before and after message processing. Use this when you need additional setup, teardown, or contextual operations (such as selecting a transactional producer from WaterDrop connection pool.

The #wrap method encompasses the consumer's entire execution flow, not just user-defined business logic. This includes:

User-Defined Logic: The custom message processing logic implemented in all actions such as #consume, #revoked, etc. methods.
Framework-Level Operations: Core functionalities such as offset management, message acknowledgment, and internal state synchronization.
Error Handling and Recovery: Ensures proper transactional rollbacks or retries in case of failures.

To implement #wrap, override it in your consumer class. The method ensures that yield is always invoked, regardless of any failures or conditions. This is critical, because skipping yield can disrupt Karafka ability to execute its internal processes, leading to inconsistencies or data loss.

In the following example, #wrap implementation is shown:

class CustomConsumer < ApplicationConsumer
  def consume
    # This will cause a backoff if no producer was available
    raise @wrap_error if @wrap_error

    # Your logic here
  end

  def wrap(_action_name)
    default_producer = producer

    begin
      # Attempt to select a producer from WaterDrop's connection pool
      WaterDrop::ConnectionPool.with do |transactional_producer|
        self.producer = transactional_producer
        yield
      end
    rescue ConnectionPool::TimeoutError => e
      # Handle scenarios where a producer isn't available
      @wrap_error = e
      yield # Ensure framework operations still execute
    ensure
      @wrap_error = false
      # Restore the original producer after execution
      self.producer = default_producer
    end
  end
end

Consuming Messages

Consuming Messages in Batches¶

Consuming Messages One At a Time¶

Accessing Topic Details¶

Setting Initial Offset Position¶

Detecting Revocation Midway¶

Consumer Persistence¶

Shutdown and Partition Revocation Handlers¶

Initial State Setup¶

Early Message Yielding (enable.partition.eof)¶

Benefits of Early Yield¶

Downsides of Early Yield¶

Configuring enable.partition.eof¶

Consuming with the Iterator API¶

Avoiding Accidental Overwriting of Consumer Instance Variables¶

Reaching the End of a Partition (EOF)¶

Enabling EOF Handling¶

Implementing EOF Handling¶

#eofed Method¶

Handling EOF in #consume Method¶

Use Cases for EOF Handling¶

Wrapping the Execution Flow¶

See Also¶

Early Message Yielding (`enable.partition.eof`)¶

Configuring `enable.partition.eof`¶

`#eofed` Method¶

Handling EOF in `#consume` Method¶