Introduction

Produces a fixed total number of records to a Kafka topic. The records are distributed evenly across the attack duration — for example, 100 records over 10 seconds produces approximately 10 records/second. The effective rate is: numberOfRecords / duration.

For rate-based production (records per second), use Produce Periodically instead.

Use Cases

• Generate a known number of messages to create consumer lag, then verify consumers can catch up.
• Produce test messages to a topic while injecting faults, and verify delivery success rate.
• Inject a burst of messages to test consumer throughput under load.

Parameters

Parameter	Description	Default
Duration	How long the producer runs. The total number of records is distributed evenly across this duration.	10s
Number of records	Total number of records to produce across the entire duration. The production rate is this value divided by the duration.
Record Key	Optional. The Kafka record key. Determines which partition receives the record based on the topic's partitioning strategy (default is round-robin when no key is set).
Record Value	Required. The message body to produce. Every record uses this same value.
Record Headers	Optional. Kafka record headers to attach to each produced record. Specified as key-value pairs.
Success Rate	Minimum percentage of records that must be produced without errors for the action to succeed. If the actual success rate drops below this threshold, the experiment fails.	100%
MaxConcurrent	Maximum number of parallel produce requests in flight at once. Min: 1, Max: 10.	5