Message Structure in Danube
A message in Danube represents the fundamental unit of data transmission between producers and consumers. Each message contains both the payload and associated metadata for proper routing and processing.
StreamMessage Structure
Core Fields
- request_id (u64): Unique identifier for tracking the message request
- msg_id (MessageID): Complex identifier containing routing and location information
- payload (Vec): The actual message content
- publish_time (u64): Timestamp when the message was published
- producer_name (String): Name of the producer that sent the message
- subscription_name (String): Name of the subscription for consumer acknowledgment routing
- attributes (HashMap
MessageID Fields
- producer_id (u64): Unique identifier for the producer within a topic
- topic_name (String): Name of the topic the message belongs to
- broker_addr (String): Address of the broker handling the message
- segment_id (u64): Unique identifier for the topic segment
- segment_offset (u64): Message position within the segment
Usage
Producer Perspective
Producers create messages by setting the payload and optional attributes. The system automatically generates and manages other fields like request_id, msg_id, and publish_time.
Consumer Perspective
Consumers receive the complete StreamMessage structure, providing access to both the message payload and all associated metadata for processing and acknowledgment handling.
Message Routing
The MessageID structure enables efficient message routing and acknowledgment handling across the Danube messaging system, ensuring messages reach their intended destinations and can be properly tracked.