Description
Summary
Implement an Iggy Source/Sink Connector that consumes data from an upstream Iggy cluster and publishes it into a downstream Iggy cluster.
This enables seamless cluster-to-cluster replication and provides the foundation for multi-region deployments, disaster recovery, active-active architectures, edge-to-cloud streaming, and hierarchical streaming topologies.
Unlike traditional application-level replication, this connector allows operators to replicate selected streams, partitions, or namespaces without requiring producers or consumers to change.
Motivation
Many production deployments require data to exist in multiple locations.
Today, applications would need to publish to multiple Iggy clusters directly, increasing complexity and coupling replication logic into business applications.
A native Iggy Source/Sink Connector would make replication a first-class capability by continuously consuming from one cluster and publishing to another while preserving ordering and delivery guarantees.
Producers
│
▼
+------------------+
| Iggy Cluster A |
| (Primary) |
+------------------+
│
Iggy Source/Sink Connector
│
▼
+------------------+
| Iggy Cluster B |
| (Replica) |
+------------------+
│
Local Consumers
The connector should behave like any other Iggy connector:
- Source = upstream Iggy
- Sink = downstream Iggy
Aspects to cover (ideal, not a comprehensive list, can be broken down into follow-up tasks after the basic version is in place):
- Iggy → Iggy replication
- configurable source streams
- configurable destination streams
- partition-aware replication
- offset checkpointing
- resumable replication
- configurable batching
- configurable compression
- TLS support
- authentication support
- metrics
- retry with backoff
- configurable parallelism
Affected area / component
Connectors
Proposed solution
No response
Alternatives considered
No response
Contribution
Good first issue
Description
Summary
Implement an Iggy Source/Sink Connector that consumes data from an upstream Iggy cluster and publishes it into a downstream Iggy cluster.
This enables seamless cluster-to-cluster replication and provides the foundation for multi-region deployments, disaster recovery, active-active architectures, edge-to-cloud streaming, and hierarchical streaming topologies.
Unlike traditional application-level replication, this connector allows operators to replicate selected streams, partitions, or namespaces without requiring producers or consumers to change.
Motivation
Many production deployments require data to exist in multiple locations.
Today, applications would need to publish to multiple Iggy clusters directly, increasing complexity and coupling replication logic into business applications.
A native Iggy Source/Sink Connector would make replication a first-class capability by continuously consuming from one cluster and publishing to another while preserving ordering and delivery guarantees.
The connector should behave like any other Iggy connector:
Aspects to cover (ideal, not a comprehensive list, can be broken down into follow-up tasks after the basic version is in place):
Affected area / component
Connectors
Proposed solution
No response
Alternatives considered
No response
Contribution
Good first issue