MongoDB Agent for Gluesync: Features & Compatibility

Index

Core principles

Gluesync’s MongoDB agent uses and implements the MongoDB Kotlin SDK to connect and perform all the operations at the node/cluster level. Gluesync also supports MongoDB HA features as well as load balancing across multiple cluster nodes, thanks to automatic node discovery.

Change data capture

This agent makes use of the Change Streams API built-in in MongoDB to catch real-time data changes from MongoDB’s database starting from Database version 3.4 and above.

Supported version

With this agent, you can target any version of MongoDB starting from 3.0.

Platform supported features

Feature	Support	Notes
Target role	✅	Delivers data to MongoDB clusters through the native Kotlin SDK with automatic node discovery and TLS support.
Source role	✅	Consumes MongoDB Change Streams from replica sets or sharded clusters to capture real-time operations.
Snapshots	✅	Entities can run full snapshots against MongoDB collections for initial loads before CDC resumes.
CDC	✅	Continuously streams oplog-derived events via Change Streams for low-latency replication.
CDC type	Change Streams (oplog)	Relies on MongoDB’s native Change Streams API introduced in 3.4+.
Bulk load	N/A	Bulk load staging tables are specific to relational targets; MongoDB applies snapshot + CDC flows directly.
Source persisted internal cache	N/A	No disk-based cache is required—state tracking lives in Core Hub metadata tables.
Target persisted internal cache	N/A	Writes are applied directly to MongoDB without intermediate file buffers.
Driver / connector	Native Kotlin SDK	Gluesync embeds the official MongoDB Kotlin SDK for connectivity and HA balancing.
Chains	N/A	Not supported being it a NoSQL database.
Truncate operation support (reading)	❌	Source agent never issues destructive commands such as collection drops or truncates.
Truncate operation support (writing)	N/A	Document collections have no TRUNCATE equivalent; deletions must be performed manually if needed.

Feature

Support

Notes

Target role

✅

Delivers data to MongoDB clusters through the native Kotlin SDK with automatic node discovery and TLS support.

Source role

✅

Consumes MongoDB Change Streams from replica sets or sharded clusters to capture real-time operations.

Snapshots

✅

Entities can run full snapshots against MongoDB collections for initial loads before CDC resumes.

CDC

✅

Continuously streams oplog-derived events via Change Streams for low-latency replication.

CDC type

Change Streams (oplog)

Relies on MongoDB’s native Change Streams API introduced in 3.4+.

Bulk load

N/A

Bulk load staging tables are specific to relational targets; MongoDB applies snapshot + CDC flows directly.

Source persisted internal cache

N/A

No disk-based cache is required—state tracking lives in Core Hub metadata tables.

Target persisted internal cache

N/A

Writes are applied directly to MongoDB without intermediate file buffers.

Driver / connector

Native Kotlin SDK

Gluesync embeds the official MongoDB Kotlin SDK for connectivity and HA balancing.

Chains

N/A

Not supported being it a NoSQL database.

Truncate operation support (reading)

❌

Source agent never issues destructive commands such as collection drops or truncates.

Truncate operation support (writing)

N/A

Document collections have no TRUNCATE equivalent; deletions must be performed manually if needed.

Change data capture

Works on any version of MongoDB starting from 3.4, to 4.2 is highly recommended due to performance optimizations around the Change Streams protocol and its indexing process.

Q&A

I have got MongoDB Realm running on the same database in which I would like to run Gluesync, is it supported? Sure, Gluesync has been specifically designed to work even within MongoDB Realm running on the same database.

I have got MongoDB running under MongoDB Atlas DBaaS, is it supported? Gluesync works great even when your MongoDB is served via MongoDB Atlas DBaaS or running in any public cloud hosting service.