Microsoft Fabric Open Mirroring Agent for Gluesync: Target Integration

Index

Core principles

Open Mirroring is the way Microsoft Fabric lets any application replicate data into a mirrored database. The application writes Parquet files into a landing zone folder in OneLake, and Fabric reads them and keeps the Delta tables of the mirrored database up to date, ready for the SQL analytics endpoint, semantic models and Spark.

This agent is such an application. It writes the change files and the small metadata files Open Mirroring expects, and Fabric does the rest: Gluesync writes no Delta files itself. Core Hub talks to OneLake over HTTPS through the Azure Data Lake Storage Gen2 API and signs in with a Microsoft Entra service principal.

Every row travels with a row marker (insert, update or delete), so inserts, updates and deletes keep the mirrored table equal to the source.

Supported version

Any Microsoft Fabric workspace with a mirrored database that uses Open Mirroring.

Platform supported features

Feature	Support	Notes
Target role	✅	Writes Parquet change files into the landing zone of a Fabric mirrored database.
Source role	❌	Current release supports Microsoft Fabric Open Mirroring only as a target.
Snapshots	✅	Snapshot files carry no row marker: Fabric loads them as plain inserts.
CDC	❌	Current release supports Microsoft Fabric Open Mirroring only as a target.
CDC type	N/A	Current release supports Microsoft Fabric Open Mirroring only as a target.
Bulk load	❌	Not supported: the agent has no bulk mode, so the bulk operations settings of an entity do not apply to it. A snapshot is always written as large Parquet files, up to Maximum row count in Parquet file rows each (see Throughput).
Database pooling	N/A	The OneLake API is stateless HTTPS: there is no connection to pool.
Source persisted internal cache	N/A	Target-only connector.
Target persisted internal cache	N/A	Files are uploaded directly to OneLake.
Driver / connector	Azure Data Lake Storage Gen2 API	HTTPS from Core Hub to OneLake, authenticated with a Microsoft Entra service principal.
Chains	N/A	Target-only connector.
Truncate operation support (reading)	N/A	Target-only connector.
Truncate operation support (writing)	❌	Open Mirroring has no truncate: the request is skipped with a warning.

Feature

Support

Notes

Target role

✅

Writes Parquet change files into the landing zone of a Fabric mirrored database.

Source role

❌

Current release supports Microsoft Fabric Open Mirroring only as a target.

Snapshots

✅

Snapshot files carry no row marker: Fabric loads them as plain inserts.

CDC

❌

Current release supports Microsoft Fabric Open Mirroring only as a target.

CDC type

N/A

Current release supports Microsoft Fabric Open Mirroring only as a target.

Bulk load

❌

Not supported: the agent has no bulk mode, so the bulk operations settings of an entity do not apply to it. A snapshot is always written as large Parquet files, up to Maximum row count in Parquet file rows each (see Throughput).

Database pooling

N/A

The OneLake API is stateless HTTPS: there is no connection to pool.

Source persisted internal cache

N/A

Target-only connector.

Target persisted internal cache

N/A

Files are uploaded directly to OneLake.

Driver / connector

Azure Data Lake Storage Gen2 API

HTTPS from Core Hub to OneLake, authenticated with a Microsoft Entra service principal.

Chains

N/A

Target-only connector.

Truncate operation support (reading)

N/A

Target-only connector.

Truncate operation support (writing)

❌

Open Mirroring has no truncate: the request is skipped with a warning.