DevReplicate
Documentation
From the first replica to a developer-ready download: configure the engine you run, define what changes, and operate the enterprise controls without losing sight of the data boundary.
What it solves
Production-shaped development data, continuous replication, consistent joins and controlled delivery.
Explore the problems →Implementing it in your organization
Assign owners, write rules, validate an isolated replica and bring approved downloads into onboarding.
Plan your rollout →Pricing
Compare the free open-source core with Enterprise at 995 EUR per product licence per year.
Compare pricing →Set up each platform
Install with native packages or portable artifacts where available, build locally or in a container, then configure PostgreSQL/Patroni, MySQL/MariaDB/Percona, Redis/Valkey topologies, or one of the five newer engines.
Open setup guide →Configure anonymization
Choose transforms, map SQL columns, Redis values, document fields or Kafka record parts, reach nested JSON paths, and safely activate new policy with initial sync or backfill.
Open anonymization guide →API authentication & metrics
Configure RS256 verification, use scoped bearer tokens, manage fields or rules, trigger backfills, and connect Prometheus to the authenticated metrics endpoint.
Open enterprise API guide →Get the latest data
Publish verified replica artifacts and let developers retrieve, resume, verify, and restore them through shell scripts, Makefiles, or Docker-based workflows.
Open download guide →The eight products
One product per engine family, one licence per product. The mechanism column explains how each product keeps a separate target current.
| Product | Engines | How it replicates |
|---|---|---|
patroni-anonymizer
|
PostgreSQL 13+ / Patroni | Consistent baseline snapshot, then logical replication; the primary is discovered through Patroni’s REST API. |
mysql-anonymizer
|
MariaDB 10.11+, MySQL 8.0+, Percona Server 8.0+ | Snapshot, then the row-based binary log. |
redis-anonymizer
|
Redis; Valkey 8.1+, including Valkey 9’s own RDB format | Initial sync, then the replication protocol (REPLCONF/PSYNC); standalone, Sentinel or Cluster. |
mongodb-anonymizer
|
MongoDB 6.0+ replica set or sharded cluster | Every collection copied once, then the deployment-wide change stream as an ordinary client; the resume token is checkpointed on the target. |
kafka-anonymizer
|
Apache Kafka 3.x (KRaft), Redpanda 24+ | Each mirrored topic consumed and produced to the same-named target topic, same partition, key, headers and timestamp; the resume position lives in a compacted state topic on the target. |
mssql-anonymizer
|
Microsoft SQL Server 2019 / 2022, an edition with CDC | Tables recreated and copied in primary-key order, then Change Data Capture polled from the recorded LSN and applied in commit order, checkpoint in the same target transaction. |
oracle-anonymizer
|
Oracle Database 19c, 21c, 23ai, 26ai | Flashback copy AS OF SCN, then LogMiner over the online and archived redo; each committed transaction applied with its checkpoint in one target transaction. |
couchdb-anonymizer
|
CouchDB 3.x on both sides |
_all_docs for the first copy, then a continuous _changes feed with _bulk_get/_bulk_docs; _id and _rev preserved, checkpoint in _local/devreplicate-state. |
Choose the correct edition
Core
Initial sync, continuous replication, configurable transforms, and backfill. Configuration is stored locally in SQLite.
Enterprise
Adds the authenticated reconfiguration API, Prometheus metrics, product-specific backup workflows, developer download endpoints and next-business-day support by e-mail.