toluaina/pgsync

Health v2 · maintenance only

Adoption not part of the score

1415 stars · 210 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

PGSync is an open-source (MIT) change data capture tool that syncs PostgreSQL, MySQL, or MariaDB data to Elasticsearch or OpenSearch in real time via logical replication. Document schemas are defined declaratively in JSON, producing denormalized nested documents with no custom ETL code.

Use cases

• keep elasticsearch in sync with postgres in real time
• sync mysql database to opensearch
• build denormalized search documents from relational tables
• change data capture from postgres wal to search engine
• index nested related tables into elasticsearch without custom code
• resume elasticsearch indexing after crashes without data loss

When to choose

• you need real-time, transactionally consistent search indexing from a relational database
• you want declarative JSON schema-driven denormalization of nested relationships
• you need fault-tolerant CDC that resumes from checkpoints
• you prefer self-hosted sync middleware with no custom ETL code

When to avoid

• you need semantic/vector search out of the box - that requires the paid PGSync Pro tier
• you want to sync to ClickHouse - that is a separate paid product (free only for single-table mirror)
• your source database does not support logical replication or you cannot grant superuser/replication privileges
• you need a fully managed hosted sync service

Facets

service · maturity active

etl search-engine database streaming cli databases developer-tools python self-hosted cli change-data-capture elasticsearch opensearch postgresql mysql mariadb logical-replication denormalization open-core search data-engineering docker linux macos

Member repositories

For agents

markdown · JSON · MCP: product_card(name="toluaina/pgsync")