Getting started

This guide runs meilisync against PostgreSQL and a local Meilisearch.

From zero to a live index

Install the package, write config.yml, start CDC. PostgreSQL is the example source — MySQL and MongoDB work the same way once extras are installed.

Checklist

1. Install meilisync[postgres] (or the extra for your source)
2. Copy config.example.yml to config.yml and fill in credentials
3. Run meilisync start and watch the first full sync, then CDC
4. Use meilisync check to compare source counts with Meilisearch

Prerequisites

• Python 3.10+ or Docker
• A Meilisearch instance
• A source database (this example uses PostgreSQL)

PostgreSQL must have logical replication enabled and the wal2json plugin installed. See PostgreSQL.

Install

pip install "meilisync[postgres]"

Or with uv:

uv pip install "meilisync[postgres]"

Write config.yml

debug: false
progress:
  type: file
  path: progress.json
source:
  type: postgres
  host: 127.0.0.1
  port: 5432
  user: postgres
  password: postgres
  database: app
meilisearch:
  api_url: http://127.0.0.1:7700
  api_key: MASTER_KEY
  insert_size: 1000
  insert_interval: 5
sync:
  - table: products
    index: products
    pk: id
    full: true
    fields:
      id:
      name:
      description:

Place the file in the working directory, or pass -c /path/to/config.yml.

Start

meilisync start

On first start, tables with full: true are copied if the Meilisearch index does not exist. Incremental replication then follows WAL changes.

Useful commands

Command	What it does
meilisync start	Full sync (if needed) + incremental CDC
meilisync refresh -t products	Rebuild an index via a tmp index + swap
meilisync check -t products	Compare source row count vs Meilisearch
meilisync version	Print the installed version

Stop CDC before refresh

Run refresh only while start is stopped, otherwise writes can race the swap.

Next: configuration reference and CLI.