# PostgreSQL to Elasticsearch sync
PGSync is a middleware for syncing data from Postgres to Elasticsearch. It allows you to keep Postgres as your source of truth data source and expose structured denormalized documents in Elasticsearch.
### Requirements
- Python 3.6+
- Postgres 9.6+
- Redis
- Elasticsearch 6.3.1+
### Postgres setup
Enable logical decoding in your Postgres setting.
You would also need to set up two parameters in your Postgres config postgresql.conf
`wal_level = logical`
`max_replication_slots = 1`
### Installation
You can install PGSync from PyPI:
$ pip install pgsync
### Config
Create a schema for the application named e.g schema.json
Example spec
### Environment variables
Setup required environment variables for the application
SCHEMA=’/path/to/schema.json’
ELASTICSEARCH_HOST=localhost ELASTICSEARCH_PORT=9200
PG_HOST=localhost PG_USER=i-am-root # this must be a postgres superuser PG_PORT=5432 PG_PASSWORD=*****
REDIS_HOST=redis REDIS_PORT=6379 REDIS_DB=0 REDIS_AUTH=*****
## Running
- Bootstrap the database (one time only)
- $ bootstrap –config schema.json
- Run pgsync as a daemon
- $ pgsync –config schema.json –daemon