/monstache-site

documentation for the monstache project

Primary LanguageJavaScriptMIT LicenseMIT

Monstache Docs

This is Monstache documentation source-code

For better readability, check the Hosted GitHub Page: https://rwynn.github.io/monstache-site

Introduction

Sync MongoDB to Elasticsearch in realtime

Monstache is a sync daemon written in Go that continously indexes your MongoDB collections into Elasticsearch. Monstache gives you the ability to use Elasticsearch to do complex searches and aggregations of your MongoDB data and easily build realtime Kibana visualizations and dashboards.

Features

  • Supports up to and including the latest versions of Elasticsearch and MongoDB

  • Single binary with a light footprint

  • Pre built Docker containers

  • Optionally filter the set of collections to sync

  • Advanced support for sharded MongoDB clusters including auto-detection of new shards

  • Direct read mode to do a full sync of collections in addition to tailing the oplog

  • Transform and filter documents before indexing using Golang plugins or JavaScript

  • Index the content of GridFS files

  • Support for hard and soft deletes in MongoDB

  • Support for propogating database and collection drops

  • Optional custom document routing in Elasticsearch

  • Stateful resume feature

  • Time machine feature to track document changes over time

  • Worker and Clustering modes for High Availability

  • Support for rfc7396 JSON merge patches

  • Systemd support

  • Optional http server to get access to liveness, stats, etc

  • See Getting Started for instructions how to get it up and running.

Links

About

Getting Started