Fuge configuration has changed if you are upgrading to fuge V2.0 read this guide upgrading.md
Fuge - The Microservice Shell
Fuge is a microservice shell and workbench. It provides an execution environment for developing microservice systems, eliminating shell hell and significantly reducing developer friction.
If you're using this module, and need help, you can:
Use npm to install globally.
npm install -g fuge
A Micoservice shell?
Yup. Fuge is a shell environment that focuses on a specific system configuration, providing an emulation environment in which services can run in development that is close to a modern container production deployment.
In order to run your system via fuge, you need to create a YAML configuration file. Fuge also integrates with docker-compose. Full documentation on configuring fuge is available here.
Full example demonstration systems are available here: https://github.com/apparatus/fuge-examples.
A simple configuration is provided below:
fuge_global: tail: true monitor: true monitor_excludes: - '**/node_modules/**' - '**/.git/**' - '*.log' myservice: type: process path: ../myservice run: 'node index.js' ports: - myservice=8000 webapp: type: process path: ../webapp run: 'npm start' ports: - webapp=3000 mongo: image: mongo type: container ports: - mongo=27017:27017
This configuration defines a front end webapp process, a single microservice and a mongodb container. The system can be started by running:
$ fuge shell <path to config file> fuge> start all
This will start the mongo container through the local docker API and also the webapp and myservice processes. Fuge will also inject configuration information into these processes as environment variables, tail the logs and watch the file system for changes, restarting the appropriate processes as you write code.
To fully explore the capabilites of Fuge, you are encouraged to run the example systems which can be found here: https://github.com/apparatus/fuge-examples.
Fuge is fully compatible with V1, V2 and V3 docker-compose formats. Fuge services can be specified entirely using docker-compose or entirely in fuge.yml or a mixture of anything in between.
Fuge can be started by running:
$ fuge shell <path to fuge config file> fuge>
The fuge command shell supports the following commands
||display a list of supported commands|
||list status of managed processes and containers|
||show information on a specific process. The
||start a process or all processes|
||stop a process and any associated watcher or all processes|
||start a process in debug mode (only supported for node.js processes)|
||run the specified shell command in the root of each service e.g.
||display dns zone information for fuge internal DNS server - requires the
||turn on watching for a process or for all processes|
||turn off watching for a process or for all processes|
||tail output for a process or for all processes|
||end tail output for a specific processes or for all processes|
||searches a process' log or all processes' logs|
||terminate all managed processes and exit the shell|
Shell pass through
In addition to the above commands, Fuge will pass through all other commands to the underlying shell, for example:
fuge> ps aux | grep -i node
Will bypass the fuge ps command and execute the underlying
fuge> netstat -an | grep -i listen
Will be passed through to the underlying shell to display a lists of open listening port and so on.
The apparatus team encourage open participation. If you feel you can help in any way, be it with documentation, examples, extra testing, or new features please get in touch.
Copyright the apparatus team 2016, Licensed under MIT.