Ekster, the XMPP IoT Registry, is an external XMPP component that can be used to register, claim and discover XMPP IoT devices.
Ekster is the Dutch word for Magpie. In the story A basket of flowers by Lilian Gask (1910) a Magpie steals a golden ring. Our Ekster component steals all of our shiny IoT gadgets and puts it in its nest, the IoT Registry.
A good start to learn more about Ekster is to read the XMPP IoT discovery extension: XEP-0347. Ekster aims to implement this XEP.
For more information about XMPP and IoT you can checkout this website.
Ekster is available under a MIT License which means that you can basically do anything you want with this code as long as you provide attribution back to us and don’t hold us liable.
Ekster is still in development. It is not used in production by us. We believe we've implemented most of of XEP-0347. If you find any bugs or misinterpretations of the XEP please let us know by creating a new issue.
Currently our roadmap for Ekster is:
- Create a more persistent solution for storing data when running Ekster on Docker
- Further testing and improving Ekster
- Validate the current XMPP IoT discovery extension.
There are various ways you can run the Ekster component. Our personal favorite is running the component with docker-compose. But you can run Ekster from the command line or as an Heroku application as well.
The preferred way to run Ekster is using docker-compose. A configuration template is available. Follow these
steps to run Ekster on a Docker host:
- First you need to add the Ekster component credentials to your XMPP server. For example on Prosody or ejabberd.
- Copy the template:
cp docker-compose-template.yml docker-compose.yml - Edit the
docker-compose.yml, change theTODOitems to the target environment. - Get Ekster up and running using the command
docker-compose up.
Please note that the database that is setup with Ekster currently does not save its data to a datastore. It is kept locally on the mongodb docker container. When you remove that container the data will also be gone.
- Clone the project from git:
git clone git@github.com:TNO-IoT/ekster.git - Go into the folder Ekster was cloned into:
cd ekster - Install nodejs
- Install coffeescript:
npm install -g coffee-script - Start Ekster using the commandline:
src/ekster.coffee
Usage: ekster.coffee [options]
Options:
-h, --help output usage information
-V, --version output the version number
-j, --jid <jid> jid for the registry
-P, --password <password> password for the component
-H, --host <hostname> hostname of the XMPP server this component connects to
-R, --no-reconnect disable reconnecting to the XMPP server when the connection is lost
-p, --port [port] the port to connect to [5347]
-b, --backend [backend] the backend to connect to [mongoose]
-B, --backendHost [backendHost] the host of the backend [localhost]
-s, --backendPort [backendPort] the port where the backend is listening [27017]
--backendOptions [] the configuration options for the backend []
-v, --verbose verbose logging
Examples:
$ ekster --help
$ ekster --jid registry.yourcompany.com --password secret --host xmpp.yourcompany.com --backend mongoose --backendHost localhost --backendPort 27017 --backendOptions '{ "user": "mongouser", "pass": "mongopass", "db": "mongodb" }'
-
Create a configuration file
.env:XMPP_JID=<your component jid> XMPP_PASSWD=<your component password> XMPP_HOST=<your xmpp server host> XMPP_PORT=<your xmpp server's component port, ie 5347> BACKEND_TYPE=mongoose BACKEND_OPTIONS=\{\"user\":\"iotregistry\",\"pass\":\"your password\",\"db\":\"registry\"\} BACKEND_HOST=<your backend server's hostname> BACKEND_PORT=<your backend server port> -
Test your configuration locally with foreman:
foreman start -
Deploy to the PaaS of choice.
The component is build in coffeescript and needs to have coffeescript installed to run. Coffeescript compiles to javascript hence to be able to run coffeescript you need to have nodejs installed on your system.
To install a development environment for Ekster please use the instructions below:
- Install nodejs
- Install coffeescript via the node package manager:
npm install coffee-script -g - Install the grunt build environment:
npm install grunt-cli -g - Install the project dependencies:
npm install --production(for development you should leave out the--production)
There is a Gruntfile with the following tasks:
default: watches the code for changes and, on change, runs the linter and test cases.grunt exec:dev: runs Ekster in development mode. For this you will change the Gruntfile according to your environment.grunt exec:production: runs Ekster in production mode: this will get you configuration from environment variables.
You can find the source code in ./src. Test scripts go into ./test. Currently the nodeunit test framework is used. Details on how to use the nodeunit test framework with coffeescript is detailed here.
Before adding sources to the git repository they should pass the coffeelint and nodeunit checks. You can run the checks manually by issuing grunt coffeelint and grunt nodeunit. Or you can run the checks automatically every time you save a file by issuing grunt watch or grunt.
Below you find a list of source files and a short description about what function they provide:
ekster.coffee- Entry script that creates the connections to the database and XMPP server and starts XMPP stanza processor.backend.coffee- Interface that backend connectors should implement.logger.coffee- Wrapper to the logger module.mongoose-backend.coffee- Backend implementation using mongoose models to write to a mongodb. Currently the default backend for Ekster.octoblu-backend.coffee- Backend implementation that uses octoblu as a backend service. Octoblu rate-limits requests and this may cause unexpected behavior with the registry. Might be outdated...presence-handler.coffee- Handles incoming presence message and friendship relations of the registry.processor.coffee- Handles incoming message for the registry. This is where the XMPP IoT discovery protocol is implemented.property.coffee- A name/value pair that a Thing uses to register META information.thing.coffee- The model of a Thing used by the processor.