Modular and configurable, simply reads data from redis keys or elasticsearch queries - it's up to you to define what they are.
- Activity graphs (numbers over time)
- Top-things-tables out of elasticsearch and other places
- Pie charts of things from elasticsearch and other places
- Raw HTML (for decoration)
- Tickers (arbitrary numbers)
Install and configure elasticsearch as your data sources.
Install memcached as your query cache.
Pop in a virtualenv and use pip + requirements.txt to set up.
pip install -f requirements
Install yarn: https://yarnpkg.com/en/docs/install
Install frontend deps:
cd dashi/static/
yarn install
Set up some configuration files:
dashi/settings/example-instance.py to dashi/settings/instance.py
dashi/settings/example-dashboard.py to dashi/settings/dashboard.py
...and customise both.
Some settings you can change:
FAKE- use fake data instead of connecting to redis, which can help when setting up.
To run on localhost:8080 just do a standard a Django dev server:
python manage.py runserver --settings=dashi.settings.instance
Make new widgets by combining renderers and sources - see blocks.py:Elaticsearch_Pie for a great exmaple.