ullala is a demo that showcase using kafka as a data pipeline and log aggregation through the ELK stack
- Download kafka command line tools:
scripts/download-kafka.sh - Run Kafka, Zookeeper, Elasticsearch and Logstash:
docker-compose up - Set kafka on your host file to point to the running kafka instance (docker-machine ip default on mac)
- Setup topics
scripts/setup-topics.sh - Install OpenCV for capturer,
brew install opencvon mac. - Symlink opencv into capturer, for mac you can run
scripts/symlink-opencv-bew.sh - Install capturer dependencies
pip install -r capturer/requirements.txt - Run Capturer:
KAFKA_HOSTS=KAFKA:9092 capturer/capturer.py - Install processor dependencies
cd processor && npm insall - Run processor:
cd processor && KAFKA_HOSTS=kafka:9092 node index.js - Install pusher dependencies
cd pusher && npm install - Run pusher `cd pusher && KAFKA_HOSTS=kafka:9092 node index.js