This is a simple project to explore the ELK stack and logspout capabilities.
logspout is not an ideal solution for production workloads, good alternatives would be fluentd and beats platform
Pre-req: docker, docker compose
Clone project and cd into elk folder
- build docker images: docker-compose build
- start the ELK stack: docker-compose up
- start logspout: ./run_logspout.sh
- start simple app: ./run_busy_box.sh
- open browser, for kibana: localhost:5601
- first time run requires to create an index pattern, under Settings -> Indices
- on the time-field name, select @timestamp and click Create button
- after that, click on the Discover top menu and explore your data
- open browser, for cadvisor: localhost:8080
To explore the sample visualizations and dashboard:
- Kibana -> Settings -> Object
- Click on Import button
- Select kibana-ui-config.json from elk/resources folder
This is a handy tool when building logstash filters: https://grokdebug.herokuapp.com