Monitoring Labs

Simple monitoring infrastructure.

Grafana/Prometheus

• grafana - http://localhost:3000/ admin:heslo https://grafana.com/docs/
• prometheus - http://localhost:9090/ https://prometheus.io/docs/
• blackbox-exporter http://localhost:9115/
• cloudprober http://localhost:9313/metrics https://cloudprober.org/
• node-exporter http://http://localhost:9100/
• cadvisor http://localhost:8080/containers/
• pushgateway http://localhost:9091/ https://github.com/prometheus/pushgateway
• statsd-exporter http://localhost:9102/

Monitoring Mixins

Basic overview at https://github.com/monitoring-mixins/docs.

• elasticsearch - https://github.com/lukas-vlcek/elasticsearch-mixin
• etcd - https://github.com/etcd-io/etcd/tree/master/Documentation/etcd-mixin
• kubernetes - https://github.com/kubernetes-monitoring/kubernetes-mixin
• nasa-swpc - https://github.com/cznewt/nasa-swpc-exporter/tree/master/nasa-swpc-mixin
• node-exporter - https://github.com/prometheus/node_exporter/tree/master/docs/node-mixin
• prometheus - https://github.com/prometheus/prometheus/tree/master/documentation/prometheus-mixin

Installing Lab

Tested to work on Ubuntu 18.04.

• Install docker ./init_docker.sh
• Install JSONNET compiler./init_jsonnet.sh
• Install JSONNET bundler ./init_jsonnet_bundler.sh
• Install monitoring mixins ./init/init_mixins.sh
• Build monitoring mixins ./init/build_mixins.sh
• Run the docker-compose environment cd ./grafana-prometheus/; make apply

Use JSONNET Utils

• Install utils ./init_jsonnet_utils.sh
• Source environment with source jsonnet-utils/venv/bin/activate
• Check Prometheus rules make prometheus_metrics
• Check Grafana metrics make grafana_metrics

Generate new mixin with Cookiecutter

• Install utils ./init_jsonnet_utils.sh.
• Source environment with source jsonnet-utils/venv/bin/activate.
• Run cookiecutter mixin-template
• Change to newly created mixin dir and type make init and make build.

Use metrics for data analytics

• Install client ./init_libmetric.sh
• Create Pandas DataFrame from Prometheus instant data make query_prometheus_instant
• Create Pandas DataFrame from Prometheus range data make query_prometheus_range