Two monitoring stacks to play with here.
Don't use either of these for production as they are wide open to abuse.
Prometheus, Grafana, Prometheus Pushgateway and the Prometheus Node Exporter
docker-compose up
I have Cockpit running on port 9090 so I mapped prometheus to 9900
Metrics can be pushed to the prometheus via the PushGateway using the test script in the scripts directory.
se-linux might be set to enforcing which can cause problems with docker volumes
Either set it to permissive or run:
chcon -Rt svirt_sandbox_file_t ./prometheus.yml
or use :z on the volume:
- ./config/prometheus.yml:/etc/prometheus/prometheus.yml:z
Influxdb, Chronograf, Telegraf, Kapacitor,
docker-compose up -f docker-compose-influx.yml up
docker-compose up -f docker-compose-elasticsearch.yml up
Latest docker image 6.5.4 fails to start on Fedora with
[1]: max virtual memory areas vm.max_map_count [65530] is too low, increase to at least [262144]
Workaround is
sysctl -w vm.max_map_count=262144
or make this change permanent by editing /etc/sysctl.conf
Image started OK on Mojave.
https://github.com/prometheus/prometheus