Monitoring of the host

[ranguard]

We want monitoring of the host instances specifically disk space on /mnt/cluster_data but RAM/CPU as well

[ssoriche]

We'll be rolling out Prometheus as part of the cluster and installing node_exporter which will monitor the underlying hosts.

[ranguard]

We might want to look at https://github.com/kubernetes/kube-state-metrics

https://github.com/grafana/grizzly interesting, rancher also

[ssoriche]

Most definitely want kube-state-metrics.

[ranguard]

Not quite so imporant as DO provides a dash boads:

Cluster https://cloud.digitalocean.com/kubernetes/clusters/490ad197-959c-4024-945b-547fc99415ef/insights_embedded?i=5289ac&period=hour&clusterObjectName=415518787%7C415518789%7C415518788

Load balancer

• Traffic: https://cloud.digitalocean.com/networking/load_balancers/1920777b-c01a-4e79-9612-1c1838a702f0/graphs?i=5289ac&period=hour
• Health: https://cloud.digitalocean.com/networking/load_balancers/1920777b-c01a-4e79-9612-1c1838a702f0?i=5289ac