Resources
In order to prevent outages, it's essential that you leverage a monitoring and alerting tool in your Kubernetes environment. Prometheus—an open-source systems monitoring and alerting toolkit—pairs particularly well with Kubernetes. In this course, learn how this toolkit integrates with Kubernetes and works to monitor distributed systems.
Logging vs. monitoring logs are qualitative =, time series-based and metrics are point-in-time data points that have some meta data around them.
Monitoring K8s There's cAdvisor, Heapster, and Prometheus. Prometheus is a pull-based time-series capture, storage, and alerting service with Kubernetes integration.
Enabling Prometheus monitoring
from promoperator.yml file - creating accounts, roles, binding. kube control create -f
[verbs: "*"] = * all verbs (can create, read, watch, update, delete)
create service endpoints - exposes tools like Prom to other apps