From DaemonSets and Static Pods to Custom Schedulers and Leader Election
In our previous article, we deconstructed the core components of the Kubernetes architecture. Now that you have a solid map of the cluster, it’s time to explore some of the more advanced ways to run workloads and control where they land.
1. DaemonSets: The “One Per Node” Workload
While a Deployment lets you specify some replicas to run somewhere in the cluster, a DaemonSet ensures that a copy of a Pod runs on all (or a specific subset of) Nodes.
This is incredibly useful for cluster-level services that need to be present on every machine.
Common Use Cases:
- Log Collectors: Running agents like Fluentd or Logstash on every node.
- Node Monitoring: Deploying monitoring agents like Prometheus Node Exporter or Datadog Agent.
- Cluster Storage: Running storage daemons like
glusterdorcephon each node.
CKA Exam Tip: You’ll be asked to deploy a Pod for a specific purpose, and you need to recognize when a DaemonSet is the right tool for the job. If the requirement is “run this on every…
