Performing maintenance on a Kubernetes node requires safely evicting running pods without causing service disruptions. Kubernetes provides the Use the following command to safely drain a node:<\/p>\n\n\n\n Instead of draining, apply a taint to prevent scheduling while allowing existing pods to run:<\/p>\n\n\n\n For managed Kubernetes services like EKS, GKE, and AKS, cluster autoscaler can handle node removal and rebalancing automatically.<\/p>\n\n\n\n Using Introduction Performing maintenance on a Kubernetes node requires safely evicting running pods without causing service disruptions. Kubernetes provides the kubectl drain command to gracefully remove a node from the cluster while ensuring workload continuity. Why Drain a Node? Draining a Node Safely Use the following command to safely drain a node: Explanation of Flags: Best […]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[9,216,1],"tags":[545,10,546],"class_list":["post-4087","post","type-post","status-publish","format-standard","hentry","category-kubernetes","category-kubernetes-basic","category-uncategorized","tag-drain-a-node","tag-kubernetes","tag-maintenance-in-kubernetes"],"yoast_head":"\nkubectl drain<\/code> command to gracefully remove a node from the cluster while ensuring workload continuity.<\/p>\n\n\n\nWhy Drain a Node?<\/h2>\n\n\n\n
\n
Draining a Node Safely<\/h2>\n\n\n\n
kubectl drain <node-name> --ignore-daemonsets --delete-emptydir-data<\/code><\/pre>\n\n\n\nExplanation of Flags:<\/h3>\n\n\n\n
\n
--ignore-daemonsets<\/code>: Ensures that daemonset-managed pods (like logging or monitoring agents) are not evicted.<\/li>\n\n\n\n--delete-emptydir-data<\/code>: Ensures that data stored in emptyDir<\/code> volumes is deleted.<\/li>\n\n\n\n--force<\/code>: Can be used to force eviction when pods have no associated controllers (use with caution).<\/li>\n<\/ul>\n\n\n\nBest Practices for Node Draining<\/h2>\n\n\n\n
\n
kubectl get pods --all-namespaces --field-selector spec.nodeName=<node-name><\/code>This helps identify which workloads will be affected.<\/li>\n\n\n\napiVersion: policy\/v1 kind: PodDisruptionBudget metadata: name: my-app-pdb spec: minAvailable: 1 selector: matchLabels: app: my-app<\/code>This prevents more than the allowed number of replicas from being evicted simultaneously.<\/li>\n\n\n\nkubectl cordon <node-name><\/code>This prevents new pods from being scheduled on the node before draining.<\/li>\n\n\n\nkubectl get pods -o wide --watch<\/code>Ensures all evictions occur without errors before proceeding.<\/li>\n\n\n\nkubectl uncordon <node-name><\/code>This allows the node to accept new pods again.<\/li>\n<\/ol>\n\n\n\nAlternative Solutions<\/h2>\n\n\n\n
Using Node Taints<\/h3>\n\n\n\n
kubectl taint nodes <node-name> key=value:NoSchedule<\/code><\/pre>\n\n\n\nUsing Cluster Autoscaler<\/h3>\n\n\n\n
Conclusion<\/h2>\n\n\n\n
kubectl drain<\/code> is a safe way to prepare a Kubernetes node for maintenance without disrupting applications. Following best practices ensures minimal downtime and workload continuity.<\/p>\n\n\n\n
\n","protected":false},"excerpt":{"rendered":"