{"id":635,"date":"2025-03-30T01:48:00","date_gmt":"2025-03-30T01:48:00","guid":{"rendered":"https:\/\/blog.ngocha.biz\/?p=635"},"modified":"2025-03-30T01:48:00","modified_gmt":"2025-03-30T01:48:00","slug":"pod-priorityclass-preemption","status":"publish","type":"post","link":"https:\/\/blog.ngocha.biz\/?p=635","title":{"rendered":"Kubernetes Pod Priority, PriorityClass and Preemption Explained"},"content":{"rendered":"

In this Kubernetes tutorial<\/a>, you will learn about important Kubernetes pod<\/a> scheduling concepts such as Pod priority, Preemption, and Pod PriorityClass<\/p>\n

What is Pod priority in Kubernetes?<\/h2>\n

Pod priority<\/strong> is a Kubernetes scheduling feature that allows Kubernetes to make scheduling decisions comparing other pods based on priority number. Let’s look at the following two main concepts in pod priority.<\/p>\n

    \n
  1. Pod Preemption<\/li>\n
  2. Pod Priority Class<\/li>\n<\/ol>\n

    Pod Preemption<\/h3>\n

    The pod preemption feature allows Kubernetes to preempt (evict) lower-priority pods from nodes when higher-priority pods are in the scheduling queue and no node resources are available.<\/p>\n

    Kubernetes Pod Priority Class<\/h3>\n

    To assign a pod a certain priority, you need a priority class.<\/p>\n

    You can set a priority for a Pod using the PriorityClass<\/code> object (non-namespaced) with a Value.<\/p>\n

    The value determines the priority. It can be 1,000,000,000 (one billion) or lower. <\/strong>Larger the number, the higher the priority.<\/p>\n

    \"Kubernetes<\/figure>\n

    The name of the priorityclass (priorityClassName<\/code>) will be used in the pod specification to set the priority.<\/p>\n

    \n
    If you don’t want the priority class to preempt the pods, you can set PreemptionPolicy: Never<\/code>. By default, Priorityclasss use PreemptLowerPriority<\/code> policy.<\/div>\n<\/div>\n
    \"Kubernetes<\/figure>\n

    Pod PriorityClass Example<\/h3>\n

    The following example has a PriorityClass object and a pod that uses the PriorityClass.<\/p>\n

    apiVersion: scheduling.k8s.io\/v1\nkind: PriorityClass\nmetadata:\n  name: high-priority-apps\nvalue: 1000000\npreemptionPolicy: PreemptLowerPriority\nglobalDefault: false\ndescription: \"Mission Critical apps.\"\n---\napiVersion: v1\nkind: Pod\nmetadata:\n  name: nginx\n  labels:\n    env: dev\nspec:\n  containers:\n  - name: web\n    image: nginx:latest\n    imagePullPolicy: IfNotPresent\n  priorityClassName: high-priority-apps<\/code><\/pre>\n
    Kubernetes System High PriorityClass<\/h3>\n
    How do you safeguard system-critical pods<\/strong> from preemption?<\/p>\n
    Well, there are two default high-priority classes set by Kubernetes<\/p>\n
      \n
    1. system-node-critical:<\/strong> This class has a value of 2000001000<\/code>. Static pods Pods like etcd, kube-apiserver, kube-scheduler and Controller manager use this priority class.<\/li>\n
    2. system-cluster-critical:<\/strong> This class has a value of 2000000000<\/code>. Addon Pods like coredns, calico controller, metrics server, etc use this Priority class.<\/li>\n<\/ol>\n
      How does Kubernetes Pod Priority & Preemption work?<\/h2>\n
        \n
      1. If a pod is deployed with PriorityClassName<\/code>, the priority admission controller gets the priority value using the PriorityClassName value.<\/li>\n
      2. If there are many pods in the scheduling queue, the scheduler arranges the scheduling order based on priority. Meaning, the scheduler places the high-priority pod ahead of low priority pods<\/li>\n
      3. Now, if there are no nodes available with resources to accommodate a higher-priority pod, the preemption logic kicks in.<\/li>\n
      4. The scheduler preempts (evicts) low priority pod from a node where it can schedule the higher-priority pod. The evicted pod gets a graceful default termination time of 30 seconds. If pods have terminationGracePeriodSeconds<\/code> set for preStop<\/code> container Lifecycle Hooks<\/a>, it overrides the default 30 seconds.<\/li>\n
      5. However, if for some reason, the scheduling requirements are not met, the scheduler goes ahead with scheduling the lower-priority pods.<\/li>\n<\/ol>\n
        \"Pod
        Click to view in HD<\/span><\/figcaption><\/figure>\n
        Now we know how kubernetes pod scheduling priority works<\/strong> with Priorityclass and preemption.<\/p>\n
        Pod Priority FAQs<\/h2>\n
        The following are the frquesntly asked questions about Pod priority.<\/p>\n
        What is Kubernetes DaemonSet Priority?<\/h3>\n
        Daemonset<\/a> has priority like any other pod. Therefore, if you want your Daemonsets to be stable and not evicted during a node resource crunch, you need to set a higher pod PriorityClass to the Daemonset.<\/p>\n