Setting Reasonable Req and Limit Without Req and Limit set, when an application’s CPU or MEM spikes, it harms other Pods on the same node, and can even cause cluster nodes to be crushed one after another. Req and Limit have four values in total. If only some of them are set, then when node resource usage reaches the threshold preset by Kubelet, Kubelet evicts Pods in the order Guaranteed > Burstable > Best-Effort

FailedCreatePodSandBox Error Error response from daemon: OCI runtime create failed: container_linux.go:380: starting container process caused: process_linux.go:402: getting the final child's pid from pipe caused: EOF: unknown Resolution Clear the cache 1 echo 3 > /proc/sys/vm/drop_caches Cause Too much memory fragmentation calico-node Keeps Restarting increase max user Error runtime: failed to

1. Background At 10:00 in the morning, a colleague asked me to run a task on the cluster, so I created a Kubernetes Job through a tool. When the tool creates a Job, it first gets all the nodes on the cluster, then binds to each node one by one to create Jobs.

1. Specify a Node via nodeSelector When Creating a Workload Add a label to the node 1 kubectl label node node2 project=A Create the workload with the specified nodeSelector 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 cat <<EOF | kubectl apply -f - apiVersion: apps/v1 kind: Deployment metadata: name: nginx-nodeselector spec: replicas: 1 selector: matchLabels: app: nginx-nodeselector template: metadata: labels: app: nginx-nodeselector spec: nodeSelector: project: A containers: - name: nginx image: nginx EOF Check the workload 1 2 3 4 kubectl get pod -o wide NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES nginx-nodeselector-7bb75b7687-7r5xk 1/1 Running 0 19s 10.