查看 k8s 某个节点,每个 pod 占用的 cpu 使用量
文章目录
#!/bin/bash
# Fetch all pods on the specified node along with their CPU usage
pods=$(kubectl get pods --all-namespaces -o=jsonpath='{range .items[*]}{.metadata.name}{"\t"}{.metadata.namespace}{"\t"}{.spec.nodeName}{"\n"}' | grep cn-hangzhou.172.16.6.12)
total_cpu_usage=0
echo "CPU_Usage(milliCPU) Pod_Name"
# Loop through each pod running on the specified node and fetch its CPU usage
while read -r pod_name namespace node_name; do
cpu_usage=$(kubectl top pod $pod_name -n $namespace --no-headers | awk '{print $2}')
cpu_usage_value=$(echo $cpu_usage | sed 's/m//') # Remove 'm' (milliCPU) unit
total_cpu_usage=$((total_cpu_usage + cpu_usage_value))
printf "%-20s %s\n" "$cpu_usage" "$pod_name"
done <<< "$pods" | sort -rn # Pipe output to sort by CPU usage in reverse numeric order
echo "使用的 CPU 总量: $total_cpu_usage milliCPU"
要使用,替换上面的 grep cn-hangzhou.172.16.6.12 部分就行了
评论
登录后参与技术讨论,分享你的经验与见解。
登录 / 注册暂无评论