指标服务器在一个节点上显示未知

Question

它恰好发生在我今天添加到集群中的 almalinux 9.0 服务器上。 k8s版本是1.19.16

NAME             CPU(cores)   CPU%   MEMORY(bytes)   MEMORY%     
k8s-dev-node1    110m         11%    2601Mi          104%             
k8s-dev-node8    160m         5%     3600Mi          56%         
k8s-dev-node9    105m         3%     2201Mi          34%         
k8s-dev-node16   <unknown>                           <unknown>               <unknown>               <unknown> 

hostNetwork 已在指标服务器部署上设置。 顶部 pod 很好，获取节点统计信息也很好。

kubectl get --raw /api/v1/nodes/k8s-dev-node16/proxy/stats/summary
{
 "node": {
  "nodeName": "k8s-dev-node16",
  "systemContainers": [
   {
    "name": "kubelet",
    "startTime": "2022-11-10T05:52:14Z",
    "cpu": {
     "time": "2022-11-10T09:42:50Z",
     "usageNanoCores": 17536218,
     "usageCoreNanoSeconds": 221415603000
    },
    "memory": {
     "time": "2022-11-10T09:42:50Z",
     "availableBytes": 8011243520,
     "usageBytes": 58892288,
     "workingSetBytes": 46686208,
     "rssBytes": 0,
     "pageFaults": 2555049,
     "majorPageFaults": 1
    }
   },
...

Answer 1

我有同样的问题。你找到解决办法了吗？