Kubernetes Worker 节点处于 NotReady 状态

Question

我一直在尝试在一组 Raspberry Pi 上设置 K8s 集群。这是我的 GitHub 页面的链接，描述了整个设置：

https://github.com/joesan/plant-infra/blob/master/pi/README.md

我现在卡在最后一步，将我的工作节点与主节点连接起来。我确实在工作节点上发出了 join 命令，但之后我检查了主节点中的节点，我看到了以下内容：

pi@k8s-master-01:~ $ kubectl get nodes
NAME            STATUS     ROLES     AGE       VERSION
k8s-master-01   Ready      master    56m       v1.9.6
k8s-worker-01   NotReady   <none>    26m       v1.9.6
k8s-worker-02   NotReady   <none>    6m        v1.9.6

问题是，我是否需要像 weave 一样在工作节点上安装容器网络？

这是来自工作节点的日志文件：

pi@k8s-worker-02:~ $ journalctl -u kubelet
-- Logs begin at Thu 2016-11-03 17:16:42 UTC, end at Tue 2018-05-01 11:35:54 UTC. --
May 01 11:27:28 k8s-worker-02 systemd[1]: Started kubelet: The Kubernetes Node Agent.
May 01 11:27:30 k8s-worker-02 kubelet[334]: I0501 11:27:30.995549     334 feature_gate.go:226] feature gates: &{{} map[]}
May 01 11:27:31 k8s-worker-02 kubelet[334]: I0501 11:27:31.005491     334 controller.go:114] kubelet config controller: starting controller
May 01 11:27:31 k8s-worker-02 kubelet[334]: I0501 11:27:31.005584     334 controller.go:118] kubelet config controller: validating combination of defaults and flags
May 01 11:27:31 k8s-worker-02 kubelet[334]: W0501 11:27:31.052134     334 cni.go:171] Unable to update cni config: No networks found in /etc/cni/net.d
May 01 11:27:31 k8s-worker-02 kubelet[334]: I0501 11:27:31.084480     334 server.go:182] Version: v1.9.6
May 01 11:27:31 k8s-worker-02 kubelet[334]: I0501 11:27:31.085670     334 feature_gate.go:226] feature gates: &{{} map[]}
May 01 11:27:31 k8s-worker-02 kubelet[334]: I0501 11:27:31.092807     334 plugins.go:101] No cloud provider specified.
May 01 11:27:31 k8s-worker-02 kubelet[334]: I0501 11:27:31.110132     334 certificate_store.go:130] Loading cert/key pair from ("/var/lib/kubelet/pki/kubelet-client.crt", "/var/lib/
May 01 11:27:39 k8s-worker-02 kubelet[334]: E0501 11:27:39.905417     334 machine.go:194] failed to get cache information for node 0: open /sys/devices/system/cpu/cpu0/cache: no suc
May 01 11:27:39 k8s-worker-02 kubelet[334]: I0501 11:27:39.911993     334 server.go:428] --cgroups-per-qos enabled, but --cgroup-root was not specified.  defaulting to /
May 01 11:27:39 k8s-worker-02 kubelet[334]: I0501 11:27:39.914203     334 container_manager_linux.go:242] container manager verified user specified cgroup-root exists: /
May 01 11:27:39 k8s-worker-02 kubelet[334]: I0501 11:27:39.914272     334 container_manager_linux.go:247] Creating Container Manager object based on Node Config: {RuntimeCgroupsName
May 01 11:27:39 k8s-worker-02 kubelet[334]: I0501 11:27:39.914895     334 container_manager_linux.go:266] Creating device plugin manager: false
May 01 11:27:39 k8s-worker-02 kubelet[334]: I0501 11:27:39.919031     334 kubelet.go:291] Adding manifest path: /etc/kubernetes/manifests
May 01 11:27:39 k8s-worker-02 kubelet[334]: I0501 11:27:39.919197     334 kubelet.go:316] Watching apiserver
May 01 11:27:39 k8s-worker-02 kubelet[334]: E0501 11:27:39.935754     334 reflector.go:205] k8s.io/kubernetes/pkg/kubelet/config/apiserver.go:47: Failed to list *v1.Pod: Get https:/
May 01 11:27:39 k8s-worker-02 kubelet[334]: E0501 11:27:39.937449     334 reflector.go:205] k8s.io/kubernetes/pkg/kubelet/kubelet.go:480: Failed to list *v1.Node: Get https://192.16
May 01 11:27:39 k8s-worker-02 kubelet[334]: E0501 11:27:39.937492     334 reflector.go:205] k8s.io/kubernetes/pkg/kubelet/kubelet.go:471: Failed to list *v1.Service: Get https://192
May 01 11:27:39 k8s-worker-02 kubelet[334]: W0501 11:27:39.948764     334 kubelet_network.go:139] Hairpin mode set to "promiscuous-bridge" but kubenet is not enabled, falling back t
May 01 11:27:39 k8s-worker-02 kubelet[334]: I0501 11:27:39.949871     334 kubelet.go:577] Hairpin mode set to "hairpin-veth"
May 01 11:27:39 k8s-worker-02 kubelet[334]: W0501 11:27:39.951008     334 cni.go:171] Unable to update cni config: No networks found in /etc/cni/net.d
May 01 11:27:39 k8s-worker-02 kubelet[334]: I0501 11:27:39.952122     334 client.go:80] Connecting to docker on unix:///var/run/docker.sock
May 01 11:27:39 k8s-worker-02 kubelet[334]: I0501 11:27:39.952976     334 client.go:109] Start docker client with request timeout=2m0s
May 01 11:27:39 k8s-worker-02 kubelet[334]: W0501 11:27:39.959045     334 cni.go:171] Unable to update cni config: No networks found in /etc/cni/net.d
May 01 11:27:39 k8s-worker-02 kubelet[334]: W0501 11:27:39.971616     334 cni.go:171] Unable to update cni config: No networks found in /etc/cni/net.d
May 01 11:27:39 k8s-worker-02 kubelet[334]: I0501 11:27:39.971765     334 docker_service.go:232] Docker cri networking managed by cni
May 01 11:27:40 k8s-worker-02 kubelet[334]: I0501 11:27:40.002411     334 docker_service.go:237] Docker Info: &{ID:25GN:65LU:UXAR:CUUY:DOQH:ST4A:IQOE:PIDR:BKYC:UVJH:LI5H:HQSG Contai
May 01 11:27:40 k8s-worker-02 kubelet[334]: I0501 11:27:40.002766     334 docker_service.go:250] Setting cgroupDriver to cgroupfs
May 01 11:27:40 k8s-worker-02 kubelet[334]: I0501 11:27:40.058142     334 remote_runtime.go:43] Connecting to runtime service unix:///var/run/dockershim.sock
May 01 11:27:40 k8s-worker-02 kubelet[334]: I0501 11:27:40.098202     334 kuberuntime_manager.go:186] Container runtime docker initialized, version: 18.04.0-ce, apiVersion: 1.37.0
May 01 11:27:40 k8s-worker-02 kubelet[334]: I0501 11:27:40.110512     334 server.go:755] Started kubelet
May 01 11:27:40 k8s-worker-02 kubelet[334]: I0501 11:27:40.112242     334 kubelet_node_status.go:273] Setting node annotation to enable volume controller attach/detach
May 01 11:27:40 k8s-worker-02 kubelet[334]: I0501 11:27:40.114014     334 server.go:129] Starting to listen on 0.0.0.0:10250
May 01 11:27:40 k8s-worker-02 kubelet[334]: E0501 11:27:40.114962     334 kubelet.go:1281] Image garbage collection failed once. Stats initialization may not have completed yet: fai
May 01 11:27:40 k8s-worker-02 kubelet[334]: I0501 11:27:40.133665     334 server.go:299] Adding debug handlers to kubelet server.
May 01 11:27:40 k8s-worker-02 kubelet[334]: E0501 11:27:40.141790     334 event.go:209] Unable to write event: 'Post https://192.168.0.101:6443/api/v1/namespaces/default/events: dia
May 01 11:27:40 k8s-worker-02 kubelet[334]: E0501 11:27:40.175654     334 container_manager_linux.go:583] [ContainerManager]: Fail to get rootfs information unable to find data for
May 01 11:27:40 k8s-worker-02 kubelet[334]: I0501 11:27:40.175765     334 fs_resource_analyzer.go:66] Starting FS ResourceAnalyzer
May 01 11:27:40 k8s-worker-02 kubelet[334]: I0501 11:27:40.176241     334 volume_manager.go:247] Starting Kubelet Volume Manager
lines 1-41

知道为什么我的工作节点显示为 NotReady 吗？

编辑：我使用 kubectl describe nodes 命令跟踪了错误：

Name:               k8s-worker-02
Roles:              <none>
Labels:             beta.kubernetes.io/arch=arm
                    beta.kubernetes.io/os=linux
                    kubernetes.io/hostname=k8s-worker-02
Annotations:        node.alpha.kubernetes.io/ttl=0
                    volumes.kubernetes.io/controller-managed-attach-detach=true
Taints:             <none>
CreationTimestamp:  Tue, 01 May 2018 11:26:50 +0000
Conditions:
  Type             Status  LastHeartbeatTime                 LastTransitionTime                Reason                       Message
  ----             ------  -----------------                 ------------------                ------                       -------
  OutOfDisk        False   Tue, 01 May 2018 11:40:17 +0000   Tue, 01 May 2018 11:26:43 +0000   KubeletHasSufficientDisk     kubelet has sufficient disk space available
  MemoryPressure   False   Tue, 01 May 2018 11:40:17 +0000   Tue, 01 May 2018 11:26:43 +0000   KubeletHasSufficientMemory   kubelet has sufficient memory available
  DiskPressure     False   Tue, 01 May 2018 11:40:17 +0000   Tue, 01 May 2018 11:26:43 +0000   KubeletHasNoDiskPressure     kubelet has no disk pressure
  Ready            False   Tue, 01 May 2018 11:40:17 +0000   Tue, 01 May 2018 11:26:43 +0000   KubeletNotReady              runtime network not ready: NetworkReady=false reason:NetworkPluginNotReady message:docker: network plugin is not ready: cni config uninitialized. WARNING: CPU hardcapping unsupported

我该如何解决这个问题？

Answer 1

1. 在主节​​点中转到您的主节点 go to /etc/cni/net.d
2. 在该文件夹中，您将找到一个 cni 配置文件
3. 复制该文件并将相同的文件上传到工作节点的 /etc/cni/net.d 中
4. 您的工作节点将在 1 到 2 分钟内准备就绪
5. 如果这不起作用，请添加评论

Answer 2

您需要尝试以下解决方案，看看其中任何一个是否有帮助

1. 检查您的防火墙状态。如果它正在运行，请停止它。
2. 检查您的 Kube-dns 状态。有时它可能已关闭或抛出一些错误
3. 尝试重新加载并重启你的 Kubelet

Answer 3

我遇到了同样的问题，并且像某些人一样，在安装在标准完全正常的设备上时，我有一个死亡之吻，所以在我将工作节点重新加入主节点之前，任何地方的项目都没有帮助 .

我的安装是在三台物理机上。一名师傅和两名工人。所有需要的重新启动。

我没想到它会起作用，但它确实起作用了。可能对你不起作用，但如果没有其他方法，也许试一试。您将需要您可能没有的加入令牌，但我将向您展示如何获取它，这样您就不必转到另一组页面并搜索它：

sudo kubeadm token list

复制TOKEN字段数据，输出是这样的（不，那不是我真实的）：

代币 ow3v08ddddgmgzfkdkdkd7 18h 2018-07-30T12:39:53-05:00 authentication,signing 由“kubeadm init”生成的默认引导令牌。 system:bootstrappers:kubeadm:default-node-token

然后在此处加入集群。主节点IP是你机器的真实IP地址：

sudo kubeadm join --token <YOUR TOKEN HASH> <MASTER_NODE_IP>:6443 --discovery-token-unsafe-skip-ca-verification

Answer 4

我设法解决了这个问题！我就是这样做的：

$sudo nano /etc/systemd/system/kubelet.service.d/10-kubeadm.conf

在那之后我只是评论了包含 KUBELET_NETWORK_ARGS 的行

之后我刚刚重新启动系统，我可以看到节点处于就绪状态！

Answer 5

您也可以按如下方式安装 Flannel CNI 插件

git clone https://github.com/containernetworking/cni
cd cni 
git checkout v0.5.2
./build.sh
cp bin/* /opt/cni/bin
mkdir -p /etc/cni/net.d

下载 kube-flannel.yml 和 kube-flannel-rbac.yml from here

然后试试这个，你的节点应该准备好了

kubectl apply -f kube-flannel.yml -f kube-flannel-rbac.yml