最近我在 ubuntu 22.04 中使用 k8s。 我已经加入了 2 个带有控制平面的工作节点。 Worker节点相关Pod已启动成功。但其他worker节点的pod还没有启动。它不断返回
ImagePullBackOff这是错误-
samrat@k8smaster01:~$ kubectl get pods -A
NAMESPACE NAME READY STATUS RESTARTS AGE
kube-system calico-kube-controllers-57758d645c-f5kn7 1/1 Running 0 67m
kube-system calico-node-2xndp 0/1 Init:ImagePullBackOff 0 12m
kube-system calico-node-m2hxw 1/1 Running 0 67m
kube-system calico-node-phbxs 1/1 Running 0 66m
kube-system coredns-76f75df574-8d75k 1/1 Running 0 135m
kube-system coredns-76f75df574-vkdlb 1/1 Running 0 135m
kube-system etcd-k8smaster01.pubalibank.com.bd 1/1 Running 31 (98m ago) 135m
kube-system kube-apiserver-k8smaster01.pubalibank.com.bd 1/1 Running 3 (98m ago) 135m
kube-system kube-controller-manager-k8smaster01.pubalibank.com.bd 1/1 Running 27 (98m ago) 135m
kube-system kube-proxy-4mk2r 0/1 ErrImagePull 0 8m21s
kube-system kube-proxy-8fzxk 1/1 Running 0 66m
kube-system kube-proxy-rt8z5 1/1 Running 1 (98m ago) 135m
kube-system kube-scheduler-k8smaster01.pubalibank.com.bd 1/1 Running 1 (98m ago) 135m
这是有问题的 pod 的事件。
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal Scheduled 9m56s default-scheduler Successfully assigned kube-system/kube-proxy-4mk2r to k8sworker02.pubalibank.com.bd
Warning Failed 9m40s kubelet Failed to pull image "registry.k8s.io/kube-proxy:v1.29.3": failed to pull and unpack image "registry.k8s.io/kube-proxy:v1.29.3": failed to resolve reference "registry.k8s.io/kube-proxy:v1.29.3": failed to do request: Head "https://registry.k8s.io/v2/kube-proxy/manifests/v1.29.3": dial tcp: lookup registry.k8s.io on 127.0.0.53:53: read udp 127.0.0.1:36562->127.0.0.53:53: i/o timeout
Warning Failed 9m9s kubelet Failed to pull image "registry.k8s.io/kube-proxy:v1.29.3": failed to pull and unpack image "registry.k8s.io/kube-proxy:v1.29.3": failed to resolve reference "registry.k8s.io/kube-proxy:v1.29.3": failed to do request: Head "https://registry.k8s.io/v2/kube-proxy/manifests/v1.29.3": dial tcp: lookup registry.k8s.io on 127.0.0.53:53: read udp 127.0.0.1:50081->127.0.0.53:53: i/o timeout
Warning Failed 8m22s kubelet Failed to pull image "registry.k8s.io/kube-proxy:v1.29.3": failed to pull and unpack image "registry.k8s.io/kube-proxy:v1.29.3": failed to resolve reference "registry.k8s.io/kube-proxy:v1.29.3": failed to do request: Head "https://registry.k8s.io/v2/kube-proxy/manifests/v1.29.3": dial tcp: lookup registry.k8s.io on 127.0.0.53:53: read udp 127.0.0.1:41785->127.0.0.53:53: i/o timeout
Normal Pulling 7m35s (x4 over 10m) kubelet Pulling image "registry.k8s.io/kube-proxy:v1.29.3"
Warning Failed 7m15s (x4 over 9m40s) kubelet Error: ErrImagePull
Warning Failed 7m15s kubelet Failed to pull image "registry.k8s.io/kube-proxy:v1.29.3": failed to pull and unpack image "registry.k8s.io/kube-proxy:v1.29.3": failed to resolve reference "registry.k8s.io/kube-proxy:v1.29.3": failed to do request: Head "https://registry.k8s.io/v2/kube-proxy/manifests/v1.29.3": dial tcp: lookup registry.k8s.io on 127.0.0.53:53: read udp 127.0.0.1:52376->127.0.0.53:53: i/o timeout
Warning Failed 7m4s (x6 over 9m40s) kubelet Error: ImagePullBackOff
Normal BackOff 4m46s (x14 over 9m40s) kubelet Back-off pulling image "registry.k8s.io/kube-proxy:v1.29.3"
控制平面resolve.conf的内容-
nameserver 127.0.0.53
options edns0 trust-ad
search .
这是有问题的节点的resolve.conf-
nameserver 127.0.0.53
options edns0 trust-ad
search .
我这是一个与 DNS 相关的问题。但无法弄清楚问题所在。预先感谢。
检查该特定节点中的容器运行时状态，如果是容器，您可以通过运行此命令来检查状态
sudo systemctl 状态containerd.service
查看containerd服务是否已启动并正在运行。
由于所有与集群相关的应用程序都作为 Pod 运行，因此节点似乎在从注册表中提取 kube-proxy 映像时遇到问题。 kube-proxy 用于在节点内创建与网络相关的内容的 ip 表。看来您已将 calico 用于 CNI（容器网络接口），希望您没有编辑该 yaml 文件。检查容器运行时。如果它没有运行，您需要删除containerd并重新安装。 Kubernetes 版本应该与每个节点（kubelet、kubeadm 和 kubectl）匹配，请尝试重新安装这些节点。您可以将版本设置为环境变量并像这样使用它
**版本=1.29.1-1.1 sudo apt-get install -y kubelet=$版本 kubeadm=$版本 kubectl=$版本 sudo apt-mark 保留 kubelet kubeadm kubectl containerd **