Failed create pod sandbox: rpc error: code = Unknown desc = NetworkPlugin cni failed to set up pod network
Failed create pod sandbox: rpc error: code = Unknown desc = NetworkPlugin cni failed to set up pod network
在 k8s(v1.10) 集群上创建 Redis POD 并且 POD 创建卡在 "ContainerCreating"
Type Reason Age From Message
---- ------ ---- ---- -------
Normal Scheduled 30m default-scheduler Successfully assigned redis to k8snode02
Normal SuccessfulMountVolume 30m kubelet, k8snode02 MountVolume.SetUp succeeded for volume "default-token-f8tcg"
Warning FailedCreatePodSandBox 5m (x1202 over 30m) kubelet, k8snode02 Failed create pod sandbox: rpc error: code = Unknown desc = NetworkPlugin cni failed to set up pod "redis_default" network: failed to find plugin "loopback" in path [/opt/loopback/bin /opt/cni/bin]
Normal SandboxChanged 47s (x1459 over 30m) kubelet, k8snode02 Pod sandbox changed, it will be killed and re-created.
确保 /etc/cni/net.d 和它的 /opt/cni/bin 朋友都存在并且正确填充了 CNI configuration files and binaries on all Nodes. For flannel specifically, one might make use of the flannel cni repo
当我使用 calico 作为 CNI 时，我遇到了类似的问题。
容器仍处于创建状态，我检查了 master 上的 /etc/cni/net.d 和 /opt/cni/bin 都存在，但不确定工作节点是否也需要。
root@KubernetesMaster:/opt/cni/bin# kubectl get pods
NAME READY STATUS RESTARTS AGE
nginx-5c7588df-5zds6 0/1 ContainerCreating 0 21m
root@KubernetesMaster:/opt/cni/bin# kubectl get nodes
NAME STATUS ROLES AGE VERSION
kubernetesmaster Ready master 26m v1.13.4
kubernetesslave1 Ready <none> 22m v1.13.4
root@KubernetesMaster:/opt/cni/bin#
kubectl describe pods
Name: nginx-5c7588df-5zds6
Namespace: default
Priority: 0
PriorityClassName: <none>
Node: kubernetesslave1/10.0.3.80
Start Time: Sun, 17 Mar 2019 05:13:30 +0000
Labels: app=nginx
pod-template-hash=5c7588df
Annotations: <none>
Status: Pending
IP:
Controlled By: ReplicaSet/nginx-5c7588df
Containers:
nginx:
Container ID:
Image: nginx
Image ID:
Port: <none>
Host Port: <none>
State: Waiting
Reason: ContainerCreating
Ready: False
Restart Count: 0
Environment: <none>
Mounts:
/var/run/secrets/kubernetes.io/serviceaccount from default-token-qtfbs (ro)
Conditions:
Type Status
Initialized True
Ready False
ContainersReady False
PodScheduled True
Volumes:
default-token-qtfbs:
Type: Secret (a volume populated by a Secret)
SecretName: default-token-qtfbs
Optional: false
QoS Class: BestEffort
Node-Selectors: <none>
Tolerations: node.kubernetes.io/not-ready:NoExecute for 300s
node.kubernetes.io/unreachable:NoExecute for 300s
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal Scheduled 18m default-scheduler Successfully assigned default/nginx-5c7588df-5zds6 to kubernetesslave1
Warning FailedCreatePodSandBox 18m kubelet, kubernetesslave1 Failed create pod sandbox: rpc error: code = Unknown desc = failed to set up sandbox container "123d527490944d80f44b1976b82dbae5dc56934aabf59cf89f151736d7ea8adc" network for pod "nginx-5c7588df-5zds6": NetworkPlugin cni failed to set up pod "nginx-5c7588df-5zds6_default" network: stat /var/lib/calico/nodename: no such file or directory: check that the calico/node container is running and has mounted /var/lib/calico/
Warning FailedCreatePodSandBox 18m kubelet, kubernetesslave1 Failed create pod sandbox: rpc error: code = Unknown desc = failed to set up sandbox container "8cc5e62ebaab7075782c2248e00d795191c45906cc9579464a00c09a2bc88b71" network for pod "nginx-5c7588df-5zds6": NetworkPlugin cni failed to set up pod "nginx-5c7588df-5zds6_default" network: stat /var/lib/calico/nodename: no such file or directory: check that the calico/node container is running and has mounted /var/lib/calico/
Warning FailedCreatePodSandBox 18m kubelet, kubernetesslave1 Failed create pod sandbox: rpc error: code = Unknown desc = failed to set up sandbox container "30ffdeace558b0935d1ed3c2e59480e2dd98e983b747dacae707d1baa222353f" network for pod "nginx-5c7588df-5zds6": NetworkPlugin cni failed to set up pod "nginx-5c7588df-5zds6_default" network: stat /var/lib/calico/nodename: no such file or directory: check that the calico/node container is running and has mounted /var/lib/calico/
Warning FailedCreatePodSandBox 18m kubelet, kubernetesslave1 Failed create pod sandbox: rpc error: code = Unknown desc = failed to set up sandbox container "630e85451b6ce2452839c4cfd1ecb9acce4120515702edf29421c123cf231213" network for pod "nginx-5c7588df-5zds6": NetworkPlugin cni failed to set up pod "nginx-5c7588df-5zds6_default" network: stat /var/lib/calico/nodename: no such file or directory: check that the calico/node container is running and has mounted /var/lib/calico/
Warning FailedCreatePodSandBox 18m kubelet, kubernetesslave1 Failed create pod sandbox: rpc error: code = Unknown desc = failed to set up sandbox container "820b919b7edcfc3081711bb78b79d33e5be3f7dafcbad29fe46b6d7aa22227aa" network for pod "nginx-5c7588df-5zds6": NetworkPlugin cni failed to set up pod "nginx-5c7588df-5zds6_default" network: stat /var/lib/calico/nodename: no such file or directory: check that the calico/node container is running and has mounted /var/lib/calico/
Warning FailedCreatePodSandBox 18m kubelet, kubernetesslave1 Failed create pod sandbox: rpc error: code = Unknown desc = failed to set up sandbox container "abbfb5d2756f12802072039dec20ba52f546ae755aaa642a9a75c86577be589f" network for pod "nginx-5c7588df-5zds6": NetworkPlugin cni failed to set up pod "nginx-5c7588df-5zds6_default" network: stat /var/lib/calico/nodename: no such file or directory: check that the calico/node container is running and has mounted /var/lib/calico/
Warning FailedCreatePodSandBox 18m kubelet, kubernetesslave1 Failed create pod sandbox: rpc error: code = Unknown desc = failed to set up sandbox container "dfeb46ffda4d0f8a434f3f3af04328fcc4b6c7cafaa62626e41b705b06d98cc4" network for pod "nginx-5c7588df-5zds6": NetworkPlugin cni failed to set up pod "nginx-5c7588df-5zds6_default" network: stat /var/lib/calico/nodename: no such file or directory: check that the calico/node container is running and has mounted /var/lib/calico/
Warning FailedCreatePodSandBox 18m kubelet, kubernetesslave1 Failed create pod sandbox: rpc error: code = Unknown desc = failed to set up sandbox container "9ae3f47bb0282a56e607779d3267127ee8b0ae1d7f416f5a184682119203b1c8" network for pod "nginx-5c7588df-5zds6": NetworkPlugin cni failed to set up pod "nginx-5c7588df-5zds6_default" network: stat /var/lib/calico/nodename: no such file or directory: check that the calico/node container is running and has mounted /var/lib/calico/
Warning FailedCreatePodSandBox 18m kubelet, kubernetesslave1 Failed create pod sandbox: rpc error: code = Unknown desc = failed to set up sandbox container "546d07f1864728b2e2675c066775f94d658e221ada5fb4ed6bf6689ec7b8de23" network for pod "nginx-5c7588df-5zds6": NetworkPlugin cni failed to set up pod "nginx-5c7588df-5zds6_default" network: stat /var/lib/calico/nodename: no such file or directory: check that the calico/node container is running and has mounted /var/lib/calico/
Normal SandboxChanged 18m (x12 over 18m) kubelet, kubernetesslave1 Pod sandbox changed, it will be killed and re-created.
Warning FailedCreatePodSandBox 3m39s (x829 over 18m) kubelet, kubernetesslave1 (combined from similar events): Failed create pod sandbox: rpc error: code = Unknown desc = failed to set up sandbox container "f586be437843537a3082f37ad139c88d0eacfbe99ddf00621efd4dc049a268cc" network for pod "nginx-5c7588df-5zds6": NetworkPlugin cni failed to set up pod "nginx-5c7588df-5zds6_default" network: stat /var/lib/calico/nodename: no such file or directory: check that the calico/node container is running and has mounted /var/lib/calico/
root@KubernetesMaster:/etc/cni/net.d#
在工作节点上，NGINX 试图启动但退出了，我不确定这里发生了什么 - 我是 kubernetes 的新手并且无法解决这个问题 -
root@kubernetesslave1:/home/ubuntu# docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
5ad5500e8270 fadcc5d2b066 "/usr/local/bin/kube…" 3 minutes ago Up 3 minutes k8s_kube-proxy_kube-proxy-f24gd_kube-system_4e2d313a-4873-11e9-a33a-06516e7d78c4_1
b1c9929ebe9e k8s.gcr.io/pause:3.1 "/pause" 3 minutes ago Up 3 minutes k8s_POD_calico-node-749qx_kube-system_4e2d8c9c-4873-11e9-a33a-06516e7d78c4_1
ceb78340b563 k8s.gcr.io/pause:3.1 "/pause" 3 minutes ago Up 3 minutes k8s_POD_kube-proxy-f24gd_kube-system_4e2d313a-4873-11e9-a33a-06516e7d78c4_1
root@kubernetesslave1:/home/ubuntu# docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
5ad5500e8270 fadcc5d2b066 "/usr/local/bin/kube…" 3 minutes ago Up 3 minutes k8s_kube-proxy_kube-proxy-f24gd_kube-system_4e2d313a-4873-11e9-a33a-06516e7d78c4_1
b1c9929ebe9e k8s.gcr.io/pause:3.1 "/pause" 3 minutes ago Up 3 minutes k8s_POD_calico-node-749qx_kube-system_4e2d8c9c-4873-11e9-a33a-06516e7d78c4_1
ceb78340b563 k8s.gcr.io/pause:3.1 "/pause" 3 minutes ago Up 3 minutes k8s_POD_kube-proxy-f24gd_kube-system_4e2d313a-4873-11e9-a33a-06516e7d78c4_1
root@kubernetesslave1:/home/ubuntu# docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
5ad5500e8270 fadcc5d2b066 "/usr/local/bin/kube…" 3 minutes ago Up 3 minutes k8s_kube-proxy_kube-proxy-f24gd_kube-system_4e2d313a-4873-11e9-a33a-06516e7d78c4_1
b1c9929ebe9e k8s.gcr.io/pause:3.1 "/pause" 3 minutes ago Up 3 minutes k8s_POD_calico-node-749qx_kube-system_4e2d8c9c-4873-11e9-a33a-06516e7d78c4_1
ceb78340b563 k8s.gcr.io/pause:3.1 "/pause" 3 minutes ago Up 3 minutes k8s_POD_kube-proxy-f24gd_kube-system_4e2d313a-4873-11e9-a33a-06516e7d78c4_1
root@kubernetesslave1:/home/ubuntu# docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
94b2994401d0 k8s.gcr.io/pause:3.1 "/pause" 1 second ago Up Less than a second k8s_POD_nginx-5c7588df-5zds6_default_677a722b-4873-11e9-a33a-06516e7d78c4_534
5ad5500e8270 fadcc5d2b066 "/usr/local/bin/kube…" 4 minutes ago Up 4 minutes k8s_kube-proxy_kube-proxy-f24gd_kube-system_4e2d313a-4873-11e9-a33a-06516e7d78c4_1
b1c9929ebe9e k8s.gcr.io/pause:3.1 "/pause" 4 minutes ago Up 4 minutes k8s_POD_calico-node-749qx_kube-system_4e2d8c9c-4873-11e9-a33a-06516e7d78c4_1
ceb78340b563 k8s.gcr.io/pause:3.1 "/pause" 4 minutes ago Up 4 minutes k8s_POD_kube-proxy-f24gd_kube-system_4e2d313a-4873-11e9-a33a-06516e7d78c4_1
root@kubernetesslave1:/home/ubuntu# docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
5ad5500e8270 fadcc5d2b066 "/usr/local/bin/kube…" 4 minutes ago Up 4 minutes k8s_kube-proxy_kube-proxy-f24gd_kube-system_4e2d313a-4873-11e9-a33a-06516e7d78c4_1
b1c9929ebe9e k8s.gcr.io/pause:3.1 "/pause" 4 minutes ago Up 4 minutes k8s_POD_calico-node-749qx_kube-system_4e2d8c9c-4873-11e9-a33a-06516e7d78c4_1
ceb78340b563 k8s.gcr.io/pause:3.1 "/pause" 4 minutes ago Up 4 minutes k8s_POD_kube-proxy-f24gd_kube-system_4e2d313a-4873-11e9-a33a-06516e7d78c4_1
root@kubernetesslave1:/home/ubuntu# docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
f72500cae2b7 k8s.gcr.io/pause:3.1 "/pause" 1 second ago Up Less than a second k8s_POD_nginx-5c7588df-5zds6_default_677a722b-4873-11e9-a33a-06516e7d78c4_585
5ad5500e8270 fadcc5d2b066 "/usr/local/bin/kube…" 4 minutes ago Up 4 minutes k8s_kube-proxy_kube-proxy-f24gd_kube-system_4e2d313a-4873-11e9-a33a-06516e7d78c4_1
b1c9929ebe9e k8s.gcr.io/pause:3.1 "/pause" 4 minutes ago Up 4 minutes k8s_POD_calico-node-749qx_kube-system_4e2d8c9c-4873-11e9-a33a-06516e7d78c4_1
ceb78340b563 k8s.gcr.io/pause:3.1 "/pause" 4 minutes ago Up 4 minutes k8s_POD_kube-proxy-f24gd_kube-system_4e2d313a-4873-11e9-a33a-06516e7d78c4_1
root@kubernetesslave1:/home/ubuntu# docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
5ad5500e8270 fadcc5d2b066 "/usr/local/bin/kube…" 5 minutes ago Up 5 minutes k8s_kube-proxy_kube-proxy-f24gd_kube-system_4e2d313a-4873-11e9-a33a-06516e7d78c4_1
b1c9929ebe9e k8s.gcr.io/pause:3.1 "/pause" 5 minutes ago Up 5 minutes k8s_POD_calico-node-749qx_kube-system_4e2d8c9c-4873-11e9-a33a-06516e7d78c4_1
ceb78340b563 k8s.gcr.io/pause:3.1 "/pause" 5 minutes ago Up 5 minutes k8s_POD_kube-proxy-f24gd_kube-system_4e2d313a-4873-11e9-a33a-06516e7d78c4_1
我也在工作节点上检查了 /etc/cni/net.d & /opt/cni/bin，它在那里 -
root@kubernetesslave1:/home/ubuntu# cd /etc/cni
root@kubernetesslave1:/etc/cni# ls -ltr
total 4
drwxr-xr-x 2 root root 4096 Mar 17 05:19 net.d
root@kubernetesslave1:/etc/cni# cd /opt/cni
root@kubernetesslave1:/opt/cni# ls -ltr
total 4
drwxr-xr-x 2 root root 4096 Mar 17 05:19 bin
root@kubernetesslave1:/opt/cni# cd bin
root@kubernetesslave1:/opt/cni/bin# ls -ltr
total 107440
-rwxr-xr-x 1 root root 3890407 Aug 17 2017 bridge
-rwxr-xr-x 1 root root 3475802 Aug 17 2017 ipvlan
-rwxr-xr-x 1 root root 3520724 Aug 17 2017 macvlan
-rwxr-xr-x 1 root root 3877986 Aug 17 2017 ptp
-rwxr-xr-x 1 root root 3475750 Aug 17 2017 vlan
-rwxr-xr-x 1 root root 9921982 Aug 17 2017 dhcp
-rwxr-xr-x 1 root root 2605279 Aug 17 2017 sample
-rwxr-xr-x 1 root root 32351072 Mar 17 05:19 calico
-rwxr-xr-x 1 root root 31490656 Mar 17 05:19 calico-ipam
-rwxr-xr-x 1 root root 2856252 Mar 17 05:19 flannel
-rwxr-xr-x 1 root root 3084347 Mar 17 05:19 loopback
-rwxr-xr-x 1 root root 3036768 Mar 17 05:19 host-local
-rwxr-xr-x 1 root root 3550877 Mar 17 05:19 portmap
-rwxr-xr-x 1 root root 2850029 Mar 17 05:19 tuning
root@kubernetesslave1:/opt/cni/bin#
AWS EKS 尚不支持 t3a、m5ad r5ad 实例
当我在 AWS EKS 上添加 PVC 时，我遇到了这个问题。
将 aws-node CNI 插件更新到最新版本解决了它 -
https://docs.aws.amazon.com/eks/latest/userguide/cni-upgrades.html
以下步骤重置了 kubernetes 集群并帮助我解决了问题。
- 全部停止运行pods
- 从集群中删除所有工作节点
- 在主节点和节点上执行 kubeadm 重置
- 启动主节点
kubeadm init --apiserver-advertise-address
- 安装 Pod 网络“WeaveNet”
kubectl apply -f "https://cloud.weave.works/k8s/net?k8s-version=$(kubectl version | base64 | tr -d '\n')&env.IPALLOC_RANGE=192.168.0.0 /16
- 将节点加入集群
- 重启所有节点
我在 GCP 上的 GKE 集群和我的抢占式节点池之一遇到了这个问题。感谢@mdaniel 检查 /etc/cni/net.d 完整性的提示，我可以使用命令 gcloud compute ssh <name of some node> --zone <zone-of-cluster> --internal-ip 通过 ssh 将问题再次重现到测试集群的节点中。然后我简单地编辑了文件 /etc/cni/net.d/10-gke-ptp.conflist 并弄乱了 "routes": [ {"dst": "0.0.0.0/0"} ] 上的值（从 0.0.0.0/0 更改为 1.0.0.0/0）。
之后，我删除了其中 运行 的 pods，它们都陷入了 ContainerCreating 状态，永远生成 kublet 事件，错误为 Failed create pod sandbox: rpc error: code...
请注意，为了进行测试，我将我的节点池设置为最多有 1 个节点。否则它将扩展一个新节点，并且 pods 将在新节点上重新创建。在我的生产事件中，节点池达到了最大节点数，因此将我的测试设置为最大 1 个节点重现了类似的情况。
从那以后，从 GKE 中删除节点解决了生产中的问题，我创建了一个 Python 脚本，列出集群上的所有事件并过滤具有关键字 "Failed create pod sandbox: rpc error: code" 的事件。然后我遍历所有事件并获取它们的 pods，然后从 pods 获取节点。最后，我遍历节点，从 Kubernetes API 和 Compute API 及其各自的 Python 客户端中删除它们。对于 Python 脚本，我使用了库：kubernetes 和 google-cloud-compute.
这是脚本的简化版本。使用前测试一下：
from kubernetes import client, config
from google.cloud.compute_v1.services.instances import InstancesClient
ERROR_KEYWORDS = [
'Failed to create pod sandbox'.lower()
]
config.load_kube_config()
v1 = client.CoreV1Api()
events_result = v1.list_event_for_all_namespaces()
filtered_events = []
# filter only the events containing ERROR_KEYWORDS
for event in events_result.items:
for error_keyword in ERROR_KEYWORDS:
if error_keyword in event.message.lower():
filtered_events.append(event)
# gets the list of pods from those events
pods_list = {}
for event in filtered_events:
try:
pod = v1.read_namespaced_pod(
event.involved_object.name,
namespace=event.involved_object.namespace
)
pod_dict = {
"name": event.involved_object.name,
"namespace": event.involved_object.namespace,
"node": pod.spec.node_name
}
pods_list[event.involved_object.name] = pod_dict
except Exception as e:
pass
# Get the nodes from those pods
broken_nodes = set()
for name, pod_dict in pods_list.items():
if pod_dict.get('node'):
broken_nodes.add(pod_dict["node"])
broken_nodes = list(broken_nodes)
# Deletes the nodes from both Kubernetes API and Compute Engine API
if broken_nodes:
broken_nodes_str = ", ".join(broken_nodes)
print(f'BROKEN NODES: "{broken_nodes_str}"')
for node in broken_nodes:
try:
api_response = v1.delete_node(node)
except Exception as e:
pass
time.sleep(30)
try:
result = gcp_client.delete(project=PROJECT_ID, zone=CLUSTER_ZONE, instance=node)
except Exception as e:
pass
kubectl drain node1 node2 --delete-local-data --force --ignore-daemonsets
刚打算在所有节点上驱逐pods，没想到一直报错的pods变成了运行。你可以尝试执行一下，希望对你有用
在 k8s(v1.10) 集群上创建 Redis POD 并且 POD 创建卡在 "ContainerCreating"
Type Reason Age From Message
---- ------ ---- ---- -------
Normal Scheduled 30m default-scheduler Successfully assigned redis to k8snode02
Normal SuccessfulMountVolume 30m kubelet, k8snode02 MountVolume.SetUp succeeded for volume "default-token-f8tcg"
Warning FailedCreatePodSandBox 5m (x1202 over 30m) kubelet, k8snode02 Failed create pod sandbox: rpc error: code = Unknown desc = NetworkPlugin cni failed to set up pod "redis_default" network: failed to find plugin "loopback" in path [/opt/loopback/bin /opt/cni/bin]
Normal SandboxChanged 47s (x1459 over 30m) kubelet, k8snode02 Pod sandbox changed, it will be killed and re-created.
确保 /etc/cni/net.d 和它的 /opt/cni/bin 朋友都存在并且正确填充了 CNI configuration files and binaries on all Nodes. For flannel specifically, one might make use of the flannel cni repo
当我使用 calico 作为 CNI 时，我遇到了类似的问题。
容器仍处于创建状态，我检查了 master 上的 /etc/cni/net.d 和 /opt/cni/bin 都存在，但不确定工作节点是否也需要。
root@KubernetesMaster:/opt/cni/bin# kubectl get pods
NAME READY STATUS RESTARTS AGE
nginx-5c7588df-5zds6 0/1 ContainerCreating 0 21m
root@KubernetesMaster:/opt/cni/bin# kubectl get nodes
NAME STATUS ROLES AGE VERSION
kubernetesmaster Ready master 26m v1.13.4
kubernetesslave1 Ready <none> 22m v1.13.4
root@KubernetesMaster:/opt/cni/bin#
kubectl describe pods
Name: nginx-5c7588df-5zds6
Namespace: default
Priority: 0
PriorityClassName: <none>
Node: kubernetesslave1/10.0.3.80
Start Time: Sun, 17 Mar 2019 05:13:30 +0000
Labels: app=nginx
pod-template-hash=5c7588df
Annotations: <none>
Status: Pending
IP:
Controlled By: ReplicaSet/nginx-5c7588df
Containers:
nginx:
Container ID:
Image: nginx
Image ID:
Port: <none>
Host Port: <none>
State: Waiting
Reason: ContainerCreating
Ready: False
Restart Count: 0
Environment: <none>
Mounts:
/var/run/secrets/kubernetes.io/serviceaccount from default-token-qtfbs (ro)
Conditions:
Type Status
Initialized True
Ready False
ContainersReady False
PodScheduled True
Volumes:
default-token-qtfbs:
Type: Secret (a volume populated by a Secret)
SecretName: default-token-qtfbs
Optional: false
QoS Class: BestEffort
Node-Selectors: <none>
Tolerations: node.kubernetes.io/not-ready:NoExecute for 300s
node.kubernetes.io/unreachable:NoExecute for 300s
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal Scheduled 18m default-scheduler Successfully assigned default/nginx-5c7588df-5zds6 to kubernetesslave1
Warning FailedCreatePodSandBox 18m kubelet, kubernetesslave1 Failed create pod sandbox: rpc error: code = Unknown desc = failed to set up sandbox container "123d527490944d80f44b1976b82dbae5dc56934aabf59cf89f151736d7ea8adc" network for pod "nginx-5c7588df-5zds6": NetworkPlugin cni failed to set up pod "nginx-5c7588df-5zds6_default" network: stat /var/lib/calico/nodename: no such file or directory: check that the calico/node container is running and has mounted /var/lib/calico/
Warning FailedCreatePodSandBox 18m kubelet, kubernetesslave1 Failed create pod sandbox: rpc error: code = Unknown desc = failed to set up sandbox container "8cc5e62ebaab7075782c2248e00d795191c45906cc9579464a00c09a2bc88b71" network for pod "nginx-5c7588df-5zds6": NetworkPlugin cni failed to set up pod "nginx-5c7588df-5zds6_default" network: stat /var/lib/calico/nodename: no such file or directory: check that the calico/node container is running and has mounted /var/lib/calico/
Warning FailedCreatePodSandBox 18m kubelet, kubernetesslave1 Failed create pod sandbox: rpc error: code = Unknown desc = failed to set up sandbox container "30ffdeace558b0935d1ed3c2e59480e2dd98e983b747dacae707d1baa222353f" network for pod "nginx-5c7588df-5zds6": NetworkPlugin cni failed to set up pod "nginx-5c7588df-5zds6_default" network: stat /var/lib/calico/nodename: no such file or directory: check that the calico/node container is running and has mounted /var/lib/calico/
Warning FailedCreatePodSandBox 18m kubelet, kubernetesslave1 Failed create pod sandbox: rpc error: code = Unknown desc = failed to set up sandbox container "630e85451b6ce2452839c4cfd1ecb9acce4120515702edf29421c123cf231213" network for pod "nginx-5c7588df-5zds6": NetworkPlugin cni failed to set up pod "nginx-5c7588df-5zds6_default" network: stat /var/lib/calico/nodename: no such file or directory: check that the calico/node container is running and has mounted /var/lib/calico/
Warning FailedCreatePodSandBox 18m kubelet, kubernetesslave1 Failed create pod sandbox: rpc error: code = Unknown desc = failed to set up sandbox container "820b919b7edcfc3081711bb78b79d33e5be3f7dafcbad29fe46b6d7aa22227aa" network for pod "nginx-5c7588df-5zds6": NetworkPlugin cni failed to set up pod "nginx-5c7588df-5zds6_default" network: stat /var/lib/calico/nodename: no such file or directory: check that the calico/node container is running and has mounted /var/lib/calico/
Warning FailedCreatePodSandBox 18m kubelet, kubernetesslave1 Failed create pod sandbox: rpc error: code = Unknown desc = failed to set up sandbox container "abbfb5d2756f12802072039dec20ba52f546ae755aaa642a9a75c86577be589f" network for pod "nginx-5c7588df-5zds6": NetworkPlugin cni failed to set up pod "nginx-5c7588df-5zds6_default" network: stat /var/lib/calico/nodename: no such file or directory: check that the calico/node container is running and has mounted /var/lib/calico/
Warning FailedCreatePodSandBox 18m kubelet, kubernetesslave1 Failed create pod sandbox: rpc error: code = Unknown desc = failed to set up sandbox container "dfeb46ffda4d0f8a434f3f3af04328fcc4b6c7cafaa62626e41b705b06d98cc4" network for pod "nginx-5c7588df-5zds6": NetworkPlugin cni failed to set up pod "nginx-5c7588df-5zds6_default" network: stat /var/lib/calico/nodename: no such file or directory: check that the calico/node container is running and has mounted /var/lib/calico/
Warning FailedCreatePodSandBox 18m kubelet, kubernetesslave1 Failed create pod sandbox: rpc error: code = Unknown desc = failed to set up sandbox container "9ae3f47bb0282a56e607779d3267127ee8b0ae1d7f416f5a184682119203b1c8" network for pod "nginx-5c7588df-5zds6": NetworkPlugin cni failed to set up pod "nginx-5c7588df-5zds6_default" network: stat /var/lib/calico/nodename: no such file or directory: check that the calico/node container is running and has mounted /var/lib/calico/
Warning FailedCreatePodSandBox 18m kubelet, kubernetesslave1 Failed create pod sandbox: rpc error: code = Unknown desc = failed to set up sandbox container "546d07f1864728b2e2675c066775f94d658e221ada5fb4ed6bf6689ec7b8de23" network for pod "nginx-5c7588df-5zds6": NetworkPlugin cni failed to set up pod "nginx-5c7588df-5zds6_default" network: stat /var/lib/calico/nodename: no such file or directory: check that the calico/node container is running and has mounted /var/lib/calico/
Normal SandboxChanged 18m (x12 over 18m) kubelet, kubernetesslave1 Pod sandbox changed, it will be killed and re-created.
Warning FailedCreatePodSandBox 3m39s (x829 over 18m) kubelet, kubernetesslave1 (combined from similar events): Failed create pod sandbox: rpc error: code = Unknown desc = failed to set up sandbox container "f586be437843537a3082f37ad139c88d0eacfbe99ddf00621efd4dc049a268cc" network for pod "nginx-5c7588df-5zds6": NetworkPlugin cni failed to set up pod "nginx-5c7588df-5zds6_default" network: stat /var/lib/calico/nodename: no such file or directory: check that the calico/node container is running and has mounted /var/lib/calico/
root@KubernetesMaster:/etc/cni/net.d#
在工作节点上，NGINX 试图启动但退出了，我不确定这里发生了什么 - 我是 kubernetes 的新手并且无法解决这个问题 -
root@kubernetesslave1:/home/ubuntu# docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
5ad5500e8270 fadcc5d2b066 "/usr/local/bin/kube…" 3 minutes ago Up 3 minutes k8s_kube-proxy_kube-proxy-f24gd_kube-system_4e2d313a-4873-11e9-a33a-06516e7d78c4_1
b1c9929ebe9e k8s.gcr.io/pause:3.1 "/pause" 3 minutes ago Up 3 minutes k8s_POD_calico-node-749qx_kube-system_4e2d8c9c-4873-11e9-a33a-06516e7d78c4_1
ceb78340b563 k8s.gcr.io/pause:3.1 "/pause" 3 minutes ago Up 3 minutes k8s_POD_kube-proxy-f24gd_kube-system_4e2d313a-4873-11e9-a33a-06516e7d78c4_1
root@kubernetesslave1:/home/ubuntu# docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
5ad5500e8270 fadcc5d2b066 "/usr/local/bin/kube…" 3 minutes ago Up 3 minutes k8s_kube-proxy_kube-proxy-f24gd_kube-system_4e2d313a-4873-11e9-a33a-06516e7d78c4_1
b1c9929ebe9e k8s.gcr.io/pause:3.1 "/pause" 3 minutes ago Up 3 minutes k8s_POD_calico-node-749qx_kube-system_4e2d8c9c-4873-11e9-a33a-06516e7d78c4_1
ceb78340b563 k8s.gcr.io/pause:3.1 "/pause" 3 minutes ago Up 3 minutes k8s_POD_kube-proxy-f24gd_kube-system_4e2d313a-4873-11e9-a33a-06516e7d78c4_1
root@kubernetesslave1:/home/ubuntu# docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
5ad5500e8270 fadcc5d2b066 "/usr/local/bin/kube…" 3 minutes ago Up 3 minutes k8s_kube-proxy_kube-proxy-f24gd_kube-system_4e2d313a-4873-11e9-a33a-06516e7d78c4_1
b1c9929ebe9e k8s.gcr.io/pause:3.1 "/pause" 3 minutes ago Up 3 minutes k8s_POD_calico-node-749qx_kube-system_4e2d8c9c-4873-11e9-a33a-06516e7d78c4_1
ceb78340b563 k8s.gcr.io/pause:3.1 "/pause" 3 minutes ago Up 3 minutes k8s_POD_kube-proxy-f24gd_kube-system_4e2d313a-4873-11e9-a33a-06516e7d78c4_1
root@kubernetesslave1:/home/ubuntu# docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
94b2994401d0 k8s.gcr.io/pause:3.1 "/pause" 1 second ago Up Less than a second k8s_POD_nginx-5c7588df-5zds6_default_677a722b-4873-11e9-a33a-06516e7d78c4_534
5ad5500e8270 fadcc5d2b066 "/usr/local/bin/kube…" 4 minutes ago Up 4 minutes k8s_kube-proxy_kube-proxy-f24gd_kube-system_4e2d313a-4873-11e9-a33a-06516e7d78c4_1
b1c9929ebe9e k8s.gcr.io/pause:3.1 "/pause" 4 minutes ago Up 4 minutes k8s_POD_calico-node-749qx_kube-system_4e2d8c9c-4873-11e9-a33a-06516e7d78c4_1
ceb78340b563 k8s.gcr.io/pause:3.1 "/pause" 4 minutes ago Up 4 minutes k8s_POD_kube-proxy-f24gd_kube-system_4e2d313a-4873-11e9-a33a-06516e7d78c4_1
root@kubernetesslave1:/home/ubuntu# docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
5ad5500e8270 fadcc5d2b066 "/usr/local/bin/kube…" 4 minutes ago Up 4 minutes k8s_kube-proxy_kube-proxy-f24gd_kube-system_4e2d313a-4873-11e9-a33a-06516e7d78c4_1
b1c9929ebe9e k8s.gcr.io/pause:3.1 "/pause" 4 minutes ago Up 4 minutes k8s_POD_calico-node-749qx_kube-system_4e2d8c9c-4873-11e9-a33a-06516e7d78c4_1
ceb78340b563 k8s.gcr.io/pause:3.1 "/pause" 4 minutes ago Up 4 minutes k8s_POD_kube-proxy-f24gd_kube-system_4e2d313a-4873-11e9-a33a-06516e7d78c4_1
root@kubernetesslave1:/home/ubuntu# docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
f72500cae2b7 k8s.gcr.io/pause:3.1 "/pause" 1 second ago Up Less than a second k8s_POD_nginx-5c7588df-5zds6_default_677a722b-4873-11e9-a33a-06516e7d78c4_585
5ad5500e8270 fadcc5d2b066 "/usr/local/bin/kube…" 4 minutes ago Up 4 minutes k8s_kube-proxy_kube-proxy-f24gd_kube-system_4e2d313a-4873-11e9-a33a-06516e7d78c4_1
b1c9929ebe9e k8s.gcr.io/pause:3.1 "/pause" 4 minutes ago Up 4 minutes k8s_POD_calico-node-749qx_kube-system_4e2d8c9c-4873-11e9-a33a-06516e7d78c4_1
ceb78340b563 k8s.gcr.io/pause:3.1 "/pause" 4 minutes ago Up 4 minutes k8s_POD_kube-proxy-f24gd_kube-system_4e2d313a-4873-11e9-a33a-06516e7d78c4_1
root@kubernetesslave1:/home/ubuntu# docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
5ad5500e8270 fadcc5d2b066 "/usr/local/bin/kube…" 5 minutes ago Up 5 minutes k8s_kube-proxy_kube-proxy-f24gd_kube-system_4e2d313a-4873-11e9-a33a-06516e7d78c4_1
b1c9929ebe9e k8s.gcr.io/pause:3.1 "/pause" 5 minutes ago Up 5 minutes k8s_POD_calico-node-749qx_kube-system_4e2d8c9c-4873-11e9-a33a-06516e7d78c4_1
ceb78340b563 k8s.gcr.io/pause:3.1 "/pause" 5 minutes ago Up 5 minutes k8s_POD_kube-proxy-f24gd_kube-system_4e2d313a-4873-11e9-a33a-06516e7d78c4_1
我也在工作节点上检查了 /etc/cni/net.d & /opt/cni/bin，它在那里 -
root@kubernetesslave1:/home/ubuntu# cd /etc/cni
root@kubernetesslave1:/etc/cni# ls -ltr
total 4
drwxr-xr-x 2 root root 4096 Mar 17 05:19 net.d
root@kubernetesslave1:/etc/cni# cd /opt/cni
root@kubernetesslave1:/opt/cni# ls -ltr
total 4
drwxr-xr-x 2 root root 4096 Mar 17 05:19 bin
root@kubernetesslave1:/opt/cni# cd bin
root@kubernetesslave1:/opt/cni/bin# ls -ltr
total 107440
-rwxr-xr-x 1 root root 3890407 Aug 17 2017 bridge
-rwxr-xr-x 1 root root 3475802 Aug 17 2017 ipvlan
-rwxr-xr-x 1 root root 3520724 Aug 17 2017 macvlan
-rwxr-xr-x 1 root root 3877986 Aug 17 2017 ptp
-rwxr-xr-x 1 root root 3475750 Aug 17 2017 vlan
-rwxr-xr-x 1 root root 9921982 Aug 17 2017 dhcp
-rwxr-xr-x 1 root root 2605279 Aug 17 2017 sample
-rwxr-xr-x 1 root root 32351072 Mar 17 05:19 calico
-rwxr-xr-x 1 root root 31490656 Mar 17 05:19 calico-ipam
-rwxr-xr-x 1 root root 2856252 Mar 17 05:19 flannel
-rwxr-xr-x 1 root root 3084347 Mar 17 05:19 loopback
-rwxr-xr-x 1 root root 3036768 Mar 17 05:19 host-local
-rwxr-xr-x 1 root root 3550877 Mar 17 05:19 portmap
-rwxr-xr-x 1 root root 2850029 Mar 17 05:19 tuning
root@kubernetesslave1:/opt/cni/bin#
AWS EKS 尚不支持 t3a、m5ad r5ad 实例
当我在 AWS EKS 上添加 PVC 时，我遇到了这个问题。
将 aws-node CNI 插件更新到最新版本解决了它 -
https://docs.aws.amazon.com/eks/latest/userguide/cni-upgrades.html
以下步骤重置了 kubernetes 集群并帮助我解决了问题。
- 全部停止运行pods
- 从集群中删除所有工作节点
- 在主节点和节点上执行 kubeadm 重置
- 启动主节点 kubeadm init --apiserver-advertise-address
- 安装 Pod 网络“WeaveNet” kubectl apply -f "https://cloud.weave.works/k8s/net?k8s-version=$(kubectl version | base64 | tr -d '\n')&env.IPALLOC_RANGE=192.168.0.0 /16
- 将节点加入集群
- 重启所有节点
我在 GCP 上的 GKE 集群和我的抢占式节点池之一遇到了这个问题。感谢@mdaniel 检查 /etc/cni/net.d 完整性的提示，我可以使用命令 gcloud compute ssh <name of some node> --zone <zone-of-cluster> --internal-ip 通过 ssh 将问题再次重现到测试集群的节点中。然后我简单地编辑了文件 /etc/cni/net.d/10-gke-ptp.conflist 并弄乱了 "routes": [ {"dst": "0.0.0.0/0"} ] 上的值（从 0.0.0.0/0 更改为 1.0.0.0/0）。
之后，我删除了其中 运行 的 pods，它们都陷入了 ContainerCreating 状态，永远生成 kublet 事件，错误为 Failed create pod sandbox: rpc error: code...
请注意，为了进行测试，我将我的节点池设置为最多有 1 个节点。否则它将扩展一个新节点，并且 pods 将在新节点上重新创建。在我的生产事件中，节点池达到了最大节点数，因此将我的测试设置为最大 1 个节点重现了类似的情况。
从那以后，从 GKE 中删除节点解决了生产中的问题，我创建了一个 Python 脚本，列出集群上的所有事件并过滤具有关键字 "Failed create pod sandbox: rpc error: code" 的事件。然后我遍历所有事件并获取它们的 pods，然后从 pods 获取节点。最后，我遍历节点，从 Kubernetes API 和 Compute API 及其各自的 Python 客户端中删除它们。对于 Python 脚本，我使用了库：kubernetes 和 google-cloud-compute.
这是脚本的简化版本。使用前测试一下：
from kubernetes import client, config
from google.cloud.compute_v1.services.instances import InstancesClient
ERROR_KEYWORDS = [
'Failed to create pod sandbox'.lower()
]
config.load_kube_config()
v1 = client.CoreV1Api()
events_result = v1.list_event_for_all_namespaces()
filtered_events = []
# filter only the events containing ERROR_KEYWORDS
for event in events_result.items:
for error_keyword in ERROR_KEYWORDS:
if error_keyword in event.message.lower():
filtered_events.append(event)
# gets the list of pods from those events
pods_list = {}
for event in filtered_events:
try:
pod = v1.read_namespaced_pod(
event.involved_object.name,
namespace=event.involved_object.namespace
)
pod_dict = {
"name": event.involved_object.name,
"namespace": event.involved_object.namespace,
"node": pod.spec.node_name
}
pods_list[event.involved_object.name] = pod_dict
except Exception as e:
pass
# Get the nodes from those pods
broken_nodes = set()
for name, pod_dict in pods_list.items():
if pod_dict.get('node'):
broken_nodes.add(pod_dict["node"])
broken_nodes = list(broken_nodes)
# Deletes the nodes from both Kubernetes API and Compute Engine API
if broken_nodes:
broken_nodes_str = ", ".join(broken_nodes)
print(f'BROKEN NODES: "{broken_nodes_str}"')
for node in broken_nodes:
try:
api_response = v1.delete_node(node)
except Exception as e:
pass
time.sleep(30)
try:
result = gcp_client.delete(project=PROJECT_ID, zone=CLUSTER_ZONE, instance=node)
except Exception as e:
pass
kubectl drain node1 node2 --delete-local-data --force --ignore-daemonsets
刚打算在所有节点上驱逐pods，没想到一直报错的pods变成了运行。你可以尝试执行一下，希望对你有用