我在这方面遇到了困难，也许答案很简单，所以如果有人知道答案，请在此处评论。
我使用以下清单创建了一个 EKS 集群。
apiVersion: eksctl.io/v1alpha5
kind: ClusterConfig
metadata:
name: test-cluster
region: us-west-2
version: "1.29"
vpc:
subnets:
public:
us-west-2a: { id: subnet-094d01de2dd2148c0 }
us-west-2b: { id: subnet-04429e132a1f42826 }
us-west-2c: { id: subnet-028a738bdafc344c6 }
nodeGroups:
- name: ng-spot
instanceType: t3.medium
labels: { role: builders }
desiredCapacity: 2
minSize: 2
maxSize: 4
volumeSize: 30
ssh:
allow: true
publicKeyName: techies
tags:
Name: ng-spot
maxPodsPerNode: 110
该集群用于测试目的，因此我使用t3.medium具有最大 pod 限制的实例110。
arun@ArunLAL555:~$ k get nodes
NAME STATUS ROLES AGE VERSION
ip-192-168-37-0.us-west-2.compute.internal Ready <none> 26m v1.29.0-eks-5e0fdde
ip-192-168-86-42.us-west-2.compute.internal Ready <none> 26m v1.29.0-eks-5e0fdde
arun@ArunLAL555:~$ kubectl get nodes -o jsonpath='{.items[*].status.allocatable.pods}{"\n"}'
110 110
这确保我可以110在每个节点上创建 pod。
arun@ArunLAL555:~$ k create deployment test-deploy --image nginx --replicas 50
deployment.apps/test-deploy created
arun@ArunLAL555:~$ k get po
NAME READY STATUS RESTARTS AGE
test-deploy-859f95ffcc-2c5k6 0/1 ContainerCreating 0 19s
test-deploy-859f95ffcc-2p9rh 1/1 Running 0 19s
test-deploy-859f95ffcc-468wm 0/1 ContainerCreating 0 18s
.
.
test-deploy-859f95ffcc-xxm7z 0/1 ContainerCreating 0 18s
test-deploy-859f95ffcc-z88x6 1/1 Running 0 19s
此处，剩余的 Pod 没有获取 IP
arun@ArunLAL555:~$ k events po test-deploy-859f95ffcc-xxm7z
1s (x5 over 55s) Warning FailedCreatePodSandBox Pod/test-deploy-859f95ffcc-m7t62 (combined from similar events): Failed to create pod sandbox: rpc error: code = Unknown desc = failed to setup network for sandbox "528eaad224c5578435db12a57a8fa7063a03423b28d57c681bab742cc8389a1a": plugin type="aws-cni" name="aws-cni" failed (add): add cmd: failed to assign an IP address to container
以下是子网及其 IP 可用性
arun@ArunLAL555:~$ aws eks describe-cluster --name test-cluster --query "cluster.resourcesVpcConfig.su
bnetIds"
[
"subnet-094d01de2dd2148c0",
"subnet-04429e132a1f42826",
"subnet-028a738bdafc344c6"
]
arun@ArunLAL555:~$ aws ec2 describe-subnets --subnet-ids subnet-094d01de2dd2148c0 subnet-04429e132a1f42826 subnet-028a738bdafc344c6 --query 'Subnets[*].[SubnetId,AvailableIpAddressCount]' --output text
subnet-028a738bdafc344c6 8167
subnet-094d01de2dd2148c0 8185
subnet-04429e132a1f42826 8168
我已经更新了 VPC CNI
arun@ArunLAL555:~$ kubectl describe daemonset aws-node --namespace kube-system | grep amazon-k8s-cni: | cut -d : -f 3
v1.16.0-eksbuild.1
arun@ArunLAL555:~$ aws eks create-addon --cluster-name test-cluster --addon-name vpc-cni --addon-version v1.17.1-eksbuild.1 \
-service> --service-account-role-arn arn:aws:iam::111122223333:role/AmazonEKSVPCCNIRole
{
"addon": {
"addonName": "vpc-cni",
"clusterName": "test-cluster",
"status": "CREATING",
"addonVersion": "v1.17.1-eksbuild.1",
"health": {
"issues": []
},
"addonArn": "arn:aws:eks:us-west-2:111122223333:addon/test-cluster/vpc-cni/fec7333d-c1fc-c2fc-1287-c14beaa883f8",
"createdAt": "2024-03-22T19:35:54.685000+05:30",
"modifiedAt": "2024-03-22T19:35:54.703000+05:30",
"serviceAccountRoleArn": "arn:aws:iam::111122223333:role/AmazonEKSVPCCNIRole",
"tags": {}
}
}
arun@ArunLAL555:~$ aws eks describe-addon --cluster-name test-cluster --addon-name vpc-cni --query addon.addonVersion --output text
v1.17.1-eksbuild.1
此后，我终止了现有实例，因为节点尚未准备好。
arun@ArunLAL555:~$ k get nodes
NAME STATUS ROLES AGE VERSION
ip-192-168-40-177.us-west-2.compute.internal NotReady <none> 86s v1.29.0-eks-5e0fdde
ip-192-168-83-11.us-west-2.compute.internal NotReady <none> 3m29s v1.29.0-eks-5e0fdde
arun@ArunLAL555:~$ k describe nodes ip-192-168-40-177.us-west-2.compute.internal
Conditions:
Type Status LastHeartbeatTime LastTransitionTime Reason Message
---- ------ ----------------- ------------------ ------ -------
MemoryPressure False Fri, 22 Mar 2024 19:45:20 +0530 Fri, 22 Mar 2024 19:44:49 +0530 KubeletHasSufficientMemory kubelet has sufficient memory available
DiskPressure False Fri, 22 Mar 2024 19:45:20 +0530 Fri, 22 Mar 2024 19:44:49 +0530 KubeletHasNoDiskPressure kubelet has no disk pressure
PIDPressure False Fri, 22 Mar 2024 19:45:20 +0530 Fri, 22 Mar 2024 19:44:49 +0530 KubeletHasSufficientPID kubelet has sufficient PID available
Ready False Fri, 22 Mar 2024 19:45:20 +0530 Fri, 22 Mar 2024 19:44:49 +0530 KubeletNotReady container runtime network not ready: NetworkReady=false reason:NetworkPluginNotReady message:Network plugin returns error: cni plugin not initialized
我想知道为什么会发生这种情况，如果有人知道答案，请评论。
- 首先，为什么即使将 Pod 限制设置为最大值，Pod 仍未获得 IP
- 二、为什么更新VPC CNI插件后节点还没准备好