未能使用kops创建k8s集群

Question

我正在尝试使用kops实用程序创建一个k8s集群，但是，我得到了以下错误。

我在route53 connectingclouds.in有一个公共宿舍

https://www.nslookup.io/domains/connectingclouds.in/dns-records/#google

kops create cluster --cloud=aws --node-count=1 --name=devk8s.connectingclouds.in --zones=ap-south-1a --master-zones=ap-south-1a --state=s3://dev-k8s-connectingclouds --dns-zone=connectingclouds.in  --node-size=t2.micro --master-size=t2.micro --master-volume-size 10 --node-volume-size 10 --networking calico 

​

​

​

​

Answer 1

您在那里看到的错误意味着控制平面无法更新您的API DNS条目。当名为dns-控制器的组件不运行时，就会发生这种情况。

您正在尝试使用非常小的实例作为您的控制飞机。小于t3.media的实例(默认为t3.media)可能无法运行控制平面组件。

Answer 2

这是对我有用的东西。希望能帮上忙。

使用kops创建k8s集群

生产-环境/工具/kops

启动/安装

先决条件

您应该拥有一个域，例如，在本例中，我拥有domainname.com

这将创建一个默认的托管区域，并包含宿主- zone -ID=XXXXXXXX。

$ dig ns domainname.com | egrep "ANSWER SECTION" -A 4
;; ANSWER SECTION:
domainname.com.           172532  IN      NS      ns-945.awsdns-54.net.
domainname.com.           172532  IN      NS      ns-1991.awsdns-56.co.uk.
domainname.com.           172532  IN      NS      ns-157.awsdns-19.com.
domainname.com.           172532  IN      NS      ns-1442.awsdns-52.org.

$ dig soa domainname.com | egrep "ANSWER SECTION" -A 2
;; ANSWER SECTION:
domainname.com.           820     IN      SOA     ns-157.awsdns-19.com. awsdns-hostmaster.amazon.com. 1 7200 900 1209600 86400

• 安装二进制

$ brew update && brew install kops

$ kops version            
Version 1.19.1

设置IAM用户

AWS_ACCESS_KEY_ID=xxxxxxxxxxxxxxxxx
AWS_SECRET_ACCESS_KEY=xxxxxxxxxxxxxxx

• 创建给键的IAM角色

aws configure

• 创建桶

$ bucket_name=k8-kops-stage-test

$ aws s3api create-bucket --bucket ${bucket_name} --region us-east-1  
{
    "Location": "/k8-kops-stage-test"
}

• 启用版本

$ aws s3api put-bucket-versioning --bucket ${bucket_name} --versioning-configuration Status=Enabled 

• 创建集群

$ export KOPS_CLUSTER_NAME=k8.domainname.com     
$ export KOPS_STATE_STORE=s3://${bucket_name}
$ kops create cluster --node-count=1 --node-size=c5.2xlarge --master-count=1 --master-size=c5.xlarge --zones=eu-west-1a --name=${KOPS_CLUSTER_NAME} --yes
.
.
I0320 14:13:03.437182   44597 create_cluster.go:713] Using SSH public key: /Users/myusername/.ssh/id_rsa.pub
.
.
kops has set your kubectl context to k8.domainname.com

Cluster is starting.  It should be ready in a few minutes.

Suggestions:
 * validate cluster: kops validate cluster --wait 10m
 * list nodes: kubectl get nodes --show-labels
 * ssh to the master: ssh -i ~/.ssh/id_rsa ubuntu@api.k8.domainname.com
 * the ubuntu user is specific to Ubuntu. If not using Ubuntu please use the appropriate user based on your OS.
 * read about installing addons at: https://kops.sigs.k8s.io/operations/addons.

• 验证集群

kops validate cluster --wait 10m
.
.

W0320 14:18:53.164348   44767 validate_cluster.go:173] (will retry): unexpected error during validation: unable to resolve Kubernetes cluster API URL dns: lookup api.k8.domainname.com: no such host
INSTANCE GROUPS
NAME                    ROLE    MACHINETYPE     MIN     MAX     SUBNETS
master-eu-west-1a       Master  c5.xlarge       1       1       eu-west-1a
nodes-eu-west-1a        Node    c5.2xlarge      1       1       eu-west-1a

NODE STATUS
NAME                                            ROLE    READY
ip-172-20-54-246.eu-west-1.compute.internal     master  True
ip-172-20-55-44.eu-west-1.compute.internal      node    True

Your cluster k8.domainname.com is ready

• 删除群集

kops delete cluster --name ${KOPS_CLUSTER_NAME} --yes

• 导出一个具有管理权限的kubeconfig，(注意，这将有一个TTL)

$ kops export kubecfg --admin --kubeconfig ~/workspace/kubeconfig --state=s3://${bucket_name}

Answer 3

将实例类型从t2.micro更改为t2.media。