K8S二进制部署详解,一文教会你部署高可用K8S集群(二)

本文涉及的产品
容器服务 Serverless 版 ACK Serverless,952元额度 多规格
容器服务 Serverless 版 ACK Serverless,317元额度 多规格
云解析 DNS,旗舰版 1个月
简介: K8S二进制部署详解,一文教会你部署高可用K8S集群(二)

K8S二进制部署详解,一文教会你部署高可用K8S集群(一)https://developer.aliyun.com/article/1496870


5.4 部署kube-controller-manager组件

#创建csr请求文件

[root@master01 work ]#cat kube-controller-manager-csr.json 
{
    "CN": "system:kube-controller-manager",
    "key": {
        "algo": "rsa",
        "size": 2048
    },
    "hosts": [
      "127.0.0.1",
      "10.10.0.10",
      "10.10.0.11",
      "10.10.0.12",
      "10.10.0.100"
    ],
    "names": [
      {
        "C": "CN",
        "ST": "Hubei",
        "L": "Wuhan",
        "O": "system:kube-controller-manager",
        "OU": "system"
      }
    ]
}


注: hosts 列表包含所有 kube-controller-manager 节点 IP;

CN 为 system:kube-controller-manager、

O 为 system:kube-controller-manager,

kubernetes 内置的 ClusterRoleBindings system:kube-controller-manager 赋予 kube-controller-manager 工作所需的权限

#生成证书

[root@master01 work ]#cfssl gencert -ca=ca.pem -ca-key=ca-key.pem -config=ca-config.json -profile=kubernetes kube-controller-manager-csr.json | cfssljson -bare kube-controller-manager
2022/10/26 14:27:54 [INFO] generate received request
2022/10/26 14:27:54 [INFO] received CSR
2022/10/26 14:27:54 [INFO] generating key: rsa-2048
2022/10/26 14:27:54 [INFO] encoded CSR
2022/10/26 14:27:54 [INFO] signed certificate with serial number 721895207641224154550279977497269955483545721536
2022/10/26 14:27:54 [WARNING] This certificate lacks a "hosts" field. This makes it unsuitable for
websites. For more information see the Baseline Requirements for the Issuance and Management
of Publicly-Trusted Certificates, v.1.1.6, from the CA/Browser Forum (https://cabforum.org);
specifically, section 10.2.3 ("Information Requirements").

#创建kube-controller-manager的kubeconfig

1.设置集群参数

[root@master01 work ]#kubectl config set-cluster kubernetes --certificate-authority=ca.pem --embed-certs=true --server=https://10.10.0.10:6443 --kubeconfig=kube-controller-manager.kubeconfig
Cluster "kubernetes" set.

[root@master01 work ]#cat kube-controller-manager.kubeconfig 
apiVersion: v1
clusters:
- cluster:
    certificate-authority-data: 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
    server: https://10.10.0.10:6443
  name: kubernetes
contexts: null
current-context: ""
kind: Config
preferences: {}
users: null


2.设置客户端认证参数

用户system:kube-controller-manager 就是kube-controller-manager-csr.json CN设置的值

[root@master01 work ]#kubectl config set-credentials system:kube-controller-manager --client-certificate=kube-controller-manager.pem --client-key=kube-controller-manager-key.pem --embed-certs=true --kubeconfig=kube-controller-manager.kubeconfig

User “system:kube-controller-manager” set.

[root@master01 work ]#cat kube-controller-manager.kubeconfig 
apiVersion: v1
clusters:
- cluster:
    certificate-authority-data: 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
    server: https://10.10.0.10:6443
  name: kubernetes
contexts: null
current-context: ""
kind: Config
preferences: {}
users:
- name: system:kube-controller-manager
  user:
    client-certificate-data: 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
    client-key-data: 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


3.设置上下文参数

[root@master01 work ]#kubectl config set-context system:kube-controller-manager --cluster=kubernetes --user=system:kube-controller-manager --kubeconfig=kube-controller-manager.kubeconfig
Context "system:kube-controller-manager" created.

4.设置当前上下文

[root@master01 work ]#kubectl config use-context system:kube-controller-manager --kubeconfig=kube-controller-manager.kubeconfig
Switched to context "system:kube-controller-manager".
[root@master01 work ]#cat kube-controller-manager.kubeconfig 
apiVersion: v1
clusters:
- cluster:
    certificate-authority-data: LS0tLS1CRUdJTiBDRVJUSUZJQ0FURS0tLS0tCk1JSUR0akNDQXA2Z0F3SUJBZ0lVUkV4RXhidjQ1ZlVBY1hCQlpENGZEZnRqZmtNd0RRWUpLb1pJaHZjTkFRRUwKQlFBd1lURUxNQWtHQTFVRUJoTUNRMDR4RGpBTUJnTlZCQWdUQlVoMVltVnBNUTR3REFZRFZRUUhFd1ZYZFdoaApiakVNTUFvR0ExVUVDaE1EYXpoek1ROHdEUVlEVlFRTEV3WnplWE4wWlcweEV6QVJCZ05WQkFNVENtdDFZbVZ5CmJtVjBaWE13SGhjTk1qSXhNREkxTURnMU5EQXdXaGNOTXpJeE1ESXlNRGcxTkRBd1dqQmhNUXN3Q1FZRFZRUUcKRXdKRFRqRU9NQXdHQTFVRUNCTUZTSFZpWldreERqQU1CZ05WQkFjVEJWZDFhR0Z1TVF3d0NnWURWUVFLRXdOcgpPSE14RHpBTkJnTlZCQXNUQm5ONWMzUmxiVEVUTUJFR0ExVUVBeE1LYTNWaVpYSnVaWFJsY3pDQ0FTSXdEUVlKCktvWklodmNOQVFFQkJRQURnZ0VQQURDQ0FRb0NnZ0VCQUxVV2lYcGZPMFZ6dTBrek8xYllaVUo2dFVnTHp2TDUKTlJwWnpoS2pNZ3pDbmZYM0pHWUM0SWpFdGoxRWNhUGdFOUErbkZ1UndCWlk0a0U3NkE1RjdhaS8rNzdiak5tMApJcWdzY3o2VmdzM0JuMEZSSXVmMlFkQm9OUmNlWWhXL011Z0ZWUlZRSUVQTjRuZGxCVUFScHpNbDk4clpxSVU5CkxTMkZ1WWRwYTRRSi8weVp3YWlYYllxazFDUk5JUkRBMUZJRDh6VEpHeXhRdmZoZ05wRGJTQkhQVlhqK3oxZW4KTU5XQzVDMGt1QWcya3IzcExHUkg3T2dMR1dHV0RuRXVwMUZKMUJONDl2V3IwTW8zQWd6TE1TcTdaRnRGYS9KYgpvN3lOWmZDZnd5NTZQTi9Rckc5eSs0aFBONHVRdG16em1YbUJYT2JJTFF2SXhEN2lKYTVWb0I4Q0F3RUFBYU5tCk1HUXdEZ1lEVlIwUEFRSC9CQVFEQWdFR01CSUdBMVVkRXdFQi93UUlNQVlCQWY4Q0FRSXdIUVlEVlIwT0JCWUUKRk4rdFZGbnBVTS9JcmMyc21qWkRYT3BlRE1VVE1COEdBMVVkSXdRWU1CYUFGTit0VkZucFVNL0lyYzJzbWpaRApYT3BlRE1VVE1BMEdDU3FHU0liM0RRRUJDd1VBQTRJQkFRQjk3NXBmODY5THN6NkdDWkdxckNyWE1GMm00aENXCmFIRGx4UWQvK3NPQU5neEsxUnV1QWZ5d2lMU2pRTGFaYmxSTnZBMWQ1cGVndXBmcUdyQXljSjE1MkpyUGQ1aE8KNkpGMmlhMWxJMS80bno4OFE1STBZcEpJc0FSUDZBTSttb2RrWjFKM3V5UHFZbFVmSnBJU1pLSlhpQStERVJCaQo1eEhjZWMyWVlPTHJqK2xTMk90eUJ4K3ZpRnlMWFU0aGkrN08zRVRwNjFjRWVKUHE0YzcvWE95UlNnMWZudFFRCkN5SC81eHFBSStSYmhXMzA4TXdVWGh1M2QxcXFyRTBoUXpOVDVoOW4zSHN6V0F3dGthQVUzOHEzMEZqZFdwYXgKeFdDcURBdyt5UElMc2JzdjdBMnVnTUkxY1o0Zk9uWmkwK1MyRHFXdnZ4UEZ6enZza013SS9BNjQKLS0tLS1FTkQgQ0VSVElGSUNBVEUtLS0tLQo=
    server: https://10.10.0.10:6443
  name: kubernetes
contexts:
- context:
    cluster: kubernetes
    user: system:kube-controller-manager
  name: system:kube-controller-manager
current-context: system:kube-controller-manager
kind: Config
preferences: {}
users:
- name: system:kube-controller-manager
  user:
    client-certificate-data: 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
    client-key-data: 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


#创建配置文件kube-controller-manager.conf

[root@master01 work ]#cat kube-controller-manager.conf 
KUBE_CONTROLLER_MANAGER_OPTS="--port=0 \
  --secure-port=10252 \
  --bind-address=127.0.0.1 \  #官方推荐绑定127.0.0.1,外界不能访问
  --kubeconfig=/etc/kubernetes/kube-controller-manager.kubeconfig \
  --service-cluster-ip-range=10.255.0.0/16 \
  --cluster-name=kubernetes \
  --cluster-signing-cert-file=/etc/kubernetes/ssl/ca.pem \
  --cluster-signing-key-file=/etc/kubernetes/ssl/ca-key.pem \
  --allocate-node-cidrs=true \
  --cluster-cidr=10.0.0.0/16 \
  --experimental-cluster-signing-duration=87600h \
  --root-ca-file=/etc/kubernetes/ssl/ca.pem \
  --service-account-private-key-file=/etc/kubernetes/ssl/ca-key.pem \
  --leader-elect=true \
  --feature-gates=RotateKubeletServerCertificate=true \
  --controllers=*,bootstrapsigner,tokencleaner \
  --horizontal-pod-autoscaler-use-rest-clients=true \
  --horizontal-pod-autoscaler-sync-period=10s \
  --tls-cert-file=/etc/kubernetes/ssl/kube-controller-manager.pem \
  --tls-private-key-file=/etc/kubernetes/ssl/kube-controller-manager-key.pem \
  --use-service-account-credentials=true \
  --alsologtostderr=true \
  --logtostderr=false \
  --log-dir=/var/log/kubernetes \
  --v=2"


#创建启动文件

[root@master01 work ]#cat kube-controller-manager.service 
[Unit]
Description=Kubernetes Controller Manager
Documentation=https://github.com/kubernetes/kubernetes
[Service]
EnvironmentFile=-/etc/kubernetes/kube-controller-manager.conf
ExecStart=/usr/local/bin/kube-controller-manager $KUBE_CONTROLLER_MANAGER_OPTS
Restart=on-failure
RestartSec=5
[Install]
WantedBy=multi-user.target


#启动服务

[root@master01 work ]#cp kube-controller-manager*.pem /etc/kubernetes/ssl/
[root@master01 work ]#cp kube-controller-manager.kubeconfig /etc/kubernetes/
[root@master01 work ]#cp kube-controller-manager.conf /etc/kubernetes/
[root@master01 work ]#cp kube-controller-manager.service /usr/lib/systemd/system/
[root@master01 work ]#
[root@master01 work ]#rsync -vaz kube-controller-manager*.pem master02:/etc/kubernetes/ssl/
sending incremental file list
kube-controller-manager-key.pem
kube-controller-manager.pem

sent 2,498 bytes  received 54 bytes  5,104.00 bytes/sec
total size is 3,180  speedup is 1.25
[root@master01 work ]#rsync -vaz kube-controller-manager*.pem master03:/etc/kubernetes/ssl/
sending incremental file list
kube-controller-manager-key.pem
kube-controller-manager.pem

sent 2,498 bytes  received 54 bytes  1,701.33 bytes/sec
total size is 3,180  speedup is 1.25
[root@master01 work ]#rsync -vaz kube-controller-manager.kubeconfig kube-controller-manager.conf master02:/etc/kubernetes/
sending incremental file list
kube-controller-manager.conf
kube-controller-manager.kubeconfig

sent 4,869 bytes  received 54 bytes  9,846.00 bytes/sec
total size is 7,575  speedup is 1.54
[root@master01 work ]#rsync -vaz kube-controller-manager.kubeconfig kube-controller-manager.conf master03:/etc/kubernetes/
sending incremental file list
kube-controller-manager.conf
kube-controller-manager.kubeconfig

sent 4,869 bytes  received 54 bytes  3,282.00 bytes/sec
total size is 7,575  speedup is 1.54
[root@master01 work ]#rsync -vaz kube-controller-manager.service master02:/usr/lib/systemd/system/
sending incremental file list
kube-controller-manager.service

sent 328 bytes  received 35 bytes  726.00 bytes/sec
total size is 325  speedup is 0.90
[root@master01 work ]#rsync -vaz kube-controller-manager.service master03:/usr/lib/systemd/system/
sending incremental file list
kube-controller-manager.service

sent 328 bytes  received 35 bytes  242.00 bytes/sec
total size is 325  speedup is 0.90


[root@master01 work ]#systemctl daemon-reload
[root@master01 work ]#
[root@master01 work ]#systemctl enable kube-controller-manager.service --now
Created symlink from /etc/systemd/system/multi-user.target.wants/kube-controller-manager.service to /usr/lib/systemd/system/kube-controller-manager.service.

[root@master02 ~ ]#systemctl daemon-reload
[root@master02 ~ ]#systemctl enable kube-controller-manager.service --now
Created symlink from /etc/systemd/system/multi-user.target.wants/kube-controller-manager.service to /usr/lib/systemd/system/kube-controller-manager.service.

[root@master03 kubernetes ]#systemctl daemon-reload
[root@master03 kubernetes ]#systemctl enable kube-controller-manager.service --now
Created symlink from /etc/systemd/system/multi-user.target.wants/kube-controller-manager.service to /usr/lib/systemd/system/kube-controller-manager.service.

端口开启表示正常

[root@master02 kubernetes ]#ss -lanptu|grep 10252
tcp    LISTEN     0      16384  127.0.0.1:10252


5.5 部署kube-scheduler组件

#创建csr请求

[root@master01 work ]#cat kube-scheduler-csr.json 
{
    "CN": "system:kube-scheduler",
    "hosts": [
      "127.0.0.1",
      "10.10.0.10",
      "10.10.0.11",
      "10.10.0.12",
      "10.10.0.100"
    ],
    "key": {
        "algo": "rsa",
        "size": 2048
    },
    "names": [
      {
        "C": "CN",
        "ST": "Hubei",
        "L": "Wuhan",
        "O": "system:kube-scheduler",
        "OU": "system"
      }
    ]
}

注: hosts 列表包含所有 kube-scheduler 节点 IP;

CN 为 system:kube-scheduler、

O 为 system:kube-scheduler,kubernetes 内置的 ClusterRoleBindings system:kube-scheduler

将赋予 kube-scheduler 工作所需的权限。#生成证书

[root@master01 work ]#cfssl gencert -ca=ca.pem -ca-key=ca-key.pem -config=ca-config.json -profile=kubernetes kube-scheduler-csr.json | cfssljson -bare kube-scheduler
2022/10/26 14:45:07 [INFO] generate received request
2022/10/26 14:45:07 [INFO] received CSR
2022/10/26 14:45:07 [INFO] generating key: rsa-2048
2022/10/26 14:45:07 [INFO] encoded CSR
2022/10/26 14:45:07 [INFO] signed certificate with serial number 635930628159443813223981479189831176983910270655
2022/10/26 14:45:07 [WARNING] This certificate lacks a "hosts" field. This makes it unsuitable for
websites. For more information see the Baseline Requirements for the Issuance and Management
of Publicly-Trusted Certificates, v.1.1.6, from the CA/Browser Forum (https://cabforum.org);
specifically, section 10.2.3 ("Information Requirements").


#创建kube-scheduler的kubeconfig

1.设置集群参数

[root@master01 work ]#kubectl config set-cluster kubernetes --certificate-authority=ca.pem --embed-certs=true --server=https://10.10.0.10:6443 --kubeconfig=kube-scheduler.kubeconfig
Cluster "kubernetes" set.

2.设置客户端认证参数

system:kube-scheduler这个用户就是kube-scheduler-csr.json 里面CN设置的用户

[root@master01 work ]#kubectl config set-credentials system:kube-scheduler --client-certificate=kube-scheduler.pem --client-key=kube-scheduler-key.pem --embed-certs=true --kubeconfig=kube-scheduler.kubeconfig
User "system:kube-scheduler" set.

3.设置上下文参数

[root@master01 work ]#kubectl config set-context system:kube-scheduler --cluster=kubernetes --user=system:kube-scheduler --kubeconfig=kube-scheduler.kubeconfig
Context "system:kube-scheduler" created.


4.设置当前上下文

[root@master01 work ]#kubectl config use-context system:kube-scheduler --kubeconfig=kube-scheduler.kubeconfig
Switched to context "system:kube-scheduler".

#创建配置文件kube-scheduler.conf

[root@master01 work ]#cat kube-scheduler.conf 
KUBE_SCHEDULER_OPTS="--address=127.0.0.1 \
--kubeconfig=/etc/kubernetes/kube-scheduler.kubeconfig \
--leader-elect=true \
--alsologtostderr=true \
--logtostderr=false \
--log-dir=/var/log/kubernetes \
--v=2"

#创建服务启动文件

[root@master01 work ]#cat kube-scheduler.service 
[Unit]
Description=Kubernetes Scheduler
Documentation=https://github.com/kubernetes/kubernetes
 
[Service]
EnvironmentFile=-/etc/kubernetes/kube-scheduler.conf
ExecStart=/usr/local/bin/kube-scheduler $KUBE_SCHEDULER_OPTS
Restart=on-failure
RestartSec=5
 
[Install]
WantedBy=multi-user.target


#启动服务

[root@master01 work ]#cp kube-scheduler*.pem /etc/kubernetes/ssl/
[root@master01 work ]#cp kube-scheduler.kubeconfig /etc/kubernetes/
[root@master01 work ]#cp kube-scheduler.conf /etc/kubernetes/
[root@master01 work ]#cp kube-scheduler.service /usr/lib/systemd/system/
[root@master01 work ]#rsync -vaz kube-scheduler*.pem master02:/etc/kubernetes/ssl/
sending incremental file list
kube-scheduler-key.pem
kube-scheduler.pem

sent 2,531 bytes  received 54 bytes  1,723.33 bytes/sec
total size is 3,159  speedup is 1.22
[root@master01 work ]#rsync -vaz kube-scheduler*.pem master03:/etc/kubernetes/ssl/
sending incremental file list
kube-scheduler-key.pem
kube-scheduler.pem

sent 2,531 bytes  received 54 bytes  1,723.33 bytes/sec
total size is 3,159  speedup is 1.22
[root@master01 work ]#rsync -vaz kube-scheduler.kubeconfig kube-scheduler.conf master02:/etc/kubernetes/
sending incremental file list
kube-scheduler.conf
kube-scheduler.kubeconfig

sent 4,495 bytes  received 54 bytes  9,098.00 bytes/sec
total size is 6,617  speedup is 1.45
[root@master01 work ]#rsync -vaz kube-scheduler.kubeconfig kube-scheduler.conf master03:/etc/kubernetes/
sending incremental file list
kube-scheduler.conf
kube-scheduler.kubeconfig

sent 4,495 bytes  received 54 bytes  9,098.00 bytes/sec
total size is 6,617  speedup is 1.45
[root@master01 work ]#rsync -vaz kube-scheduler.service master02:/usr/lib/systemd/system/
sending incremental file list
kube-scheduler.service

sent 301 bytes  received 35 bytes  672.00 bytes/sec
total size is 293  speedup is 0.87
[root@master01 work ]#rsync -vaz kube-scheduler.service master03:/usr/lib/systemd/system/
sending incremental file list
kube-scheduler.service

sent 301 bytes  received 35 bytes  224.00 bytes/sec
total size is 293  speedup is 0.87


[root@master01 work ]#systemctl daemon-reload
[root@master01 work ]#systemctl enable kube-scheduler.service --now


[root@master02 kubernetes ]#systemctl daemon-reload
[root@master02 kubernetes ]#systemctl enable kube-scheduler.service --now

[root@master03 kubernetes ]#systemctl daemon-reload
[root@master03 kubernetes ]#systemctl enable kube-scheduler.service --now


端口开启表示正常

[root@master01 kubernetes ]#ss -lanptu|grep 10251
tcp    LISTEN     0      16384  127.0.0.1:10251                 *:*                   users:(("kube-scheduler",pid=823,fd=8))


kubeadm安装也是绑定到127.0.0.1

5.6 导入离线镜像压缩包

#把pause-cordns.tar.gz上传到node01节点,手动解压

[root@node01 ~ ]#docker load -i pause-cordns.tar.gz 
225df95e717c: Loading layer [==================================================>]  336.4kB/336.4kB
96d17b0b58a7: Loading layer [==================================================>]  45.02MB/45.02MB
Loaded image: k8s.gcr.io/coredns:1.7.0
ba0dae6243cc: Loading layer [==================================================>]  684.5kB/684.5kB
Loaded image: k8s.gcr.io/pause:3.2
[root@node01 ~ ]#
[root@node01 ~ ]#
[root@node01 ~ ]#docker images 
REPOSITORY           TAG       IMAGE ID       CREATED       SIZE
k8s.gcr.io/coredns   1.7.0     bfe3a36ebd25   2 years ago   45.2MB
k8s.gcr.io/pause     3.2       80d28bedfe5d   2 years ago   683kB

5.7 部署kubelet组件


kubelet: 每个Node节点上的kubelet定期就会调用API Server的REST接口报告自身状态,

API Server接收这些信息后,将节点状态信息更新到etcd中。

kubelet也通过API Server监听Pod信息,从而对Node机器上的POD进行管理,如创建、删除、更新Pod


控制节点不允许调度pod,pod都会被调度到工作节点,所以kubelet只需要部署到工作节点,启动


以下操作在master01上操作


创建kubelet-bootstrap.kubeconfig
[root@master01 kubernetes ]#cd /data/work/
[root@master01 work ]#BOOTSTRAP_TOKEN=$(awk -F “,” ‘{print $1}’ /etc/kubernetes/token.csv)
[root@master01 work ]#rm -r kubelet-bootstrap.kubeconfig
rm: cannot remove ‘kubelet-bootstrap.kubeconfig’: No such file or directory


1.设置集群参数

[root@master01 work ]#kubectl config set-cluster kubernetes --certificate-authority=ca.pem --embed-certs=true --server=https://10.10.0.10:6443 --kubeconfig=kubelet-bootstrap.kubeconfig
Cluster “kubernetes” set.


2.设置客户端认证参数

[root@master01 work ]#kubectl config set-credentials kubelet-bootstrap --token=${BOOTSTRAP_TOKEN} --kubeconfig=kubelet-bootstrap.kubeconfig
User “kubelet-bootstrap” set.


3.设置上下文参数

[root@master01 work ]#kubectl config set-context default --cluster=kubernetes --user=kubelet-bootstrap --kubeconfig=kubelet-bootstrap.kubeconfig
Context “default” created.


4.设置当前上下文


[root@master01 work ]#kubectl config use-context default --kubeconfig=kubelet-bootstrap.kubeconfig
Switched to context “default”.
[root@master01 work ]#cat kubelet-bootstrap.kubeconfig 
apiVersion: v1
clusters:
- cluster:
    certificate-authority-data: LS0tLS1CRUdJTiBDRVJUSUZJQ0FURS0tLS0tCk1JSUR0akNDQXA2Z0F3SUJBZ0lVUkV4RXhidjQ1ZlVBY1hCQlpENGZEZnRqZmtNd0RRWUpLb1pJaHZjTkFRRUwKQlFBd1lURUxNQWtHQTFVRUJoTUNRMDR4RGpBTUJnTlZCQWdUQlVoMVltVnBNUTR3REFZRFZRUUhFd1ZYZFdoaApiakVNTUFvR0ExVUVDaE1EYXpoek1ROHdEUVlEVlFRTEV3WnplWE4wWlcweEV6QVJCZ05WQkFNVENtdDFZbVZ5CmJtVjBaWE13SGhjTk1qSXhNREkxTURnMU5EQXdXaGNOTXpJeE1ESXlNRGcxTkRBd1dqQmhNUXN3Q1FZRFZRUUcKRXdKRFRqRU9NQXdHQTFVRUNCTUZTSFZpWldreERqQU1CZ05WQkFjVEJWZDFhR0Z1TVF3d0NnWURWUVFLRXdOcgpPSE14RHpBTkJnTlZCQXNUQm5ONWMzUmxiVEVUTUJFR0ExVUVBeE1LYTNWaVpYSnVaWFJsY3pDQ0FTSXdEUVlKCktvWklodmNOQVFFQkJRQURnZ0VQQURDQ0FRb0NnZ0VCQUxVV2lYcGZPMFZ6dTBrek8xYllaVUo2dFVnTHp2TDUKTlJwWnpoS2pNZ3pDbmZYM0pHWUM0SWpFdGoxRWNhUGdFOUErbkZ1UndCWlk0a0U3NkE1RjdhaS8rNzdiak5tMApJcWdzY3o2VmdzM0JuMEZSSXVmMlFkQm9OUmNlWWhXL011Z0ZWUlZRSUVQTjRuZGxCVUFScHpNbDk4clpxSVU5CkxTMkZ1WWRwYTRRSi8weVp3YWlYYllxazFDUk5JUkRBMUZJRDh6VEpHeXhRdmZoZ05wRGJTQkhQVlhqK3oxZW4KTU5XQzVDMGt1QWcya3IzcExHUkg3T2dMR1dHV0RuRXVwMUZKMUJONDl2V3IwTW8zQWd6TE1TcTdaRnRGYS9KYgpvN3lOWmZDZnd5NTZQTi9Rckc5eSs0aFBONHVRdG16em1YbUJYT2JJTFF2SXhEN2lKYTVWb0I4Q0F3RUFBYU5tCk1HUXdEZ1lEVlIwUEFRSC9CQVFEQWdFR01CSUdBMVVkRXdFQi93UUlNQVlCQWY4Q0FRSXdIUVlEVlIwT0JCWUUKRk4rdFZGbnBVTS9JcmMyc21qWkRYT3BlRE1VVE1COEdBMVVkSXdRWU1CYUFGTit0VkZucFVNL0lyYzJzbWpaRApYT3BlRE1VVE1BMEdDU3FHU0liM0RRRUJDd1VBQTRJQkFRQjk3NXBmODY5THN6NkdDWkdxckNyWE1GMm00aENXCmFIRGx4UWQvK3NPQU5neEsxUnV1QWZ5d2lMU2pRTGFaYmxSTnZBMWQ1cGVndXBmcUdyQXljSjE1MkpyUGQ1aE8KNkpGMmlhMWxJMS80bno4OFE1STBZcEpJc0FSUDZBTSttb2RrWjFKM3V5UHFZbFVmSnBJU1pLSlhpQStERVJCaQo1eEhjZWMyWVlPTHJqK2xTMk90eUJ4K3ZpRnlMWFU0aGkrN08zRVRwNjFjRWVKUHE0YzcvWE95UlNnMWZudFFRCkN5SC81eHFBSStSYmhXMzA4TXdVWGh1M2QxcXFyRTBoUXpOVDVoOW4zSHN6V0F3dGthQVUzOHEzMEZqZFdwYXgKeFdDcURBdyt5UElMc2JzdjdBMnVnTUkxY1o0Zk9uWmkwK1MyRHFXdnZ4UEZ6enZza013SS9BNjQKLS0tLS1FTkQgQ0VSVElGSUNBVEUtLS0tLQo=
    server: https://10.10.0.10:6443
  name: kubernetes
contexts:
- context:
    cluster: kubernetes
    user: kubelet-bootstrap
  name: default
current-context: default
kind: Config
preferences: {}
users:
- name: kubelet-bootstrap
  user: {}

授权:

[root@master01 work ]#kubectl create clusterrolebinding kubelet-bootstrap --clusterrole=system:node-bootstrapper --user=kubelet-bootstrap
clusterrolebinding.rbac.authorization.k8s.io/kubelet-bootstrap created

#创建配置文件kubelet.json

“cgroupDriver”: "systemd"要和docker的驱动一致。

address替换为自己node01的IP地址。

[root@master01 work ]#cat kubelet.json 
{
  "kind": "KubeletConfiguration",
  "apiVersion": "kubelet.config.k8s.io/v1beta1",
  "authentication": {
    "x509": {
      "clientCAFile": "/etc/kubernetes/ssl/ca.pem"
    },
    "webhook": {
      "enabled": true,
      "cacheTTL": "2m0s"
    },
    "anonymous": {
      "enabled": false
    }
  },
  "authorization": {
    "mode": "Webhook",
    "webhook": {
      "cacheAuthorizedTTL": "5m0s",
      "cacheUnauthorizedTTL": "30s"
    }
  },
  "address": "10.10.0.14",
  "port": 10250,
  "readOnlyPort": 10255,
  "cgroupDriver": "systemd",
  "hairpinMode": "promiscuous-bridge",
  "serializeImagePulls": false,
  "featureGates": {
    "RotateKubeletClientCertificate": true,
    "RotateKubeletServerCertificate": true
  },
  "clusterDomain": "cluster.local.",
  "clusterDNS": ["10.255.0.2"]
}
[root@master01 work ]#cat kubelet.service 
[Unit]
Description=Kubernetes Kubelet
Documentation=https://github.com/kubernetes/kubernetes
After=docker.service
Requires=docker.service
[Service]
WorkingDirectory=/var/lib/kubelet
ExecStart=/usr/local/bin/kubelet \
  --bootstrap-kubeconfig=/etc/kubernetes/kubelet-bootstrap.kubeconfig \  #第一次启动用这个文件,后续都用--kubeconfig
  --cert-dir=/etc/kubernetes/ssl \
  --kubeconfig=/etc/kubernetes/kubelet.kubeconfig \    #这个是自动生成的
  --config=/etc/kubernetes/kubelet.json \
  --network-plugin=cni \
  --pod-infra-container-image=k8s.gcr.io/pause:3.2 \
  --alsologtostderr=true \
  --logtostderr=false \
  --log-dir=/var/log/kubernetes \
  --v=2
Restart=on-failure
RestartSec=5
 
[Install]
WantedBy=multi-user.target

#注: –hostname-override:显示名称,集群中唯一

–network-plugin:启用CNI

–kubeconfig:空路径,会自动生成,后面用于连接apiserver

–bootstrap-kubeconfig:首次启动向apiserver申请证书

–config:配置参数文件

–cert-dir:kubelet证书生成目录

–pod-infra-container-image:管理Pod网络容器的镜像

#注:kubelete.json配置文件address改为各个节点的ip地址,在各个work节点上启动服务

[root@node01 ~ ]#mkdir /etc/kubernetes/ssl -p

[root@master01 work ]#scp kubelet-bootstrap.kubeconfig kubelet.json node01:/etc/kubernetes/
kubelet-bootstrap.kubeconfig                                                                                           100% 2107     3.7MB/s   00:00    
kubelet.json

[root@master01 work ]#scp ca.pem node01:/etc/kubernetes/ssl/

ca.pem


[root@master01 work ]#scp kubelet.service node01:/usr/lib/systemd/system/

kubelet.service


#启动kubelet服务

这个目录最好先删一下 /var/lib/kubelet,如果存在之前的会有冲突


[root@node01 ~ ]#mkdir /var/lib/kubelet

[root@node01 ~ ]#mkdir /var/log/kubernetes

[root@node01 system ]#systemctl daemon-reload 
[root@node01 system ]#
[root@node01 system ]#
[root@node01 system ]#
[root@node01 system ]#systemctl enable kubelet.service --now
Created symlink from /etc/systemd/system/multi-user.target.wants/kubelet.service to /usr/lib/systemd/system/kubelet.service.

确认kubelet服务启动成功后,接着到master01节点上Approve一下bootstrap请求。

master节点执行如下命令可以看到一个worker节点发送了一个 CSR 请求:

[root@master01 work ]#kubectl get csr
NAME                                                   AGE   SIGNERNAME                                    REQUESTOR           CONDITION
node-csr-auJz6SMLC0bu-hO2MXiibM-syMll0YQHZ1kAXyo3gaM   32s   kubernetes.io/kube-apiserver-client-kubelet   kubelet-bootstrap   Pending


批准请求:

[root@master01 work ]#kubectl certificate approve node-csr-auJz6SMLC0bu-hO2MXiibM-syMll0YQHZ1kAXyo3gaM
certificatesigningrequest.certificates.k8s.io/node-csr-auJz6SMLC0bu-hO2MXiibM-syMll0YQHZ1kAXyo3gaM approved
[root@master01 work ]#kubectl get csr
NAME                                                   AGE   SIGNERNAME                                    REQUESTOR           CONDITION
node-csr-auJz6SMLC0bu-hO2MXiibM-syMll0YQHZ1kAXyo3gaM   11m   kubernetes.io/kube-apiserver-client-kubelet   kubelet-bootstrap   Approved,Issued
[root@master01 work ]#kubectl get nodes
NAME     STATUS     ROLES    AGE   VERSION
node01   NotReady   <none>   34s   v1.20.7

#注意:STATUS是NotReady表示还没有安装网络插件


二进制安装 kubectl get nodes 看不到控制节点,只能看到工作节点


默认主节点是没部署kubelet的,不具备服务调度能力。

一般生产上都是这么做的。如果想让主节点可具备调度能力,

可以像worker节点一样去在master上部署一下kubelet相关服务,也就可以get no看到了

node节点动态生成一些证书:

[root@node01 ~ ]#cd /etc/kubernetes/
[root@node01 kubernetes ]#ll
total 12
-rw------- 1 root root 2148 Oct 27 08:44 kubelet-bootstrap.kubeconfig
-rw-r--r-- 1 root root  800 Oct 27 08:44 kubelet.json
-rw------- 1 root root 2277 Oct 27 08:49 kubelet.kubeconfig
drwxr-xr-x 2 root root  138 Oct 27 08:49 ssl
[root@node01 kubernetes ]#cd ssl/
[root@node01 ssl ]#ll
total 16
-rw-r--r-- 1 root root 1346 Oct 27 08:44 ca.pem
-rw------- 1 root root 1212 Oct 27 08:49 kubelet-client-2022-10-27-08-49-42.pem
lrwxrwxrwx 1 root root   58 Oct 27 08:49 kubelet-client-current.pem -> /etc/kubernetes/ssl/kubelet-client-2022-10-27-08-49-42.pem
-rw-r--r-- 1 root root 2237 Oct 27 08:47 kubelet.crt
-rw------- 1 root root 1675 Oct 27 08:47 kubelet.key


5.8 部署kube-proxy组件

[root@master01 work ]#kubectl get svc
NAME         TYPE        CLUSTER-IP   EXTERNAL-IP   PORT(S)   AGE
kubernetes   ClusterIP   10.255.0.1   <none>        443/TCP   22h


10.255.0.1在外面是访问不了的,只能在iptables,ipvs规则中使用,需要kube-proxy来生成规则

#创建csr请求

[root@master01 work ]#cat kube-proxy-csr.json 
{
  "CN": "system:kube-proxy",
  "key": {
    "algo": "rsa",
    "size": 2048
  },
  "names": [
    {
      "C": "CN",
      "ST": "Hubei",
      "L": "Wuhan",
      "O": "k8s",
      "OU": "system"
    }
  ]
}


生成证书

[root@master01 work ]#cfssl gencert -ca=ca.pem -ca-key=ca-key.pem -config=ca-config.json -profile=kubernetes kube-proxy-csr.json | cfssljson -bare kube-proxy
2022/10/27 08:52:02 [INFO] generate received request
2022/10/27 08:52:02 [INFO] received CSR
2022/10/27 08:52:02 [INFO] generating key: rsa-2048
2022/10/27 08:52:02 [INFO] encoded CSR
2022/10/27 08:52:02 [INFO] signed certificate with serial number 621761669559249301102184867130469483006128723356
2022/10/27 08:52:02 [WARNING] This certificate lacks a "hosts" field. This makes it unsuitable for
websites. For more information see the Baseline Requirements for the Issuance and Management
of Publicly-Trusted Certificates, v.1.1.6, from the CA/Browser Forum (https://cabforum.org);
specifically, section 10.2.3 ("Information Requirements").
[root@master01 work ]#ll kube-proxy*
-rw-r--r-- 1 root root 1005 Oct 27 08:52 kube-proxy.csr
-rw-r--r-- 1 root root  212 Oct 25 15:39 kube-proxy-csr.json
-rw------- 1 root root 1679 Oct 27 08:52 kube-proxy-key.pem
-rw------- 1 root root 6238 Oct 27 08:54 kube-proxy.kubeconfig
-rw-r--r-- 1 root root 1391 Oct 27 08:52 kube-proxy.pem
-rw-r--r-- 1 root root  297 Oct 25 15:39 kube-proxy.yaml

#创建kubeconfig文件,安全上下文,确认kube-proxy 跟哪个集群交互

设置集群参数

[root@master01 work ]#kubectl config set-cluster kubernetes --certificate-authority=ca.pem --embed-certs=true --server=https://10.10.0.10:6443 --kubeconfig=kube-proxy.kubeconfig
Cluster “kubernetes” set.


设置客户端认证参数

[root@master01 work ]#kubectl config set-credentials kube-proxy --client-certificate=kube-proxy.pem --client-key=kube-proxy-key.pem --embed-certs=true --kubeconfig=kube-proxy.kubeconfig
User “kube-proxy” set.


设置上下文参数

[root@master01 work ]#kubectl config set-context default --cluster=kubernetes --user=kube-proxy --kubeconfig=kube-proxy.kubeconfig
Context “default” created.


–user=kube-proxy 是在kube-proxy-csr.json中设置的CN用户


如果OU那里设置的是system,可以–user那里可以直接使用后面的用户名,与加上system:效果一样


设置当前上下文

[root@master01 work ]#kubectl config use-context default --kubeconfig=kube-proxy.kubeconfig
Switched to context “default”.

#创建kube-proxy配置文件

[root@master01 work ]#cat kube-proxy.yaml 
apiVersion: kubeproxy.config.k8s.io/v1alpha1
bindAddress: 10.10.0.14                 #node节点ip
clientConnection:
  kubeconfig: /etc/kubernetes/kube-proxy.kubeconfig
clusterCIDR: 10.10.0.0/24               #这里要使用物理机的网段
healthzBindAddress: 10.10.0.14:10256    #node节点ip
kind: KubeProxyConfiguration 
metricsBindAddress: 10.10.0.14:10249    #node节点ip  采集指标
mode: "ipvs"                            #使用ipvs转发模式


#创建服务启动文件

[root@master01 work ]#cat kube-proxy.service 
[Unit]
Description=Kubernetes Kube-Proxy Server
Documentation=https://github.com/kubernetes/kubernetes
After=network.target
 
[Service]
WorkingDirectory=/var/lib/kube-proxy
ExecStart=/usr/local/bin/kube-proxy \
  --config=/etc/kubernetes/kube-proxy.yaml \
  --alsologtostderr=true \
  --logtostderr=false \
  --log-dir=/var/log/kubernetes \
  --v=2
Restart=on-failure
RestartSec=5
LimitNOFILE=65536
 
[Install]
WantedBy=multi-user.target
[root@master01 work ]#scp  kube-proxy.kubeconfig kube-proxy.yaml node01:/etc/kubernetes/
kube-proxy.kubeconfig                                                                                                  100% 6238     5.9MB/s   00:00    
kube-proxy.yaml                                                                                                        100%  282   418.8KB/s   00:00    
[root@master01 work ]#scp  kube-proxy.service node01:/usr/lib/systemd/system/
kube-proxy.service

#启动服务,在node节点创建目录

[root@node01 kubernetes ]#mkdir -p /var/lib/kube-proxy

[root@node01 kubernetes ]#systemctl daemon-reload
[root@node01 kubernetes ]#systemctl enable kube-proxy.service --now
Created symlink from /etc/systemd/system/multi-user.target.wants/kube-proxy.service to /usr/lib/systemd/system/kube-proxy.service.
[root@node01 kubernetes ]#systemctl status kube-proxy.service 
● kube-proxy.service - Kubernetes Kube-Proxy Server
   Loaded: loaded (/usr/lib/systemd/system/kube-proxy.service; enabled; vendor preset: disabled)
   Active: active (running) since Thu 2022-10-27 09:39:33 CST; 1min 22s ago
     Docs: https://github.com/kubernetes/kubernetes
 Main PID: 12737 (kube-proxy)
    Tasks: 7
   Memory: 57.5M
   CGroup: /system.slice/kube-proxy.service
           └─12737 /usr/local/bin/kube-proxy --config=/etc/kubernetes/kube-proxy.yaml --alsologtostderr=true --logtostderr=false --log-dir=/var/log/ku...

Oct 27 09:39:35 node01 kube-proxy[12737]: I1027 09:39:35.136193   12737 shared_informer.go:240] Waiting for caches to sync for endpoint slice config
Oct 27 09:39:35 node01 kube-proxy[12737]: I1027 09:39:35.136212   12737 config.go:315] Starting service config controller
Oct 27 09:39:35 node01 kube-proxy[12737]: I1027 09:39:35.136214   12737 shared_informer.go:240] Waiting for caches to sync for service config
Oct 27 09:39:35 node01 kube-proxy[12737]: I1027 09:39:35.136432   12737 reflector.go:219] Starting reflector *v1beta1.EndpointSlice (15m0s) f...ry.go:134
Oct 27 09:39:35 node01 kube-proxy[12737]: I1027 09:39:35.136441   12737 reflector.go:219] Starting reflector *v1.Service (15m0s) from k8s.io/...ry.go:134
Oct 27 09:39:35 node01 kube-proxy[12737]: I1027 09:39:35.172077   12737 service.go:275] Service default/kubernetes updated: 1 ports
Oct 27 09:39:35 node01 kube-proxy[12737]: I1027 09:39:35.236331   12737 shared_informer.go:247] Caches are synced for service config
Oct 27 09:39:35 node01 kube-proxy[12737]: I1027 09:39:35.236505   12737 shared_informer.go:247] Caches are synced for endpoint slice config
Oct 27 09:39:35 node01 kube-proxy[12737]: I1027 09:39:35.236532   12737 proxier.go:1036] Not syncing ipvs rules until Services and Endpoints ...om master
Oct 27 09:39:35 node01 kube-proxy[12737]: I1027 09:39:35.236928   12737 service.go:390] Adding new service port "default/kubernetes:https" at...1:443/TCP
Hint: Some lines were ellipsized, use -l to show in full.

5.9 部署calico组件

#解压离线镜像压缩包


#把calico.tar.gz上传到node01节点,手动解压


[root@node01 ~ ]#docker load -i calico.tar.gz


#把calico.yaml文件上传到master01上的的/data/work目录


Calico组件的安装


Calico:https://www.projectcalico.org/

https://docs.projectcalico.org/getting-started/kubernetes/self-managed-onprem/onpremises


curl https://docs.projectcalico.org/manifests/calico.yaml -O 原版本


点击Manifest下载:


curl https://projectcalico.docs.tigera.io/manifests/calico.yaml -O 新版本


点击requirements 可以查看calico支持的kubernetes的版本


修改calico.yaml的以下位置,把注释取消,改成自己配置的podSubnet网段

- name: CALICO_IPV4POOL_CIDR

value: “10.0.0.0/16”


这个如果是二进制安装,必须改,否则生成的pod 的ip地址默认是 192.168.0.0/16网段的


使用默认的:

[root@master01 work ]#vim calico.yaml

# - name: CALICO_IPV4POOL_CIDR
# value: “192.168.0.0/16”
[root@master01 ~ ]#kubectl get pods -owide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
nginx 1/1 Running 0 55s 192.168.196.130 node01


修改后:


[root@master01 work ]#vim calico.yaml
- name: CALICO_IPV4POOL_CIDR
value: “10.0.0.0/16”
[root@master01 work ]#kubectl get pods -owide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
nginx 1/1 Running 0 2m9s 10.0.196.130 node01
[root@master01 work ]#kubectl apply -f calico.yaml 

[root@master01 work ]#kubectl get pods -n kube-system -owide
NAME                                       READY   STATUS    RESTARTS   AGE   IP             NODE     NOMINATED NODE   READINESS GATES
calico-kube-controllers-6949477b58-dpwmn   1/1     Running   0          24s   10.0.196.129   node01   <none>           <none>
calico-node-rmkbm                          1/1     Running   0          24s   10.10.0.14     node01   <none>           <none>

calico-node-rmkbm 分配ip的

calico-kube-controllers-6949477b58-dpwmn 做网络策略的

[root@master01 work ]#kubectl get nodes
NAME     STATUS   ROLES    AGE   VERSION
node01   Ready    <none>   62m   v1.20.7

报错分析:

[root@master03 kubernetes ]#kubectl get pods -n kube-system -owide
NAME                                       READY   STATUS    RESTARTS   AGE     IP             NODE       NOMINATED NODE   READINESS GATES
calico-kube-controllers-6949477b58-cr5lh   1/1     Running   0          5h44m   10.0.196.129   node01     <none>           <none>
calico-node-49kj4                          1/1     Running   6          35m     10.10.0.11     master02   <none>           <none>
calico-node-bgv4n                          1/1     Running   5          35m     10.10.0.10     master01   <none>           <none>
calico-node-bjsmm                          1/1     Running   0          5h44m   10.10.0.14     node01     <none>           <none>
calico-node-t2ljz                          0/1     Running   1          8m46s   10.10.0.12     master03   <none>           <none>
coredns-7bf4bd64bd-572z9                   1/1     Running   0          5h31m   10.0.196.131   node01     <none>           <none>


master03的calico组件未就绪


K8S集群Calico网络组件报错BIRD


calico/node is not ready: BIRD is not ready: BGP not established with


Warning Unhealthy 8s kubelet Readiness probe failed: 2022-10-27 08:00:12.256 [INFO][374] confd/health.go 180: Number of node(s) with BGP peering established = 0

calico/node is not ready: BIRD is not ready: BGP not established with 10.10.0.10,10.10.0.11,10.10.0.14


产生该报错的原因是节点上出现了冲突的网卡,将有问题的网卡删除后,Calico组件的pod就能正常启动了异常网卡命名都是以br开头的,

[root@master03 kubernetes ]#ip link 
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
2: ens33: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP mode DEFAULT group default qlen 1000
    link/ether 00:0c:29:77:30:7c brd ff:ff:ff:ff:ff:ff
3: tunl0@NONE: <NOARP,UP,LOWER_UP> mtu 1480 qdisc noqueue state UNKNOWN mode DEFAULT group default qlen 1000
    link/ipip 0.0.0.0 brd 0.0.0.0
4: docker0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state DOWN mode DEFAULT group default 
    link/ether 02:42:c2:48:1f:bc brd ff:ff:ff:ff:ff:ff
5: br-8763acf3655c: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP mode DEFAULT group default 
    link/ether 02:42:da:e8:e8:55 brd ff:ff:ff:ff:ff:ff

就是这个br-8763acf3655c

将br开头的网卡删除:

[root@master03 kubernetes ]#ip link delete br-8763acf3655c

重新生成calico,成功:
[root@master03 kubernetes ]#kubectl get pods -n kube-system -owide
NAME                                       READY   STATUS    RESTARTS   AGE     IP             NODE       NOMINATED NODE   READINESS GATES
calico-kube-controllers-6949477b58-5rqgc   1/1     Running   2          20m     10.0.241.64    master01   <none>           <none>
calico-node-g48ng                          1/1     Running   0          7m53s   10.10.0.12     master03   <none>           <none>
calico-node-nb4g4                          1/1     Running   0          20m     10.10.0.11     master02   <none>           <none>
calico-node-ppcqd                          1/1     Running   0          20m     10.10.0.10     master01   <none>           <none>
calico-node-tsrd4                          1/1     Running   0          20m     10.10.0.14     node01     <none>           <none>
coredns-7bf4bd64bd-572z9                   1/1     Running   0          5h57m   10.0.196.131   node01     <none>           <none>

或者:

估计是没用发现实际真正的网卡

解决方法

/*
调整calicao 网络插件的网卡发现机制,修改IP_AUTODETECTION_METHOD对应的value值。
官方提供的yaml文件中,ip识别策略(IPDETECTMETHOD)没有配置,即默认为first-found,这会导致一个网络异常的ip作为nodeIP被注册,
从而影响node-to-node mesh。我们可以修改成can-reach或者interface的策略,尝试连接某一个Ready的node的IP,以此选择出正确的IP。
*/

// calico.yaml 文件添加以下二行
- name: IP_AUTODETECTION_METHOD
value: “interface=ens.*” # ens 根据实际网卡开头配置

// 配置如下
- name: CLUSTER_TYPE
value: “k8s,bgp”
- name: IP_AUTODETECTION_METHOD
value: “interface=ens.*”
#或者 value: “interface=ens160”
# Auto-detect the BGP IP address.
- name: IP
value: “autodetect”
# Enable IPIP
- name: CALICO_IPV4POOL_IPIP
value: “Always”

5.10 部署coredns组件

coreDNS 就是DNS服务器,将域名解析成ip

会根据svc创建FQDN 全限定域名 svcname.namespace.svc.cluster.local

镜像就是之前解压的pause-cordns.tar.gz

[root@master01 work ]#cat coredns.yaml 
apiVersion: v1
kind: ServiceAccount
metadata:
  name: coredns
  namespace: kube-system
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
metadata:
  labels:
    kubernetes.io/bootstrapping: rbac-defaults
  name: system:coredns
rules:
  - apiGroups:
    - ""
    resources:
    - endpoints
    - services
    - pods
    - namespaces
    verbs:
    - list
    - watch
  - apiGroups:
    - discovery.k8s.io
    resources:
    - endpointslices
    verbs:
    - list
    - watch
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
  annotations:
    rbac.authorization.kubernetes.io/autoupdate: "true"
  labels:
    kubernetes.io/bootstrapping: rbac-defaults
  name: system:coredns
roleRef:
  apiGroup: rbac.authorization.k8s.io
  kind: ClusterRole
  name: system:coredns
subjects:
- kind: ServiceAccount
  name: coredns
  namespace: kube-system
---
apiVersion: v1
kind: ConfigMap
metadata:
  name: coredns
  namespace: kube-system
data:
  Corefile: |
    .:53 {
        errors
        health {
          lameduck 5s
        }
        ready
        kubernetes cluster.local in-addr.arpa ip6.arpa {
          fallthrough in-addr.arpa ip6.arpa
        }
        prometheus :9153
        forward . /etc/resolv.conf {
          max_concurrent 1000
        }
        cache 30
        loop
        reload
        loadbalance
    }
---
apiVersion: apps/v1
kind: Deployment
metadata:
  name: coredns
  namespace: kube-system
  labels:
    k8s-app: kube-dns
    kubernetes.io/name: "CoreDNS"
spec:
  # replicas: not specified here:
  # 1. Default is 1.
  # 2. Will be tuned in real time if DNS horizontal auto-scaling is turned on.
  strategy:
    type: RollingUpdate
    rollingUpdate:
      maxUnavailable: 1
  selector:
    matchLabels:
      k8s-app: kube-dns
  template:
    metadata:
      labels:
        k8s-app: kube-dns
    spec:
      priorityClassName: system-cluster-critical
      serviceAccountName: coredns
      tolerations:
        - key: "CriticalAddonsOnly"
          operator: "Exists"
      nodeSelector:
        kubernetes.io/os: linux
      affinity:
         podAntiAffinity:
           preferredDuringSchedulingIgnoredDuringExecution:
           - weight: 100
             podAffinityTerm:
               labelSelector:
                 matchExpressions:
                   - key: k8s-app
                     operator: In
                     values: ["kube-dns"]
               topologyKey: kubernetes.io/hostname
      containers:
      - name: coredns
        image: coredns/coredns:1.7.0
        imagePullPolicy: IfNotPresent
        resources:
          limits:
            memory: 170Mi
          requests:
            cpu: 100m
            memory: 70Mi
        args: [ "-conf", "/etc/coredns/Corefile" ]
        volumeMounts:
        - name: config-volume
          mountPath: /etc/coredns
          readOnly: true
        ports:
        - containerPort: 53
          name: dns
          protocol: UDP
        - containerPort: 53
          name: dns-tcp
          protocol: TCP
        - containerPort: 9153
          name: metrics
          protocol: TCP
        securityContext:
          allowPrivilegeEscalation: false
          capabilities:
            add:
            - NET_BIND_SERVICE
            drop:
            - all
          readOnlyRootFilesystem: true
        livenessProbe:
          httpGet:
            path: /health
            port: 8080
            scheme: HTTP
          initialDelaySeconds: 60
          timeoutSeconds: 5
          successThreshold: 1
          failureThreshold: 5
        readinessProbe:
          httpGet:
            path: /ready
            port: 8181
            scheme: HTTP
      dnsPolicy: Default
      volumes:
        - name: config-volume
          configMap:
            name: coredns
            items:
            - key: Corefile
              path: Corefile
---
apiVersion: v1
kind: Service
metadata:
  name: kube-dns
  namespace: kube-system
  annotations:
    prometheus.io/port: "9153"
    prometheus.io/scrape: "true"
  labels:
    k8s-app: kube-dns
    kubernetes.io/cluster-service: "true"
    kubernetes.io/name: "CoreDNS"
spec:
  selector:
    k8s-app: kube-dns
  clusterIP: 10.255.0.2
  ports:
  - name: dns
    port: 53
    protocol: UDP
  - name: dns-tcp
    port: 53
    protocol: TCP
  - name: metrics
    port: 9153
    protocol: TCP
[root@master01 work ]#kubectl apply -f coredns.yaml 
serviceaccount/coredns created
clusterrole.rbac.authorization.k8s.io/system:coredns created
clusterrolebinding.rbac.authorization.k8s.io/system:coredns created
configmap/coredns created
deployment.apps/coredns created
service/kube-dns created
[root@master01 work ]#kubectl get svc -n kube-system
NAME       TYPE        CLUSTER-IP   EXTERNAL-IP   PORT(S)                  AGE
kube-dns   ClusterIP   10.255.0.2   <none>        53/UDP,53/TCP,9153/TCP   2m6s

calico和coredns都是以pod方式运行的

4.查看集群状态

[root@master01 work ]#kubectl get nodes
NAME     STATUS   ROLES    AGE    VERSION
node01   Ready    <none>   102m   v1.20.7

5.测试k8s集群部署tomcat服务

#把tomcat.tar.gz和busybox-1-28.tar.gz上传到node01,手动解压

[root@node01 ~ ]#docker load -i busybox-1-28.tar.gz 
432b65032b94: Loading layer [==================================================>]   1.36MB/1.36MB
Loaded image: busybox:1.28
[root@node01 ~ ]#docker load -i tomcat.tar.gz 
f1b5933fe4b5: Loading layer [==================================================>]  5.796MB/5.796MB
9b9b7f3d56a0: Loading layer [==================================================>]  3.584kB/3.584kB
edd61588d126: Loading layer [==================================================>]  80.28MB/80.28MB
48988bb7b861: Loading layer [==================================================>]   2.56kB/2.56kB
8e0feedfd296: Loading layer [==================================================>]  24.06MB/24.06MB
aac21c2169ae: Loading layer [==================================================>]  2.048kB/2.048kB
Loaded image: tomcat:8.5-jre8-alpine


在master01执行:

[root@master01 work ]#cat tomcat.yaml 
apiVersion: v1  #pod属于k8s核心组v1
kind: Pod  #创建的是一个Pod资源
metadata:  #元数据
  name: demo-pod  #pod名字
  namespace: default  #pod所属的名称空间
  labels:
    app: myapp  #pod具有的标签
    env: dev      #pod具有的标签
spec:
  containers:      #定义一个容器,容器是对象列表,下面可以有多个name
  - name:  tomcat-pod-java  #容器的名字
    ports:
    - containerPort: 8080
    image: tomcat:8.5-jre8-alpine   #容器使用的镜像
    imagePullPolicy: IfNotPresent
  - name: busybox
    image: busybox:latest
    command:  #command是一个列表,定义的时候下面的参数加横线
    - "/bin/sh"
    - "-c"
    - "sleep 3600"

[root@master01 work ]#kubectl apply -f tomcat.yaml

pod/demo-pod created

[root@master01 work ]#kubectl get pods -owide
NAME       READY   STATUS    RESTARTS   AGE   IP             NODE     NOMINATED NODE   READINESS GATES
demo-pod   2/2     Running   0          17s   10.0.196.132   node01   <none>           <none>

创建个service讲服务代理出去:

[root@master01 work ]#cat tomcat-service.yaml 
apiVersion: v1
kind: Service
metadata:
  name: tomcat
spec:
  type: NodePort
  ports:
    - port: 8080
      nodePort: 30080
  selector:
    app: myapp
    env: dev


[root@master01 work ]#kubectl apply -f tomcat-service.yaml

service/tomcat created

[root@master01 work ]#kubectl get svc
NAME         TYPE        CLUSTER-IP       EXTERNAL-IP   PORT(S)          AGE
kubernetes   ClusterIP   10.255.0.1       <none>        443/TCP          23h
tomcat       NodePort    10.255.199.247   <none>        8080:30080/TCP   12s


在浏览器访问 http://node01的ip:30080

验证可以正常展示页面

6.验证cordns是否正常

[root@master01 work ]#kubectl run busybox --image busybox:1.28 --restart=Never --rm -it busybox -- sh
If you don't see a command prompt, try pressing enter.
/ # ping www.baidu.com
PING www.baidu.com (14.215.177.39): 56 data bytes
64 bytes from 14.215.177.39: seq=0 ttl=127 time=5.189 ms
64 bytes from 14.215.177.39: seq=1 ttl=127 time=17.240 ms
^C
--- www.baidu.com ping statistics ---
2 packets transmitted, 2 packets received, 0% packet loss
round-trip min/avg/max = 5.189/11.214/17.240 ms


/ # nslookup kubernetes.default.svc.cluster.local
Server:    10.255.0.2
Address 1: 10.255.0.2 kube-dns.kube-system.svc.cluster.local

Name:      kubernetes.default.svc.cluster.local
Address 1: 10.255.0.1 kubernetes.default.svc.cluster.local



/ # nslookup tomcat.default.svc.cluster.local
Server:    10.255.0.2
Address 1: 10.255.0.2 kube-dns.kube-system.svc.cluster.local

Name:      tomcat.default.svc.cluster.local
Address 1: 10.255.199.247 tomcat.default.svc.cluster.local

#注意:

busybox要用指定的1.28版本,不能用最新版本,最新版本,nslookup会解析不到dns和ip,报错如下:

/ # nslookup kubernetes.default.svc.cluster.local
Server:   10.255.0.2
Address:  10.255.0.2:53
*** Can't find kubernetes.default.svc.cluster.local: No answer
*** Can't find kubernetes.default.svc.cluster.local: No answer

10.255.0.2 就是我们coreDNS的clusterIP,说明coreDNS配置好了。

解析内部Service的名称,是通过coreDNS去解析的。

6.安装keepalived+nginx实现k8s apiserver高可用

把epel.repo上传到xianchaomaster1的/etc/yum.repos.d目录下,这样才能安装keepalived和nginx

把epel.repo传到xianchaomaster2、xianchaomaster3、node01上


总结一下,各个节点安装nginx和keepalived


三个master节点。直接用epel源安装:

yum install nginx keepalived -y


所有Master节点配置nginx(详细配置参考nginx文档,所有Master节点的nginx配置相同):http模块外面下方添加,stream模块与http模块时平行的

[root@master01 work ]#cat /etc/nginx/nginx.conf
# For more information on configuration, see:
#   * Official English Documentation: http://nginx.org/en/docs/
#   * Official Russian Documentation: http://nginx.org/ru/docs/

user nginx;
worker_processes auto;
error_log /var/log/nginx/error.log;
pid /run/nginx.pid;

# Load dynamic modules. See /usr/share/doc/nginx/README.dynamic.
include /usr/share/nginx/modules/*.conf;

events {
    worker_connections 1024;
}

http {
    log_format  main  '$remote_addr - $remote_user [$time_local] "$request" '
                      '$status $body_bytes_sent "$http_referer" '
                      '"$http_user_agent" "$http_x_forwarded_for"';

    access_log  /var/log/nginx/access.log  main;

    sendfile            on;
    tcp_nopush          on;
    tcp_nodelay         on;
    keepalive_timeout   65;
    types_hash_max_size 4096;

    include             /etc/nginx/mime.types;
    default_type        application/octet-stream;

    # Load modular configuration files from the /etc/nginx/conf.d directory.
    # See http://nginx.org/en/docs/ngx_core_module.html#include
    # for more information.
    include /etc/nginx/conf.d/*.conf;



    server {
        listen       80;
        listen       [::]:80;
        server_name  _;
        root         /usr/share/nginx/html;

        # Load configuration files for the default server block.
        include /etc/nginx/default.d/*.conf;

        error_page 404 /404.html;
        location = /404.html {
        }

        error_page 500 502 503 504 /50x.html;
        location = /50x.html {
        }
    }

# Settings for a TLS enabled server.
#
#    server {
#        listen       443 ssl http2;
#        listen       [::]:443 ssl http2;
#        server_name  _;
#        root         /usr/share/nginx/html;
#
#        ssl_certificate "/etc/pki/nginx/server.crt";
#        ssl_certificate_key "/etc/pki/nginx/private/server.key";
#        ssl_session_cache shared:SSL:1m;
#        ssl_session_timeout  10m;
#        ssl_ciphers HIGH:!aNULL:!MD5;
#        ssl_prefer_server_ciphers on;
#
#        # Load configuration files for the default server block.
#        include /etc/nginx/default.d/*.conf;
#
#        error_page 404 /404.html;
#            location = /40x.html {
#        }
#
#        error_page 500 502 503 504 /50x.html;
#            location = /50x.html {
#        }
#    }

}

    stream {

    upstream apiserver {
        server 10.10.0.10:6443 max_fails=3 fail_timeout=30s;
        server 10.10.0.11:6443 max_fails=3 fail_timeout=30s;
        server 10.10.0.12:6443 max_fails=3 fail_timeout=30s;
     }

    server {
        listen 7443;
        proxy_connect_timeout 2s;
        proxy_timeout 900s;
        proxy_pass apiserver;

     }

}


启动nginx如果报错:

[root@master01 work ]#systemctl status nginx.service
● nginx.service - The nginx HTTP and reverse proxy server
   Loaded: loaded (/usr/lib/systemd/system/nginx.service; disabled; vendor preset: disabled)
   Active: failed (Result: exit-code) since Thu 2022-10-27 10:55:16 CST; 12s ago
  Process: 3019 ExecStartPre=/usr/sbin/nginx -t (code=exited, status=1/FAILURE)
  Process: 3017 ExecStartPre=/usr/bin/rm -f /run/nginx.pid (code=exited, status=0/SUCCESS)

Oct 27 10:55:16 master01 systemd[1]: Starting The nginx HTTP and reverse proxy server...
Oct 27 10:55:16 master01 nginx[3019]: nginx: [emerg] unknown directive "stream" in /etc/nginx/nginx.conf:37
Oct 27 10:55:16 master01 nginx[3019]: nginx: configuration file /etc/nginx/nginx.conf test failed
Oct 27 10:55:16 master01 systemd[1]: nginx.service: control process exited, code=exited status=1
Oct 27 10:55:16 master01 systemd[1]: Failed to start The nginx HTTP and reverse proxy server.
Oct 27 10:55:16 master01 systemd[1]: Unit nginx.service entered failed state.
Oct 27 10:55:16 master01 systemd[1]: nginx.service failed.

就需要安装stream模块,并不是所有的版本都有现成的,没有的话需要编译,目前epel源的1.20版本有

yum install nginx-mod-stream -y

[root@master01 work ]#systemctl enable nginx.service --now
Created symlink from /etc/systemd/system/multi-user.target.wants/nginx.service to /usr/lib/systemd/system/nginx.service.

[root@master02 yum.repos.d ]#systemctl enable nginx.service --now
Created symlink from /etc/systemd/system/multi-user.target.wants/nginx.service to /usr/lib/systemd/system/nginx.service.


[root@master03 yum.repos.d ]#systemctl enable nginx.service --now
Created symlink from /etc/systemd/system/multi-user.target.wants/nginx.service to /usr/lib/systemd/system/nginx.service.

keepalived配置:

所有Master节点配置KeepAlived,配置不一样,注意区分 公有云不支持keepalived

主备三个地方配置不一样

router_id

state

priority

其他都完全一样

主keepalived-master01

[root@master01 keepalived ]#cat keepalived.conf 
! Configuration File for keepalived

global_defs {
   router_id 10.10.0.10
}

vrrp_script chk_nginx {
   script "/etc/keepalived/check_nginx.sh"
   interval 2
   weight -20
}

vrrp_instance VI_1 {
    state MASTER
    interface ens33           # 修改为实际网卡名
    virtual_router_id 51      # VRRP 路由 ID实例,每个实例是唯一的 ,如果运行多个keepalived,该id不能重复
    priority 100              # 优先级,备服务器设置 90 
    advert_int 1              # 指定VRRP 心跳包通告间隔时间,默认1秒 
  nopreempt                 #默认是抢占模式,主节点改成非抢占模式,防止VIP来回漂移
    authentication {
        auth_type PASS
        auth_pass 1111
    }
    virtual_ipaddress {
  10.10.0.100
    }
    track_script {
        chk_nginx
    }
}


如果nginx死掉,会导致用户请求失败,但keepalived并不会进行切换,需要写脚本检查nginx存活状态,

如果确认nginx已死掉,则停掉本机keepalived,已确保keepalived能够正常切换

[root@master01 keepalived ]#cat check_nginx.sh 
#!/bin/bash
while true;do
nginxpid=$(ps -C nginx --no-header|wc -l) &> /dev/null
if [ $nginxpid -eq 0 ];then
        systemctl start nginx
        echo "nginx已重启!"
        sleep 5
        nginxpid=$(ps -C nginx --no-header|wc -l) &> /dev/null
        if [ $nginxpid -eq 0 ];then
          systemctl stop keepalived
          echo "该机已出现故障!"
          exit 1
        fi
fi
        sleep 5
        echo "正常运行中!"
done

给脚本加执行权限

[root@master01 keepalived ]#chmod +x check_nginx.sh

备-master02配置:

[root@master02 keepalived ]#cat keepalived.conf 
! Configuration File for keepalived

global_defs {
   router_id 10.10.0.11
}

vrrp_script chk_nginx {
   script "/etc/keepalived/check_nginx.sh"
   interval 2
   weight -20
}

vrrp_instance VI_1 {
    state BACKUP
    interface ens33
    virtual_router_id 51
    priority 90
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass 1111
    }
    virtual_ipaddress {
  10.10.0.100
    }
    track_script {
        chk_nginx
    }
}

备-master03配置:

[root@master03 keepalived ]#cat keepalived.conf 
! Configuration File for keepalived

global_defs {
   router_id 10.10.0.12
}

vrrp_script chk_nginx {
   script "/etc/keepalived/check_nginx.sh"
   interval 2
   weight -20
}

vrrp_instance VI_1 {
    state BACKUP
    interface ens33
    virtual_router_id 51
    priority 90
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass 1111
    }
    virtual_ipaddress {
  10.10.0.100
    }
    track_script {
        chk_nginx
    }
}

测试keepalived

停掉master01上的nginx。vip会漂移到master02或master03

经测试,会漂移

更改node节点连接VIP

目前所有的Worker Node组件连接都还是master01 Node,如果不改为连接VIP走负载均衡器,那么Master还是单点故障。

因此接下来就是要改所有Worker Node(kubectl get node命令查看到的节点)组件配置文件,

由原来10.10.0.10修改为10.10.0.100(VIP)。在所有Worker Node执行:

[root@node01 ~]# sed -i 's#10.10.0.10:6443#10.10.0.100:7443#' /etc/kubernetes/kubelet-bootstrap.kubeconfig

[root@node01 ~]# sed -i 's#10.10.0.10:6443#10.10.0.100:7443#' /etc/kubernetes/kubelet.kubeconfig

[root@node01 ~]# sed -i 's#10.10.0.10:6443#10.10.0.100:7443#' /etc/kubernetes/kube-proxy.yaml

[root@node01 ~]# sed -i 's#10.10.0.10:6443#10.10.0.100:7443#' /etc/kubernetes/kube-proxy.kubeconfig

[root@node01 ~]# systemctl restart kubelet kube-proxy

每个master节点:

[root@master02 ~ ]#sed -i 's/10.10.0.10:6443/10.10.0.100:7443/g' /root/.kube/config
[root@master03 keepalived ]#sed -i 's/10.10.0.10:6443/10.10.0.100:7443/g' /root/.kube/config
[root@master01 work ]#sed -i 's/10.10.0.10:6443/10.10.0.100:7443/g' /root/.kube/config


[root@master03 ~ ]#sed -i 's/10.10.0.10:6443/10.10.0.100:7443/g' /etc/kubernetes/kube-scheduler.kubeconfig
[root@master03 ~ ]#sed -i 's/10.10.0.10:6443/10.10.0.100:7443/g' /etc/kubernetes/kube-controller-manager.kubeconfig

[root@master02 ~ ]#sed -i 's/10.10.0.10:6443/10.10.0.100:7443/g' /etc/kubernetes/kube-scheduler.kubeconfig
[root@master02 ~ ]#sed -i 's/10.10.0.10:6443/10.10.0.100:7443/g' /etc/kubernetes/kube-controller-manager.kubeconfig

[root@master01 ~ ]#sed -i 's/10.10.0.10:6443/10.10.0.100:7443/g' /etc/kubernetes/kube-scheduler.kubeconfig
[root@master01 ~ ]#sed -i 's/10.10.0.10:6443/10.10.0.100:7443/g' /etc/kubernetes/kube-controller-manager.kubeconfig


[root@master03 ~ ]#systemctl restart kube-scheduler.service kube-controller-manager.service 
[root@master02 ~ ]#systemctl restart kube-scheduler.service kube-controller-manager.service 
[root@master01 ~ ]#systemctl restart kube-scheduler.service kube-controller-manager.service 


[root@master03 ~ ]#kubectl cluster-info
Kubernetes control plane is running at https://10.10.0.100:7443
CoreDNS is running at https://10.10.0.100:7443/api/v1/namespaces/kube-system/services/kube-dns:dns/proxy

To further debug and diagnose cluster problems, use 'kubectl cluster-info dump'.

这样高可用集群就安装好了


目前kubectl get nodes 只能看到工作节点,若想能看到控制节点。则需要在控制节点将kubelet kube-proxy流程也走一遍

并且注意修改kubelet.json kubelet-bootstrap.kubeconfig kube-proxy.yaml kube-proxy.kubeconfig 中的ip


将master节点打上标签:

[root@master01 work ]#kubectl label node master01 node-role.kubernetes.io/controlplane=true
node/master01 labeled
[root@master01 work ]#kubectl label node master02 node-role.kubernetes.io/controlplane=true
node/master02 labeled
[root@master01 work ]#kubectl label node master03 node-role.kubernetes.io/controlplane=true
node/master03 labeled
[root@master01 work ]#kubectl label node master01 node-role.kubernetes.io/etcd=true
node/master01 labeled
[root@master01 work ]#kubectl label node master02 node-role.kubernetes.io/etcd=true
node/master02 labeled
[root@master01 work ]#kubectl label node master03 node-role.kubernetes.io/etcd=true

工作节点打标签

[root@master01 work ]#kubectl get nodes
NAME       STATUS   ROLES               AGE     VERSION
master01   Ready    controlplane,etcd   67m     v1.20.7
master02   Ready    controlplane,etcd   67m     v1.20.7
master03   Ready    controlplane,etcd   67m     v1.20.7
node01     Ready    worker              7h42m   v1.20.7
[root@master01 work ]#kubectl get nodes
NAME       STATUS   ROLES               AGE     VERSION
master01   Ready    controlplane,etcd   67m     v1.20.7
master02   Ready    controlplane,etcd   67m     v1.20.7
master03   Ready    controlplane,etcd   67m     v1.20.7
node01     Ready    worker              7h42m   v1.20.7

7.将master节点打上污点,禁止调度

[root@master01 work ]#kubectl taint node master01 master01=null:NoSchedule
node/master01 tainted
[root@master01 work ]#kubectl taint node master02 master02=null:NoSchedule
node/master02 tainted
[root@master01 work ]#kubectl taint node master03 master03=null:NoSchedule
node/master03 tainted

好了,以上就是二进制文件搭建高可用K8S集群的全部流程,希望对大家学习工作中提供帮助,ღ( ´・ᴗ・` )比心。

相关实践学习
通过Ingress进行灰度发布
本场景您将运行一个简单的应用,部署一个新的应用用于新的发布,并通过Ingress能力实现灰度发布。
容器应用与集群管理
欢迎来到《容器应用与集群管理》课程,本课程是“云原生容器Clouder认证“系列中的第二阶段。课程将向您介绍与容器集群相关的概念和技术,这些概念和技术可以帮助您了解阿里云容器服务ACK/ACK Serverless的使用。同时,本课程也会向您介绍可以采取的工具、方法和可操作步骤,以帮助您了解如何基于容器服务ACK Serverless构建和管理企业级应用。 学习完本课程后,您将能够: 掌握容器集群、容器编排的基本概念 掌握Kubernetes的基础概念及核心思想 掌握阿里云容器服务ACK/ACK Serverless概念及使用方法 基于容器服务ACK Serverless搭建和管理企业级网站应用
相关文章
|
1月前
|
Kubernetes 持续交付 Docker
利用 Docker 和 Kubernetes 实现微服务部署
【10月更文挑战第2天】利用 Docker 和 Kubernetes 实现微服务部署
|
1月前
|
Prometheus Kubernetes 监控
k8s部署针对外部服务器的prometheus服务
通过上述步骤,您不仅成功地在Kubernetes集群内部署了Prometheus,还实现了对集群外服务器的有效监控。理解并实施网络配置是关键,确保监控数据的准确无误传输。随着监控需求的增长,您还可以进一步探索Prometheus生态中的其他组件,如Alertmanager、Grafana等,以构建完整的监控与报警体系。
121 60
|
1月前
|
Prometheus Kubernetes 监控
k8s部署针对外部服务器的prometheus服务
通过上述步骤,您不仅成功地在Kubernetes集群内部署了Prometheus,还实现了对集群外服务器的有效监控。理解并实施网络配置是关键,确保监控数据的准确无误传输。随着监控需求的增长,您还可以进一步探索Prometheus生态中的其他组件,如Alertmanager、Grafana等,以构建完整的监控与报警体系。
210 62
|
20天前
|
Kubernetes 关系型数据库 MySQL
Kubernetes入门:搭建高可用微服务架构
【10月更文挑战第25天】在快速发展的云计算时代,微服务架构因其灵活性和可扩展性备受青睐。本文通过一个案例分析,展示了如何使用Kubernetes将传统Java Web应用迁移到Kubernetes平台并改造成微服务架构。通过定义Kubernetes服务、创建MySQL的Deployment/RC、改造Web应用以及部署Web应用,最终实现了高可用的微服务架构。Kubernetes不仅提供了服务发现和负载均衡的能力,还通过各种资源管理工具,提升了系统的可扩展性和容错性。
61 3
|
1月前
|
NoSQL 关系型数据库 Redis
高可用和性能:基于ACK部署Dify的最佳实践
本文介绍了基于阿里云容器服务ACK,部署高可用、可伸缩且具备高SLA的生产可用的Dify服务的详细解决方案。
|
1月前
|
Kubernetes Cloud Native 流计算
Flink-12 Flink Java 3分钟上手 Kubernetes云原生下的Flink集群 Rancher Stateful Set yaml详细 扩容缩容部署 Docker容器编排
Flink-12 Flink Java 3分钟上手 Kubernetes云原生下的Flink集群 Rancher Stateful Set yaml详细 扩容缩容部署 Docker容器编排
74 0
|
1月前
|
Kubernetes 网络协议 安全
[kubernetes]二进制方式部署单机k8s-v1.30.5
[kubernetes]二进制方式部署单机k8s-v1.30.5
|
6月前
|
Kubernetes 网络协议 应用服务中间件
K8S二进制部署实践-1.15.5
K8S二进制部署实践-1.15.5
81 0
|
6月前
|
Kubernetes 调度 Docker
Kubernetes高可用集群二进制部署(五)kubelet、kube-proxy、Calico、CoreDNS
Kubernetes高可用集群二进制部署(五)kubelet、kube-proxy、Calico、CoreDNS
Kubernetes高可用集群二进制部署(五)kubelet、kube-proxy、Calico、CoreDNS
|
2月前
|
存储 Kubernetes 负载均衡
CentOS 7.9二进制部署K8S 1.28.3+集群实战
本文详细介绍了在CentOS 7.9上通过二进制方式部署Kubernetes 1.28.3+集群的全过程,包括环境准备、组件安装、证书生成、高可用配置以及网络插件部署等关键步骤。
415 3
CentOS 7.9二进制部署K8S 1.28.3+集群实战

推荐镜像

更多