在線云原生邊緣計(jì)算KubeEdge安裝配置（二）

這篇具有很好參考價(jià)值的文章主要介紹了在線云原生邊緣計(jì)算KubeEdge安裝配置（二）。希望對(duì)大家有所幫助。如果存在錯(cuò)誤或未考慮完全的地方，請(qǐng)大家不吝賜教，您也可以點(diǎn)擊"舉報(bào)違法"按鈕提交疑問。

1. K8S集群部署，可以參考如下博客

請(qǐng)安裝k8s集群，centos安裝k8s集群
請(qǐng)安裝k8s集群，ubuntu安裝k8s集群

請(qǐng)安裝kubeedge cloudcore　centos安裝K8S

2.安裝kubEedge

2.1 編輯kube-proxy使用ipvs代理

kubectl edit configmaps kube-proxy -n kube-system #修改kube-proxy

#大約在40多行
     37     ipvs:
     38       excludeCIDRs: null
     39       minSyncPeriod: 0s
     40       scheduler: ""
     41       strictARP: true  #這里由原來的false修改為true
     42       syncPeriod: 0s
     43       tcpFinTimeout: 0s
     44       tcpTimeout: 0s
     45       udpTimeout: 0s
     46     kind: KubeProxyConfiguration
     47     metricsBindAddress: ""
     48     mode: "ipvs"  #這里原來是空的修改為ipvs
     49     nodePortAddresses: null
     50     oomScoreAdj: null
     51     portRange: ""
     52     showHiddenMetricsForVersion: ""
     53     winkernel:
     54       enableDSR: false
     55       forwardHealthCheckVip: false
     56       networkName: ""
     57       rootHnsEndpointName: ""
     58       sourceVip: ""

kubectl rollout restart daemonset kube-proxy -n kube-system
kubectl get pods -n kube-system

2.2 基礎(chǔ)服務(wù)提供，負(fù)載均衡器metallb

#無網(wǎng)絡(luò)情況下，先上傳kubeedge.tar 和 metallb-native.yaml文件
docker load -i kubeedge.tar  #導(dǎo)入鏡像

#有網(wǎng)絡(luò)可以直接拉去鏡像
docker pull quay.io/metallb/controller:v0.13.5
docker pull quay.io/metallb/speaker:v0.13.5
wget https://raw.githubusercontent.com/metallb/metallb/v0.13.5/config/manifests/metallb-native.yaml

kubectl apply -f metallb-native.yaml
kubectl get pods -n metallb-system

[root@k8s-master01 ~]# kubectl get pods -n metallb-system
NAME                         READY   STATUS    RESTARTS   AGE
controller-8d9cf599f-m4x27   1/1     Running   0          23s
speaker-zm7w4                1/1     Running   0          23s
speaker-zww5l                1/1     Running   0          23s
[root@k8s-master01 ~]#

#具體參考官網(wǎng)地址：https://metallb.universe.tf/configuration/#layer-2-configuration

#創(chuàng)建IP地址資源池，原則是跟主機(jī)在同一個(gè)網(wǎng)段，沒有使用的地址，根據(jù)自己項(xiàng)目的規(guī)模我合理預(yù)留了21個(gè)IP
cat > first-ippool.yaml << EOF
apiVersion: metallb.io/v1beta1
kind: IPAddressPool
metadata:
  name: first-pool
  namespace: metallb-system
spec:
  addresses:
  - 192.168.186.200-192.168.186.220
EOF

kubectl apply -f first-ippool.yaml
kubectl get ipaddresspools -n metallb-system

[root@k8s-master01 ~]# kubectl get ipaddresspools -n metallb-system
NAME         AGE
first-pool   2m1s
[root@k8s-master01 ~]#

#開啟二層轉(zhuǎn)發(fā)，實(shí)現(xiàn)k8s集群節(jié)點(diǎn)外訪問
cat > l2forward.yaml << EOF
apiVersion: metallb.io/v1beta1
kind: L2Advertisement
metadata:
  name: example
  namespace: metallb-system
EOF

kubectl apply -f l2forward.yaml
kubectl get l2advertisements -n metallb-system

#二層轉(zhuǎn)發(fā)案例測(cè)試
cat > nginx.yaml << EOF
apiVersion: apps/v1
kind: Deployment
metadata:
  name: nginx
spec:
  replicas: 1
  selector:
    matchLabels:
      app: nginx
  template:
    metadata:
      labels:
        app: nginx
    spec:
      containers:
      - image: nginx
        name: nginx
        imagePullPolicy: IfNotPresent
---
apiVersion: v1
kind: Service
metadata:
  name: nginx-svc
spec:
  type: LoadBalancer
  ports:
  - port: 80
    targetPort: 80
  selector:
    app: nginx
EOF

kubectl apply -f nginx.yaml

[root@k8s-master01 ~]# kubectl get pods,svc #svc的ip從地址池中分配的,如果是云服務(wù)器則是公網(wǎng)IP,公網(wǎng)必須帶端口訪問
NAME                         READY   STATUS    RESTARTS   AGE
pod/nginx-5f5c64f949-bhpvf   1/1     Running   0          50s

NAME                 TYPE           CLUSTER-IP     EXTERNAL-IP       PORT(S)        AGE
service/kubernetes   ClusterIP      10.96.0.1      <none>            443/TCP        5h43m
service/nginx-svc    LoadBalancer   10.99.41.146   192.168.186.200   80:30293/TCP   50s
[root@k8s-master01 ~]#

在線云原生邊緣計(jì)算KubeEdge安裝配置（二）,kubernetes,docker容器相關(guān),云原生,邊緣計(jì)算,人工智能

測(cè)試通過

kubectl delete -f nginx.yaml  #刪除無用的pod和svc

yum install iptables-services -y #安裝iptables后期有用
systemctl start  iptables.service
systemctl enable  iptables.service
systemctl status  iptables.service

3. 對(duì)于kubeEdge概述

KubeEdge由云和邊緣組成。它建立在Kubernetes之上，為聯(lián)網(wǎng)應(yīng)用部署和云與邊緣之間的元數(shù)據(jù)同步提供核心基礎(chǔ)設(shè)施支持。所以如果我們想要設(shè)置KubeEdge，我們需要設(shè)置Kubernetes集群(可以使用現(xiàn)有的集群)，云端和邊緣端在cloud side, 我們需要安裝

Docker
Kubernetes cluster
cloudcore

在 edge side, 我們需要安裝

Docker
MQTT （配置可以選用，不是一定需要）
edgecore

在線云原生邊緣計(jì)算KubeEdge安裝配置（二）,kubernetes,docker容器相關(guān),云原生,邊緣計(jì)算,人工智能文章來源地址http://www.zghlxwxcb.cn/news/detail-728070.html

3.部署kubeedge cloudcore(主節(jié)點(diǎn))

3.1 獲取keadm工具

wget https://github.com/kubeedge/kubeedge/releases/download/v1.10.1/keadm-v1.12.1-linux-amd64.tar.gz

tar -zxvf keadm-v1.12.1-linux-amd64.tar.gz
cp keadm-v1.12.1-linux-amd64/keadm/keadm  /usr/local/bin/
keadm version

3.2 部署cloudcore

#在master上部署
#設(shè)置云端  使用keadm初始化，安裝cloudcore
docker pull kubeedge/cloudcore:v1.14.2  
docker pull  kubeedge/iptables-manager:v1.14.2

keadm init --advertise-address=192.168.186.200 --set iptablesManager.mode="external" --profile version=v1.12.1 #這里的IP是負(fù)載均衡器metallb分配的IP，如果是云服務(wù)器則用公網(wǎng)

#以上參數(shù)分別為master節(jié)點(diǎn)地址和要安裝的KubeEdge版本。

#keadm reset  #安裝錯(cuò)誤可以重置
keadm gettoken  #獲取token

[root@k8s-ke-cloud docker]# kubectl get deployment -n kubeedge #上述多此執(zhí)行后依然不行
NAME        READY   UP-TO-DATE   AVAILABLE   AGE    CONTAINERS   IMAGES                       SELECTOR
cloudcore   0/1     1            0           7m2s   cloudcore    kubeedge/cloudcore:v1.12.1   k8s-app=kubeedge,kubeedge=cloudcore
[root@k8s-ke-cloud docker]# kubectl delete  deployment -n kubeedge  #刪除這個(gè)deployment
error: resource(s) were provided, but no name was specified
[root@k8s-ke-cloud docker]# kubectl delete  deployment  cloudcore -n kubeedge 
deployment.apps "cloudcore" deleted
[root@k8s-ke-cloud docker]#

[root@k8s-master01 ~]# keadm init --advertise-address=192.168.186.200 --set iptablesManager.mode="external" --profile version=v1.12.1
Kubernetes version verification passed, KubeEdge installation will start...
CLOUDCORE started
=========CHART DETAILS=======
NAME: cloudcore
LAST DEPLOYED: Mon Oct  9 15:25:36 2023
NAMESPACE: kubeedge
STATUS: deployed
REVISION: 1
[root@k8s-master ~]# keadm gettoken
7da3a4434a154f267a7ba42727a0f038bbca5ae0d8c7adc739bc32591574b818.eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJleHAiOjE2OTY5OTAxNjd9.htL9plVcKIgVdGuRdb4cp2gviQVol_Jiwa_RNE3ieVI[root@k8s-master ~]#

kubectl get pods -n kubeedge
kubectl get svc -n kubeedge
kubectl get deployment -n kubeedge

[root@k8s-master ~]#kubectl get pods  -n kubeedge
NAME                           READY   STATUS             RESTARTS      AGE
cloud-iptables-manager-rpxm8   1/1     Running            0             7m37s
cloudcore-5876c76687-ff9zv     1/1     Running            0     7m37s
[root@k8s-master ~]# kubectl get svc   -n kubeedge
NAME        TYPE        CLUSTER-IP       EXTERNAL-IP   PORT(S)                                             AGE
cloudcore   ClusterIP   10.109.201.171   <none>        10000/TCP,10001/TCP,10002/TCP,10003/TCP,10004/TCP   7m40s
[root@k8s-master ~]# kubectl get deployment    -n kubeedge
NAME        READY   UP-TO-DATE   AVAILABLE   AGE
cloudcore   1/1     1            0           7m43s
[root@ecs-2ef1-0001 kubeedge]#

[root@cloudcorer ~]# kubectl get ns
NAME                   STATUS   AGE
default                Active   58d
kube-node-lease        Active   58d
kube-public            Active   58d
kube-system            Active   58d
kubeedge               Active   8m44s  #在這個(gè)空間中中
kubernetes-dashboard   Active   58d
metallb-system         Active   57d
[root@cloudcorer ~]# kubectl get pods,svc -n kubeedge
NAME                               READY   STATUS    RESTARTS   AGE
pod/cloud-iptables-manager-8kh6q   1/1     Running   0          109m
pod/cloudcore-5876c76687-tkkzx     1/1     Running   0          109m

NAME                TYPE        CLUSTER-IP       EXTERNAL-IP   PORT(S)                                             AGE
service/cloudcore   ClusterIP   10.101.140.171   <none>        10000/TCP,10001/TCP,10002/TCP,10003/TCP,10004/TCP   109m
[root@cloudcorer ~]#

kubectl edit svc cloudcore -n kubeedge #修改大概在51行 
     49   selector:
     50     k8s-app: kubeedge
     51     kubeedge: cloudcore
     52   sessionAffinity: None
     53   type: LoadBalancer  #修改后的
     54 status:
     55   loadBalancer: {}

[root@k8s-master ~]# kubectl get pods,svc -n kubeedge
NAME                               READY   STATUS    RESTARTS   AGE
pod/cloud-iptables-manager-8kh6q   1/1     Running   0          110m
pod/cloudcore-5876c76687-tkkzx     1/1     Running   0          110m

NAME                TYPE           CLUSTER-IP       EXTERNAL-IP       PORT(S)                                                                           AGE
service/cloudcore   LoadBalancer   10.101.140.171   192.168.186.200   10000:31953/TCP,10001:32329/TCP,10002:31154/TCP,10003:30807/TCP,10004:31038/TCP   110m
[root@k8s-master ~]#

kubectl apply -f metrics-server.yaml  #安裝metrics-server

[root@k8s-master kubernetes]# kubectl get pods -n kube-system -o wide #metrics-server不正常
NAME                                       READY   STATUS    RESTARTS   AGE     IP                NODE         NOMINATED NODE   READINESS GATES
calico-kube-controllers-67bb5696f5-sfhs6   1/1     Running   0          4h3m    10.244.235.195    k8s-master   <none>           <none>
calico-node-2w8jn                          1/1     Running   0          4h3m    192.168.186.128   k8s-master   <none>           <none>
calico-node-vt8g4                          1/1     Running   0          4h3m    192.168.186.129   k8s-node1    <none>           <none>
coredns-7f6cbbb7b8-q8z29                   1/1     Running   0          4h5m    10.244.235.193    k8s-master   <none>           <none>
coredns-7f6cbbb7b8-qc8d2                   1/1     Running   0          4h5m    10.244.235.194    k8s-master   <none>           <none>
etcd-k8s-master                            1/1     Running   0          4h5m    192.168.186.128   k8s-master   <none>           <none>
kube-apiserver-k8s-master                  1/1     Running   0          4h5m    192.168.186.128   k8s-master   <none>           <none>
kube-controller-manager-k8s-master         1/1     Running   0          4h5m    192.168.186.128   k8s-master   <none>           <none>
kube-proxy-2vt4s                           1/1     Running   0          3h52m   192.168.186.129   k8s-node1    <none>           <none>
kube-proxy-99x2t                           1/1     Running   0          3h52m   192.168.186.128   k8s-master   <none>           <none>
kube-scheduler-k8s-master                  1/1     Running   0          4h5m    192.168.186.128   k8s-master   <none>           <none>
metrics-server-5f85c44dcd-kpnpk            1/1     Running   0          4m4s    10.244.36.85      k8s-node1    <none>           <none>    #如果狀態(tài)和READY不正確，就執(zhí)行如下的語句
[root@k8s-master kubernetes]#

kubectl patch deploy metrics-server -n kube-system --type='json' -p='[{"op":"add","path":"/spec/template/spec/containers/0/args/-","value":"--kubelet-insecure-tls"}]'  #metrics-server容器不正常情況下，才修改為不認(rèn)證，正常就不要執(zhí)行這句

#修改 kube-proxy 和 kube-flannel 的節(jié)點(diǎn)親和性，不調(diào)度到邊緣節(jié)點(diǎn)
kubectl patch daemonset kube-proxy -n kube-system -p '{"spec": {"template": {"spec": {"affinity": {"nodeAffinity": {"requiredDuringSchedulingIgnoredDuringExecution": {"nodeSelectorTerms": [{"matchExpressions": [{"key": "node-role.kubernetes.io/edge", "operator": "DoesNotExist"}]}]}}}}}}}'

kubectl patch daemonset kube-flannel-ds -n kube-flannel -p '{"spec": {"template": {"spec": {"affinity": {"nodeAffinity": {"requiredDuringSchedulingIgnoredDuringExecution": {"nodeSelectorTerms": [{"matchExpressions": [{"key": "node-role.kubernetes.io/edge", "operator": "DoesNotExist"}]}]}}}}}}}'

kubectl get daemonset -n kube-system | grep -v NAME | awk '{print $1}' | xargs -n 1 kubectl patch daemonset -n kube-system --type='json' -p='[{"op": "replace", "path": "/spec/template/spec/affinity", "value":{"nodeAffinity":{"requiredDuringSchedulingIgnoredDuringExecution":{"nodeSelectorTerms":[{"matchExpressions":[{"key":"node-role.kubernetes.io/edge","operator":"DoesNotExist"}]}]}}}}]'

kubectl get daemonset -n metallb-system | grep -v NAME | awk '{print $1}' | xargs -n 1 kubectl patch daemonset -n metallb-system --type='json' -p='[{"op": "replace", "path": "/spec/template/spec/affinity", "value":{"nodeAffinity":{"requiredDuringSchedulingIgnoredDuringExecution":{"nodeSelectorTerms":[{"matchExpressions":[{"key":"node-role.kubernetes.io/edge","operator":"DoesNotExist"}]}]}}}}]'

kubectl get daemonset -n calico-system | grep -v NAME | awk '{print $1}' | xargs -n 1 kubectl patch daemonset -n calico-system --type='json' -p='[{"op": "replace", "path": "/spec/template/spec/affinity", "value":{"nodeAffinity":{"requiredDuringSchedulingIgnoredDuringExecution":{"nodeSelectorTerms":[{"matchExpressions":[{"key":"node-role.kubernetes.io/edge","operator":"DoesNotExist"}]}]}}}}]'

至此cloudcore部署完畢

以下是邊緣節(jié)點(diǎn)的部署方式

4. 部署edgecore（邊緣節(jié)點(diǎn)）

[root@k8s-master ~]# kubectl get nodes
NAME         STATUS   ROLES                  AGE     VERSION
k8s-master   Ready    control-plane,master   4h12m   v1.22.6
k8s-node1    Ready    <none>                 4h9m    v1.22.6
[root@k8s-master ~]#

４.1 獲取keadm工具

#注意邊緣節(jié)點(diǎn)配置都很低，只需要運(yùn)行業(yè)務(wù)容器
#邊緣節(jié)點(diǎn)只需要安裝docker就可以了
hostnamectl set-hostname edge-node1  && bash

wget https://github.com/kubeedge/kubeedge/releases/download/v1.10.1/keadm-v1.12.1-linux-amd64.tar.gz

tar -zxvf keadm-v1.12.1-linux-amd64.tar.gz
cp keadm-v1.12.1-linux-amd64/keadm/keadm  /usr/local/bin/
keadm version

#提前拉去鏡像
docker pull kubeedge/installation-package:v1.12.1
docker pull kubeedge/pause:3.1
docker pull eclipse-mosquitto:1.6.15

# 在云節(jié)點(diǎn)獲取 token
keadm gettoken

# 使用 token 將邊緣節(jié)點(diǎn)加入集群
keadm join --cloudcore-ipport=192.168.186.200:10000 --token=23bacb0d0fee0b137f821d7f6b100039dee72b198aef159ad7aca5c63f2ad778.eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJleHAiOjE2OTcwNDA5ODF9.siZ4CuljbcNtCmOwgOGS9ZGc8FNIejKQh3q0feiRCI4 --kubeedge-version=v1.12.1 --cgroupdriver systemd

＃如果添加失敗了，請(qǐng)?jiān)趀dgecoe中執(zhí)行
＃docker rm -f mqtt   && rm -fr /etc/kubeedge/
＃keadm join --token=$TOKEN --cloudcore-ipport=$SERVER --kubeedge-version=1.12.1

[root@edge-node1 ~]# keadm join --cloudcore-ipport=192.168.186.200:10000 --token=23bacb0d0fee0b137f821d7f6b100039dee72b198aef159ad7aca5c63f2ad778.eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJleHAiOjE2OTcwNDA5ODF9.siZ4CuljbcNtCmOwgOGS9ZGc8FNIejKQh3q0feiRCI4 --kubeedge-version=v1.12.1 --cgroupdriver systemd
I0526 22:43:35.338932   19494 command.go:845] 1. Check KubeEdge edgecore process status
I0526 22:43:35.508735   19494 command.go:845] 2. Check if the management directory is clean
I0526 22:43:35.509314   19494 join.go:100] 3. Create the necessary directories
I0526 22:43:35.516006   19494 join.go:176] 4. Pull Images
Pulling kubeedge/installation-package:v1.12.1 ...
Pulling eclipse-mosquitto:1.6.15 ...
Pulling kubeedge/pause:3.1 ...
I0526 22:43:35.521861   19494 join.go:176] 5. Copy resources from the image to the management directory
I0526 22:43:36.445496   19494 join.go:176] 6. Start the default mqtt service
I0526 22:43:37.119023   19494 join.go:100] 7. Generate systemd service file
I0526 22:43:37.120245   19494 join.go:100] 8. Generate EdgeCore default configuration
I0526 22:43:37.120309   19494 join.go:230] The configuration does not exist or the parsing fails, and the default configuration is generated
W0526 22:43:37.232157   19494 validation.go:71] NodeIP is empty , use default ip which can connect to cloud.
I0526 22:43:37.243148   19494 join.go:100] 9. Run EdgeCore daemon
I0526 22:43:37.824194   19494 join.go:317] 
I0526 22:43:37.824216   19494 join.go:318] KubeEdge edgecore is running, For logs visit: journalctl -u edgecore.service -xe
[root@edge-node1 ~]#

journalctl -u edgecore -n 50   #edge端查看日志

iptables -t nat -A OUTPUT -p tcp --dport 10350 -j DNAT --to 114.115.220.147:10003 #這個(gè)IP114.115.220.147是cloudcorer的

到了這里，關(guān)于在線云原生邊緣計(jì)算KubeEdge安裝配置（二）的文章就介紹完了。如果您還想了解更多內(nèi)容，請(qǐng)?jiān)谟疑辖撬阉鱐OY模板網(wǎng)以前的文章或繼續(xù)瀏覽下面的相關(guān)文章，希望大家以后多多支持TOY模板網(wǎng)！

国产无码综合区,色欲AV无码国产永久播放,无码天堂亚洲国产AV,国产日韩欧美女同一区二区

Toy模板網(wǎng)

在線云原生邊緣計(jì)算KubeEdge安裝配置（二）

1. K8S集群部署，可以參考如下博客

2.安裝kubEedge

2.1 編輯kube-proxy使用ipvs代理

2.2 基礎(chǔ)服務(wù)提供，負(fù)載均衡器metallb

測(cè)試通過

3. 對(duì)于kubeEdge概述

3.部署kubeedge cloudcore(主節(jié)點(diǎn))

3.1 獲取keadm工具

3.2 部署cloudcore

至此cloudcore部署完畢

以下是邊緣節(jié)點(diǎn)的部署方式

4. 部署edgecore（邊緣節(jié)點(diǎn)）

４.1 獲取keadm工具

覺得文章有用就打賞一下文章作者

支付寶掃一掃打賞

微信掃一掃打賞

支付寶掃一掃領(lǐng)取紅包，優(yōu)惠每天領(lǐng)

二維碼1

二維碼2

在線云原生邊緣計(jì)算KubeEdge安裝配置（二）

1. K8S集群部署，可以參考如下博客

2.安裝kubEedge

2.1 編輯kube-proxy使用ipvs代理

2.2 基礎(chǔ)服務(wù)提供，負(fù)載均衡器metallb

測(cè)試通過

3. 對(duì)于kubeEdge概述

3.部署kubeedge cloudcore(主節(jié)點(diǎn))

3.1 獲取keadm工具

3.2 部署cloudcore

至此cloudcore部署完畢

以下是邊緣節(jié)點(diǎn)的部署方式

4. 部署edgecore（邊緣節(jié)點(diǎn)）

４.1 獲取keadm工具

相關(guān)文章

覺得文章有用就打賞一下文章作者

支付寶掃一掃打賞

微信掃一掃打賞

支付寶掃一掃領(lǐng)取紅包，優(yōu)惠每天領(lǐng)

二維碼1

二維碼2

1. K8S集群部署，可以參考如下博客

2.2 基礎(chǔ)服務(wù)提供，負(fù)載均衡器metallb

支付寶掃一掃領(lǐng)取紅包，優(yōu)惠每天領(lǐng)