[cloud native learning notes] kubernetes practice command

---

Preface

I introduced some information about k8s Basic theoretical knowledge of cluster , This time, we will deepen our understanding of theory in combination with practice .

---

One 、Kubernetes install

Given the limitations of the learning environment , This installation Minikube.minikube It's local Kubernetes, Focus on making Kubernetes Easy to learn and develop .k8s Commonly used in linux In the system , Because it is only used for learning and familiarity k8s command , This time, it will be based on the system used , install windows Version of Minikube. The official tutorial

1. download .exe Installation package Installation .
2. Add the installation directory to the system environment variable Path. The default installation directory is C:\Program Files\Kubernetes\Minikube
3. Need to be advanced Install well docker, The same is windows Version and launch .
4. stay cmd Run command in minikube start To initialize and start minikube.
5. Can be used later minikube stop To stop the cluster

Minikube Will start a virtual machine ,Kubernetes The cluster runs on a virtual machine .minikube with dashboard service , For visual monitoring and viewing minikube situation . In the new cmd Running on the terminal minikube dashboard To activate the function .

Two 、 Cluster interaction

adopt kubectl Command can be root Kubernetes Interaction .kubectl The common format of the command is ：kubectl action resources . The function is to perform the specified action on the specified resource . Execute the run command kubectl You can get all the information about resources and actions . Official documents There are also detailed instructions .

1、 Check the cluster status

see kubectl Is the installation successful , It is generally reflected by the view version ：

kubectl version

You'll see... At the same time client and server Two version numbers . among ,client Corresponding kubectl Version of ,server Corresponding Master Installed on Kubernetes Version of .

View the details of the cluster ：kubectl cluster-info
View cluster node：kubectl get nodes

2、 About deploying applications

adopt kubectl run The command can run the specified image in the cluster . For deploying images , In terms of launching applications , What is actually commonly used is not kubectl run, It is kubectl creat perhaps kubectl apply.apply Use relatively more .

kubectl creat： Create a new resource with configuration , Since the resource name should be unique in the namespace , When repeated , Will report a mistake .
kubectl apply： Apply configuration to resources . If the resource does not exist , Create a new resource . When repeated , If the configuration is modified, the resource will be updated , Otherwise, the resources will not change .

Use here kubectl run Run the application deployment ：

kubectl run kubernetes-bootcamp --image=jocatalin/kubernetes-bootcamp:v1 --port=8080

Check pod state running Success after ：

kubectl get pod -A | grep kubernetes-bootcamp
default                kubernetes-bootcamp                         1/1     Running   0             28m

Applications are deployed in clusters , If the application services are not exposed outside the cluster , So use url You can't access the service directly .
Direct access will prompt to reject the link ：

curl http://localhost:8001/version
curl: (7) Failed to connect to localhost port 8001 after 2203 ms: Connection refused

kubectl proxy Command can create a proxy , Forward communications to a cluster wide private network . The agent can pass control-C End , And no output will be displayed at run time , You can run the agent in other terminal windows .
Revisit ：

curl http://localhost:8001/version
{
    
  "major": "1",
  "minor": "23",
  "gitVersion": "v1.23.1",
  "gitCommit": "86ec240af8cbd1b60bcc4c03c20da9b98005b92e",
  "gitTreeState": "clean",
  "buildDate": "2021-12-16T11:34:54Z",
  "goVersion": "go1.17.5",
  "compiler": "gc",
  "platform": "linux/amd64"
}

Here, you can access our deployed applications through the forwarding provided by the cluster ：

$ curl http://localhost:8001/api/v1/namespaces/default/pods/kubernetes-bootcamp/proxy/
Hello Kubernetes bootcamp! | Running on: kubernetes-bootcamp | v=1

3、 ... and 、 About the characteristics of the cluster

• Automatic load balancing ： Automatically assign requests to each pod To respond to .
• Service scaling ： Update applied pod Copy number
• Service version upgrade ： The image used by the update application is used for upgrading .
  have access to kubectl rollout status deployments/xxx Check out the upgrade process , Is to replace each... One by one pod.
  have access to kubectl rollout undo deployments/xxx Quickly go back to the previous version .
• Common commands for troubleshooting cluster resources are ：
  kubectl get ： List resources
  kubectl describe ： Show details about the resource
  kubectl logs ： Print log
  kubectl exec ： Go to the container and execute the command