This article records the usage information of common components of Kubernetes. Please correct any errors.

Pod

Pod is a container running environment that contains a group of containers and enables each container to share namespace, network, cgroups, data volumes and other resources with each other. Pod is just a container running environment, not a thread. It will always exist unless it is deleted. Two forms of pod.

• A pod with only one container: This pod can be considered as a wrapper of the container
• pod of a group of containers: it aggregates a group of tightly coupled containers that need to share resources

pod controller

Manage pod components

• Deployment
• ReplicaSet
• ReplicationController
• StatefulSet
• DaemonSet

Pod templates

The description of creating a pod tells the controller how to create a pod and what state you want the pod to run in k8s.

apiVersion: batch/v1 
kind: Job
metadata:
  name: hello
spec:
  template:
    # This is the pod template
    spec:
      containers:
      - name: hello
        image: busybox
        command: ['sh', '-c', 'echo "Hello, Kubernetes!" && sleep 3600']
      restartPolicy: OnFailure
    # The pod template ends here

Pod life cycle

Start - > pending
If at least one master node is started successfully - > running
If any container fails to start - > failed, otherwise it is successful - > succeeded

The pod will only be called once and will remain in the storage node unless the pod is stopped or removed. If the node hangs up, it will be deleted after exceeding the time limit. Pod creates a new one every time. With the same lifecycle is volume.

Init container

https://kubernetes.io/docs/concepts/workloads/pods/init-containers/

pod cluster distribution constraints (topology constraints)

https://kubernetes.io/docs/concepts/workloads/pods/pod-topology-spread-constraints/

ReplicaSet

A ReplicaSet's purpose is to maintain a stable set of replica Pods running at any given time. As such, it is often used to guarantee the availability of a specified number of identical Pods.

This means that a certain number of replica sets of pods are always running to achieve high availability of pods. y generally, ReplicaSet is rarely used directly.

• Replicas: number of replicas expected by pod
• selector: Label used to filter the target Pod
• Template: the Pod template used to create a new Pod when the number of copies of the Pod is less than the expected number

apiVersion: apps/v1
kind: ReplicaSet
metadata:
  name: frontend
  labels:
    app: guestbook
    tier: frontend
spec:
  # modify replicas according to your case
  replicas: 3
  selector:
    matchLabels:
      tier: frontend
  template:
    metadata:
      labels:
        tier: frontend
    spec:
      containers:
      - name: php-redis
        image: gcr.io/google_samples/gb-frontend:v3

Deployments

A Deployment provides declarative updates for Pods and ReplicaSets.
You describe a desired state in a Deployment, and the Deployment Controller changes the actual state to the desired state at a controlled rate. You can define Deployments to create new ReplicaSets, or to remove existing Deployments and adopt all their resources with new Deployments.

Create a deployment

apiVersion: apps/v1
kind: Deployment
metadata:
  name: nginx-deployment
  labels:
    app: nginx
spec:
  replicas: 3
  selector:
    matchLabels:
      app: nginx
  template:
    metadata:
      labels:
        app: nginx
    spec:
      containers:
      - name: nginx
        image: nginx:1.14.2
        ports:
        - containerPort: 80

Compared with ReplicaSet, the biggest upgrade is that you can not only create replicas, but also view the status of deployment.

kubectl get deployments
kubectl rollout status deployment/nginx-deployment

update deployment

kubectl set image deployment/nginx-deployment nginx=nginx:1.16.1 --record

Rolling Back a Deployment

Sometimes there may be a problem with the deployed pod. At this time, you need to roll back to the previous version.
View version information

kubectl rollout history deployment.v1.apps/nginx-deployment  #View version information
kubectl describe deployment    # View current deployment information
kubectl rollout history deployment.v1.apps/nginx-deployment --revision=2   #View version details
kubectl rollout undo deployment.v1.apps/nginx-deployment  #Fallback to previous version
kubectl rollout undo deployment.v1.apps/nginx-deployment --to-revision=2    #Fallback to the specified version

! [insert picture description here]( https://img-blog.csdnimg.cn/img_convert/51ff93353c7c893ac8baccf91b536753.png

scale

kubectl scale deployment.v1.apps/nginx-deployment --replicas=10
kubectl autoscale deployment.v1.apps/nginx-deployment --min=10 --max=15 --cpu-percent=80

Labels and Selectors

https://kubernetes.io/docs/concepts/overview/working-with-objects/labels/

Namespace

kubectl get pods -n kube-system
kubectl get ns
kubectl get pods --all-namespaces #Find pod s under all namespaces

Create namespace myns namespace yaml

apiVersion: v1
kind: Namespace
metadata:
  name: myns