If you want to deploy large applications on Docker, the first thing is to solve the network problem, and there are a series of complex problems, including cluster management, load balancing, etc. Container Choreography deployment tools are needed to address these issues.
Container Choreography deployment tool
Container management tools can be the basis of the complete container management, but the container is not only simple application deployment, you can use the container to complete the enterprise deployment, the application of more complex when the need for many application systems for deployment, you need more sophisticated tools to complete the container run the choreography, this is the container arrangement deployment tool.
Container Choreographer deployment tools are:
(1) Docker three swordsmen: Docker machine, Docker compose and Docker swarm
(2) the mesos + marathon
(3) the kubernetes
Introduction to Kubernetes
Kubernetes, called K8S for short because of the eight letters between K and S, is the open source version of Borg system within Google and the mainstream container choreography deployment tool at present.
K8s is a portable and extensible open source platform used to manage containerized applications on multiple hosts. It makes the deployment of containerized applications simple and efficient, and provides a mechanism for application deployment, planning, updating, and maintenance. It can also automate the deployment and scaling of applications.
As shown in the figure above, THE K8S cluster is the Master and Worker mode.
Kube-apiserver, Kube-Controller-Mansger, kube-Scheduler, etcd, etc. They are used to receive client requests, control the cluster and resource object control center, and monitor container health, resource scheduling, and resource object data storage.
There are kubelet, Kube-proxy and Docker on Worker node, which are used to manage Pod and Pod container respectively and report node resource information to Master regularly, realize transparent proxy and load balancing of Service, and run container.
Iii. Main functions of K8S
(1) Automatic packing
Automatically deploy an application container based on the resource configuration requirements of the container on the application running environment
(2) Self-repair
When the container fails, it is restarted
When deployed nodes go down, containers are redeployed and rescheduled
If a container fails to pass the monitoring check, the container is shut down. Services are not provided until the container runs properly.
(3) Horizontal expansion
You can expand or reduce the capacity of an application container by using simple commands, user interface (UI), or based on the CPU usage
(4) Service discovery
There is no need to use additional service discovery mechanisms, k8S has its own service discovery and load balancing
(5) Rolling update
You can update the applications running in the application container at a time or in batches based on the application changes
(6) Version rollback
You can roll back historical versions of applications running in the application container based on the application deployment
(7) Key and configuration management
Without the need to rebuild the image, you can deploy and update keys and application configurations, similar to hot deployment
(8) Storage orchestration
Automatic storage system mounting and application, especially for stateful applications to implement data persistence
The storage system can be a local directory, network storage, or public cloud storage service.
Four, quick installation
A, preparation,
1. Disable SELINUX so that the container can read the host file system. Restart takes effect
Vim /etc/selinux/config Change selinux =enforcing to selinux =disabledCopy the codeNote: All machines are subject to change
2. Close the swap partition
Kubernetes cluster deployment must disable the swap partition, otherwise an error will be reported, restart the effect (to do this, I crashed two VMS)
vim /etc/fstab
Copy the codeComment out the line with swap
Run the free -m command to check whether the server is disabled. 0 indicates that the server is disabled
Note: All machines are subject to change
3. Add bridge filtering
(1) Add bridge filtering and address forwarding
vim /etc/sysctl.d/k8s.conf
net.bridge.bridge-ng-call-ip6tables = 1
net.bridge.bridge-ng-call-iptables = 1
net.ipv4.ip_forword = 1
Copy the code(2) Run the following command to load the BR_netfilter module
modprobe br_netfilter
Copy the code(3) Run the following command to check whether the file is loaded
lsmod | grep br_netfilte
Copy the code(4) Use the following command to load the bridge filtering file
sysctl -p /etc/sysctl.d/k8s.conf
Copy the codeNote: All machines are subject to change
4. Enable IPVS
(1) Install ipset and ipvsadm
yum -y install ipset ipvsadm
Copy the code(2) Add modules to be loaded
vim /etc/sysconfig/modules/ipvs.modules
#! /bin/bash
modprobe -- ip_vs
modprobe -- ip_vs_rr
modprobe -- ip_vs_wrr
modprobe -- ip_vs_sh
modprobe -- nf_conntrack_ipv4
Copy the code(3) Run the above script
sh /etc/sysconfig/modules/ipvs.modules
Copy the codeCheck whether it is loaded:
lsmod | grep ip_vs
Copy the codeNote: All machines are subject to change
5. Install Docker
— > docker installation
After installation, you need to change the Docker configuration file
vim /etc/docker/daemon.json
#Add the following line configuration
{ "exec-opts":["native.cgroupdriver=systemd"]}
Copy the codeNote: All machines are subject to change
Install the K8S cluster
1. Components need to be installed
Kubeadm kubelet kubectl Initialize cluster, manage cluster, etc. Used to receive API-server instructions, manage cluster command line management tool for POD life cycle
2. Set the Aliyun YUM source
vim /etc/yum.repos.d/k8s.repo
[kubernetes]
name=Kubernetes
baseurl=https://mirrors.aliyun.com/kubernetes/yum/repos/kubernetes-el7-x86_64
enabled=1
gpgcheck=1
repo_gpgcheck=1
gpgkey=https://mirrors.aliyun.com/kubernetes/yum/doc/yum-key.gpg
https://mirrors.aliyun.com/kubernetes/yum/doc/rpm-package-key.gpg
Copy the codeCheck out the latest version of Kubeadm
yum list | grep kubeadm
Copy the codeNote: All machines are subject to change
3. Install components
The installed version is 1.16.0. Change the version as required
Yum -y install kubeadm-1.16.0-0 kubelet-1.16.0-0 kubectl-1.16.0-0Copy the codeNote: All machines must be installed
4. Configure Kubelet
vim /etc/sysconfig/kubelet
KUBELET_EXTRA_ARGS="--cgroup-driver=systemd"
Copy the codeConfigure automatic startup after startup:
systemctl enable kubelet
Copy the codeNote: All machines are subject to change
5. View the image to be downloaded
The image required on the Master machine can be viewed with the following command:
kubeadm config images list
Copy the code
You can download it using the following script:
#! /bin/bashImg_list = 'gotok8s/kube - apiserver: v1.16.0 gotok8s/kube - controller - manager: v1.16.0 gotok8s/kube - the scheduler: v1.16.0 Gotok8s /kube-proxy:v1.16.0 GOtok8S/Pause :3.1 GoTok8s/ETCD :3.3.15-0 Gotok8S/CoreDNS :1.6.2 '
#Pull the mirror
for img in ${img_list}
do
docker pull $img
done
#Use the Docker tag to re-mark
for img in ${img_list}
do
docker tag $img k8s.gcr.io${img:7}
done
#Delete unnecessary mirrors
for img in ${img_list}
do
docker rmi $img
done
Copy the codeBecause k8S.gcr. IO warehouse needs scientific Internet to download, so GOtoK8S warehouse is used, and then re-marking is carried out
The following image is required on the Worker node
K8s. GCR. IO/kube - proxy: v1.16.0 k8s. GCR. IO/pause: 3.1Copy the codeDocker image import/export command
#exportDocker save-o kube-proxy.tar k8s.gcr. IO /kube-proxy:v1.16.0#The import
docker load -i kube-proxy.tar
Copy the code6. Initialize the cluster
Run the following command on the master node, including specifying the kubernetes version and the current host IP address
Kubeadm init - kubernetes - version = v1.16.0 - apiserver - advertise - address = 192.168.197.100Copy the codeCluster initialization and certificate creation are performed.
If the following information is displayed, the initialization is successful, including copying configuration files and adding nodes to the cluster.
Your Kubernetes control-plane has initialized successfully! To start using your cluster, you need to run the following as a regular user: mkdir -p $HOME/.kube sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config sudo chown $(id -u):$(id -g) $HOME/.kube/config You should now deploy a pod network to the cluster. Run "kubectl apply -f [podnetwork].yaml" with one of the options listed at: https://kubernetes.io/docs/concepts/cluster-administration/addons/ Then you can join any number of worker nodes by running the following on each as root: Kubeadm join 192.168.197.1001:6443 --token cmx34b.juizw6tp9ptlgg9i \ --discovery-token-ca-cert-hash sha256:77661093886eb76ffa7595e200a4ce2a5b20f02c164f4946956dff16d941a1e7Copy the code(1) create “.kube” folder in root directory
mkdir -p $HOME/.kube
sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
sudo chown $(id -u):$(id -g) $HOME/.kube/config
Copy the code(2) Run the following command to install weave
kubectl apply -f "https://cloud.weave.works/k8s/net? k8s-version=$(kubectl version | base64 | tr -d '\n')"Copy the code(3) Add worker nodes to the cluster
Run the following command on the workder node
Kubeadm join 192.168.197.1001:6443 --token cmx34b.juizw6tp9ptlgg9i \ --discovery-token-ca-cert-hash sha256:77661093886eb76ffa7595e200a4ce2a5b20f02c164f4946956dff16d941a1e7Copy the code7. Verify that the cluster is available
#Obtaining a Cluster Node
kubectl get nodes
Copy the code
#View the cluster health status
kubectl cluster-info
Copy the code
There is also a way to install: binary file installation K8S cluster, more trouble, free to see how to do.
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
I’m Liusy, a programmer who likes to work out.
Welcome to pay attention to wechat public number [Liusy01], exchange Java technology and fitness, get more dry goods, get Java advanced dry goods, become a Java god together.
Here we go. Keep an eye on it.