Kubernetes/Install Master and nodes
This example is based on Ubuntu 16 LTS
Updates
- Updated version of Kubernetes binary to 1.13.10-00 to align with AWS version. The commands output haven't been reflected, but all worked.
Install binaries
#Docker gpg key curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo apt-key add - #Docker repository sudo add-apt-repository "deb [arch=amd64] https://download.docker.com/linux/ubuntu $(lsb_release -cs) stable" #Kubernetes gpg key curl -s https://packages.cloud.google.com/apt/doc/apt-key.gpg | sudo apt-key add - #Kubernetes repository cat << EOF | sudo tee /etc/apt/sources.list.d/kubernetes.list deb https://apt.kubernetes.io/ kubernetes-xenial main EOF # Install software sudo apt-get update sudo apt-get install -y docker-ce=18.06.1~ce~3-0~ubuntu kubelet=1.13.10-00 kubeadm=1.13.10-00 kubectl=1.13.10-00 # Set packages at the current versions so they won't autoupdate sudo apt-mark hold docker-ce kubelet kubeadm kubectl # Enable iptables bridge call, then enable immediately echo "net.bridge.bridge-nf-call-iptables=1" | sudo tee -a /etc/sysctl.conf sudo sysctl -p
Initialize a cluster
Run only on the master to initialize the cluster
sudo kubeadm init --pod-network-cidr=10.100.0.0/16 I0705 06:23:54.675905 24293 version.go:237] remote version is much newer: v1.15.0; falling back to: stable-1.13 [init] Using Kubernetes version: v1.13.7 [preflight] Running pre-flight checks [preflight] Pulling images required for setting up a Kubernetes cluster [preflight] This might take a minute or two, depending on the speed of your internet connection [preflight] You can also perform this action in beforehand using 'kubeadm config images pull' [kubelet-start] Writing kubelet environment file with flags to file "/var/lib/kubelet/kubeadm-flags.env" [kubelet-start] Writing kubelet configuration to file "/var/lib/kubelet/config.yaml" [kubelet-start] Activating the kubelet service [certs] Using certificateDir folder "/etc/kubernetes/pki" [certs] Generating "front-proxy-ca" certificate and key [certs] Generating "front-proxy-client" certificate and key [certs] Generating "etcd/ca" certificate and key [certs] Generating "etcd/peer" certificate and key [certs] etcd/peer serving cert is signed for DNS names [kubemaster.acme.com localhost] and IPs [172.31.115.255 127.0.0.1 ::1] [certs] Generating "etcd/healthcheck-client" certificate and key [certs] Generating "apiserver-etcd-client" certificate and key [certs] Generating "etcd/server" certificate and key [certs] etcd/server serving cert is signed for DNS names [kubemaster.acme.com localhost] and IPs [172.31.115.255 127.0.0.1 ::1] [certs] Generating "ca" certificate and key [certs] Generating "apiserver" certificate and key [certs] apiserver serving cert is signed for DNS names [kubemaster.acme.com kubernetes kubernetes.default kubernetes.default.svc kubernetes.default.svc.cluster.local] and IPs [10.96.0.1 172.31.115.255] [certs] Generating "apiserver-kubelet-client" certificate and key [certs] Generating "sa" key and public key [kubeconfig] Using kubeconfig folder "/etc/kubernetes" [kubeconfig] Writing "admin.conf" kubeconfig file [kubeconfig] Writing "kubelet.conf" kubeconfig file [kubeconfig] Writing "controller-manager.conf" kubeconfig file [kubeconfig] Writing "scheduler.conf" kubeconfig file [control-plane] Using manifest folder "/etc/kubernetes/manifests" [control-plane] Creating static Pod manifest for "kube-apiserver" [control-plane] Creating static Pod manifest for "kube-controller-manager" [control-plane] Creating static Pod manifest for "kube-scheduler" [etcd] Creating static Pod manifest for local etcd in "/etc/kubernetes/manifests" [wait-control-plane] Waiting for the kubelet to boot up the control plane as static Pods from directory "/etc/kubernetes/manifests". This can take up to 4m0s [apiclient] All control plane components are healthy after 20.002150 seconds [uploadconfig] storing the configuration used in ConfigMap "kubeadm-config" in the "kube-system" Namespace [kubelet] Creating a ConfigMap "kubelet-config-1.13" in namespace kube-system with the configuration for the kubelets in the cluster [patchnode] Uploading the CRI Socket information "/var/run/dockershim.sock" to the Node API object "kubemaster.acme.com" as an annotation [mark-control-plane] Marking the node kubemaster.acme.com as control-plane by adding the label "node-role.kubernetes.io/master=''" [mark-control-plane] Marking the node kubemaster.acme.com as control-plane by adding the taints [node-role.kubernetes.io/master:NoSchedule] [bootstrap-token] Using token: xkcoul.0i2m*******ockj [bootstrap-token] Configuring bootstrap tokens, cluster-info ConfigMap, RBAC Roles [bootstraptoken] configured RBAC rules to allow Node Bootstrap tokens to post CSRs in order for nodes to get long term certificate credentials [bootstraptoken] configured RBAC rules to allow the csrapprover controller automatically approve CSRs from a Node Bootstrap Token [bootstraptoken] configured RBAC rules to allow certificate rotation for all node client certificates in the cluster [bootstraptoken] creating the "cluster-info" ConfigMap in the "kube-public" namespace [addons] Applied essential addon: CoreDNS [addons] Applied essential addon: kube-proxy Your Kubernetes master 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/ You can now join any number of machines by running the following on each node as root: kubeadm join 172.31.115.255:6443 --token xkcoul.0i2m*******ockj --discovery-token-ca-cert-hash sha256:808*******6a
Set up local kubeconfig:
mkdir -p $HOME/.kube sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config sudo chown $(id -u):$(id -g) $HOME/.kube/config
Create Flannel CNI network overlay to allow nodes to communicate with each other
kubectl apply -f https://raw.githubusercontent.com/coreos/flannel/bc79dd1505b0c8681ece4de4c0d86c5cd2643275/Documentation/kube-flannel.yml clusterrole.rbac.authorization.k8s.io/flannel created clusterrolebinding.rbac.authorization.k8s.io/flannel created serviceaccount/flannel created configmap/kube-flannel-cfg created daemonset.extensions/kube-flannel-ds-amd64 created daemonset.extensions/kube-flannel-ds-arm64 created daemonset.extensions/kube-flannel-ds-arm created daemonset.extensions/kube-flannel-ds-ppc64le created daemonset.extensions/kube-flannel-ds-s390x created
Join worker nodes
Run on worker nodes - join worker nodes to the master
sudo kubeadm join 172.31.115.255:6443 --token xkcoul.0i2m*******ockj --discovery-token-ca-cert-hash sha256:808*******6a [preflight] Running pre-flight checks [discovery] Trying to connect to API Server "172.31.115.255:6443" [discovery] Created cluster-info discovery client, requesting info from "https://172.31.115.255:6443" [discovery] Requesting info from "https://172.31.115.255:6443" again to validate TLS against the pinned public key [discovery] Cluster info signature and contents are valid and TLS certificate validates against pinned roots, will use API Server "172.31.115.255:6443" [discovery] Successfully established connection with API Server "172.31.115.255:6443" [join] Reading configuration from the cluster... [join] FYI: You can look at this config file with 'kubectl -n kube-system get cm kubeadm-config -oyaml' [kubelet] Downloading configuration for the kubelet from the "kubelet-config-1.13" ConfigMap in the kube-system namespace [kubelet-start] Writing kubelet configuration to file "/var/lib/kubelet/config.yaml" [kubelet-start] Writing kubelet environment file with flags to file "/var/lib/kubelet/kubeadm-flags.env" [kubelet-start] Activating the kubelet service [tlsbootstrap] Waiting for the kubelet to perform the TLS Bootstrap... [patchnode] Uploading the CRI Socket information "/var/run/dockershim.sock" to the Node API object "kubeworker1.acme.com" as an annotation This node has joined the cluster: * Certificate signing request was sent to apiserver and a response was received. * The Kubelet was informed of the new secure connection details. Run 'kubectl get nodes' on the master to see this node join the cluster.
Verify nodes joined the cluster
$ kubectl get nodes NAME STATUS ROLES AGE VERSION kubemaster.acme.com Ready master 16m v1.13.5 kubeworker1.acme.com Ready <none> 2m45s v1.13.5 kubeworker2.acme.com Ready <none> 2m39s v1.13.5
Highly Available Kubernetes Cluster
In the output below you can see out-of-box K8s installation. Note that not all components are duplicated. But even we douplicate components not all can run in active mode. Eg. because Controller Manger and Scheduler are constantly watching for a cluster events only one can be in active mode, the rest would be in standby mode.
#View cluster components kubectl get pods -o custom-columns=POD:metadata.name,NODE:spec.nodeName --sort-by spec.nodeName -n kube-system POD NODE coredns-86c58d9df4-cdl5a kube-master.acme.com coredns-86c58d9df4-csxca kube-master.acme.com etcd-kube-master.acme.com kube-master.acme.com kube-apiserver-kube-master.acme.com kube-master.acme.com kube-controller-manager-kube-master.acme.com kube-master.acme.com kube-scheduler-kube-master.acme.com kube-master.acme.com kube-flannel-ds-amd64-cwd74 kube-master.acme.com kube-proxy-z264w kube-master.acme.com kube-proxy-fxl6f kube-worker-1.acme.com kube-flannel-ds-amd64-c7hva kube-worker-1.acme.com kube-flannel-ds-amd64-c5p9a kube-worker-2.acme.com kube-proxy-jtbwm kube-worker-2.acme.com #View details of components kubectl get endpoints kube-scheduler -n kube-system -o yaml
References
- Creating Highly Available Kubernetes Clusters with kubeadm
- Highly Available Topologies in Kubernetes
- Operating a Highly Available etcd Cluster
Testing cluster end-to-end
kubectl run nginx --image=nginx #deployment test, run eg. nginx kubectl get deployments #access a pod directly using port-forwarding by creating local listener on port 8080 that forwards traffic to the pod's port 80. kubectl port-forward $pod_name 8080:80& #Check a response from the 'nginx' pod directly. Run from a master or where kubectl is installed curl --head http://127.0.0.1:8080 kubectl logs $pod_name #view pod's logs kubectl exec -it nginx -- nginx -v #run a commands directly on a container #Create a service by exposing port 80 of the nginx deployment kubectl expose deployment nginx --port 80 --type NodePort #List the services in your cluster, you can see the service is exposed on port <tt>31839</tt> kubectl get service nginx -o wide NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE SELECTOR nginx NodePort 10.110.225.169 <none> 80:31839/TCP 2m26s run=nginx #Check a response from the service curl -I localhost:$node_port #run on a worker node HTTP/1.1 200 OK Server: nginx/1.17.1 Date: Sun, 07 Jul 2019 07:36:00 GMT Content-Type: text/html Content-Length: 612 Last-Modified: Tue, 25 Jun 2019 12:19:45 GMT Connection: keep-alive ETag: "5d121161-264" Accept-Ranges: bytes #Check information about nodes kubectl get nodes kubectl describe nodes
Upgrade
kubeadm
allows us to upgrade a cluster components in controlled smart way it means the upgrade process will verify what components can be upgraded. Output examples are based on Amazon EKS, but EKS does not allow access for Control Place access, thus cannot see API Server or kube-controller.
#Verify different components versions kubectl version --short #check version for a API client (kubelet) and API server Client Version: v1.15.0 Server Version: v1.12.6-eks-d69f1b kubectl get pods -n kube-system kube-controller-manager-<podUID> -o yaml |grep image: #version of the scheduler and controller manager image: k8s.gcr.io/kube-controller-manager:v1.12.10 image: k8s.gcr.io/kube-controller-manager:v1.12.10 kubectl get pods -n kube-system #kube-controller pod #Set the VERSION and ARCH you want to upgrade to export VERSION=v1.13.8; export ARCH=amd64 #Download Kubernetes binaries curl -sSL https://dl.k8s.io/release/${VERSION}/bin/linux/${ARCH}/kubeadm > kubeadm sudo install -o root -g root -m 0755 ./kubeadm /usr/bin/kubeadm #install kubeadm sudo kubeadm version sudo kubeadm upgrade plan #plan upgrade [preflight] Running pre-flight checks. [upgrade] Making sure the cluster is healthy: [upgrade/config] Making sure the configuration is correct: [upgrade/config] Reading configuration from the cluster... [upgrade/config] FYI: You can look at this config file with 'kubectl -n kube-system get cm kubeadm-config -oyaml' [upgrade] Fetching available versions to upgrade to [upgrade/versions] Cluster version: v1.12.10 [upgrade/versions] kubeadm version: v1.13.8 I0711 07:32:05.071864 17589 version.go:237] remote version is much newer: v1.15.0; falling back to: stable-1.13 [upgrade/versions] Latest stable version: v1.13.8 [upgrade/versions] Latest version in the v1.12 series: v1.12.10 Components that must be upgraded manually after you have upgraded the control plane with 'kubeadm upgrade apply': COMPONENT CURRENT AVAILABLE Kubelet 3 x v1.12.2 v1.13.8 Upgrade to the latest stable version: COMPONENT CURRENT AVAILABLE API Server v1.12.10 v1.13.8 Controller Manager v1.12.10 v1.13.8 Scheduler v1.12.10 v1.13.8 Kube Proxy v1.12.10 v1.13.8 CoreDNS 1.2.2 1.2.6 Etcd 3.2.24 3.2.24 You can now apply the upgrade by executing the following command: kubeadm upgrade apply v1.13.8 #Upgrade kubeadm upgrade apply v1.13.8 #apply upgrade #You can check differences between the old and new manifests if you got it saved before upgrade sudo vi /etc/kubernetes/manifests/kube-controller-manager.yaml #current #Download the latest version of kubelet: curl -sSL https://dl.k8s.io/release/${VERSION}/bin/linux/${ARCH}/kubelet > kubelet sudo install -o root -g root -m 0755 ./kubelet /usr/bin/kubelet #install kublet sudo systemctl restart kubelet.service #restart is required kubectl get nodes -w #check nodes versions #Upgrade version of kubelet on all worker nodes now
Controlling nodes during upgrade
- if a node downtime is <5 minutes a node controller will restart a pod on a node but if it's greater then it will delete the pod
- so upgrading a node, you should check pod running on it, then evict them
kubectl drain <node> --ignore-daemonsets kubectl get nodes #the drained node will be in 'Ready,SchedulingDisabled' status #the node now can be upgraded kubectl uncordon <node> #bring back to participate in a cluster kubectl delete <node> #delete node from a cluster, remember to drain it first
Add new node to a cluster
On master node create connection token
sudo kubeadm token generate #below is a token-name not a secret! 123abc.abcabcabcabc sudo kubeadm token list sudo kubeadm token create 123abc.abcabcabcabc --ttl 2h --print-join-command #Run join command on the new worker-node after you install all required packages: etc. Docker sudo kuebadm join <IP>:6443 --token 123abc.abcabcabcabc --discovery-token-ca-cert-hash sha256:****
References
- kubeadm upgrade K8s docs
- Maintenance on a Node K8s docs