 Kubernetes is an open-source “container management tool” which automates

container deployment, container scaling & load balancing.

 It schedules, run and manages isolated containers which are running on
“virtual/physical [on-prem]/cloud machines”.
 All top cloud providers support Kubernetes.

History: -
 Google developed an internal system called ‘borg’ [latex named as omega]
to deploy and manage thousands of google application & service on their
clusters.
 In 2014, google introduced Kubernetes an open-source plat form written in
“Go-lang” & later donated to CNCF. [Cloud native computing &
foundation]
Cloud based k8s service: -
 Amazon EKS = Elastic Kubernetes Service.
 AKS = Azure Kubernetes Service
 GKE = Google Kubernetes Service
K8s installation tool: -
 Minikube = It is used to run the K8s locally.
 Kubeadm = It is a single-node k8s cluster that you can use locally in your
own development and it cannot be used in production as it is a 1 node
machine.
Limitations od Docker Swarm: -
Containers cannot communicate with each other.
Auto Scaling & Load balancing was not possible.
Containers had to be managed carefully.
Features of k8s: -
a. Orchestration = Clustering of any number of containers running in different
networks.

Clustering
b. Auto-Scaling = Vertical and Horizontal is possible

c. Load Balancing
d. Platform Independent (virtual/cloud/on-prem)
e. Fault Tolerance (Node / Pod failure)
f. Roll back (Going back to previous versions)
g. Health & Monitoring of containers

Difference between K8s and Docker Swarn: -

Features Kubernetes Docker Swarm
Installation and cluster Complicated and time Fast and Easy
config consuming
Supports K8s can work with almost Work with docker only
 all container types like
Docker, Container
GUI GUI is available GUI is not available
Data Volume Data shared with Data can be shared with
containers in same pod any other container
Updates & Rollback Process scheduling to Progressive updates of
maintain service while service health monitoring
updating through out the update
Auto-Scaling Support vertical and Not support autoscaling
Horizontal auto-scaling
Logging and Inbuilt tool present for Used 3rd party tools like
Monitoring monitoring Splunk

Types of Clusters: -
a. Single Master Single Node

b. Single Master Number of Nodes

c. Number of Masters & Number of Nodes

Kubernetes Architecture

Role of K8s Master: -

 K8s cluster contains containers running / VM instances / Cloud Instances.
 K8s designates one or more of these as master and all other as a worker.
 The master is new going to run set of K8s processes. These processes will
ensure smooth functioning of cluster. These processes are called “Control
Plane”.
 Can be multi-master for high availability.
 Master runs control plane to run cluster smoothly.

Kube-APIServer: -
 This api-server interacts directly with user.
[i.e.., we apply (.yml) / (json) manifest to kube-apiserver]
 Kube-APIServer is a front-end of control-plane.
 This Kube-APIServer is meant to scale automatically as per load.

Etcd: -
 Contains details of POD.
 Stores metadata and status of cluster.
 Etcd is consistent and high-available store [key-value store].
 Source of touch for cluster state [info about state of cluster].
etcd features: -
a. Fully Replicated
b. Secure
c. Fast

Kube-Scheduler: -
 When users make request for the creation & management of PODs, Kube-
Scheduler is going to take action on the request.
 A scheduler watches for newly created Pods that have a no worker node
assigned for every pod that the scheduler discovers, the scheduler becomes
responsible for finding best node for that pod to run on.
 Scheduler gets the information for hardware configuration from
configuration files and schedules the pods on nodes accordingly.
 Handles POD creation & management.
 Kube-Scheduler match / assign any node to create and run pods.

Controller-Manager: -
 It always makes sure that [ actual state = desired state]
Two possible choices for controller manager
a. If K8s on cloud, then it will be cloud-controller-manager.
b. If K8s on non-cloud, then it will be kube-controller-manager.
 Components on master that runs controller
a. Node-Controller = For checking the cloud provider to determine if a node
has been detected in the cloud after it stops responding.
b. Router-Controller = Responsible for setting up network routes on your
cloud.
c. Service-Controller = Responsible for load balancers on your cloud against
services of type load balancer.
d. Volume-Controller = For creating, attaching and maintaining volumes and
interacting with the cloud providers to orchestrate volume.
Kubelet: -
 Agent running on the node.
 Listens to Kubernetes master [Ex: - Pod Creation request]
 Use port 10255
 Send successful/failed reports to master

Container Engine: -
 Works with kubelet
 Pulling Images
 Start / Stop Containers
 Exposing containers on ports specified in manifest.

Kube – Proxy: -
 Assign IP address to each pod.
 It is required to assign IP Address to pods [dynamic]
 Kube-Proxy runs on each node & this make sure that each pod will get its
own unique IP Address.

Pods: -
 Consists of one or more containers with tightly coupled.
 Smallest unit in K8s and in K8s, the control unit is POD, not Container.
 POD is a group of one or more containers that are deployed together on the
same host.
  A cluster is a group of nodes.
 One POD will contain one Container.
 Cannot start container without a POD.
 K8s will only know about POD’s.
 K8s does not know about Containers.
 POD runs on node, which is controlled by Master.
 A cluster will have at least one master and worker node.

POD Limitations: -
 Autoscaling is not possible
 POD Crashes.
Important Commands: -
 kubectl = Single Cloud
 kubeadn = On-Prem
 kubefed = Federated [Mixing of (cloud + on-prem)]