ReplicaSet, how it manages Pods in k8s cluster.

In the previous post we have discussed about namespace, how namespace is used to manage resources in kubernetes cluster.
In this post, we are going to discuss about replicaset.

ReplicaSet

ReplicaSet is a process that run multiple instances of Pods. It constantly monitors the status of Pods and if any one fails or terminates then it restores by creating new instance of Pod and by deleting old one.

Creating ReplicaSet using manifest file (yaml)

~ vim ReplicaSet.yaml

apiVersion: apps/v1
kind: ReplicaSet  
metadata:
  name: replica-set
spec:
  replicas: 3
  selector:
    matchLabels:
      app: backend
  template:
    metadata:
      labels:
        app: backend
    spec:
      containers:
      - name: app-container
        image: nginx

Lets apply,

~ kubectl apply -f ReplicaSet.yaml

  replicaset.apps/replica-set created

View ReplicaSet

Similar to namespace and pods, we can get all the replicaset with command:

~ kubectl get replicaset

or

~ kubectl get rs

Output

NAME          DESIRED   CURRENT   READY   AGE

replica-set   3         3         3       4m31s

In above manifest file, we have mentioned replicas:3 so that it runs 3 instances of Pods with label app:backend.
Lets see,

~ kubectl get pods --show-labels

Output

NAME              READY STATUS   RESTARTS AGE    LABELS

replica-set-85d7k 1/1   Running  0        8m22s  app=backend
replica-set-nm4dr 1/1   Running  0        8m22s  app=backend
replica-set-q2thz 1/1   Running  0        8m22s  app=backend

How Pods are associated with ReplicaSet and How ReplicaSet manage that Pods ?

As we have seen in above manifest file, We have Pods with label app:backend and in ReplicaSet there is option matchLabels with app:backend. It means that pods are associated with replicaset with the help of one of kubernetes feature known as label. The pods having label app:bakend are only associated with the replicaset replica-set that we have created using manifest file. As per our requirement, we can easily scale, upgrade pods with the help of replicaset.

Scaling application

An application can be scaled up and down depending on the situations in two ways.

Method 1: By updating value of replicas in the configuration file and apply the changes.

Lets update the value of replicas to 5 in configuraton file.

Apply above changes,

~ kubectl apply -f ReplicaSet.yaml

  replicaset.apps/replica-set configured

Lets check the pods, we will see

NAME              READY STATUS   RESTARTS  AGE   LABELS

replica-set-85d7k 1/1   Running  0         115m  app=backend
replica-set-nm4dr 1/1   Running  0         115m  app=backend
replica-set-q2thz 1/1   Running  0         115m  app=backend
replica-set-stdmp 1/1   Running  0         13s   app=backend
replica-set-t9zmt 1/1   Running  0         13s   app=backend

Similarly we can scale down by assigning the lower value to the replicas in the configuration file.

Method 2: By using CLI

We can scale up or down our application by using kubectl command. Lets scale down the application with CLI kubectl scale --replicas=2 replicaset <replicaset-name>.

~ kubectl scale --replicas=2 replicaset replica-set

Lets check the pods again,

NAME              READY STATUS   RESTARTS  AGE   LABELS

replica-set-85d7k 1/1   Running  0         121m  app=backend
replica-set-nm4dr 1/1   Running  0         121m  app=backend

Edit ReplicaSet

We can edit replicaset with command:

~ kubectl edit replicaset <replicaset-name>

Delete ReplicaSet

We can delete replicaset with command:

~ kubectl delete replicaset <replicaset-name>

We have successfully managed, scaled our application using replicaset.
In next, we will be discussing on deployment object.