Technical Stuff

Some times we are facing problem in understanding replicationcontroller in kubernetes.

Today we'll drill down to one step for understanding kubernetes replication controller.

Sample replication controller file:

Overall view

1. Replication controller make insures that number of specified pods running in the cluster OR homogeneous set of pods are always up and available.
2. use a replication controller even if your application requires only a single pod.
3. If there are too many pods, it will kill some. If there are too few, the replication controller will start more.

Syntax of myapp-rc.yaml file

Before going ahead it's very important to understand syntax and terminology of replicationcontroller file.

1. apiVersion,kind,metadata,spec will come in one alignment as green colour in ablove file
2. metadata can contain name,label of replicationcontroller
3. all the specification of replication controller will come under spec section
4. spec mainly contains three information replicas,selector,template
5. number of replicas means number of pods which cluster will contain each time.
6. selector means pod will contain only those containers whose label we have mentioned in spec.template.metadata.labels

      7.template mainly contains two keywords metadata and spec 

Labels on the Replication Controller

The replication controller can itself have labels (.metadata.labels). Typically, you would set these the same as the .spec.template.metadata.labels; if .metadata.labels is not specified then it is defaulted to .spec.template.metadata.labels. However, they are allowed to be different, and the .metadata.labels do not affec the behavior of the replication controller.

Pod Selector

Multiple Replicas

You can specify how many pods should run concurrently by setting .spec.replicas to the number of pods you would like to have running concurrently. The number running at any time may be higher or lower, such as if the replicas was just increased or decreased, or if a pod is gracefully shutdown, and a replacement starts early.

If you do not specify .spec.replicas, then it defaults to 1.

$apt-get update

$apt-get install ssh curl vim

2.  Docker installation

sudo apt-get update

sudo apt-key adv --keyserver hkp://p80.pool.sks-keyservers.net:80 --recv-keys 58118E89F3A912897C070ADBF76221572C52609D

sudo apt-get update

sudo apt-get purge lxc-docker

sudo apt-cache policy docker-engine

sudo apt-get install docker-engine -y

3.ssh-keygen -t rsa

4.Copy the ssh id_rsa key locally ///  optional for passwordless auth

$ ssh-copy-id -i /root/.ssh/id_rsa.pub 127.0.0.1

5. In case this command fails please use this alternative solution in order to add the key //optional

$ cat /root/.ssh/id_rsa.pub >> /root/.ssh/authorized_keys

6.Validate the password-less ssh-login // optional passwordless auth test

$ ssh root@127.0.0.1
root@virtual-machine:~$ exit
logout
Connection to 127.0.0.1 closed

7. Get the Kubernetes release bundle from the official github repository

$ wget https://github.com/GoogleCloudPlatform/kubernetes/releases/download/v1.0.1/kubernetes.tar.gz

8. Untar the Kubernetes bundle in the same directory

$ tar -xvf kubernetes.tar.gz

We will build the binaries of Kubernetes code specifically for ubuntu cluster

9. Execute the following shell script

$ cd kubernetes/cluster/ubuntu  

 $ ./build.sh

10 .Configure the cluster information by editing only the following parameters of the filecluster/ubuntu/config-default.sh in the editor of your choice.

$ cd
$ vi kubernetes/cluster/ubuntu/config-default.sh
export nodes="root@127.0.0.1  root@172.16.1.8"
export roles="ai i"
export NUM_MINIONS=${NUM_MINIONS:-2}

export FLANNEL_NET=172.16.0.0/16

NOTE :- for multiple node need to edit export roles=("ai" "i" "i")   /// Script BUG

/* Only update the above mentioned information in the file, rest of the configuration will remain as it is. The first variable nodes defines all the cluster nodes, in our case same machine will be configured as master and node so it contains only one entry.The role below “ai” specifies that same machine will act as master, “a” stands for master and “i” stands for node.

Now, we will be starting the cluster with the following command; */

11. startup cluster configuration

$ cd kubernetes/cluster
$ KUBERNETES_PROVIDER=ubuntu ./kube-up.sh

Cluster validation succeeded
Done, listing cluster services:

Note :-  if validations

Kubernetes master is running at http://127.0.0.1:8080

12 .We will add kubectl binary to path in order to manage the Kubernetes cluster.

$ export PATH=$PATH:~/kubernetes/cluster/ubuntu/binaries //  to get command access

Now, we will validate if the K8s cluster created above is properly configured;

$ kubectl get nodes

NAME        LABELS                             STATUS
127.0.0.1   kubernetes.io/hostname=127.0.0.1   Ready

13. The Kubernetes UI can be accessed at the following address https://127.0.0.1:8080/ui . Suppose in case it is not accessible please run the following services and try again

// for UI configuration //optional

$ kubectl create -f addons/kube-ui/kube-ui-rc.yaml --namespace=kube-system
$ kubectl create -f addons/kube-ui/kube-ui-svc.yaml --namespace=kube-system

14 TO container monitor uI // Cadvisor

http://127.0.0.1:4194/containers/ // it can be ip of master instend of 127.0.0.1

15 To check other resource through UI

http://127.0.0.1:8080/ // and selected options

TO check kubernetes pod list

#kubernetes get pod

To get kubernetes services

#kubernetes get services

Run a container nginx (name my-nginx)  and running on 80 port(without replica)

#kubectl run-container my-nginx --image=nginx --port=80

TO create a container with yaml

#kubectl create -f replication.yaml

Replication/Scaling

Kubernetes container replication:-

1.pull image nginx at docker

#docker pull nginx

3.create a container with replication

# kubectl run my-nginxtest --image=nginx --replicas=2 --port=80

4.to check replication node

#kubectl get rc

5. to check replication node in POD

#kubectl get pod

6. create a load balancer for my-nginx container  

#kubectl expose rc my-nginxtest --port=80 --create-external-load-balancer

7.to check container full detials including IP

#kubectl describe pod my-nginxtest-h91jo

8. to stop replication service

Scaling replication

scaling our nginx replication

#kubectl scale --current-replicas=2 --replicas=3 replication controllers my-nginxtest

Note:if anybody facing some issue then plz contact me using comment section and provide suggestion