Namespaces in Kubernetes

We may virtually divide our Kubernetes cluster according on the needs of the team, the applications, and the environment by using namespace.

The entire Kubernetes cluster can be divided into numerous virtual sections using namespace. We may divide your Kubernetes cluster into different teams, environments, applications, or according to any other particular needs you may have.

Kube-System Namespace

Upon installation, the cluster automatically creates two namespaces called Kube-system and default namespaces. Consequently, the Kubernetes system-related pods like API server, control manager, etcd, scheduler, Kube-proxy, make up the Kube-system namespace if your Kubernetes cluster had been configured using the Kubeadm tool.

Why we need namespace?

In namespace, we build and manage Kubernetes objects. We don’t want a third party to interfere with our application in real time. We therefore desire to have our own private location at this time and at this point, it is not good to use default namespace. For instance, let’s say we’re making the namespaces dev and prod. Resources inside of these two namespaces can thus be individually generated and controlled by doing this. For instance, it doesn’t effect the other when we create two identical resources with the same names in Dev and Prod, respectively.

Some cluster-level resources, like nodes, are shared across namespaces and are thus global throughout the cluster. The network isolation is the additional factor. Consider the scenario where you wish to completely isolate the dev and prod namespaces, which means that no pod in the dev namespace should speak to any pod in the prod namespace. If so, and to achieve this proper network solution need to be installed on the cluster. Namespaces are therefore efficient ways to coordinate amongst many applications, teams, and environments on a broad scale.

kubectl create namespace <namespace-name>

vi ns.yaml

2. Create a namespace called “prod” using Imperatively

3. Create a pod object called nginx with image as nginx in production namespace

4. Display the pods in “prod” namespace

Remember the default namespace is default. So to the set the namespace as default, say for example, if you want to set “prod” namespace as default where you can permanently save the namespace for all subsequent namespaces in the context.

5. Change the default namespace to a namespace called “dev”

a. You can see the namespace by using below command

In above screenshot, you can see no namespace is set

b. You can set the namespace by setting the context

From now on, all the kubectl commands will use “dev” as the namespace in the context. All the pods, deployments and other resources will be created in dev namespace.

6. Now, change back the namespace from dev to “default”

7. Delete namespaces created. Remember when you delete namespace all resources created under that namespace gets deleted

Related posts:

The Bank for International Settlements has taken a look under the hood of the decentralized finance (DeFi) monster in a new working paper and presented a DeFi stack reference model (DSR) to…

The second largest cause of irritation to the entirety of humankind after Joffrey Baratheon from GOT are the relatives who are ready to take every ounce of credit for your success by recounting…

New founders and talent building within Alchemist is the fuel that powers our community. This reality organically emerged out of the chaos of no plan. In less than half a year we’ve built three very…