Kubernetes namespaces divide cluster resources for organization and isolation. They act as virtual clusters within a single physical cluster, separating resources like services and pods, enabling per-namespace access control via RBAC, and allowing resource quotas to prevent overuse. Common use cases include separating dev/staging/production, isolating by team, or managing multi-tenant setups. To use, create a namespace with kubectl create namespace, specify it in resource definitions, or set the current context with kubectl config set-context. Not all resources are namespace-scoped—nodes and persistent volumes remain cluster-wide. While not always needed, namespaces simplify management in shared environments.

A namespace in Kubernetes is a way to divide cluster resources between multiple users or projects. Think of it like a virtual cluster within a single physical cluster — it helps organize and isolate resources without needing separate clusters for each team or app.

Why Namespaces Matter

If you're managing a shared Kubernetes cluster, especially with multiple teams or apps, namespaces are super useful. They let you keep things separated — so the dev team's stuff doesn't accidentally mess with production apps, for example.

Here’s how they help:

• Resource separation – Each namespace can have its own set of services, pods, and configs.
• Access control – You can apply different RBAC rules per namespace.
• Resource limits – Set quotas so one team doesn’t take up all the CPU or memory.

It’s not full isolation (like separate clusters would be), but for most internal setups, it’s more than enough.

When to Use Namespaces

You don’t always need them — if you’re running a small setup or just messing around, they might add unnecessary complexity. But in real-world environments, they come in handy.

Common use cases:

• Separate dev, staging, and production environments
• Isolate workloads by team or department
• Manage multi-tenant setups (like SaaS platforms)

Also, some tools and services assume you're using namespaces. For example, service meshes like Istio often rely on them to manage traffic between services.

How to Work with Namespaces

Using namespaces is pretty straightforward. You can create one like this:

kubectl create namespace dev

Then, when deploying resources, just specify the namespace:

metadata:
  name: my-app
  namespace: dev

Or, if you want to set your current context to a specific namespace:

kubectl config set-context --current --namespace=dev

Some quick tips:

• The default namespace is where everything goes if you don’t specify one
• System-critical components live in kube-system
• There’s also kube-public for resources visible to all

Just remember, not all resources are namespace-scoped. Things like nodes or persistent volumes exist cluster-wide.

That’s the basics of Kubernetes namespaces. Not too complicated once you get the hang of it, but definitely something easy to overlook at first.

The above is the detailed content of What is a namespace in Kubernetes?. For more information, please follow other related articles on the PHP Chinese website!