Overview of kubectl

The kubectl command line tool lets you control Kubernetes clusters. For configuration, kubectl looks for a file named config in the $HOME/.kube directory. You can specify other kubeconfig files by setting the KUBECONFIG environment variable or by setting the --kubeconfig flag.

This overview covers kubectl syntax, describes the command operations, and provides common examples. For details about each command, including all the supported flags and subcommands, see the kubectl reference documentation. For installation instructions see installing kubectl.

Syntax

Use the following syntax to run kubectl commands from your terminal window:

kubectl [command] [TYPE] [NAME] [flags]

where command, TYPE, NAME, and flags are:

• command: Specifies the operation that you want to perform on one or more resources, for example create, get, describe, delete.

• TYPE: Specifies the resource type. Resource types are case-insensitive and you can specify the singular, plural, or abbreviated forms. For example, the following commands produce the same output:

  kubectl get pod pod1
  kubectl get pods pod1
  kubectl get po pod1
  

• NAME: Specifies the name of the resource. Names are case-sensitive. If the name is omitted, details for all resources are displayed, for example kubectl get pods.

  When performing an operation on multiple resources, you can specify each resource by type and name or specify one or more files:

  • To specify resources by type and name:

    • To group resources if they are all the same type: TYPE1 name1 name2 name<#>.
      Example: kubectl get pod example-pod1 example-pod2

    • To specify multiple resource types individually: TYPE1/name1 TYPE1/name2 TYPE2/name3 TYPE<#>/name<#>.
      Example: kubectl get pod/example-pod1 replicationcontroller/example-rc1

  • To specify resources with one or more files: -f file1 -f file2 -f file<#>

    • Use YAML rather than JSON since YAML tends to be more user-friendly, especially for configuration files.
      Example: kubectl get -f ./pod.yaml

• flags: Specifies optional flags. For example, you can use the -s or --server flags to specify the address and port of the Kubernetes API server.
  

Caution: Flags that you specify from the command line override default values and any corresponding environment variables.

If you need help, run kubectl help from the terminal window.

In-cluster authentication and namespace overrides

By default kubectl will first determine if it is running within a pod, and thus in a cluster. It starts by checking for the KUBERNETES_SERVICE_HOST and KUBERNETES_SERVICE_PORT environment variables and the existence of a service account token file at /var/run/secrets/kubernetes.io/serviceaccount/token. If all three are found in-cluster authentication is assumed.

To maintain backwards compatibility, if the POD_NAMESPACE environment variable is set during in-cluster authentication it will override the default namespace from the service account token. Any manifests or tools relying on namespace defaulting will be affected by this.

POD_NAMESPACE environment variable

If the POD_NAMESPACE environment variable is set, cli operations on namespaced resources will default to the variable value. For example, if the variable is set to seattle, kubectl get pods would return pods in the seattle namespace. This is because pods are a namespaced resource, and no namespace was provided in the command. Review the output of kubectl api-resources to determine if a resource is namespaced.

Explicit use of --namespace <value> overrides this behavior.

How kubectl handles ServiceAccount tokens

If:

• there is Kubernetes service account token file mounted at /var/run/secrets/kubernetes.io/serviceaccount/token, and
• the KUBERNETES_SERVICE_HOST environment variable is set, and
• the KUBERNETES_SERVICE_PORT environment variable is set, and
• you don't explicitly specify a namespace on the kubectl command line then kubectl assumes it is running in your cluster. The kubectl tool looks up the namespace of that ServiceAccount (this is the same as the namespace of the Pod) and acts against that namespace. This is different from what happens outside of a cluster; when kubectl runs outside a cluster and you don't specify a namespace, the kubectl command acts against the default namespace.

Operations

The following table includes short descriptions and the general syntax for all of the kubectl operations:

To learn more about command operations, see the kubectl reference documentation.

Resource types

The following table includes a list of all the supported resource types and their abbreviated aliases.

(This output can be retrieved from kubectl api-resources, and was accurate as of Kubernetes 1.19.1.)

Output options

Use the following sections for information about how you can format or sort the output of certain commands. For details about which commands support the various output options, see the kubectl reference documentation.

Formatting output

The default output format for all kubectl commands is the human readable plain-text format. To output details to your terminal window in a specific format, you can add either the -o or --output flags to a supported kubectl command.

Syntax

kubectl [command] [TYPE] [NAME] -o <output_format>

Depending on the kubectl operation, the following output formats are supported:

Example

In this example, the following command outputs the details for a single pod as a YAML formatted object:

kubectl get pod web-pod-13je7 -o yaml

Remember: See the kubectl reference documentation for details about which output format is supported by each command.

Custom columns

To define custom columns and output only the details that you want into a table, you can use the custom-columns option. You can choose to define the custom columns inline or use a template file: -o custom-columns=<spec> or -o custom-columns-file=<filename>.

Examples

Inline:

kubectl get pods <pod-name> -o custom-columns=NAME:.metadata.name,RSRC:.metadata.resourceVersion

Template file:

kubectl get pods <pod-name> -o custom-columns-file=template.txt

where the template.txt file contains:

NAME          RSRC
metadata.name metadata.resourceVersion

The result of running either command is similar to:

NAME           RSRC
submit-queue   610995

Server-side columns

kubectl supports receiving specific column information from the server about objects. This means that for any given resource, the server will return columns and rows relevant to that resource, for the client to print. This allows for consistent human-readable output across clients used against the same cluster, by having the server encapsulate the details of printing.

This feature is enabled by default. To disable it, add the --server-print=false flag to the kubectl get command.

Examples

To print information about the status of a pod, use a command like the following:

kubectl get pods <pod-name> --server-print=false

The output is similar to:

NAME       AGE
pod-name   1m

Sorting list objects

To output objects to a sorted list in your terminal window, you can add the --sort-by flag to a supported kubectl command. Sort your objects by specifying any numeric or string field with the --sort-by flag. To specify a field, use a jsonpath expression.

Syntax

kubectl [command] [TYPE] [NAME] --sort-by=<jsonpath_exp>

Example

To print a list of pods sorted by name, you run:

kubectl get pods --sort-by=.metadata.name

Examples: Common operations

Use the following set of examples to help you familiarize yourself with running the commonly used kubectl operations:

kubectl apply - Apply or Update a resource from a file or stdin.

# Create a service using the definition in example-service.yaml.
kubectl apply -f example-service.yaml

# Create a replication controller using the definition in example-controller.yaml.
kubectl apply -f example-controller.yaml

# Create the objects that are defined in any .yaml, .yml, or .json file within the <directory> directory.
kubectl apply -f <directory>

kubectl get - List one or more resources.

# List all pods in plain-text output format.
kubectl get pods

# List all pods in plain-text output format and include additional information (such as node name).
kubectl get pods -o wide

# List the replication controller with the specified name in plain-text output format. Tip: You can shorten and replace the 'replicationcontroller' resource type with the alias 'rc'.
kubectl get replicationcontroller <rc-name>

# List all replication controllers and services together in plain-text output format.
kubectl get rc,services

# List all daemon sets in plain-text output format.
kubectl get ds

# List all pods running on node server01
kubectl get pods --field-selector=spec.nodeName=server01

kubectl describe - Display detailed state of one or more resources, including the uninitialized ones by default.

# Display the details of the node with name <node-name>.
kubectl describe nodes <node-name>

# Display the details of the pod with name <pod-name>.
kubectl describe pods/<pod-name>

# Display the details of all the pods that are managed by the replication controller named <rc-name>.
# Remember: Any pods that are created by the replication controller get prefixed with the name of the replication controller.
kubectl describe pods <rc-name>

# Describe all pods
kubectl describe pods

Note: The kubectl get command is usually used for retrieving one or more resources of the same resource type. It features a rich set of flags that allows you to customize the output format using the -o or --output flag, for example. You can specify the -w or --watch flag to start watching updates to a particular object. The kubectl describe command is more focused on describing the many related aspects of a specified resource. It may invoke several API calls to the API server to build a view for the user. For example, the kubectl describe node command retrieves not only the information about the node, but also a summary of the pods running on it, the events generated for the node etc.

kubectl delete - Delete resources either from a file, stdin, or specifying label selectors, names, resource selectors, or resources.

# Delete a pod using the type and name specified in the pod.yaml file.
kubectl delete -f pod.yaml

# Delete all the pods and services that have the label '<label-key>=<label-value>'.
kubectl delete pods,services -l <label-key>=<label-value>

# Delete all pods, including uninitialized ones.
kubectl delete pods --all

kubectl exec - Execute a command against a container in a pod.

# Get output from running 'date' from pod <pod-name>. By default, output is from the first container.
kubectl exec <pod-name> -- date

# Get output from running 'date' in container <container-name> of pod <pod-name>.
kubectl exec <pod-name> -c <container-name> -- date

# Get an interactive TTY and run /bin/bash from pod <pod-name>. By default, output is from the first container.
kubectl exec -ti <pod-name> -- /bin/bash

kubectl logs - Print the logs for a container in a pod.

# Return a snapshot of the logs from pod <pod-name>.
kubectl logs <pod-name>

# Start streaming the logs from pod <pod-name>. This is similar to the 'tail -f' Linux command.
kubectl logs -f <pod-name>

kubectl diff - View a diff of the proposed updates to a cluster.

# Diff resources included in "pod.json".
kubectl diff -f pod.json

# Diff file read from stdin.
cat service.yaml | kubectl diff -f -

Examples: Creating and using plugins

Use the following set of examples to help you familiarize yourself with writing and using kubectl plugins:

# create a simple plugin in any language and name the resulting executable file
# so that it begins with the prefix "kubectl-"
cat ./kubectl-hello

#!/bin/sh

# this plugin prints the words "hello world"
echo "hello world"

With a plugin written, let's make it executable:

chmod a+x ./kubectl-hello

# and move it to a location in our PATH
sudo mv ./kubectl-hello /usr/local/bin
sudo chown root:root /usr/local/bin

# You have now created and "installed" a kubectl plugin.
# You can begin using this plugin by invoking it from kubectl as if it were a regular command
kubectl hello

hello world

# You can "uninstall" a plugin, by removing it from the folder in your
# $PATH where you placed it
sudo rm /usr/local/bin/kubectl-hello

In order to view all of the plugins that are available to kubectl, use the kubectl plugin list subcommand:

kubectl plugin list

The output is similar to:

The following kubectl-compatible plugins are available:

/usr/local/bin/kubectl-hello
/usr/local/bin/kubectl-foo
/usr/local/bin/kubectl-bar

kubectl plugin list also warns you about plugins that are not executable, or that are shadowed by other plugins; for example:

sudo chmod -x /usr/local/bin/kubectl-foo # remove execute permission
kubectl plugin list

The following kubectl-compatible plugins are available:

/usr/local/bin/kubectl-hello
/usr/local/bin/kubectl-foo
  - warning: /usr/local/bin/kubectl-foo identified as a plugin, but it is not executable
/usr/local/bin/kubectl-bar

error: one plugin warning was found

You can think of plugins as a means to build more complex functionality on top of the existing kubectl commands:

cat ./kubectl-whoami

The next few examples assume that you already made kubectl-whoami have the following contents:

#!/bin/bash

# this plugin makes use of the `kubectl config` command in order to output
# information about the current user, based on the currently selected context
kubectl config view --template='{{ range .contexts }}{{ if eq .name "'$(kubectl config current-context)'" }}Current user: {{ printf "%s\n" .context.user }}{{ end }}{{ end }}'

Running the above command gives you an output containing the user for the current context in your KUBECONFIG file:

# make the file executable
sudo chmod +x ./kubectl-whoami

# and move it into your PATH
sudo mv ./kubectl-whoami /usr/local/bin

kubectl whoami
Current user: plugins-user

What's next

• Start using the kubectl commands.

• To find out more about plugins, take a look at the example cli plugin.