Using Advanced kubectl Techniques
1. Using JSONPath Queries
| JSONPath Expression | Description | Example Output |
|---|---|---|
| $.items[*].metadata.name | Retrieves names of all items in the list | ["pod1" "pod2" "pod3"] |
| $.items[0].spec.containers[*].name | Gets names of containers from the first item | ["nginx" "redis"] |
| $.items[*].status.phase | Retrieves status phase of all items | ["Running" "Pending" "Succeeded"] |
| $.items[*].metadata.labels.app | Extracts app label from metadata of all items | ["frontend" "backend" "database"] |
Example: Filtering Pods by Label
kubectl get pods -o jsonpath='{.items[*].metadata.name}' --selector=app=frontend
Example: Getting Container Image Names
kubectl get pods -o jsonpath='{.items[*].spec.containers[*].image}'
Note: JSONPath expressions must be enclosed in single quotes to avoid shell interpretation.
2. Using Custom Columns
| Field | Description | Example |
|---|---|---|
| .metadata.name | Name of the resource | my-pod |
| .spec.containers[0].image | First container's image | nginx:latest |
| .status.phase | Current phase of pod | Running |
| .metadata.creationTimestamp | Time resource was created | 2025-04-01T12:34:56Z |
Example: Custom Columns with kubectl get
kubectl get pods -o custom-columns=NAME:.metadata.name,IMAGE:.spec.containers[0].image,STATUS:.status.phase
Note: Custom columns support JSONPath expressions for flexible field selection.
3. Formatting Output
| Format | Description | Usage |
|---|---|---|
| json | JSON format output | kubectl get pods -o json |
| yaml | YAML format output | kubectl get pods -o yaml |
| name | Names only | kubectl get pods -o name |
| wide | Additional fields | kubectl get pods -o wide |
| custom-columns | Custom column layout | kubectl get pods -o custom-columns=NAME:.metadata.name,STATUS:.status.phase |
Example: Custom Columns Output
kubectl get pods -o custom-columns=NAME:.metadata.name,STATUS:.status.phase,NAMESPACE:.metadata.namespace</code>
</pre>
</div>
</div>
<div class="example">
<h2 id="example-json-output">Example: JSON Output<a href="#example-json-output" class="toc__headingLink" aria-label="Link to this section"><svg width="16" height="16" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round"><path d="M10 13a5 5 0 0 0 7.54.54l3-3a5 5 0 0 0-7.07-7.07l-1.72 1.71"></path><path d="M14 11a5 5 0 0 0-7.54-.54l-3 3a5 5 0 0 0 7.07 7.07l1.71-1.71"></path></svg></a></h2>
<div class="codeBlock" data-language="bash">
<pre><code>kubectl get pod my-pod -o json</code>
</pre>
</div>
</div>
<h2 id="4-using-field-selectors">4. Using Field Selectors<a href="#4-using-field-selectors" class="toc__headingLink" aria-label="Link to this section"><svg width="16" height="16" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round"><path d="M10 13a5 5 0 0 0 7.54.54l3-3a5 5 0 0 0-7.07-7.07l-1.72 1.71"></path><path d="M14 11a5 5 0 0 0-7.54-.54l-3 3a5 5 0 0 0 7.07 7.07l1.71-1.71"></path></svg></a></h2>
<table>
<tbody>
<tr>
<th>Field Selector</th>
<th>Description</th>
<th>Examples</th>
</tr>
<tr>
<td class="property">metadata.name</td>
<td>Selects resources by name</td>
<td><code>name=my-pod</code></td>
</tr>
<tr>
<td class="property">metadata.namespace</td>
<td>Selects resources by namespace</td>
<td><code>namespace=default</code></td>
</tr>
<tr>
<td class="property">spec.nodeName</td>
<td>Selects pods by the node they are scheduled on</td>
<td><code>spec.nodeName=node-1</code></td>
</tr>
<tr>
<td class="property">status.phase</td>
<td>Selects pods by their current phase</td>
<td><code>status.phase=Running</code></td>
</tr>
<tr>
<td class="property">status.podIP</td>
<td>Selects pods by their IP address</td>
<td><code>status.podIP=10.244.0.5</code></td>
</tr>
</tbody>
</table>
<div class="example">
<h2 id="example-filtering-pods-by-status-phase">Example: Filtering Pods by Status Phase<a href="#example-filtering-pods-by-status-phase" class="toc__headingLink" aria-label="Link to this section"><svg width="16" height="16" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round"><path d="M10 13a5 5 0 0 0 7.54.54l3-3a5 5 0 0 0-7.07-7.07l-1.72 1.71"></path><path d="M14 11a5 5 0 0 0-7.54-.54l-3 3a5 5 0 0 0 7.07 7.07l1.71-1.71"></path></svg></a></h2>
<div class="codeBlock" data-language="bash">
<pre><code>kubectl get pods --field-selector=status.phase=Running
Note: Field selectors can be combined using comma separation, e.g.,
--field-selector=metadata.namespace=default,status.phase=Running.
5. Using Label Selectors
| Selector Type | Description | Examples |
|---|---|---|
| Equality-based Selectors | Match labels with exact key-value pairs | environment=production, app=nginx |
| Set-based Selectors | Match labels using set operations (in, notIn, exists, doesNotExist) | environment in (production, staging), version notIn (v1.0, v1.1) |
| Multiple Label Selectors | Combine multiple selectors with comma (AND logic) | environment=production,app=nginx |
Example: Label Selectors in kubectl
kubectl get pods -l environment=production
kubectl get pods -l 'environment in (production, staging)'
kubectl get pods -l 'version notIn (v1.0, v1.1)'
kubectl get pods -l environment=production,app=nginx
Note: Label selectors are used in various kubectl commands like
get, delete, describe, and rollout.
Warning: Ensure labels are consistently applied across resources to avoid unexpected selection behavior.
6. Performing Bulk Operations
| Operation | Description | Flags | Notes |
|---|---|---|---|
| kubectl apply | Applies configuration from files or directories | --all |
Applies to all resources in the directory |
| kubectl patch | Updates resources with JSON patches | --type, --patch |
Useful for bulk updates to labels or annotations |
| kubectl edit | Edits resources in the default editor | --output |
Can be used in scripts for bulk editing |
7. Using kubectl patch
The kubectl patch command allows you to update resources in your Kubernetes cluster by applying changes directly to the resource's configuration. It supports different patch strategies and is useful for making targeted updates without replacing the entire resource.
| Option | Values | Description |
|---|---|---|
| --type | strategic, merge, json |
Specifies the patch strategy to use. Default is strategic. |
| --patch | <json-patch> |
The JSON patch or merge patch to apply. Required for json and merge strategies. |
| --local | true, false |
If set to true, patches a local file instead of applying to the cluster. |
Example: Patching a Deployment's Replicas
kubectl patch deployment my-deployment -p '{"spec":{"replicas":5}}'Example: Patching with Strategic Merge
kubectl patch deployment my-deployment --type strategic -p '{"spec":{"template":{"spec":{"containers":[{"name":"app","image":"nginx:1.25"}]}}}}'
Note: The
--type strategic is the default and supports more complex structures like lists. For simple key-value updates, merge or json can be used.
Warning: Patching resources can cause unintended side effects if not done carefully. Always verify the patch before applying it in production.
8. Creating Resources from stdin
Use kubectl create or kubectl apply with the -f - flag to create Kubernetes resources from stdin. This is useful for piping resource definitions or applying configurations dynamically.
| Command | Description | Example |
|---|---|---|
| kubectl create -f - | Creates resources from stdin | |
| kubectl apply -f - | Applies resources from stdin, updating existing ones | cat config.yaml | kubectl apply -f - |
Example: Creating a Pod from stdin
echo 'apiVersion: v1
kind: Pod
metadata:
name: example-pod
spec:
containers:
- name: nginx
image: nginx:latest
ports:
- containerPort: 80' | kubectl create -f -
Note: Ensure the input is valid YAML or JSON for Kubernetes resources.
9. Using kubectl diff
The kubectl diff command allows you to compare the current state of a resource with its desired state as defined in a configuration file. This is particularly useful for identifying changes before applying them.
| Option | Values | Description |
|---|---|---|
| kubectl diff | -f <filename> |
Compare the current state of a resource with the desired state from a file |
| kubectl diff | --namespace <namespace> |
Specify the namespace to compare resources in |
| kubectl diff | --output <format> |
Specify the output format (e.g., json, yaml) |
Note: The
kubectl diff command requires the kubectl plugin to be installed and configured properly.
Warning: Ensure that the configuration file used with
kubectl diff is up-to-date to avoid misleading comparisons.
10. Using kubectl wait
The kubectl wait command is a powerful utility that allows you to wait for a Kubernetes resource to reach a specified condition. It's particularly useful in automation scripts and CI/CD pipelines where you need to ensure that resources are in a particular state before proceeding.
| Option | Values | Description |
|---|---|---|
| --for | condition=Ready, jsonpath={.status.phase}=Running |
Specifies the condition to wait for. Can be a condition name or a JSONPath expression. |
| --timeout | 30s, 5m |
Timeout for the wait command. Defaults to 0 (no timeout). |
| --output | json, yaml |
Output format for the resource once it reaches the desired state. |
Example: Waiting for a Deployment to be Available
kubectl wait --for=condition=Available deployment/my-deployment --timeout=2m
Note: The
--for=condition=Ready option is commonly used for pods and other resources that support readiness conditions.
Warning: If the timeout is exceeded,
kubectl wait will return a non-zero exit code. This can be used in scripts to handle timeouts gracefully.
11. Creating Dry Runs
| Flag | Description | Values |
|---|---|---|
| --dry-run | Simulates the action without actually performing it | client, server |
| -o yaml | Output format as YAML | yaml, json |
Note:
--dry-run=client is faster and does not require server interaction, while --dry-run=server validates against the API server.
Warning: Dry runs do not execute the actual command, so they cannot be used to test runtime behavior or side effects.
12. Using kubectl explain
The kubectl explain command provides detailed information about Kubernetes resources, their fields, and their structure. It's particularly useful for understanding the schema of resources without having to consult external documentation.
| Resource | Field | Description | Example |
|---|---|---|---|
| apiVersion | v1 |
API version of the resource | apiVersion: v1 |
| kind | Pod |
Type of Kubernetes resource | kind: Pod |
| metadata.name | my-pod |
Name of the resource | |
Note: The
kubectl explain command can be used with any Kubernetes resource, including custom resources defined by CRDs.
Warning: If a resource is not installed or is not available in the current cluster,
kubectl explain will return an error.