k8s-utils

A collection of Kubernetes command-line utilities built with Python and Typer.

Installation

This project uses uv for package management. Make sure you have uv and uvx installed.

Just testing it out:

# Install dependencies
uvx k8s-helpers --help

Installation

uv tool install k8s-helpers

This installs the k8s-helpers command and the shorthand k8sh

Commands

podcount

Count non-daemonset pods per node in the current Kubernetes context.

# Use current context
k8sh podcount --help

# Specify a different context
k8sh podcount --context my-context

# Show nodegroup and zone labels, and sort by nodegroup
k8sh podcount \
--node-labels "eks.amazonaws.com/nodegroup" \
--node-labels "topology.kubernetes.io/zone" \
--sort "eks.amazonaws.com/nodegroup" \
--show-taints

This command:

• Retrieves all pods and nodes in the cluster
• Excludes pods managed by DaemonSets
• Displays a table showing pod count per node
• Shows total non-DaemonSet pod count

delpods

Delete pods matching certain statuses across namespaces.

# Delete pods with specific statuses across all namespaces
k8sh delpods --all-namespaces --status ImagePullBackOff --status CrashLoopBackOff

# Preview what would be deleted (dry run)
k8sh delpods -A --status CrashLoopBackOff --dry-run

# Delete in specific namespace without confirmation
k8sh delpods -n default --status Error --yes

# Use specific context
k8sh delpods -c my-cluster -A --status ImagePullBackOff

Available arguments:

• --all-namespaces, -A: Search pods in all namespaces
• --namespace, -n: Specify a namespace (default: current context namespace)
• --status: Pod status to match (can be specified multiple times)
• --dry-run: Show what would be deleted without deleting
• --yes, -y: Skip confirmation prompt
• --context, -c: Kubernetes context to use

resources

Analyze CPU and memory consumption per node (excluding DaemonSet pods).

# Show all nodes
k8sh resources

# Analyze single node (faster for large clusters)
k8sh resources --node my-node-name

# Include pod breakdown
k8sh resources -p

# Sort by memory requests
k8sh resources --sort mem-req

Shows requests, limits, allocatable capacity, and actual usage (if metrics-server is available).

node

Analyze resource waste for pods on a specific node.

# Analyze waste on a node
k8sh node my-node-name

# Sort by CPU waste
k8sh node my-node-name --sort cpu-waste

Shows actual usage vs requests, with color-coded waste percentages (green=efficient, red=wasteful).