Kubetail

Kubetail is a real-time logging dashboard for Kubernetes (browser/terminal)

Demo: https://www.kubetail.com/demo

Introduction

Kubetail is a general-purpose logging dashboard for Kubernetes, optimized for tailing logs across multi-container workloads in real-time. With Kubetail, you can view logs from all the containers in a workload (e.g. Deployment or DaemonSet) merged into a single, chronological timeline, delivered to your browser or terminal.

The primary entry point for Kubetail is the kubetail CLI tool, which can launch a local web dashboard on your desktop or stream raw logs directly to your terminal. Behind the scenes, Kubetail uses your cluster's Kubernetes API to fetch logs directly from your cluster, so it works out of the box without needing to forward your logs to an external service first. Kubetail also uses your Kubernetes API to track container lifecycle events in order to keep your log timeline in sync as containers start, stop or get replaced. This makes it easy to follow logs seamlessly as user requests move from one ephemeral container to another across services.

Our goal is to build the most powerful, user-friendly logging platform for Kubernetes and we'd love your input. If you notice a bug or have a suggestion please create a GitHub Issue or send us an email ([email protected])!

Features

• Clean, easy-to-use interface
• View log messages in real-time
• Filter logs by:
  • Workload (e.g. Deployment, CronJob, StatefulSet)
  • Absolute or relative time range
  • Node properties (e.g. availability zone, CPU architecture, node ID)
  • Grep
• Uses your Kubernetes API to retrieve log messages so data never leaves your possession (private by default)
• Web dashboard can be installed on desktop or in cluster
• Switch between multiple clusters (Desktop-only)

Quickstart [Desktop]

Install

You can install the Kubetail CLI tool (kubetail) by downloading the latest release binary or by using a package manager:

# Homebrew
brew install kubetail

# Krew
kubectl krew install kubetail

# Snap
sudo snap install kubetail

# Winget
winget install kubetail

# Chocolatey
choco install kubetail

# Scoop
scoop install kubetail

# MacPorts
sudo port install kubetail

# Ubuntu/Mint (apt)
sudo add-apt-repository ppa:kubetail/kubetail
sudo apt update && sudo apt install kubetail-cli

# Fedora/CentOS/RHEL/Amazonlinux/Mageia (copr)
dnf copr enable kubetail/kubetail
dnf install kubetail

# SUSE (zypper)
zypper addrepo 'https://download.opensuse.org/repositories/home:/kubetail/$releasever/' kubetail
zypper refresh && zypper install kubetail-cli

# Alpine (apk)
apk add kubetail --repository=https://dl-cdn.alpinelinux.org/alpine/edge/testing

# Arch Linux (AUR)
yay -S --noconfirm kubetail-cli

# Gentoo (GURU)
ACCEPT_KEYWORDS="~$(portageq envvar ARCH)" emerge dev-util/kubetail

# Nix (Flake)
nix profile add github:kubetail-org/kubetail-nix

# Nix (Classic)
nix-env -i -f https://github.com/kubetail-org/kubetail-nix/archive/refs/heads/main.tar.gz

# asdf
asdf plugin add kubetail https://github.com/kubetail-org/asdf-kubetail.git
asdf install kubetail latest

You can also use our install script:

curl -sS https://www.kubetail.com/install.sh | bash

Usage

Start the web dashboard using the serve command:

kubetail serve

This will open the dashboard at http://localhost:7500. You can also view logs in your terminal using the logs command:

kubetail logs -f deployments/my-app

See the documentation for a full list of commands. Have fun tailing your logs!

Quickstart [Cluster]

Install

To install Kubetail in your cluster using Helm, first add our chart repository, then install the "kubetail" chart:

helm repo add kubetail https://kubetail-org.github.io/helm-charts/
helm install kubetail kubetail/kubetail --namespace kubetail-system --create-namespace

For more information on how to configure the helm chart, see the chart's values.yaml file.

If you prefer, you can also install Kubetail in your cluster by applying the latest manifest file:

kubectl create namespace kubetail-system
kubectl apply -f https://github.com/kubetail-org/helm-charts/releases/latest/download/kubetail-clusterauth.yaml

Usage

To access the web dashboard you can expose it as an ingress using the chart or you can use your usual access methods such as kubectl port-forward:

kubectl port-forward -n kubetail-system svc/kubetail-dashboard 8080:8080

Visit http://localhost:8080. Have fun tailing your logs!

Quickstart [Docker]

To run the Kubetail CLI tool (kubetail) inside a docker container you can use the kubetail-cli docker image.

To use the image with docker run you need to mount your local .kube/config file into a kubetail-cli container then use normal CLI commands:

docker run --rm -it \
  -v ~/.kube/config:/root/.kube/config:ro \
  kubetail/kubetail-cli logs \
  --kubeconfig /root/.kube/config \
  -f \
  deployments/my-app

You can also use the sh entrypoint to start an interactive session:

docker run --rm -it \
  -v ~/.kube/config:/root/.kube/config:ro \
  --entrypoint sh \
  kubetail/kubetail-cli

If you prefer, you can also run kubetail in a container using these options:

• Docker Compose
• Kubernetes Pod

You can also embed the binary in your custom docker builds like this:

FROM ubuntu:24.04

# Copy kubetail binary from the kubetail-cli image
COPY --from=kubetail/kubetail-cli /usr/local/bin/kubetail /usr/local/bin/kubetail

# Your application code and setup here
# ...

Docker Registries

Our docker images are available at these registry endpoints:

* Docker Hub - docker.io/kubetail/kubetail-*
* GitHub - ghcr.io/kubetail-org/kubetail-*

The Helm chart defaults to pulling images from ghcr.io to avoid Docker Hub rate limits.

Minikube

As of minikube v1.36.0, you can install Kubetail as an addon:

minikube addons enable kubetail

Once the Kubetail pods are running in the cluster you can access the dashboard via a service:

minikube service -n kubetail-system kubetail-dashboard

Documentation

Visit our full documentation at https://www.kubetail.com

Roadmap and Status

This is our high-level plan for the Kubetail project, in order:

	Step	Status
1	Real-time container logs	✅
2	Real-time search and polished user experience	🛠️
3	Real-time system logs (e.g. systemd, k8s events)	🔲
4	Basic customizability (e.g. colors, time formats)	🔲
5	Message parsing and metrics	🔲
6	Historic data (e.g. log archives, metrics timeseries)	🔲
7	Kubetail API and developer-facing client libraries	🔲
N	World Peace	🔲

And here are some additional details:

Real-time container logs

Users can view the container logs from the pods currently running inside their clusters quickly and easily using a web dashboard. Users can view container logs organized by workloads and follow log messages as ephemeral containers get created and deleted. They can also narrow their viewing window by timestamp and filter logs by source properties such as region, zone and node.

Real-time search and polished user experience

Users can install Kubetail easily on their desktops and in their clusters. By default, Kubetail will use only the Kubernetes API to fetch basic data such as running workloads and container logs. If a user wants more advanced functionality they can install Kubetail custom services in their cluster (i.e. "Kubetail Cluster API" and "Kubetail Cluster Agent", collectively known as the "Kubetail API") and gain access to features such as log search, log file sizes and last event timestamps. The entire experience of installing, upgrading and uninstalling the Kubetail API is very polished and users are able to view their logs with equally powerful tools in the browser and the terminal using the Kubetail web dashboard and CLI tool.

Real-time system logs

Users who install the Kubetail API gain immediate access to their node-level logs (e.g. systemd) and cluster-level logs (e.g. kubernetes events) and view them in an integrated interface that shows their container logs in context with other system information such as CPU utilization, memory usage and disk space. System logs are viewable in real-time, in the same merged timeline with other logs. Users can filter system logs by timestamp and source properties.

Basic customizability

Users can fully customize their Kubetail experience when using the web dashboard and CLI tool by modifying their user settings. The user settings are modifiable by hand using a config file or via the dashboard UI. The experience is very polished and works seamlessly across upgrades that may add/remove/modify user settings. Users have the option to sync their settings across multiple devices.

Development

Repository Structure

This monorepo contains the following modules:

• Kubetail CLI (modules/cli)
• Kubetail Cluster API (modules/cluster-api)
• Kubetail Cluster Agent (crates/cluster_agent)
• Kubetail Dashboard (modules/dashboard)

It also contains the source code for the Kubetail Dashboard's frontend:

• Dashboard UI (dashboard-ui)

Setting up the Development Environment

Dependencies

• Go
• pnpm
• Tilt
• Tilt-compatible cluster (e.g. minikube, kind, docker-desktop)
• ctlptl (optional)

Next steps

1. Create a Tilt-compatible Kubernetes Dev Cluster:

# minikube
ctlptl apply -f hack/ctlptl/minikube.yaml

# kind
ctlptl apply -f hack/ctlptl/kind.yaml

# docker-desktop
ctlptl apply -f hack/ctlptl/docker-desktop.yaml

2. Start the dev environment:

tilt up

3. Start the Dashboard server:

cd modules/dashboard
go run cmd/main.go -c hack/config.yaml

4. Run the Dashboard UI locally:

cd dashboard-ui
pnpm install
pnpm dev

Now access the dashboard at http://localhost:5173.

Optimize Development Environment for Rust (Optional)

By default, the dev environment compiles "release" builds of the Rust components when you run run tilt up. If you want to iterate more quickly, you can have Tilt compile the rust code locally using "debug" builds instead.

Dependencies

• rustup
• protobuf

Next steps

First, install the Rust target required for your architecture:

# x86_64
rustup target add x86_64-unknown-linux-musl

# aarch64
rustup target add aarch64-unknown-linux-musl

Next, install the tools required by Rust cross compiler:

# macOS (Homebrew)
brew install FiloSottile/musl-cross/musl-cross

# Linux (Ubuntu)
apt-get install musl-tools

On macOS, add this to your ~/.cargo/config.toml file:

[target.x86_64-unknown-linux-musl]
linker = "x86_64-linux-musl-gcc"

[target.aarch64-unknown-linux-musl]
linker = "aarch64-linux-musl-gcc"

Finally, to use the local compiler, just run Tilt using using the KUBETAIL_DEV_RUST_LOCAL env flag:

KUBETAIL_DEV_RUST_LOCAL=true tilt up

Build

CLI Tool

To build the Kubetail CLI tool executable (kubetail), run the following command:

make

When the build process finishes you can find the executable in the local bin/ directory.

Dashboard

To build a docker image for a production deployment of the Kubetail Dashboard server, run the following command:

docker build -f build/package/Dockerfile.dashboard -t kubetail-dashboard:latest .

Cluster API

To build a docker image for a production deployment of the Kubetail Cluster API server, run the following command:

docker build -f build/package/Dockerfile.cluster-api -t kubetail-cluster-api:latest .

Cluster Agent

To build a docker image for a production deployment of the Kubetail Cluster Agent, run the following command:

docker build -f build/package/Dockerfile.cluster-agent -t kubetail-cluster-agent:latest .

Get Involved

We're building the most user-friendly, cost-effective, and secure logging platform for Kubernetes and we'd love your contributions! Here's how you can help:

• UI/UX design
• React frontend development
• Reporting issues and suggesting features

Reach us at [email protected], or join our Discord server or Slack channel.