logo
search
EXPAND ALL
  • Home

CLI (Recommended)

Pixie's CLI is the fastest and easiest way to install and manage your Pixie installation.

Prerequisites

  • Review Pixie's requirements to make sure that your Kubernetes cluster is supported.

  • Determine if you already have Operator Lifecycle Manager (OLM) deployed to your cluster, possibly to the default olm namespace. Pixie uses the Kubernetes Operator pattern to manage its Vizier, which handles data collection and query execution (see the Architecture diagram). The OLM is used to install, update and manage the Vizier Operator.

1. Install the Pixie CLI

You can install the CLI in one of four ways:

Using the install script (easiest)

# Copy and run command to install the Pixie CLI.
bash -c "$(curl -fsSL https://withpixie.ai/install.sh)"

Using Homebrew

brew install pixie

Directly downloading the binary

Assets

Download the correct binary for your operating system and make it executable:

# Download the latest Pixie linux binary.
curl -o px https://storage.googleapis.com/pixie-dev-public/cli/latest/cli_linux_amd64
# (Optional) Check the signature matches.
curl -o px_checksum https://storage.googleapis.com/pixie-dev-public/cli/latest/cli_linux_amd64.sha256
cat px_checksum
sha256sum px
# Make it executable.
chmod +x px
# (Optional) Move it to another location that has executables.
mv px /usr/local/bin

Using Docker

alias px="docker run -i --rm -v ${HOME}/.pixie:/root/.pixie pixielabs/px"

You may validate the signature on the docker image by verifying the image with cosign.

Using Debian package

Download the assets:

Install the Pixie package:

# Install Pixie .deb package.
dpkg -i pixie-px.x86_64.deb

Using RPM

Download the assets:

Install the Pixie package:

# Install Pixie .rpm package.
rpm -i pixie-px.x86_64.rpm

2. Deploy Pixie

# List Pixie deployment options.
px deploy --help
# Deploy the Pixie Platform in your K8s cluster (No OLM present on cluster).
export PX_CLOUD_ADDR=getcosmic.ai
px deploy
# Deploy the Pixie Platform in your K8s cluster (OLM already exists on cluster).
export PX_CLOUD_ADDR=getcosmic.ai
px deploy --deploy_olm=false
# Deploy Pixie with a specific memory limit (2Gi is the default, 1Gi is the minimum recommended)
export PX_CLOUD_ADDR=getcosmic.ai
px deploy --pem_memory_limit=1Gi

Pixie will deploy pods to the pl, px-operator, and olm(if deploying the OLM) namespaces.

More Deploy Options

For more deploy options that you can specify to configure Pixie, refer to our deploy options.

This site uses cookies to provide you with a better user experience. By using Pixie, you consent to our use of cookies.