Last modified November 27, 2024

Usage data collection in 'kubectl gs'

Since kubectl-gs version 4.3.0, Giant Swarm collects anonymous usage data about kubectl gs command execution. This allows us to understand better which commands are used how widely and how frequently, and to better prioritize our development efforts.

You are free to opt out of this data collection. However, we appreciate your support and ensure that we only collect non-identifyable information and keep data collection to a minimum.

Details we collect

Whenever a command is executed (this does not include calling a command with -h or --help), we collect the following details:

• Name of the command executed
• kubectl-gs version
• Name of the operating system
• Processor architecture (e.g. amd64, arm64)
• The library and version used to submit the data (e. g. telemetrydeck-go/0.0.1)
• A generated user identifier hash (more details below)
• A random UUID as a session identifier. Currently this is unique per command execution. We may change this in the future to correlate several data submissions to a single session.

User identifier hash

We submit a hash that is supposed to be distinct to each user using kubectl-gs, to allow counting users. It is generated based on the following data and SHA256-hashed before submission:

• Operating system
• Processor architecture
• Host name
• OS user ID
• OS group ID
• OS user name
• The MAC adresses of network interfaces

For more details, see the source code.

Disabling usage data collection

To disable the data collection, please set the following environment variable to any value:

KUBECTL_GS_TELEMETRY_OPTOUT

For example, in your Bash profile, you might add the following:

export KUBECTL_GS_TELEMETRY_OPTOUT=1

Data storage

Data is submitted to TelemetryDeck, who use Microsoft Azure servers in Amsterdam, Netherlands.

For more information, see the TelemetryDeck privacy FAQ.