Lifecycle CLI telemetry
Understand the pseudonymous command-usage event sent by lfc, its destination, and how to opt out.
By default, the lfc CLI attempts to send one usage event to the configured
Lifecycle deployment. The event helps deployment owners understand command
reliability and adoption.
What the CLI sends
Each event can contain:
- a random identifier for the CLI installation
- the command path, such as
builds list - names of flags passed on the command line
- duration, success or error outcome, and process exit code
- error class, HTTP status, and stable Lifecycle error code when available
- CLI version, Node.js runtime version, operating-system platform, and architecture
The telemetry event does not contain flag values, positional arguments, command output, repository content, or user identity.
The install identifier is pseudonymous. It lets a deployment count recurring CLI installations. It is not a name or account ID.
Where it goes
The CLI posts to its own configured Lifecycle deployment:
It does not send this usage event directly to a shared external analytics service. A telemetry failure does not change the command result.
Opt out
Set LFC_TELEMETRY_DISABLED to a non-empty value other than 0 or false:
Set it in the environment where lfc runs and in CI. This keeps the setting for
that caller.
Retention and access
Each deployment controls retention. The CLI contract does not specify retention. Get this information from your Lifecycle operator:
- how long the deployment keeps CLI events
- which administrators can see aggregate statistics
- if the deployment sends events to a different observability system
- how your privacy policy handles the installation identifier
For application workload labels sent through your platform’s Datadog pipeline, see Datadog application telemetry.