Skip to main content

Configuration Basics

The corridor is configured through two environment files at the root of the smart-corridor package: .env (corridor image versions and the dashboard port) and .env.hub, which defines the units the stack serves and how they behave. Face Matcher keeps its own configuration inside the vendored face-matcher/ folder and in Station — leave that folder untouched and change Face Matcher settings the Face Matcher way. Corridor changes follow the usual Compose cycle: edit the file, then recreate the affected service. This page covers the settings every deployment touches; the long tail is in Advanced Configuration.

Units: what this stack serves​

A unit is a physical corridor or gate lane. Each unit gets an ID, a type, and its cameras:

FOUNDATION_UNITS_0_ID=corridor-a
FOUNDATION_UNITS_0_TYPE=CORRIDOR # or EGATE
FOUNDATION_UNITS_0_CAMERAS=<Face Matcher camera IDs, comma-separated>

The camera IDs come from Face Matcher — register the streams there first (Cameras in Station, or see Add Cameras), then list the resulting IDs on the unit. Every event the Hub publishes carries the unit ID, which is how consumers tell lanes apart.

Watchlists and clearance​

Watchlists live in Face Matcher; the Hub needs to know which of them mean "allowed" so it can raise identification.clearance.green instead of a plain match. Create the watchlists in Station, note their IDs, and list the allowed ones in VPP_ADAPTER_ALLOWED_WATCHLISTS as described in Add People to Watchlists. Matches on any other watchlist raise red clearance events automatically.

Storage​

Decide explicitly whether the stack persists events. STORAGE_ENABLED=false stores nothing at rest (the GDPR-friendly mode); true enables the event store with configurable retention TTLs for events and images, the latter kept in the shared S3 store. The defaults and the full retention model are documented in Event Storage & Retention.

What is not here​

Detection sensitivity, matching threshold, liveness, camera stream parameters and the Station UI are Face Matcher settings: Detection and Matching Settings, Camera Settings Reference and Station Settings.