Engine only
Sidecar, models, CLI, systemd unit, installer, and backup helper. No desktop UI.
Linux server deployment
Run the BSS attendance engine on an Ubuntu server without a desktop, Tauri UI, tray, or display server. The engine handles RTSP cameras, face matching, local event storage, and BSS sync. The local API stays bound to 127.0.0.1.
Sidecar, models, CLI, systemd unit, installer, and backup helper. No desktop UI.
bss-attendancectl wraps activation, cameras, resources, workers, diagnostics, and backups.
Binds to localhost by default. Live view is accessed through an SSH tunnel.
OS: Ubuntu 22.04/24.04 x86_64
Resources: 4 GB RAM minimum, 3 GB free disk
Network: outbound HTTPS to BSS and local RTSP reachability
GPU: optional NVIDIA GPU; CPU fallback supported
Access: root for install; SSH tunnel for remote live view
Camera: employer-approved and public-IP allowlisted in BSS
tar -xzf BSS-Attendance-<version>-Headless-Linux-x86_64.tar.gz
cd bss-attendance
sudo ./install-headless.shsudo -u bss /opt/bss-attendance/bss-attendancectl --home /var/lib/bss-attendance statusbss-attendancectl activate request --phone 9999999999
bss-attendancectl activate verify --phone 9999999999 --otp 123456
bss-attendancectl activate complete --token <token> --name "NOIDA server"bss-attendancectl camera test --rtsp "rtsp://user:pass@camera/stream"
bss-attendancectl camera add --label "Main Gate" --site BSS... --rtsp "rtsp://user:pass@camera/stream"ssh -N -L 8765:127.0.0.1:8765 user@server
# then open http://127.0.0.1:8765/stream/1sudo -u bss /opt/bss-attendance/bss-attendancectl --home /var/lib/bss-attendance backup --output /var/backups/bss-attendanceData is stored in /var/lib/bss-attendance. Keep the old folder when changing storage; back it up before upgrades.
If the attendance engine runs in the cloud but the cameras stay on the client’s local network, the client must provide a site-to-cloud tunnel. This is a client-managed deployment.
Read the site-tunnel guide8765 to the internet.