deployment-scripts

Scripts and deployment configurations for deployment of SCV2 software.

Overview

All containers should be run via Docker compose. Docker compose is the preferred way to handle deployments. That is to say, it is not recommended to use the individual deployment scripts.

There are a handful of profiles that can be enabled/disabled to run all or only a subset of services. The profiles are listed below, and are explained in the update script.

Notes for deployment

Quick start

Prerequisites

• Host OS must be Linux or MacOS
• Docker installed, with a recent version of docker compose. Docker 28.x.x + docker compose 2.35+ confirmed to work.
  • Check with docker --version and docker compose version

Build script

Run the build.sh script to enable/disable compose profiles and set environment variables.

This will create a .settings, .env, and the compiled docker-compose.yml file. .settings records the set of enabled profiles and project name. .env records the list of environment variables that have been changed from their defaults. docker-compose.yml is the generated compose output file, with all profiles and environment variables baked-in.

Flags:

• Pass -q to re-generate the docker-compose.yml, using existing settings in .env and .settings, with no user prompts.

Update script

Run the update.sh script to pull the latest docker images and update the compose.

You will be prompted to 'reconfigure'. Pressing 'y' on this will invoke the build.sh script.

This script will attempt to run a migration script to ensure correct permissions are set for any docker volumes.

Flags

• Pass -q to skip all user prompts.
• Pass --pull to force-pull all latest images referenced in the compose file.

Tools

Record & Stitch Video

1. Record Video

Records a raw RTSP stream from a specific camera to disk. This should be run as a detached container to avoid blocking your terminal during long recordings.

Command:

docker compose run -d --rm record_video <camera_id> <duration>

Arguments:

• <camera_id>: The specific camera ID (e.g., cam_01) as it is defined in the realtime container
• <duration>: Recording length. Accepts seconds (300) or shorthand (10m, 1h, 2.5d).

Output:

• Videos are saved to: ~/scv2/videos/<YYYY-MM-DD>/<camera_id>/
• Files are automatically made writable by the host user.

---

2. Stitch Videos

Scans a specific date folder, stitches all video segments for every camera into single files, and deletes the source directories by default.

Command:

docker compose run --rm stitch_videos <date_string> [options]

Arguments:

• <date_string>: The date folder to process (e.g., 2025-12-18). Can be an absolute path to a date folder, if it isn't located at ~/scv2/videos/<YYYY-MM-DD>/
• --keep-source: (Optional) If passed, the source directories and raw segments will not be deleted after stitching.

Output:

• Stitched files are saved to: ~/scv2/videos/<YYYY-MM-DD>/<camera_id>-<date>.mp4
• Files are automatically made writable by the host user.

Certbot

TODO: Document me!

Advanced usage

Backup realtime, auditgui and rdb configs

To easily backup configs from realtime, auditgui and rdb containers, you can run

./scripts/backup/backup_auditgui_realtime_rdb.sh

This will save backup tar files of each container's volumes in ~/scv2/backups. The realtime backup only includes each camera's /config and /resources folder (excludes /resources/backgrounds)