SunPeek Documentation#

Contents#

Core:

Pipline Status Test Coverage Supported Python GitLab contributors Open Issues

WebUI:

Pipline Status GitLab contributors Open Issues

About SunPeek#

SunPeek implements a dynamic, in situ test methodology for large solar thermal plants, packaged as an open source software application and python library. It also includes the first open source implementation of the ISO 24194 procedure for checking the performance of solar thermal collector fields.

Full documentation is at https://docs.sunpeek.org

SunPeek was originally developed as part of the HarvestIT research project, see https://www.collector-array-test.org

A Web Application and a Python Library#

SunPeek is available as both a complete, containerised web application - intended to make the ongoing monitoring of one or several solar thermal plants simple and intuitive - and as a python library, for use by researchers and for building into other tools. To install the python library, simply run pip install sunpeek. To set up the web application, see below.

License#

Except where specifically noted otherwise, SunPeek is made available under the GNU Lesser General Public License. This means that you can use the software, copy it, redistribute it and include it in other software, including commercial, proprietary software, for free, as long as you abide by the terms of the GNU GPL, with the exceptions provided by the LGPL. In particular, if you redistribute a modified version of the software, you must make the source code of your modifications available, and if you include the software in another piece of software or physical product, you must give users notice that SunPeek is used, and inform them where to obtain a copy of the SunPeek source code and license.

Note that the SunPeek WebUI, is covered by a separate licence, the BSD-3-Clause, see: https://opensource.org/licenses/BSD-3-Clause

SunPeek is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

Running SunPeek#

Prerequisites#

Ideally, a modern linux environment, although running on Windows and MacOS is also possible, with at least:
- 5GB free disk space (this is needed for unpacking the application, after installation it will use around 2 GB)
- A 7th Gen i5 or better processor.
- 8GB RAM
Docker and Docker compose installed (see below), version 2.20 or newer.

Get Docker#

In order to provide a consistent environment and allow SunPeek to work across a wide range of install environments, it is provided as a set of Docker images (essentially, very lightweight virtual machines). A docker compose configuration is also provided, if you are installing sunpeek on a single machine, this is probably what you want to use, follow the links below for instructions depending on your environment.

Linux#

To install docker on Linux go to https://docs.docker.com/engine/install/ select the relevant platform and follow the instructions.

Windows 10 and 11#

On desktop windows, the easiest way to get Docker is to install docker desktop.

Windows server#

To install the Docker Engine on Windows Server, see this guide from Microsoft

Using the easy setup GUI or scripts#

For convenience, a bash script for Linux or a simple graphical utility for use on Windows is provided. These will do some basic configuration for running SunPeek on a single machine, optionally accessible over a local network. Docker is still required.

Note

If you have previously set up SunPeek using the default configuration, you must first remove all stored data by running the command docker volume rm harvestit_hit_postgres_data in a terminal/command prompt, this will also remove uploaded data. You do not need to do this to update the software, see Upgrading to a new version of SunPeek

On Linux#

In the location you want to store sunpeek configuration, run curl https://gitlab.com/sunpeek/sunpeek/-/raw/main/deploy/quick-setup.sh?inline=false -o quick-setup.sh,
Run quick-setup.sh (usually just with the command ./quick-setup.sh) and enter the url which sunpeek can be accessed at when prompted.
Run docker compose up -d
After at most 2 minutes (usually a few seconds), the web UI should be accessible at http://localhost, or the url set in step 2.

On Windows#

Download this file, and unzip it to a temporary location.
Run sunpeek_easy_installer.exe
You should then get a small window with 2 fields. You must select a location to store the configuration files for the application, if you are running the application only for access from the local machine, leave the default in the url field.
Click setup.
Once the window closes open the directory you specified, and double-click the start.bat file. A command prompt will open to display the startup process.
Docker compose will download the required application components, this may take several minutes, once you see all components listed as ‘started’ or ‘healthy’ you can close the command prompt.
Open a browser and go to the url specified in the setup tool, probably http://localhost to see the web-UI for the tool.
To stop the application, assuming no other processes are running under docker on your machine, simply shut down the docker engine. If you are using Docker Desktop, this can be done by right-clicking the Docker icon the system try and selecting Quit

Advanced Configuration#

Danger

It is strongly recommended that you don’t make SunPeek accessible from the public internet. At present there are NO built-in access controls

Configuration is via environment variables, which can be set by any configuration management system you use, however the default setup uses .env files. To deploy the application on a single host, only the external URL needs to be set. The easy installer does this for you, or you can set the value of HIT_API_BASE_URL in the ui.env file to <external.url>/api/v1. Other configuration variables are documented at docs.sunpeek.org/configuration-variables.html

Note

This setup is designed to deploy all containers on a single machine where docker compose is running. The configuration for the Traefik reverse proxy is stored in a directory which is bind mounted to the container. For other deployment approaches (e.g. using Kubernetes), a more appropriate Traefik dynamic configuration provider should be selected.

Database Backends#

SunPeek supports SQLite and PostgreSQL as backend databases for storing plant configurations, collector and fluid types etc. For single host installs with small number of plants, SQLite should be sufficient and this is therefore the default configuration. To use the PostgreSQL backend you can add HIT_DB_TYPE=postgresql to api.env, and set the values of the HIT_DB_PW and POSTGRES_PASSWORD variables in the api.env and db.env files to the same random, unique password string. The docker compose setup includes an optional postgres service, which can be started along with the other containers with docker compose --profile postgres up

Upgrading to a new version of SunPeek#

Updates to SunPeek are accomplished by pulling newer versions of the docker images used to run the application. These are used to create new containers in place of the old ones. If you used the default configuration, a persistent docker volume called sunpeek_postgres_data will have been created, this should avoid data loss, however keeping backups is always recommended. The update process is as follows, and is the same on all operating systems:

Open a terminal/command prompt in the configuration folder selected during setup.
Run docker compose pull to download the latest images
Run docker compose up -d to recreate any containers which have updated images.

Technical Details - What does compose do?#

Docker Compose is a tool for orchestrating docker containers, to create applications made up of several docker containers. When the HarvestIT application is started with the default docker-compose file, the following things happen:

Compose checks if each of the images defined in the compose file, is available with the correct tag locally, if not it pulls them from the relevant registry
A virtual network is created, for the containers to communicate with each other. This is segregated from the host machine’s main network interfaces.
The database container (using image timescale/timescaledb:latest-pg14), is started, with a healthcheck defined. Alongside this container, a Docker Volume is created called hit_postgres_data, which is mapped to the default data directory in the database container, to ensure that database data is persisted when the containers are recreated (e.g. during an update).
The reverse proxy container (using image traefik:v2.8) is started. As well as being attached to the virtual network created by docker compose, this has port 80 exposed to the host, so that it can be accessed at localhost/ or from an extrnal connection. This routes web requests to either the web-ui or api containers, depending on the path in the request URL. It can also be configured to terminate TLS (HTTPS) encrypted connections and obtain certificates automatically, to secure connections to the application.
Compose waits until the database container reports a “healthy” status, then starts the api container (using image sunpeek:latest), this is the main HarvestIT application.
The harvestit container runs a database initialization scripts to get the database ready.
Once the api container has started, the webui container is started.

Development#

Roadmap#

For information on the project roadmap, see https://gitlab.com/sunpeek/sunpeek-governance/-/wikis/Roadmap. The Roadmap is maintained by the Steering Committee, based on input from the community, recieved primarily via feature request issues and discussions at industry events.

Contributing#

SunPeek is developed as an open source project, with contributions gladly accepted from interested members of the community. For information on contributing to sunpeek, see CONTRIBUTING.md, for developer documentation see https://docs.sunpeek.org/developing.html

Maintainers and Steering Committee#

The overall direction of the project is managed by a steering committee, who appoint the maintainers, details can be found in the Governance project