Deployment and Maintenance / Raspberry Pi Ubuntu Deployment

This document mainly describes how to run this system on a Raspberry Pi with Ubuntu Server (64bit) installed.

Always use the latest version of DataFlux Func for operations

It is recommended to connect the Raspberry Pi to the network via a wired connection during operations

This article is based on the Raspberry Pi 4B 8GB version. The Raspberry Pi 3B is too weak to run this system.

You must use the 64-bit Ubuntu Server image for Raspberry Pi

1. Preparation

Before installing DataFlux Func on a Raspberry Pi, some preparation is required.

1.1 Flashing the Raspberry Pi SD Card

It is recommended to use a clean official image to flash the SD card before installing DataFlux Func.

When flashing the SD card, you can choose from the following:

1. Use the official Raspberry Pi Imager tool to flash

   • Official tool download page

2. Directly download the official Ubuntu Server 64-bit image and flash it with a third-party tool

   • Official image download page
   • Etcher flashing tool

After flashing is complete, insert the SD card into the Raspberry Pi and power it on.

1.2 Initializing the System

On first boot of Ubuntu for Raspberry Pi, log in using the default username and password:

Username	Password
ubuntu	ubuntu

After successful login, you will be asked to change the password

1.3 Configuring Wi-Fi Connection (Optional)

If you want to connect to the network via Wi-Fi, follow the steps below:

Open the configuration file

sudo vim /etc/netplan/50-cloud-init.yaml

Assume your Wi-Fi SSID is my-wifi and the password is my-wifi-password. Then the configuration should be modified as follows:

network:
    ethernets:
        eth0:
            dhcp4: true
            optional: true
    # Add Wi-Fi configuration here
    wifis:
        wlan0:
            dhcp4: true
            access-points:
                my-wifi:
                    password: my-wifi-password
    version: 2

Apply the new configuration

sudo netplan apply

Restart the Raspberry Pi

sudo reboot

Use cURL to verify that the network connection is successful.

curl -L bing.com

2. Installing DataFlux Func

The process of installing DataFlux Func on a Raspberry Pi is basically the same as on a regular server.

2.1 Download

The download operation is the same as on other platforms. Use the following command; the Script will detect the current environment architecture and download the ARM version resources.

bash -c "$(curl -fsSL docs.dataflux-func.com/download)"

If you need to download the DataFlux Func installation package for Raspberry Pi on a PC, you need to specify --aarch64 in the download command

bash -c "$(curl -fsSL docs.dataflux-func.com/download)" -- --arch=aarch64

2.2 Installation

After the download is complete, enter the download directory and run the following command to install:

sudo bash run-portable.sh

2.3 Initializing the System

After installation is complete, use a browser to open the DataFlux Func initialization page and proceed.

• Visit http://{Raspberry Pi IP}:8088

2.4 Confirming Installation

After installation is complete, log in to the system, go to 'Management / About', and you will see that 'Architecture' is aarch64:

Subsequent operations are no different from DataFlux Func installed on a regular server.

X. Appendix

This section records content related to the Raspberry Pi for reference.

X.1 Maximizing Raspberry Pi Performance

Open the /boot/firmware/usercfg.txt file

sudo vim /boot/firmware/usercfg.txt

Add the following content as needed:

force_turbo=0  # Prevent CPU from running at maximum frequency
arm_freq=2100  # Overclock CPU to 2.1Ghz (default is 1.5Ghz)
over_voltage=6 # Raise voltage to level 6

Even without the above settings, it works normally, and a simple metal case is sufficient for passive cooling

These settings involve overclocking. Under full load, you cannot rely solely on a metal case for passive cooling; at least one fan is required

More aggressive configurations can further improve Raspberry Pi performance, but they will void the warranty

If overclocking damages your Raspberry Pi, the author of this article is not responsible. Think twice before proceeding

X.2 Raspberry Pi Stress Test Program

You can use the Raspberry Pi stress test tool to test the stability of the Raspberry Pi (especially after overclocking).

Clone the repository

git clone https://gitee.com/sujivin/rpi-cpu-stress.git

The original author's GitHub repository address is: github.com/xukejing/rpi-cpu-stress

Add execute permission

cd rpi-cpu-stress
chmod +x stress.sh

Start the stress test

sudo ./stress.sh

Then, in the terminal, you can see the following output:

a72 freq: 2100000
temp: 50634

a72 freq: 2100000
temp: 54530

a72 freq: 2100000
temp: 55991

a72 freq: 2100000
temp: 56478

• a72 freq: 2100000: indicates the current CPU frequency is 2.1Ghz
• temp: 50634: indicates the temperature is 50.634 degrees Celsius

Do not keep the Raspberry Pi at high temperatures (above 80 degrees Celsius) for a long time