Repository to control displays using Adafruit Mini PiTFT (https://www.adafruit.com/product/4393)
This software was tested under python 3.10. Please create a virtual environmnet
conda create -n pitft_env python==3.10 pip
conda activate pitft_env
pip install -r requirements.txt
(pitft_env) ~/pitft_display $ python3 monitor_stats.py
If you want scripts in this repository to run in the background, follow the instructions below.
Important
Run which python in order to get the proper execution path
Tip
The steps below are detailed for the stat monitor using a python instance under /home/pi/miniforge3/envs/pitft_env/bin/python and a script located at /home/pi/pitft_display/monitor_stats.py, modify for your use if changing the script to run.
Using this following configuration file:
[Unit]
Description=Display service to run a stat monitor using pitft display
After=multi-user.target
[Service]
Type=idle
ExecStart=/home/pi/miniforge3/envs/pitft_env/bin/python /home/pi/pitft_display/monitor_stats.py
Restart=on-failure
[Install]
WantedBy=multi-user.target
-
Save the Configuration:
- Open a terminal.
- Use a text editor (like nano or vim) to create a new service file:
sudo nano /etc/systemd/system/pitft_monitor_stats.service- Copy and paste the modified service configuration above into the editor.
- Save and close the editor (CTRL+X, Y, Enter for nano).
-
Enable the Service:
sudo systemctl enable pitft_monitor_stats
- Start the Service:
sudo systemctl start pitft_monitor_stats
- Check the Service Status (or visually inspect the monitor):
sudo systemctl status pitft_monitor_stats
- Try rebooting and checking if it executes properly or you get error logs
sudo journalctl -u pitft_monitor_stats
If things go wrong, this service can be somewhat hungry in terms of disk usage. You should check
journalctl --disk-usage
It is recommended that you
journalctl --vacuum-size=50M # or other size
journalctl --vacuum-size=1d # or any other period
You can also choose to keep the sizes of logs small permanently. It's recommended that you use as script to do this (especially if you are running multiple Rpi). Assuming you have ssh access to the with username and password, and you can read ip (df['ip']) addresses from each of the raspberry pis in the network (df['computer_name']). This script assumes a pattern of 'raspberry' and converts all to lower case, adjust as necessary.
def configure_journalctl(df, username, password):
ssh = paramiko.SSHClient()
ssh.set_missing_host_key_policy(paramiko.AutoAddPolicy())
raspberries = df[df['computer_name'].str.lower().str.contains("raspberry", na=False)]
print(f"found these raspberries ({len(raspberries.index)})")
print(raspberries)
for index, row in raspberries.iterrows():
ip = row['ip'] # Assuming IP is stored in the YAML file
try:
print(f"Trying IP {ip}")
ssh.connect(ip, username=username, password=password, timeout=10)
# Get current disk usage of journalctl
stdin, stdout, stderr = ssh.exec_command("journalctl --disk-usage")
current_usage = stdout.read().decode().strip()
print(f"Current journalctl disk usage on {row['computer_name']} ({ip}): {current_usage}")
# Execute configuration commands
commands = [
"sudo sed -i '/SystemMaxUse=/c\\SystemMaxUse=50M' /etc/systemd/journald.conf",
"sudo systemctl restart systemd-journald",
"sudo journalctl --rotate",
"sudo journalctl --vacuum-size=50M",
"journalctl --disk-usage" # Check disk usage again after changes
]
for command in commands:
stdin, stdout, stderr = ssh.exec_command(command)
print(f"Executed on {row['computer_name']} ({ip}): {command}")
print(stdout.read().decode() + stderr.read().decode())
ssh.close()
except Exception as e:
print(f"Failed to connect or execute on {row['computer_name']} ({ip}): {e}")