I have a Flask app that I'd like to deploy and start using Ansible. I've already got my playbook setup to install all the dependencies needed but I'm having trouble getting Ansible to start my application. I've used command and shell both of which do startup the Flask app, but they block Ansible from exiting and Flask doesn't popup in it's own terminal window, which makes it difficult to visually debug and see what Flask is doing.
#Current playbook; both command and shell act the same way
tasks:
- command: python3 /home/user/Desktop/Flask/app.py
- shell: python3 /home/user/Desktop/Flask/app.py
Q: How can I have Ansible startup a Flask script in it's own terminal window?
If you have gnu screen installed on this system then you can use it to background tasks. I use this to run a shell script asynchronously as the deploy user so that I can log in later and see how it's doing:
- name: Invoke reset script
command: /usr/bin/screen -d -m sudo -u deploy /usr/local/bin/do-reset.sh -i -y reset
async: True
poll: 0
The -d -m parameters tell screen to start up in detached mode, and the async and poll settings tell Ansible to background the command and forget about it.
Related
As my first batch script, I want to run my backend and frontend with just one script. My backend is in python with Django and my frontend is in react. I, therefore, have to processes I need to run. But I don't know how to run them in parallel as I need to start a virtual environment in python first.
I don't understand how it would log the server logs. Is there an option to make two terminal windows or a new terminal tab? Or would it log in to one terminal together?
#!/bin/bash
DJANGODIR=/Users/usr/Desktop/programming/app/django
REACTDIR=/Users/urs/Desktop/programming/app/react
# START REACT FRONTEND
echo Starting React Server
# Option 1
# cd $REACTDIR
# npm start
# Option 2
# start app from any directory without cd into it
# npm --prefix $REACTDIR run start
# START DJANGO BACKEND
echo Starting Django Server
# cd $DJANGODIR
# Start virtual environment
# source venv/bin/activate
# Start Server
# python3 manage.py runserver
# Activating and running in one command
# source venv/bin/activate python3 manage.py runserver
# Here is where I'm not knowing how to run them in parallel
source venv/bin/activate python3 manage.py runserver & npm --prefix $REACTDIR run start
Have you tried with the pipe operator? Pipe operator explained
I currently am running a program through Thonny, and I want to make the pi autoboot that program whenever it turns on. I currently have the pi 4, and the code is run on Python3. Have tried many ways to autoboot such as using rc local, and bash rc, but none seem to work.
I recommend using the autostart file.
Edit the following file:
sudo nano /etc/xdg/lxsession/LXDE-pi/autostart
The default should look something like:
#lxpanel --profile LXDE-pi
#pcmanfm --desktop --profile LXDE-pi
#xscreensaver -no-splash
Add the command to run your script utilizing the # prefix and save. The new autostart file should look like:
#lxpanel --profile LXDE-pi
#pcmanfm --desktop --profile LXDE-pi
#xscreensaver -no-splash
#python3 /home/pi/your_script.py
More info: https://www.raspberrypi.org/forums/viewtopic.php?t=294014
By using things like .bashrc, your program will only get executed when you open a command prompt or terminal.
You can use systemd services:
create a file your_service.service in /etc/systemd/system/ (you need to be root, so use sudo). The filename will be the name of your service, without the .service extension (in this case your service will be named your_service)
put this in it (without the comments):
[Unit]
Description=A short description of your service
After=network.target # <-- put this if your script needs network, otherwise you can omit this line
StartLimitIntervalSec=0
[Service]
Type=simple
Restart=always # use this to always restart the script if the process ends. If your script only needs to run once at startup and is not a daemon, don't use this.
RestartSec=1
User=root # user account used to run the service
ExecStart=python3 /path/to/your/script.py
[Install]
WantedBy=multi-user.target
Then, you can start your script with this command:
sudo systemctl start your_service
You can finally enable the service to be run automatically at startup with this command:
sudo systemctl enable your_service
You can find more documentation / tutorials on systemd services and on all the available options in the following links:
https://medium.com/#benmorel/creating-a-linux-service-with-systemd-611b5c8b91d6
https://www.digitalocean.com/community/tutorials/how-to-use-systemctl-to-manage-systemd-services-and-units
https://www.freedesktop.org/software/systemd/man/systemd.service.html
https://wiki.archlinux.org/title/Systemd
I have a modified AWS basicPubSub function to transfer data to the AWS IoT core, I want the script to run at start-up.
I have added this script into, make it executable and updated the init.d
/etc/init.d
chmod 755 LOMAWS.sh
sudo update-rc.d LOMAWS.sh defaults
But the script does not start, how can I make it run from start up?
clear
echo "LOM AWS Script starting"
cd /home/pi/Documents/awsiot/aws-iot-device-sdk-python/samples/basicPubSub
sudo python basicPubSub.py -e "XXXXXXXX-ats.iot.us-east-2.amazonaws.com" -r root_CA.crt -c XXXXXXXX-certificate.pem.crt -k XXXXXXX-private.pem.key
Have you tried UserData?
By default, user data scripts and cloud-init directives run only during the first boot cycle when an instance is launched
I have a python script. Script have selenium with Chrome and go to a website, take data and put in CSV file.
This is a very long work.
I put the script on the server. And run. All work.
But I need script work in the background.
chmod +x createdb.py
nohup python ./createdb.py &
And I see
(env)$ nohup ./createdb.py &
[1] 32257
(env)$ nohup: ignoring input and appending output to 'nohup.out'
Press Enter.
(env)$ nohup ./createdb.py &
[1] 32257
(env)$ nohup: ignoring input and appending output to 'nohup.out'
[1]+ Exit 1 nohup ./createdb.py
Then it runs and immediately writes errors to the file, that Chrome did not start or there was no click.
I want to remind you that if you start without nohup, then everything will work.
What am I doing wrong? How to run a script?
Thank you very much.
You could create a background daemon (service)
You taged Ubuntu 16.04 it means you got systemd, for more information on how to set it up, please visit this link
create a file called <my_service>.system
and put it there: /etc/systemd/system
you systemd unit could look like this:
[Unit]
Description=my service
After=graphical.target
[Service]
Type=simple
WorkingDirectory=/my_dir
ExecStart=python my_script.py
[Install]
WantedBy=multi-user.target
then all you have to do is, reload systemd manage and start your service:
sudo systemctl daemon-reload
sudo systemctl myservice start
You can use the screen command, it works perfectly.
Here is a very good link: https://www.rackaid.com/blog/linux-screen-tutorial-and-how-to/
You can use a simple command, from the env directory:
(env)$ python /path/to/createdb.py > logger.txt 2>&1 &
This will help for storing the program logs in a defined file called "logger.txt"
Using ubuntu's 16.04 crontab and #reboot to run python3 script. The script runs properly on reboot as I see the logged output. However, my script's os.system command is not running. It runs fine if ran outside of crontab. My scripts are all executable.
crontab -l output:
SHELL=/bin/bash
#reboot nohup /usr/bin/python3 -u /home/path/scheduler.py >> /path/log.out &
scheduler.py code:
#...(check if web server is running...if not restart)
os.system('nohup /usr/bin/python3 -u /path/webserver/main.py &')
print('this function ran')
When I logged the output of the os.system command , there was no output.
As a side note, I am running python schedule commands to check the general health of a webserver. crontab doesn't seem to be the right tool for this so I just use crontab to start my python scheduler on reboot.
I am using flask as the webserver, and would use gunicorn and systemctrl if I could get it to work... but it didn't so this is my workaround.
The point is that, the command called by os.system is not in default path.
For example, tcpdump is not in /usr/bin/.
So, you can solve the problem by adding the full path of the command.
I was facing the same issue when we try to run python script directly in crontab it just by passes the os.system() commands.
Make launcher.sh:
#!bin/bash
cd /home/pi/
sudo python example.py
Then, make your script executable:
chmod 755 launcher.sh
And at last, add your script to crontab:
crontab -e
and add this line at the end:
#reboot sh /home/pi/launcher.sh
(I set the program to run at each reboot)