2 practical ways to use the Python subprocess module

In my previous article, I introduced the Python subprocess module, explained how to use it, and described when to use it. This article will review two different use cases for the subprocess library: running simple Bash commands and running Bash scripts.

[ Download the Bash cheat sheet to become more efficient at the command line. ]

Use subprocess to run simple Bash commands

You can use the subprocess module to run simple Bash commands such as mkdir and ls. This exercise will use Python to parse a text file and create directories based on the folder contents.

First, clone the subprocess_demo repository into your terminal: 

git clone https://github.com/nicolenlama/subprocess_demo.git

Then cd to usecase1.

There should be two files: directories.txt and create_directories.py. The directories.txt file contains the names of your soon-to-be-generated folders. The create_directories.py file reads the contents of directories.txt and creates the directories using the subprocess module and mkdir. I'll review the contents of the script.

The first line in the script imports the subprocess library:

import subprocess

Next, it loops through the directory names in the directories.txt file and uses the mkdir command in the run method to create the directories.

with open("directories.txt", "r") as directories:
   for dirs in directories:
     subprocess.run("mkdir ./{0}".format(dirs),shell=True, capture_output=True)

Run the ls command to verify that the script created all the directories. You should see automation, backup, development, production, and testing directories.

subprocess.run("ls",shell=True)

Use subprocess to run a Bash script

You can also run a Bash script with the subprocess library. This example uses the usecase2 folder in the subprocess_demo repo. The Bash script in this example (check_ceph.sh) accesses the backend Ceph cluster for OpenShift Data Foundation via rsh and runs several Ceph commands to determine the cluster's health and architecture.

[ Do you know the difference between Red Hat OpenShift and Kubernetes? ]

If you do not have an active OpenShift Data Foundation cluster, you can run the Python script using the -i output.txt flag in the terminal. The Python script, ceph_status_check.py, uses subprocess.run("check_ceph.sh") to obtain the output of the Bash script and then process the text to generate critical insights about the cluster. I highly encourage you to open the script in the subprocess_demo repo to understand how to use Python to parse the output from the subprocess.

To run the Python script, enter the following command:

python3 ceph_status_check.py

Or, if you do not have a running OpenShift Cluster, run:

python3 ceph_status_check.py -i output.txt 

If the code ran successfully, four metrics appear in the terminal: ceph status, available capacity, number of osds, and number of pgs.

Give it a try

There you have it! You have just examined two different use cases for the subprocess module.

For a deeper dive, please see the subprocess documentation and Jose Vicente Nunez's How to write an Ansible plugin to create inventory files on Enable Sysadmin.