Execute shell commands in Python
I’m currently studying penetration testing and Python programming. I just want to know how I would go about executing a Linux command in Python. The commands I want to execute are:
echo 1 > /proc/sys/net/ipv4/ip_forward iptables -t nat -A PREROUTING -p tcp --destination-port 80 -j REDIRECT --to-port 8080 If I just use print in Python and run it in the terminal will it do the same as executing it as if you was typing it yourself and pressing Enter ?
6 Answers 6
You can use os.system() , like this:
os.system('echo 1 > /proc/sys/net/ipv4/ip_forward') os.system('iptables -t nat -A PREROUTING -p tcp --destination-port 80 -j REDIRECT --to-port 8080') Better yet, you can use subprocess’s call, it is safer, more powerful and likely faster:
from subprocess import call call('echo "I like potatos"', shell=True) Or, without invoking shell:
If you want to capture the output, one way of doing it is like this:
import subprocess cmd = ['echo', 'I like potatos'] proc = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE) o, e = proc.communicate() print('Output: ' + o.decode('ascii')) print('Error: ' + e.decode('ascii')) print('code: ' + str(proc.returncode)) I highly recommend setting a timeout in communicate , and also to capture the exceptions you can get when calling it. This is a very error-prone code, so you should expect errors to happen and handle them accordingly.
@binarysubstrate, deprecated as in not supported or not available? I’ve been recently working on machine with 2.7 (not by choice), and os.system still works.
With Python 3.4 the shell=True has to be stated otherwise the call command will not work. By default call will try to open a file specified by the string unless the shell=True is set. It also looks like that in Python 3.5 call is replaced with run
The first command simply writes to a file. You wouldn’t execute that as a shell command because python can read and write to files without the help of a shell:
with open('/proc/sys/net/ipv4/ip_forward', 'w') as f: f.write("1") The iptables command is something you may want to execute externally. The best way to do this is to use the subprocess module.
import subprocess subprocess.check_call(['iptables', '-t', 'nat', '-A', 'PREROUTING', '-p', 'tcp', '--destination-port', '80', '-j', 'REDIRECT', '--to-port', '8080']) Note that this method also does not use a shell, which is unnecessary overhead.
import os os.system("your command here") This isn’t the most flexible approach; if you need any more control over your process than «run it once, to completion, and block until it exits», then you should use the subprocess module instead.
As a general rule, you’d better use python bindings whenever possible (better Exception catching, among other advantages.)
For the echo command, it’s obviously better to use python to write in the file as suggested in @jordanm’s answer.
For the iptables command, maybe python-iptables (PyPi page, GitHub page with description and doc) would provide what you need (I didn’t check your specific command).
This would make you depend on an external lib, so you have to weight the benefits. Using subprocess works, but if you want to use the output, you’ll have to parse it yourself, and deal with output changes in future iptables versions.
How to Execute Linux Commands in Python
Linux is one of the most popular operating systems used by software developers and system administrators. It is open-source, free, customizable, very robust, and adaptable. Making it an ideal choice for servers, virtual machines (VMs), and many other use cases.
Therefore, it is essential for anyone working in the tech industry to know how to work with Linux because it is used almost everywhere. In this tutorial, we are going to look at how we can automate and run Linux commands in Python.
Table of contents
Prerequisites
Introduction
Python has a rich set of libraries that allow us to execute shell commands.
A naive approach would be to use the os library:
import os cmd = 'ls -l' os.system(cmd) The os.system() function allows users to execute commands in Python. The program above lists all the files inside a directory. However, we can’t read and parse the output of the command.
In some commands, it is imperative to read the output and analyze it. The subprocess library provides a better, safer, and faster approach for this and allows us to view and parse the output of the commands.
| OS | subprocess |
|---|---|
| os.system function has been deprecated. In other words, this function has been replaced. | The subprocess module serves as a replacement to this and Python officially recommends using subprocess for shell commands. |
| os.system directly executes shell commands and is susceptible to vulnerabilities. | The subprocess module overcomes these vulnerabilities and is more secure. |
| The os.system function simply runs the shell command and only returns the status code of that command. | The subprocess module returns an object that can be used to get more information on the output of the command and kill or terminate the command if necessary. This cannot be done in the os module. |
Although you can execute commands using the OS module, the subprocess library provides a better and newer approach and is officially recommended. Therefore, we are going to use subprocess in this tutorial. This documentation explores the motivation behind creating this module.
Building an application to ping servers
Let’s use the subprocess library to write a script that pings multiple servers to see whether they are reachable or not. This would be a good use case when you have multiple hosts, servers, or VMs(AWS ec2 instances) and want to check if they are up and running without any problems.
A simple solution is to just ping these servers and see if they respond to the request. However, when you have a considerable amount of machines, it will be extremely tedious and time-consuming to manually ping them. A better approach is to use Python to automate this process.
Code
According to the official documentation, the subprocess module allows you to spawn new processes, connect to their input/output/error pipes, and obtain their return codes.
This module intends to replace several older modules and functions. The subprocess library has a class called Popen() that allows us to execute shell commands and get the output of the command.
Create a Python file and add the following code. We also need to create a file called “servers.txt”, where we can add a list of all the servers we need to ping. The Python script will read from this file and ping each server listed in it.
I have added 4 servers, out of which two exist and the other two do not. Only the servers that exist can be “pinged”.
import subprocess def ping(servers): # The command you want to execute cmd = 'ping' # send one packet of data to the host # this is specified by '-c 1' in the argument list outputlist = [] # Iterate over all the servers in the list and ping each server for server in servers: temp = subprocess.Popen([cmd, '-c 1', server], stdout = subprocess.PIPE) # get the output as a string output = str(temp.communicate()) # store the output in the list outputlist.append(output) return outputlist if __name__ == '__main__': # Get the list of servers from the text file servers = list(open('servers.txt')) # Iterate over all the servers that we read from the text file # and remove all the extra lines. This is just a preprocessing step # to make sure there aren't any unnecessary lines. for i in range(len(servers)): servers[i] = servers[i].strip('\n') outputlist = ping(servers) # Uncomment the following lines to print the output of successful servers # print(outputlist) 
As you can see in the output, we get the message “name or service not known” for the two servers that did not exist.
In the program above, the ping() function takes a list of servers and returns the output of each running ping command on each server. If a server is unreachable, it displays an output saying “ping: somethingthatdoesntexist: Name or service not known”.
The Popen() is a constructor method of the Popen class and takes in the following arguments:
- A list of commands and any additional options these commands might require. For example, the ls command can be used with ‘-l’ option. To execute the ls -l command, the argument list would look like this: [‘ls’, ‘-l’] . The commands are specified as strings. In the example above, we use the ping command with the option -c 1 so that it only sends one packet of data, and the server replies with a single packet . Without this limit, the command would run forever until an external process stops it.
- The stdout argument is optional and can be used to set where you want the subprocess to display the output. By default, the output is sent to the terminal. However, if you don’t want to dump a large output onto the terminal, you can use subprocess.PIPE to send the output of one command to the next. This corresponds to the | option in Linux.
- The stderr argument is also optional and is used to set where you want the errors to be displayed. By default, it sends the errors to the terminal. Since we need to get a list of servers that cannot be reached, we don’t need to change this. The servers that cannot be reached (error) will be displayed to us on the terminal.
The output of the command is stored in a variable called temp . The communicate() function allows us to read the output and the str function can be used to convert it to a string. Once we get the output, we can parse it to extract only the essential details or just display it as it is. In this example, I am storing the output in a list for future use.
Conclusion
In conclusion, automation is one of the hottest topics in the industry, and almost every company is investing huge amounts of money to automate various manual tasks. In this tutorial, we explored the process of automatically running and analyzing Linux commands on multiple hosts using Python.
An old way of doing this is by using shell scripts. However, using Python gives developers more power and control over the execution and output of the commands. Now that you have understood the basics of executing Linux commands, you can go ahead and experiment with different commands and build more complex and robust applications.
How can I execute an os/shell command from python [duplicate]
For say in terminal I did cd Desktop you should know it moves you to that directory, but how do I do that in python but with use Desktop with raw_input(«») to pick my command?
3 Answers 3
The following code reads your command using raw_input, and execute it using os.system()
import os if __name__ == '__main__': while True: exec_cmd = raw_input("enter your command:") os.system(exec_cmd) As mentioned here: stackoverflow.com/questions/35843054/… — when executing this code, with «cd /home» — the directory is changed while python is running. After python process ends, you receive the original shell with the original current-directory.
To go with your specific example, you’d do the following:
import os if __name__ == "__main__": directory = raw_input("Please enter absolute path: ") old_dir = os.getcwd() #in case you need your old directory os.chdir(directory) I’ve used this technique before in some directory maintenance functions I’ve written and it works. If you want to run shell commands more generally you’d something like:
import subprocess if __name__ == "__main__": command_list = raw_input("").split(" ") ret = subprocess(command_list) #from here you can check ret if you need to But beware with this method. The system here has no knowledge about whether it’s passing a valid command, so it’s likely to fail and miss exceptions. A better version might look like:
import subprocess if __name__ == "__main__": command_kb = < "cd": True, "ls": True #etc etc >command_list = raw_input("").split(" ") command = command_list[0] if command in command_kb: #do some stuff here to the input depending on the #function being called pass else: print "Command not supported" return -1 ret = subprocess(command_list) #from here you can check ret if you need to This method represents a list of supported commands. You can then manipulate the list of args as needed to verify it’s a valid command. For instance, you can check if the directory you’re about to cd exists and return an error to the user if not. Or you can check if the path name is valid, but only when joined by an absolute path.