Using Python to handle return code from Shell scripts

Question

I have a Shell script that handles Satellite server and Redhat IDM registrations. I don't have rights to update the script. The -i argument in the command below is for IDM registration and -s for Satellite server registration.

/usr/local/sbin/new-clone.sh -i aws -s aws-prod

Error handling is done as follows: Satellite registration:

 if [ "${RETURN_VALUE}" -ne 0 ]; then
 echo -e "\n\nSatellite registration failed. Please correct and re-run this script.\n\n"
 exit 2
 fi

IDM registration:

idm_failed ()
{
 echo -e "\n- IDM registration failed to 1ドル. This script did not complete. Please check network connection between this system and 1ドル servers. Re-run this script after troubleshooting. Exiting.."
 exit 2
}

I am executing the Shell script from Python as follows. The server.execute_script command is proprietary to a COTS application.

registration_command = "/usr/local/sbin/new-clone.sh -i aws -s aws-pro"
join_script = """#!/bin/bash
{}
yum clean all
yum -y upgrade 
systemctl reboot && exit 0
""".format(registration_command)
 try:
 server.execute_script(script_contents=join_script, runas_username='ec2-user', run_with_sudo=True,timeout=1200)
 except:
 logger.info('Failed with SEC satellite or IDM')

I want to update the logic in this try-catch statement so that it's more specific to whether the issue was with IDM registration or the Satellite registration. Since both these functions have a return code of 2, I was wondering it's possible to use the output of the echo command to implement. I would love to hear from the community on what makes sense here.

Please stay safe and be kind.

• Jongware: Does this answer your question? Calling an external command from Python

Answer 1

I think that the subprocess module would be perfect for you. If you run:

import subprocess
res = subprocess.run('your command',shell=True, capture_output=True, check=True)
# try to run it with 'ls -l' and check the returned response.

You will be able to get the returncode, stdout, stderr using: res.returncode, res.stdout, res.stderr and if the command fails, it will throw exception.

• Raj: Unfortunately, I have to rely on server.execute_script to execute the command. This command handles authentication to the remote server. server.execute_script returns the output from the from the script's execution. So, is it possible to parse it in Python?
• Gabio: Could you share the output please?
• Raj: Here it is: ``` ➤ bash new-clone.sh -i aws -s aws-prod - Registering to AWS IDM servers. - IDM registration failed to SP IDM. This script did not complete. Please check network connection between this system and SP IDM servers. Re-run this script after troubleshooting. Exiting.. ➤ bash new-clone.sh -s aws-prod Satellite registration failed. Please correct and re-run this script. ```

Answer 2

You could change one of the exit codes and have it printed with:

 try:
 server.execute_script(script_contents=join_script, runas_username='ec2-user', run_with_sudo=True,timeout=1200)
 except Exception as e:
 logger.info('Failed with error:\n{}'.format(e))

Normally this will print the error message with the exit code.

• Raj: Unfortunately, I can't update the Shell script since it's not owned by my team.