Skip to main content

Harbin, China

Python SSH Logger

We want to build a small python program that allows us to log server states over the network by SSH. We start with creating 3 environment files that hold the server ip address, the SSH user login and the commands that we have to send to the server to get the information:

  1. serverip.env
  2. userlogin.env
  3. commands.env
  • ip_file_valid.py : Ask for serverip.env file location and read IP address
  • ip_addr_valid.py : Check if Server IP is a valid IP and does not belong to a restricted range.
  • ip_reach.py : Check if IP address can be pinged,
  • ssh_connect.py : Verify userlogin.env and commands.env and use Paramiko to connect to your server and send commands via SSHv2.
  • create_threads.py create parallel threads for each SSH connection (only useful if you contact more than 1 server at once)

The command I want to use is the top -n 1 that will give me an overview over the server load:

Python Network Logger

To extract the CPU load from the server reply, I am going to use the following regular expression - not that we cannot prepend our RegEx with r to get the RAW string, but have to use b to handle the response that Python calls byte-like:

# Searching for the CPU utilization value within the output of 'show processes top once'
cpu = re.search(b'(%Cpu\(s\): ) (.+?)(us)', server_response)
# cpu = server_response

# Extracting the second group, which matches the actual value of the CPU utilization and decoding to the UTF-8 format from the binary data type
utilization = cpu.group(2).decode('utf-8')
# utilization = cpu.decode('utf-8')

# Printing the CPU utilization value to the screen
# print(utilization)

# Opening the CPU utilization text file and appending the results
with open('E:\\python-ssh-logger\\cpu-load.txt', 'a') as f:
    # f.write('{},{}\n'.format(str(datetime.datetime.now()), utilization))
    f.write(utilization + '\n')

A good way to develop a fitting regular expression for your task is to test it on regex101.com:

Python Network Logger

Once the txt file with the CPU utilization is created, we can use matplotlib to plot the results in graph.py (You might have to install the library first python -m pip install matplotlib).

Running both programs parallel shows us the real-time CPU utilization of our server:

Python Network Logger