Skip to main content

Remote command execution

Need​

Prevention of unauthorized remote command execution

Context​

  • Usage of C# 7.0 for modern language features and enhancements
  • Usage of Microsoft.AspNetCore.Mvc for building web applications with ASP.NET Core
  • Usage of System.Diagnostics for logging and performance monitoring in .NET development

Description​

Non compliant code​

public class CommandController : Controller
{
    [HttpGet]
    public IActionResult ExecuteCommand(string cmd)
    {
        var process = new Process()
        {
            StartInfo = new ProcessStartInfo
            {
                FileName = "/bin/bash",
                Arguments = cmd,
                RedirectStandardOutput = true,
                UseShellExecute = false,
                CreateNoWindow = true,
            }
        };
        process.Start();
        string result = process.StandardOutput.ReadToEnd();
        process.WaitForExit();
        return Ok(result);
    }
}

The above code represents a controller in an ASP.NET Core application that takes an input string cmd from the user and executes it as a bash command.

The vulnerability in this code is the ExecuteCommand method that takes a user-supplied string and directly uses it as an argument for a new process start info. This is a clear example of a Remote Command Execution vulnerability.

In this scenario, an attacker could potentially execute arbitrary commands on the server by simply passing them through the cmd parameter. For example, they could pass something like ; rm -rf / to delete all files in the root directory or ; cat /etc/passwd to read sensitive information.

This vulnerability exists because the application does not sanitize or validate the cmd input in any way before using it. This means that special characters like ; are not neutralized and can be used to chain multiple commands together.

Steps​

  • Validate and sanitize user input before using it to build OS commands.
  • Use parameterized queries or prepared statements to prevent command injection.
  • Avoid using user input directly in command arguments or parameters.
  • Implement input validation to restrict the type and format of user input.
  • Implement a whitelist approach to only allow specific characters or patterns in user input.
  • Implement proper error handling to prevent sensitive information leakage.
  • Regularly update and patch the underlying operating system and dependencies to mitigate potential vulnerabilities.

Compliant code​

public class CommandController : Controller
{
    private readonly Regex _regex = new Regex(@"^[a-zA-Z0-9\\s]*$");

    [HttpGet]
    public IActionResult ExecuteCommand(string cmd)
    {
        if (string.IsNullOrEmpty(cmd) || !_regex.IsMatch(cmd))
        {
            return BadRequest("Invalid command.");
        }

        var process = new Process()
        {
            StartInfo = new ProcessStartInfo
            {
                FileName = "/bin/bash",
                Arguments = cmd,
                RedirectStandardOutput = true,
                UseShellExecute = false,
                CreateNoWindow = true,
            }
        };
        try
        {
            process.Start();
            string result = process.StandardOutput.ReadToEnd();
            process.WaitForExit();
            return Ok(result);
        }
        catch (Exception ex)
        {
            // Log the exception
            return StatusCode(500, "An error occurred while executing the command.");
        }
    }
}

The updated code includes several measures to mitigate the risk of remote command execution:

  1. Input Validation: The code checks if the command is null or empty, and if it matches a regular expression that only allows alphanumeric characters and spaces. This prevents the injection of special characters that could modify the OS command.

  2. Error Handling: The code includes a try-catch block to handle any exceptions that occur during the execution of the command. This prevents the leakage of sensitive information in case of an error.

  3. Response Codes: The code returns appropriate HTTP response codes based on the result of the command execution. A 400 (Bad Request) status is returned for invalid commands, and a 500 (Internal Server Error) status is returned in case of an error.

  4. Whitelisting: The regular expression used for input validation acts as a whitelist, only allowing specific characters in the command.

Remember to regularly update and patch the underlying operating system and dependencies to mitigate potential vulnerabilities.

References​

Last updated on