As the EU Cyber Resilience Act (CRA) has already entered into force in 2024, and the first legal requirements must be fulfilled starting in September 2026, we are now explicitly collecting tips, tricks, and best practices related to the topic here.

• Treat all external input as untrusted
• Prefer simple and explicit architectures
• Fail safely and log security-relevant failures
• Minimize attack surface and unnecessary dependencies
• Apply the principle of least privilege

• Never hardcode:

  • passwords
  • API keys
  • certificates
  • tokens
  • database credentials
• Do not store secrets in:

  • block diagram constants
  • typedef defaults
  • test VIs
  • screenshots
  • source-controlled config files
• Do not log sensitive information

• Prefer encrypted communication (TLS, HTTPS, SSH)
• Avoid plaintext credentials and insecure protocols
• Validate certificates and remote peers
• Explicitly document:

  • timeout behavior
  • reconnect strategy
  • retry handling

• Use parameterized queries whenever possible
• Avoid dynamically concatenating SQL strings from user input
• Validate and sanitize externally provided data before database operations
• Restrict database permissions to the minimum required access
• Never store database credentials directly in source code
• Log database failures without exposing sensitive query data
• Explicitly handle connection loss and timeout behavior

• Disable VI Server if not required
• Restrict network access and permissions
• Never expose unrestricted VI Server access on production systems

• Only load trusted plugins or VIs
• Avoid loading VIs from user-writable directories
• Use strict connector pane contracts and versioning

• Treat all command line input as untrusted
• Avoid constructing shell commands from unchecked user input
• Document all external tool dependencies

• Validate file paths and filenames
• Avoid path traversal vulnerabilities
• Restrict writable and executable locations

• Encapsulate queues, references, and communication resources
• Avoid exposing internal module resources
• Validate message payloads and typedef compatibility
• Keep security-related blocking operations outside the EHL

• Log security-relevant events:

  • failed authentication
  • rejected certificates
  • malformed packets
  • permission violations
• Logs should include:

  • timestamps
  • module names
  • severity
  • connection information
• Avoid excessive logging in real-time loops

• Remove development tooling from production systems
• Avoid unnecessary administrator privileges
• Keep dependencies updated
• Prefer signed installers and executables