Skip to main content Skip to docs navigation
View on GitHub

Session Hijacking

Session hijacking occurs when an attacker takes over a valid session between two computers. The attacker steals a valid session ID in order to break into the system and snoop data.

On this page

Session Hijacking

Most authentication occurs only at the start of a TCP session. In TCP session hijacking, an attacker gains access by taking over a TCP session between two machines in mid session.

The attacker sniffs and accesses a legitimate session id from a user interacting with a web server, then uses that session identifier to spoof the session between the regular user and the server to exploit the user's session and access the server directly.

Session hijacking occurs because

  • no account lockout for invalid session IDs
  • weak session-ID generation algorithm
  • insecure handling
  • indefinite session expiration time
  • short session IDs
  • transmission in plain text

Session hijacking process

  1. Sniff , that is perform a man-in-the-middle (MITM) attack, place yourself between victim and server.
  2. Monitor packets flowing between server and user.
  3. Break the victim machine's connection.
  4. Take control of the session.
  5. Inject new packets to the server using the Victim's Session ID.

Protection against session hijacking

  • create a secure communication channel with SSH (secure shell)
  • pass authentication cookies over HTTPS connection
  • implement logout functionality so the user can end the session
  • generate the session ID after successful login
  • pass encrypted data between the users and the web server
  • use a string or long random number as a session key

See also

Updated on April 20, 2024 by Datarist.