A spoofing attack is different from a hijack in that an attacker is not actively taking another user offline to perform the attack. he pretends to be another user or machine to gain access.
The early record of a session hijacking is perhaps the Morris Worm episode that affected nearly 6000 computers on the ARPANET in 1988. This was ARPANET's first automated network security incident. Robert T. Morris wrote a program that would connect to another computer, find and use one of several vulnerabilities to copy itself to that second computer, and begin to run the copy of itself at the new location. Both the original code and the copy would then repeat these actions in an infinite loop to other computers on the ARPANET.
Though this has found reference time and again in the context of worms and denial of service, the basic working of the Morris worm was based on the discovery that the security of a TCP/IP connection rested in the sequence numbers and that it was possible to predict them
Blind IP spoofing involves predicting the sequence numbers that the victimized host will send in order to create a connection which appears to originate from the host. Before exploring blind spoofing further, let us take a look at sequence number prediction.
TCP sequence numbers are used to provide flow control and data integrity for TCP sessions. Every byte in a TCP session has a unique sequence number. Moreover, every TCP segment provides the sequence number of the initial byte (ISN), as part of the segment header. The initial sequence number does not start at zero for each session. Instead, the participants specify initial sequence numbers as part of the handshake process-a different ISN for each direction-and begin numbering the bytes sequentially from there.
Blind IP spoofing relies on the attacker's ability to predict sequence numbers as he is unable to sniff the communication between the two hosts by virtue of not being on the same network segment. He cannot spoof a trusted host on a different network and see the reply packets because the packets are not routed back to him. He cannot resort to ARP cache poisoning as well because routers do not route ARP broadcasts across the Internet. As he is not able to see the replies he is forced to anticipate the responses from the victim and prevent the host from sending a RST to the victim. The attacker then injects himself into the communication by predicting what sequence number the remote host is expecting from the victim. This is used extensively to exploit the trust relationships between users and remote machines, these services include NFS, telnet, IRC, etc.
IP spoofing is relatively easy to accomplish. The only pre-requisite on part of the attacker is to have root access on a machine in order to create raw packets. In order to establish a spoofed connection the attacker must know what sequence numbers are being used. Therefore, IP spoofing forces the attacker to have to predict the next sequence number.
The attacker can use "blind" hijacking, to send a command, but can never see the response. However, a common command would be to set a password allowing access from somewhere else on the net. The attack became famous when Kevin Mitnick used it to hack into Tsutomu Shimomura's computer network. The attack exploited the trust that Shimomura's machines had with the other network. By SYN flooding the trusted host, Mitnick was able to establish a short connection which was then used to gain access through traditional methods.
With Hijacking an attacker is taking over an existing session, which means he is relying on the legitimate user to make a connection and authenticate. Then take over the session.
With IP Spoofing there is no need to guess the sequence number since there is no session currently open with that IP address. The traffic would get back to the attacker only by using source routing. This is where the attacker tells the network how to route the output and input from a session, and he simply sniffs it from the network as it passes by him. Source routing is an IP option used today mainly by network managers to check connectivity. Normally, when an IP packet leaves a system, its path is controlled by the routers and their current configuration. Source routing provides a means to override the control of the routers.
When an attacker uses captured, reverse engineered or brute forced authentication tokens to take over the control of a legitimate user's session while he is in session, the session is said to be hijacked. Due to this attack, the legitimate user may loose access or be deprived of the normal functionality of the session to the attacker, who now acts with the user's privileges.
Most authentications occur at the beginning of a TCP session, this makes it possible for the attacker to gain access to a target machine. A popular method attackers adopt is to use source-routed IP packets. This allows an attacker to become a part of the target - host conversation by deceiving the IP packets to pass through his system. The attacker can also carry out the classic man-in-the-middle attack using a sniffing program to monitor the conversation.
In TCP session hijacking, a familiar aspect of the attacks is the carrying out of a denial-of-service (DoS) attack against the target / host to prevent it from responding by either forcing the machine to crash, or against the network connection to result in a heavy packet loss (e.g. SYN flood).
Session hijacking is even more difficult than IP address spoofing. In session hijacking, John would seek to insert himself into a session that Jane already had set up with \\Mail. John would wait until Jane established a session, then knock her off the air by some means and pick up the session as though he was her. As before, John would send a scripted set of packets to \\Mail but would not be able to see the responses. To do this, he would need to know the sequence number in use when he hijacked the session, which could be calculated knowing the ISN and the number of packets that have been exchanged.
Successful session hijacking is extremely difficult and only possible when a number of factors are under the attacker's control. Knowledge of the ISN would be the least of John's challenges. For instance, he would need a way to knock Jane off the air at will. He also would need a way to know the exact status of Jane's session at the moment he mounted his attack. Both of these require that John have far more knowledge about and control over the session than normally would be possible.
However, IP address spoofing attacks can only be successful if IP addresses are used for authentication. An attacker cannot perform IP address spoofing or session hijacking if per-packet integrity checking is executed. Similarly, neither IP address spoofing nor session hijacking are possible if the session uses encryption such as SSL or PPTP, as the attacker will not be able to participate in the key exchange. Therefore the essential requirements to hijack non-encrypted TCP communications can be listed as: Presence of non-encrypted session oriented traffic, ability to recognize TCP sequence numbers and predict the next sequence number (NSN) and capability to spoof a hosts MAC or IP address to receive communications which are not destined for the attackers host. If the attacker is on the local segment, they can sniff and predict the ISN+1 number and have the traffic routed back to them by poisoning the ARP cache.
Tags: Session hijacking