| Name |
TCP ACK Scan |
|
| Likelyhood of attack |
Typical severity |
| Low |
Low |
|
| Summary |
An adversary uses TCP ACK segments to gather information about firewall or ACL configuration. The purpose of this type of scan is to discover information about filter configurations rather than port state. This type of scanning is rarely useful alone, but when combined with SYN scanning, gives a more complete picture of the type of firewall rules that are present. When a TCP ACK segment is sent to a closed port, or sent out-of-sync to a listening port, the RFC 793 expected behavior is for the device to respond with a RST. Getting RSTs back in response to a ACK scan gives the attacker useful information that can be used to infer the type of firewall present. Stateful firewalls will discard out-of-sync ACK packets, leading to no response. When this occurs the port is marked as filtered. When RSTs are received in response, the ports are marked as unfiltered, as the ACK packets solicited the expected behavior from a port. When combined with SYN techniques an attacker can gain a more complete picture of which types of packets get through to a host and thereby map out its firewall rule-set. ACK scanning, when combined with SYN scanning, also allows the adversary to analyze whether a firewall is stateful or non-stateful (described in notes). TCP ACK Scans are somewhat faster and more stealthy than other types of scans but often requires rather sophisticated analysis by an experienced person. A skilled adversary may use this method to map out firewall rules, but the results of ACK scanning will be less useful to a novice. |
| Prerequisites |
The adversary requires logical access to the target network. ACK scanning requires the use of raw sockets, and thus cannot be performed from some Windows systems (Windows XP SP 2, for example). On Unix and Linux, raw socket manipulations require root privileges. |
| Execution Flow |
| Step |
Phase |
Description |
Techniques |
| 1 |
Experiment |
An adversary sends TCP packets with the ACK flag set and that are not associated with an existing connection to target ports. |
|
| 2 |
Experiment |
An adversary uses the response from the target to determine the port's state. If a RST packet is received the target port is either closed or the ACK was sent out-of-sync. If no response is received, the target is likely using a stateful firewall. |
|
|
| Solutions | |
| Related Weaknesses |
|
CWE ID
|
Description
|
| CWE-200 |
Exposure of Sensitive Information to an Unauthorized Actor |
|
| Related CAPECS |
|
CAPEC ID
|
Description
|
| CAPEC-300 |
An adversary uses a combination of techniques to determine the state of the ports on a remote target. Any service or application available for TCP or UDP networking will have a port open for communications over the network. Although common services have assigned port numbers, services and applications can run on arbitrary ports. Additionally, port scanning is complicated by the potential for any machine to have up to 65535 possible UDP or TCP services. The goal of port scanning is often broader than identifying open ports, but also give the adversary information concerning the firewall configuration. Depending upon the method of scanning that is used, the process can be stealthy or more obtrusive, the latter being more easily detectable due to the volume of packets involved, anomalous packet traits, or system logging. Typical port scanning activity involves sending probes to a range of ports and observing the responses. There are four port statuses that this type of attack aims to identify: open, closed, filtered, and unfiltered. For strategic purposes it is useful for an adversary to distinguish between an open port that is protected by a filter vs. a closed port that is not protected by a filter. Making these fine grained distinctions is requires certain scan types. Collecting this type of information tells the adversary which ports can be attacked directly, which must be attacked with filter evasion techniques like fragmentation, source port scans, and which ports are unprotected (i.e. not firewalled) but aren't hosting a network service. An adversary often combines various techniques in order to gain a more complete picture of the firewall filtering mechanisms in place for a host. |
|