Getty Images
What to know about UDP vulnerabilities and security
UDP is a simple protocol, but it has inherent vulnerabilities that make it prone to attacks, such as limited packet verification, IP spoofing and DDoS attacks.
Despite its ubiquity in computer networking, User Datagram Protocol is susceptible to security vulnerabilities and attacks.
UDP is a simple protocol because it doesn't require connection setup or acknowledgement exchanges to send data packets to their destinations. It just transfers the packet and doesn't know if the data reaches the destination or drops off somewhere along the path.
Applications that require a quick request and response, such as DNS, Dynamic Host Configuration Protocol (DHCP), audio and video, typically use UDP. Those applications can't detect whether UDP receives the request, but they aren't seriously affected by lost packets -- perhaps some audio static or video flickers.
But this limited packet verification subjects UDP to vulnerabilities that don't affect other networking protocols, such as TCP. For example, TCP must go through a connection setup process before it responds to incoming data packets from the other end of a connection. It ignores incoming data packets that aren't from a connection. In contrast, UDP applications respond to any received request because UDP doesn't use an established connection.
Common UDP attacks
Bad actors might use port scan attacks to gauge UDP services as a potential target. A port scan attack sends packets to a host and uses its replies to learn about the system and find vulnerabilities. UDP services could also be susceptible to hacking if they have an exploit or a bug that enables remote access and overflow.
DoS and DDoS attacks can disrupt UDP and other protocols, like TCP. Malicious actors create DDoS attacks by inserting malware on a system -- sometimes thousands. Attackers use the inserted software on all infected systems to bombard the one under attack. Replies go to the infected system and are discarded, but the incoming packet has contributed to a DoS attack.
Attackers might also use IP spoofing to insert an invented source address into packets used in the attack. The system under attack responds whether or not the address belongs to an existing system. It is difficult to protect against IP spoofing because an attacker might use many spoofed addresses. A filter that depends on detecting a high volume from a specific source address might not be able to filter out the spoofed addresses.
How to defend against UDP attacks
A single attack might be enough to slow down useful processing. The volume of legitimate traffic and the expected volume of attacks can determine the type of defense. For example, some websites might attract a greater number of attacks, as well as attacks that are more intense. Protection against a distributed attack requires some type of external filtering.
A number of options are available to defend against UDP vulnerabilities, including the following:
- Enterprises can protect valuable information by setting up VPNs to legitimate request sources.
- Implement incoming packet inspection and filtering in a VM that runs in the same server as the VM executing the application. The server containing both VMs has a finite amount of processing capability, and running packet analysis and filtering in a VM still slows down the application.
- Firewall vendors include packet filtering for some types of attacks but often include other types of filtering. While a firewall can stop low-level attacks, more intensive attacks, including distributed attacks, can overwhelm a firewall.
- Use intrusion detection and removal products to eliminate certain incoming attacks.
- Applications can execute in cloud environments that have configured protection services. Applications that use SaaS receive some type of protection by the cloud provider.
- Content delivery networks (CDNs) also provide protection for geographically distributed systems served by a website. CDNs are used by frequently accessed web servers to speed up high-volume sites. Each CDN system has a copy of the website. A request to one of these sites redirects to the nearest CDN. DDoS attacks are also directed to the nearest CDN server, so each CDN server receives a fraction of the attack, making it easier to defend more effectively.
Attacks against UDP services will undoubtedly continue. As new attack methods might emerge, new defenses can help protect against them. Even with existing defenses, attacks require network staff to look out for vulnerabilities and implement appropriate defenses.
David B. Jacobs has more than 30 years of networking industry experience. He has managed leading-edge network development projects and consulted for Fortune 500 companies, as well as software startups.
