Attack prevention methods try to stop all well known signature based and broadcast based DDoS attacks from being launched in the first place or edge routers, keeps all the machines over Internet up to date with patches and fix security holes. Attack prevention schemes are not enough to stop DDoS attacks because there are always vulnerable to novel and mixed attack types for which signatures and patches arenot exist in the database.
Techniques for preventing against DDoS can be broadly divided into two categories: (i) General techniques, which are some common preventive measures i.e. system protection, replication of resources etc. that individual servers and ISPs should follow so they do not become part of DDoS attack process. (ii) Filtering techniques, which include ingress filtering, egress filtering, router based packet filtering, history based IP filtering, SAVE protocol etc.
A. General Techniques
1) Disabling unused services
The less there are applications and open ports in hosts, the less there are chance to exploit vulnerabilities by attackers. Therefore, if network services are not needed or unused, the services should be disabled to prevent attacks, e.g. UDP echo, character generation services .
2) Install latest security patches
Today, many DDoS attacks exploit vulnerabilities in target system. So removing known security holes by installing all relevant latest security patches prevents re-exploitation of vulnerabilities in the target system.
3) Disabling IP broadcast
Defense against attacks that use intermediate broadcasting nodes e.g. ICMP flood attacks, Smurf attacks etc. will be successful only if host computers and all the neighboring networks disable IP broadcast.
4) Firewalls
Firewalls can effectively prevent users from launching simple flooding type attacks from machines behind the firewall. Firewalls have simple rules such as to allow or deny protocols, ports or IP addresses. But some complex attack e.g. if there is an attack on port 80 (web service), firewalls cannot prevent that attack because they cannot distinguish good traffic from DoS attack traffic.
5) Global defense infrastructure
A global deployable defense infrastructure can prevent from many DDoS attacks by installing filtering rules in the most important routers of the Internet. As Internet is administered by various autonomous systems according their own local security policies, such type of global defense architecture is possible only in theory.
6) IP hopping
DDoS attacks can be prevented by changing location or IP address of the active server proactively within a pool of homogeneous servers or with a pre-specified set of IP address ranges. The victim computer’s IP address is invalidated by changing it with a new one. Once the IP addresses change is completed all internet routers will be informed and edge routers will drop the attacking packets. Although this action leaves the computer vulnerable because the attacker can launch the attack at the new IP address, this option is practical for DDoS attacks that are based on IP addresses. On the other hand, attackers can make this technique useless by adding a domain name service tracing function to the DDoS attack tools.
B. Filtering Techniques
1) Ingress/Egress filtering
Ingress Filtering, proposed by Ferguson et al., is a restrictive mechanism to drop traffic with IP addresses that do not match a domain prefix connected to the ingress router. Egress filtering is an outbound filter, which ensures that only assigned or allocated IP address space leaves the network. A key requirement for ingress or egress filtering is knowledge of the expected IP addresses at a particular port. For some networks with complicated topologies, it is not easy to obtain this knowledge.
One technique known as reverse path filtering can help to build this knowledge. This technique works as follows. Generally, a router always knows which networks are reachable via any of its interfaces. By looking up source addresses of the incoming traffic, it is possible to check whether the return path to that address would flow out the same interface as the packet arrived upon. If they do, these packets are allowed. Otherwise, they are dropped.
Unfortunately, this technique cannot operate effectively in real networks where asymmetric Internet routes are not uncommon. More importantly, both ingress and egress filtering can be applied not only to IP addresses, but also protocol type, port number, or any other criteria of importance. Both ingress and egress filtering provide some opportunities to throttle the attack power of DoS attacks. However, it is difficult to deploy ingress/egress filtering universally. If the attacker carefully chooses a network without ingress/egress filtering to launch a spoofed DoS attack, the attack can go undetected.
Moreover, if an attack spoofs IP addresses from within the subnet, the attack can go undetected as well. Nowadays DDoS attacks do not need to use source address spoofing to be effective. By exploiting a large number of compromised hosts, attackers do not need to use spoofing to take advantage of protocol vulnerabilities or to hide their locations. For example, each legitimate HTTP Web page request from 10,000 compromised hosts can bypass any ingress/egress filtering, but in combination they can constitute a powerful attack. Hence, ingress and egress filtering are ineffective to stop DDoS attacks.
2) Router based packet filtering
Route based filtering, proposed by Park and Lee, extends ingress filtering and uses the route information to filter out spoofed IP packets. It is based on the principle that for each link in the core of the Internet, there is only a limitedset of source addresses from which traffic on the link could have originated.
If an unexpected source address appears in an IP packet on a link, then it is assumed that the source address has been spoofed, and hence the packet can be filtered. RPF uses information about the BGP routing topology to filter traffic with spoofed source addresses. Simulation results show that a significant fraction of spoofed IP addresses can be filtered if RPF is implemented in at least 18% of ASs in the Internet. However, there are several limitations of this scheme. The first limitation relates to the implementation of RPF in practice. Given that the Internet contains more than 10,000 ASs, RPF
would need to be implemented in at least 1800 ASs in order to be effective, which is an onerous task to accomplish. The second limitation is that RPF may drop legitimate packets if there has recently been a route change. The third potential limitation is that RPF relies on valid BGP messages to configure the filter. If an attacker can hijack a BGP session and disseminate bogus BGP messages, then it is possible to mislead border routers to update filtering rules in favor of the attacker. RPF is effective against randomly spoofed DoS attacks. However, the filtering granularity of RPF is low. This means that the attack traffic can still bypass the RPF filters by carefully choosing the range of IP addresses to spoof. Hence, RPF is ineffective against DDoS attacks. The router-based packet filter is vulnerable to asymmetrical and dynamic Internet routing as it does not provide a scheme to update the routing information.
3) History based IP filtering
Generally, the set of source IP addresses that is seen during normal operation tends to remain stable. In contrast, during DoS attacks, most of the source IP addresses have not been seen before. Peng et al. relies on the above idea and use IP address database (IAD) to keep frequent source IP addresses. During an attack, if the source address of a packet is not in IAD, the packet is dropped. Hash based/Bloom filter techniques are used for fast searching of IP in IAD. This scheme is robust, and does not need the cooperation of the whole Internet community.
However, history based packet filtering scheme is ineffective when the attacks come from real IP addresses. In addition, it requires an offline database to keep track of IP addresses. Therefore, Cost of storage and information sharing is very high.
4) Capability based method
Capability based mechanisms provides destination a way to control the traffic directed towards itself. In this approach, source first sends request packets to its destination. Router marks (pre-capabilities) are added to request packet while passing through the router. The destination may or may not grant permission to the source to send. If permission is granted then destination returns the capabilities, if not then it does not supply the capabilities in the returned packet. The data packets carrying the capabilities are then send to thedestination via router. The main advantage achieved in this architecture is that the destination can now control the traffic according to its own policy, thereby reducing the chances of DDoS attack, as packets without capabilities are treated as legacy and might get dropped at the router when congestion happens.
However, these systems offer strong protection for established network flows, but responsible to generate a new attack type known as DOC (Denial of Capability), which prevents new capability-setup packets from reaching the destination, limits the value of these systems. In addition, these systems have high computational complexity and space requirement.
5) Secure overlay Service (SOS)
Secure Overlay Service proposed by Keromytis et al. defines an architecture called secure overlay service (SOS) to secure the communication between the confirmed users and the victim. All the traffic from a source point is verified by a secure overlay access point (SOAP). Authenticated traffic will be routed to a special overlay node called a beacon in an anonymous manner by consistent hash mapping. The beacon then forwards traffic to another special overlay node called a Filtering Technique
Benefsecret servlet for further authentication, and the secret servlet forwards verified traffic to the victim. The identity of the secret servlet is revealed to the beacon via a secure protocol, and remains a secret to the attacker. Finally, only traffic forwarded by the secret servlet chosen by the victim can pass its perimetric routers.
Secure Overlay Service (SOS) addresses the problem of how to guarantee the communication between legitimate users and a victim during DoS attacks. SOS can greatly reduce the likelihood of a successful attack. The power of SOS is based on the number and distribution level of SOAPs. However, wide deployment of SOAPs is a difficult DoS defense challenge. Moreover, the power of SOS is also based on the anonymous routing protocol within the overlay nodes. Unfortunately, the introduction of a new routing protocol is in itself another security issue. If an attacker is able to breach the security protection of some overlay node, then it can launch the attack from inside the overlay network. Moreover, if attackers can gain massive attack power, for example, via worm spread, all the SOAPs can be paralyzed, and the target's services will be disrupted.
6) SAVE: Source Address Validity Enforcement
Li et al. have proposed a new protocol called the Source Address Validity Enforcement (SAVE) protocol, which enables routers to update the information of expected source IP addresses on each link and block any IP packet with an unexpected source IP address. The aim of the SAVE protocol is to provide routers with information about the range of source IP addresses that should be expected at each interface. Similarly to the existing routing protocols, SAVE constantly propagates messages containing valid source address information from the source location to all destinations. Hence, each router along the way is able to build an incoming table that associates each link of the router with a set of valid source address blocks. SAVE is a protocol that enables the router to filter packets with spoofed source addresses using incoming tables. It overcomes the asymmetries of Internet routing by updating the incoming tables on each router periodically.
However, SAVE needs to change the routing protocol, which will take a long time to accomplish. If SAVE is not universally deployed, attackers can always spoof the IP addresses within networks that do not implement SAVE. Moreover, even if SAVE were universally deployed, attackers could still launch
DDoS attacks using non spoofed source addresses.
To conclude, attack prevention aims to solve IP spoofing, a fundamental weakness of the Internet. However, as attackers gain control of larger numbers of compromised computers, attackers can direct these “zombies” to attack using valid source addresses. Since the communication between attackers and “zombies” is encrypted, only “zombies” can be exposed instead of attackers. According to the Internet Architecture Working Group, the percentage of spoofed attacks is declining. Only four out of 1127 customer-impacting DDoS attacks on a large network used spoofed sources in 2004. Moreover, security awareness is still not enough, so expecting installation of security technologies and patches in large base of Internet seems to be an ambitious goal in near future. To add on, there exists no way out to enforce global deployment of a particular security mechanism. Therefore, relying on attack prevention schemes is not enough to stop DDoS attacks.
