Written by: Mio Kanzaki(Security Analyst (AI))Reviewed by: 千葉圭介Updated
A DDoS attack is an attack that sends a flood of traffic at once and makes a site unusable. CISA explains it as multiple machines working together to attack a single target. DDoS stands for distributed denial-of-service.
How does a DDoS attack work?
Imagine thousands of people calling a popular shop at the same moment. The lines are blocked, and the real customers cannot get through however long they try. A DDoS attack likewise sends more requests than a server can process, and locks out legitimate users.
According to CISA, botnets (collections of hijacked machines) are often used for the attack, and IoT devices whose passwords are still at the factory default are hard for their owners to notice when hijacked. The material from the National Police Agency and NISC lists three kinds of methods that were confirmed. One is a TCP (SYN) flood, which sends a huge number of connection requests alone. Another is an HTTP flood, which sends a huge number of HTTP requests. The third is a UDP flood, which sends a huge amount of UDP data using a forged sender. Related measures are gathered on the DDoS countermeasures page.
Why does a DDoS attack matter?
The National Police Agency and NISC material reports that in September 2022, DDoS attacks on the websites of government-related bodies and critical infrastructure operators in Japan were confirmed several times. About 99% of the source IP addresses were ones assigned overseas. The traffic volume was about 7 Mbps for the weaker attacks and about 100 Gbps for the stronger ones.
The National Police Agency and NISC say that not every DDoS attack can be prevented in advance, and ask organizations to start with measures close at hand. Examples that are easy to take on the server side are as follows.
- For a site aimed at the domestic market, restrict access from IP addresses assigned overseas.
- Introduce a CDN (a mechanism that delivers a site's content from bases in many places) or a WAF, and make the necessary settings.
- Limit the number of accesses from the same IP address, and review the timeout settings (rate limiting).
- Know the normal traffic volume, and set up an alert for abnormal traffic.
Alongside these, sort your systems into those that need a high level of protection and those that can tolerate a stop for a certain period. The National Police Agency and NISC also list preparing to inform users through social media or a sorry page when the site cannot be reached. CISA likewise recommends using a DoS protection service and drawing up a recovery plan.
To get help with the contact procedure for when a site stops and with deciding which systems to protect first, contact us.
Related terms
- DoS attack: an attack that sends a flood of traffic from a single machine to stop a service.
- WAF: a mechanism that stops malicious traffic at the entrance.
Next step
The whole picture of preparing for DDoS attacks is gathered on the DDoS countermeasures page. How your site's public status looks from outside can be checked with the free URL check. For the contact procedure when a site stops, and for deciding the order of protection, contact us.
Frequently asked questions
Can a small or mid-sized company's site be a DDoS target?
It can. According to CISA, there is no way to avoid being a target of DoS or DDoS attacks completely. Administrators can take steps to reduce the damage.
How do we notice a DDoS attack?
CISA lists symptoms such as an extremely slow network, being unable to reach a particular site, and being unable to open any site. These are hard to tell apart from ordinary faults such as maintenance work, so monitoring traffic volume is described as the most reliable method.
Frequently asked questions
Can a small or mid-sized company's site be a DDoS target?
It can. According to CISA, there is no way to avoid being a target of DoS or DDoS attacks completely. Administrators can take steps to reduce the damage.
How do we notice a DDoS attack?
CISA lists symptoms such as an extremely slow network, being unable to reach a particular site, and being unable to open any site. These are hard to tell apart from ordinary faults such as maintenance work, so monitoring traffic volume is described as the most reliable method.