• Proceedings of the Korean Information Science Society Conference
• /
• 2004.10a
• /
• pp.643-645
• /
• 2004

네트워크가 복잡해지면서 다양한 형태의 위험에 노출된다. 일반적인 보안 솔루션으로 사용하는 방화벽(Firewall)이나 침입탐지시스템(IDS)은 허가받지 않은 외부의 접속이나 알려진 공격만을 차단하는 단순하고 수동적인 시스템이다. 이에 반해 허니팟은 웹서버와 같은 실제 Front-End 시스템과 유사하거나, 밀접한 관련을 갖고 직접적으로 반응하므로 신뢰성 높은 실시간 정보를 얻을 수 있어서 관리자가 다양한 위협에 능동적이고 효과적으로 대응할 수 있다. 실제로 웜이나 DRDoS(Distributed Reflection DoS)등 수분만에 네트워크를 점유하는 자동화 공격과 함께 시스템 구성 취약점을 파고들거나 계정 획득을 통한 DB서버 등의 Back-End 시스템에 대한 수동 공격이 혼재한다. 따라서 시스템 전반적인 관리의 중요성이 강조되고 있다. 본 논문에서는 그간의 실험결과를 바탕으로 가상 머신으로 허니팟을 구성하고 특성별로 포트를 분리하여 관리하는 허니넷을 제안하고자 한다 이를 통해 1) 유연한 보안 시스템 구성이 가능하고 2) 관리 효율이 높아지며 3) 하드웨어 도입 비용 절감을 통해 시스템의 TCO(Totai Cost of Ownership)를 감소시키는 효과를 기대할 수 있다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.143-145
• /
• 2022

With the spread of the Internet around the world, devices connected to the Internet are gradually increasing. In addition, the number of distributed reflection service attacks (DrDoS), an attack that maliciously requests large responses by deceiving IPs as if the attacker was a victim, using vulnerabilities in application protocols such as DNS, NTP, and CLDAP, is increasing rapidly. It is believed that the security threat of distributed reflection service attacks will not disappear unless ISPs establish appropriate countermeasures to IP Spoofing. Therefore, this paper describes the security threat and response status of distributed reflection service attacks based on IP Spoofing.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.20 no.1
• /
• pp.109-118
• /
• 2024

Recently, a social issue has arisen involving RDDoS attacks following the sending of threatening emails to security administrators of companies and institutions. According to a report published by the Korea Internet & Security Agency and the Ministry of Science and ICT, survey results indicate that DDoS attacks are increasing. However, the top response in the survey highlighted the difficulty in countering DDoS attacks due to issues related to security personnel and costs. In responding to DDoS attacks, administrators typically detect anomalies through traffic monitoring, utilizing security equipment and programs to identify and block attacks. They also respond by employing DDoS mitigation solutions offered by external security firms. However, a challenge arises from the initial failure in early response to DDoS attacks, leading to frequent use of detection and mitigation measures. This issue, compounded by increased costs, poses a problem in effectively countering DDoS attacks. In this paper, we propose a system that creates detection rules, periodically collects traffic using mail detection and IDS, notifies administrators when rules match, and Based on predefined threshold, we use IPS to block traffic or DDoS mitigation. In the absence of DDoS mitigation, the system sends urgent notifications to administrators and suggests that you apply for and use of a cyber shelter or DDoS mitigation. Based on this, the implementation showed that network traffic was reduced from 400 Mbps to 100 Mbps, enabling DDoS response. Additionally, due to the time and expense involved in modifying detection and blocking rules, it is anticipated that future research could address cost-saving through reduced usage of DDoS mitigation by utilizing artificial intelligence for rule creation and modification, or by generating rules in new ways.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.108-110
• /
• 2021

Since 2015, attacks using the IoT protocol have been continuously reported. Among various IoT protocols, attackers attempt DDoS attacks using SSDP(Simple Service Discovery Protocol), and as statistics of cyber shelters, Korea has about 1 million open SSDP servers. Vulnerable SSDP servers connected to the Internet can generate more than 50Gb of traffic and the risk of attack increases gradually. Until recently, distributed denial of service attacks and distributed reflective denial of service attacks have been a security issue. Accordingly, the purpose of this study is to analyze the request packet of the existing SSDP protocol to identify an amplification attack and to avoid a response when an amplification attack is suspected, thereby preventing network load due to the occurrence of a large number of response packets due to the role of traffic reflection amplification.