A Study on the Security of Online Judge System Applied Sandbox Technology

Abstract

The majority of programming courses currently employ online judge systems as lesson materials. Online judge systems are becoming more common as the number of courses and persons studying computer science and information engineering grows. At the same time, there is an increase in the number of attacks against online judge systems; for example, Denial-Of-Service attacks, whose goal is to disrupt the target system by exhausting resources and blocking ordinary users from using the service normally. As a result, preventing attacks on online judge systems is becoming increasingly crucial. This research investigates and organizes these attack techniques, as well as develops a threat model for the online judge system by the STRIDE threat model approach, which provides a way to classify attacks into six categories. This research also designs code analysis rules and implements a code analysis tool. This tool can assist developers in analyzing the existing online judge system to determine whether the judge system is vulnerable to attack and dealing with vulnerability as soon as feasible to improve the judge system’s security. After enhancing the security of online judge systems, the system can foster trust and reliability among users as benefits.

Design of an Online Programming Platform and a Study on Learners’ Testing Ability

“…Due to security considerations [31,32], learners' programs are submitted to the server for execution and results retrieval. To prevent the submission of malicious programs or those with infinite loops that could potentially harm the server, we implemented Docker container technology [33].…”