Biblio

Network security is of vital importance, and Information Technology admins must always be vigilant. But they often lack the expertise and skills required to harden the network properly, in with the emergence of security threats. The router plays a significant role in maintaining operational security for an organization. When it comes to information security, information security professionals mainly focus on protecting items such as firewalls, virtual private networks, etc. Routers are the foundation of any network's communication method, which means all the network information passes through the routers, making them a desirable target. The proposed automation of the router security hardening solution will immediately improve the security of routers and ensure that they are updated and hardened with minimal human intervention and configuration changes. This is specially focused on small and medium-sized organizations lacking workforce and expertise on network security and will help secure the routers with less time consumption, cost, and increased efficiency. The solution consists of four primary functions, initial configuration, vulnerability fixing, compliance auditing, and rollback. These focus on all aspects of router security in a network, from its configuration when it is initially connected to the network to checking its compliance errors, continuously monitoring the vulnerabilities that need to be fixed, and ensuring that the behavior of the devices is stable and shows no abnormalities when it comes to configuration changes.

Context-dependent applications are relatively complex due to their multiple variations caused by context activation, especially in the presence of unanticipated adaptation. Testing these systems is challenging, as it is hard to reproduce the same execution environments. Therefore, a software failure caused by bugs is no exception. This paper presents a rollback mechanism to recover from software failures as part of a role-based runtime with support for unanticipated adaptation. The mechanism performs checkpoints before each adaptation and employs specialized sensors to detect bugs resulting from recent configuration changes. When the runtime detects a bug, it assumes that the bug belongs to the latest configuration. The runtime rolls back to the recent checkpoint to recover and subsequently notifies the developer to fix the bug and re-applying the adaptation through unanticipated adaptation. We prototype the concept as part of our role-based runtime engine LyRT and demonstrate the applicability of the rollback recovery mechanism for unanticipated adaptation in erroneous situations.