Clusters for high availability: a primer of HP-UX solutions
Clusters for high availability: a primer of HP-UX solutions
Storage: high-availability file server with heartbeat
Sys Admin - Storage
Enabling secure on-line DNS dynamic update
ACSAC '00 Proceedings of the 16th Annual Computer Security Applications Conference
A High-Availability Clustering Architecture with Data Integrity Guarantees
CLUSTER '01 Proceedings of the 3rd IEEE International Conference on Cluster Computing
Using the domain name system for system break-ins
SSYM'95 Proceedings of the 5th conference on USENIX UNIX Security Symposium - Volume 5
SSYM'95 Proceedings of the 5th conference on USENIX UNIX Security Symposium - Volume 5
A survey of software aging and rejuvenation studies
ACM Journal on Emerging Technologies in Computing Systems (JETC) - Special Issue on Reliability and Device Degradation in Emerging Technologies and Special Issue on WoSAR 2011
Hi-index | 0.00 |
Domain Name Systems (DNS) provide the mapping between easily remembered host names and their IP addresses. While domain name information is typically created and updated off-line, dynamic DNS updates allow clients to manage domain names online, in real time. The current secure DNS standards (DNSSEC) require private keys to be kept online to sign dynamic updates, leaving private keys subject to network-based attacks.In this work, we develop a secure implementation framework of DNS servers that voids the above requirement. Our approach, called Self-Cleansing Intrusion Tolerance (SCIT), strengthens DNSSEC through hardware redundancy. Our system uses a highly integrated cluster of DNS servers that constantly rotates the role of individual servers, handles one-server failures gracefully, confines the damages of successful intrusion to a limited time, and digitally signs dynamic updates by a clean server using the DNS zone key while keeping the key offline at all times. It is our belief that the availability and integrity of critical communications infrastructure, such as DNS, far outweigh the costs of hardware redundancy.In this paper, we present (1) the architecture of SCIT DNS clusters that achieves the above goals, (2) a secure Cluster Coordination Protocol (CCP) that servers in the cluster use to coordinate role changes without ever opening a port, and (3) the designs of our ongoing SCIT DNS cluster prototype. Preliminary experiences of our prototype show that role rotation and self-cleansing cycles are in the range of minutes, restricting the damages of even undetected but successful attacks to short time windows.