R.I.P.
π»
Ghosted
Characterizing Metastable Faults and Failures
May 31, 2026 Β· Grace Period Β· π SOSP 2026
Authors
Ali Farahbakhsh, Qingjie Lu, Lorenzo Alvisi, Andreas Haeberlen, Robbert Van Renesse
arXiv ID
2606.00942
Category
cs.OS: Operating Systems
Cross-listed
cs.DC
Citations
0
Venue
SOSP 2026
Abstract
Metastable failures are hard to detect, prevent, and mitigate. During a metastable failure, a system exhibits self-sustaining bad behavior even in the absence of adversarial conditions. Prior work focuses on symptoms and has portrayed metastable failures as instances of self-sustaining overload. This characterization leaves the underlying failure causes and dynamics unknown, and does not account for metastable failures that do not manifest as overload. We present the first causal characterization of metastable failures by identifying their origin in metastable faults, i.e., structural destabilizing cycles of interaction among systems components that, in isolation, are stabilizing. Metastable failures arise when scheduling decisions let these destabilizing interactions gain the upper hand over the individual components' stabilizing tendencies. We then derive a methodology to predict metastable failures, and to build metastable-fault-tolerant (MFT) systems. We apply our methodology to three case studies, showcasing the generality of our results.
Community Contributions
Found the code? Know the venue? Think something is wrong? Let us know!
π Similar Papers
In the same crypt β Operating Systems
R.I.P.
π»
Ghosted
Occlum: Secure and Efficient Multitasking Inside a Single Enclave of Intel SGX
R.I.P.
π»
Ghosted
LazyFP: Leaking FPU Register State using Microarchitectural Side-Channels
R.I.P.
π»
Ghosted
SGX-LKL: Securing the Host OS Interface for Trusted Execution
R.I.P.
π»
Ghosted
Optimal Virtual Cluster-based Multiprocessor Scheduling
R.I.P.
π»
Ghosted