{
  "slug": "metastable-failure",
  "term": "Metastable failure",
  "aliases": [
    "metastable system failure"
  ],
  "definition": "A metastable failure is a persistently degraded system state sustained by feedback after its initiating disturbance has ended.",
  "note": "Separate the trigger from the mechanism maintaining low useful throughput; the absence of a new hardware fault does not establish recovery. [bronson-metastable]\n\nThe OSDI 2022 cache experiment used a one-hour recovery observation window, not proof of permanent non-recovery. Its incident study is not a population failure-rate estimate. [huang-metastable]",
  "related": [
    "retry-amplification",
    "io-timeout",
    "fail-slow"
  ],
  "citations": [
    {
      "id": "bronson-metastable",
      "title": "Bronson et al. Metastable Failures in Distributed Systems. HotOS 2021; DOI 10.1145/3458336.3465286, sections 1-3.",
      "url": "https://sigops.org/s/conferences/hotos/2021/papers/hotos21-s11-bronson.pdf",
      "accessed": "2026-10-20"
    },
    {
      "id": "huang-metastable",
      "title": "Huang et al. Metastable Failures in the Wild. OSDI 2022, pp. 73-90, sections 2.1, 4 and 5.3.",
      "url": "https://www.usenix.org/system/files/osdi22-huang-lexiang.pdf",
      "accessed": "2026-10-20"
    }
  ],
  "status": "unknown",
  "advantages": [],
  "limitations": [],
  "supersedes": [],
  "supersededBy": [],
  "firstSeen": null,
  "lastSeen": null,
  "updated": "2026-10-20"
}