{
  "id": "redis/redis-sentinel-down-after-master-failover",
  "signature": "SENTINEL: failover aborted - no reachable quorum after master change",
  "signature_zh": "哨兵：故障转移中止 - 主节点变更后无法达到仲裁数",
  "regex": "SENTINEL: failover aborted - no reachable quorum after master change",
  "domain": "redis",
  "category": "runtime_error",
  "subcategory": null,
  "root_cause": "After a master failover, the new master's configuration epochs are inconsistent, causing sentinel instances to lose quorum and abort the failover process.",
  "root_cause_type": "generic",
  "root_cause_zh": "在主节点故障转移后，新主节点的配置纪元不一致，导致哨兵实例失去仲裁并中止故障转移过程。",
  "versions": [
    {
      "version": "redis 6.2",
      "introduced": null,
      "deprecated": null,
      "removed": null,
      "behavior_change": null,
      "status": "active"
    },
    {
      "version": "redis 7.0",
      "introduced": null,
      "deprecated": null,
      "removed": null,
      "behavior_change": null,
      "status": "active"
    },
    {
      "version": "redis 7.2",
      "introduced": null,
      "deprecated": null,
      "removed": null,
      "behavior_change": null,
      "status": "active"
    },
    {
      "version": "redis 7.4",
      "introduced": null,
      "deprecated": null,
      "removed": null,
      "behavior_change": null,
      "status": "active"
    }
  ],
  "os_specific": {},
  "dead_ends": [
    {
      "action": "Restarting all sentinel instances simultaneously",
      "why_fails": "Simultaneous restart can cause all sentinels to lose state, delaying failover detection and potentially causing split-brain scenarios.",
      "fail_rate": 0.75,
      "condition": "",
      "sources": []
    },
    {
      "action": "Manually setting the old master back as leader",
      "why_fails": "The old master may be unreachable or have stale data; forcing it back can cause data loss and inconsistency with replicas.",
      "fail_rate": 0.95,
      "condition": "",
      "sources": []
    },
    {
      "action": "Increasing sentinel quorum to 5",
      "why_fails": "Higher quorum makes it harder to reach consensus, especially in degraded networks, and does not fix the epoch mismatch.",
      "fail_rate": 0.8,
      "condition": "",
      "sources": []
    }
  ],
  "workarounds": [
    {
      "action": "Manually reset the sentinel state for the affected master: `SENTINEL RESET mymaster` on each sentinel. Then monitor with `SENTINEL MASTER mymaster` to verify quorum.",
      "success_rate": 0.85,
      "how": "Manually reset the sentinel state for the affected master: `SENTINEL RESET mymaster` on each sentinel. Then monitor with `SENTINEL MASTER mymaster` to verify quorum.",
      "condition": "",
      "sources": []
    },
    {
      "action": "Force a new failover by running `SENTINEL FAILOVER mymaster` on the leader sentinel after ensuring all sentinels agree on the new master's configuration.",
      "success_rate": 0.8,
      "how": "Force a new failover by running `SENTINEL FAILOVER mymaster` on the leader sentinel after ensuring all sentinels agree on the new master's configuration.",
      "condition": "",
      "sources": []
    },
    {
      "action": "Check and synchronize sentinel configuration files across all nodes, ensuring `sentinel monitor` and `sentinel auth-pass` are identical, then restart sentinels one by one.",
      "success_rate": 0.9,
      "how": "Check and synchronize sentinel configuration files across all nodes, ensuring `sentinel monitor` and `sentinel auth-pass` are identical, then restart sentinels one by one.",
      "condition": "",
      "sources": []
    }
  ],
  "workarounds_zh": [
    "Manually reset the sentinel state for the affected master: `SENTINEL RESET mymaster` on each sentinel. Then monitor with `SENTINEL MASTER mymaster` to verify quorum.",
    "Force a new failover by running `SENTINEL FAILOVER mymaster` on the leader sentinel after ensuring all sentinels agree on the new master's configuration.",
    "Check and synchronize sentinel configuration files across all nodes, ensuring `sentinel monitor` and `sentinel auth-pass` are identical, then restart sentinels one by one."
  ],
  "transition_graph": {
    "leads_to": [],
    "preceded_by": [],
    "frequently_confused_with": []
  },
  "official_doc_url": "https://redis.io/docs/latest/operate/oss_and_stack/management/sentinel/",
  "official_doc_section": null,
  "error_code": "SENTINEL",
  "verification_tier": "ai_generated",
  "confidence": 0.83,
  "fix_success_rate": 0.85,
  "resolvable": "partial",
  "first_seen": "2024-06-20",
  "last_confirmed": "2024-06-01",
  "last_updated": "2024-06-01",
  "evidence_count": 1,
  "tags": [],
  "locale": "en",
  "aliases": []
}