{
  "schema_version": 1,
  "demo_name": "gpu_load_step_pcie_downtrain",
  "data_origin": "synthetic",
  "scope_note": "All events, measurements, and failure IDs are synthetic and public-safe. They illustrate a correlation and closure model; they are not employer logs, customer data, or proof of a production AI-server deployment.",
  "summary": {
    "event_count": 28,
    "case_count": 3,
    "closed_count": 3,
    "open_count": 0,
    "first_ts_ms": 0,
    "last_ts_ms": 7000,
    "duration_ms": 7000
  },
  "cases": [
    {
      "case_id": "case_power_pcie",
      "case_title": "GPU load-step rail droop causes PCIe downtrain",
      "correlation_id": "F-2026-001",
      "status": "closed",
      "event_count": 12,
      "duration_ms": 3100,
      "first_symptom": "GPU0 core rail droop exceeds validation threshold during load step.",
      "closure": "Regression passes cold boot, warm reboot, load-step, and thermal soak coverage.",
      "domains": [
        "BMC",
        "PCIe",
        "Power",
        "Validation",
        "Workload"
      ],
      "events": [
        {
          "ts_ms": 0,
          "case_id": "case_power_pcie",
          "case_title": "GPU load-step rail droop causes PCIe downtrain",
          "event_type": "workload_marker",
          "source": "host",
          "workload": "boot",
          "phase": "start",
          "message": "System enters validation boot sequence."
        },
        {
          "ts_ms": 1180,
          "case_id": "case_power_pcie",
          "case_title": "GPU load-step rail droop causes PCIe downtrain",
          "event_type": "rail_sample",
          "source": "pmbus",
          "rail_id": "GPU0_CORE",
          "voltage_v": 0.842,
          "nominal_v": 0.85,
          "current_a": 168.4,
          "state": "nominal",
          "message": "GPU0 core rail is stable before workload step."
        },
        {
          "ts_ms": 1800,
          "case_id": "case_power_pcie",
          "case_title": "GPU load-step rail droop causes PCIe downtrain",
          "event_type": "workload_marker",
          "source": "host",
          "workload": "gpu_stress",
          "phase": "load_step",
          "message": "GPU stress workload transitions from idle to high-current phase."
        },
        {
          "ts_ms": 1806,
          "case_id": "case_power_pcie",
          "case_title": "GPU load-step rail droop causes PCIe downtrain",
          "event_type": "rail_sample",
          "source": "pmbus",
          "rail_id": "GPU0_CORE",
          "voltage_v": 0.781,
          "nominal_v": 0.85,
          "current_a": 312.8,
          "state": "degraded",
          "correlation_id": "F-2026-001",
          "message": "GPU0 core rail droop exceeds validation threshold during load step."
        },
        {
          "ts_ms": 1812,
          "case_id": "case_power_pcie",
          "case_title": "GPU load-step rail droop causes PCIe downtrain",
          "event_type": "bmc_event",
          "source": "bmc",
          "severity": "warning",
          "code": "PWR_GPU0_UV_WARN",
          "correlation_id": "F-2026-001",
          "message": "BMC records PMBus undervoltage warning on GPU0_CORE."
        },
        {
          "ts_ms": 1826,
          "case_id": "case_power_pcie",
          "case_title": "GPU load-step rail droop causes PCIe downtrain",
          "event_type": "pcie_event",
          "source": "host",
          "link_id": "CPU0-SW0-GPU0",
          "negotiated_speed": "Gen5",
          "negotiated_width": "x16",
          "state": "training_retry",
          "correlation_id": "F-2026-001",
          "message": "PCIe link enters training retry window after rail droop."
        },
        {
          "ts_ms": 1872,
          "case_id": "case_power_pcie",
          "case_title": "GPU load-step rail droop causes PCIe downtrain",
          "event_type": "pcie_event",
          "source": "host",
          "link_id": "CPU0-SW0-GPU0",
          "negotiated_speed": "Gen4",
          "negotiated_width": "x16",
          "state": "downtrained",
          "correlation_id": "F-2026-001",
          "message": "Link downtrains after repeated training retries."
        },
        {
          "ts_ms": 1930,
          "case_id": "case_power_pcie",
          "case_title": "GPU load-step rail droop causes PCIe downtrain",
          "event_type": "validation_event",
          "source": "lab",
          "test": "oscilloscope_capture",
          "verdict": "fail",
          "closure": "open",
          "correlation_id": "F-2026-001",
          "message": "Scope capture confirms rail droop overlaps PCIe training retry window."
        },
        {
          "ts_ms": 2200,
          "case_id": "case_power_pcie",
          "case_title": "GPU load-step rail droop causes PCIe downtrain",
          "event_type": "validation_event",
          "source": "lab",
          "test": "mitigation_plan",
          "verdict": "reviewed",
          "closure": "open",
          "correlation_id": "F-2026-001",
          "message": "Mitigation selected: add bulk capacitance, tune compensation, extend OCP blanking, and guard retimer reset release."
        },
        {
          "ts_ms": 2920,
          "case_id": "case_power_pcie",
          "case_title": "GPU load-step rail droop causes PCIe downtrain",
          "event_type": "rail_sample",
          "source": "pmbus",
          "rail_id": "GPU0_CORE",
          "voltage_v": 0.836,
          "nominal_v": 0.85,
          "current_a": 316.2,
          "state": "nominal",
          "correlation_id": "F-2026-001",
          "message": "Retest rail droop stays within target after mitigation."
        },
        {
          "ts_ms": 2965,
          "case_id": "case_power_pcie",
          "case_title": "GPU load-step rail droop causes PCIe downtrain",
          "event_type": "pcie_event",
          "source": "host",
          "link_id": "CPU0-SW0-GPU0",
          "negotiated_speed": "Gen5",
          "negotiated_width": "x16",
          "state": "stable",
          "correlation_id": "F-2026-001",
          "message": "PCIe link trains at expected speed and width after mitigation."
        },
        {
          "ts_ms": 3100,
          "case_id": "case_power_pcie",
          "case_title": "GPU load-step rail droop causes PCIe downtrain",
          "event_type": "validation_event",
          "source": "lab",
          "test": "regression_matrix",
          "verdict": "pass",
          "closure": "closed",
          "correlation_id": "F-2026-001",
          "message": "Regression passes cold boot, warm reboot, load-step, and thermal soak coverage."
        }
      ]
    },
    {
      "case_id": "case_retimer_refclk",
      "case_title": "Retimer reset race causes Gen5 training retries",
      "correlation_id": "F-2026-002",
      "status": "closed",
      "event_count": 8,
      "duration_ms": 700,
      "first_symptom": "PCIe link retries during Gen5 training on warm reboot.",
      "closure": "Warm reboot regression closes with stable Gen5 training over repeated cycles.",
      "domains": [
        "BMC",
        "PCIe",
        "Validation",
        "Workload"
      ],
      "events": [
        {
          "ts_ms": 4000,
          "case_id": "case_retimer_refclk",
          "case_title": "Retimer reset race causes Gen5 training retries",
          "event_type": "workload_marker",
          "source": "host",
          "workload": "warm_reboot",
          "phase": "start",
          "message": "Warm reboot validation begins for fabric stability check."
        },
        {
          "ts_ms": 4060,
          "case_id": "case_retimer_refclk",
          "case_title": "Retimer reset race causes Gen5 training retries",
          "event_type": "bmc_event",
          "source": "bmc",
          "severity": "info",
          "code": "RST_RETIMER_RELEASE",
          "correlation_id": "F-2026-002",
          "message": "BMC releases retimer reset shortly after reference clock lock indication."
        },
        {
          "ts_ms": 4074,
          "case_id": "case_retimer_refclk",
          "case_title": "Retimer reset race causes Gen5 training retries",
          "event_type": "pcie_event",
          "source": "host",
          "link_id": "CPU0-SW1-GPU3",
          "negotiated_speed": "Gen5",
          "negotiated_width": "x16",
          "state": "training_retry",
          "correlation_id": "F-2026-002",
          "message": "PCIe link retries during Gen5 training on warm reboot."
        },
        {
          "ts_ms": 4108,
          "case_id": "case_retimer_refclk",
          "case_title": "Retimer reset race causes Gen5 training retries",
          "event_type": "pcie_event",
          "source": "host",
          "link_id": "CPU0-SW1-GPU3",
          "negotiated_speed": "Gen4",
          "negotiated_width": "x16",
          "state": "downtrained",
          "correlation_id": "F-2026-002",
          "message": "Link falls back to Gen4 after retry budget is exhausted."
        },
        {
          "ts_ms": 4140,
          "case_id": "case_retimer_refclk",
          "case_title": "Retimer reset race causes Gen5 training retries",
          "event_type": "validation_event",
          "source": "lab",
          "test": "refclk_reset_capture",
          "verdict": "fail",
          "closure": "open",
          "correlation_id": "F-2026-002",
          "message": "Capture shows retimer reset release is too close to refclk lock settle time."
        },
        {
          "ts_ms": 4300,
          "case_id": "case_retimer_refclk",
          "case_title": "Retimer reset race causes Gen5 training retries",
          "event_type": "validation_event",
          "source": "lab",
          "test": "mitigation_plan",
          "verdict": "reviewed",
          "closure": "open",
          "correlation_id": "F-2026-002",
          "message": "Mitigation selected: add 25 ms guard time after refclk lock before retimer reset release."
        },
        {
          "ts_ms": 4620,
          "case_id": "case_retimer_refclk",
          "case_title": "Retimer reset race causes Gen5 training retries",
          "event_type": "pcie_event",
          "source": "host",
          "link_id": "CPU0-SW1-GPU3",
          "negotiated_speed": "Gen5",
          "negotiated_width": "x16",
          "state": "stable",
          "correlation_id": "F-2026-002",
          "message": "PCIe link trains at Gen5 x16 after reset guard change."
        },
        {
          "ts_ms": 4700,
          "case_id": "case_retimer_refclk",
          "case_title": "Retimer reset race causes Gen5 training retries",
          "event_type": "validation_event",
          "source": "lab",
          "test": "warm_reboot_regression",
          "verdict": "pass",
          "closure": "closed",
          "correlation_id": "F-2026-002",
          "message": "Warm reboot regression closes with stable Gen5 training over repeated cycles."
        }
      ]
    },
    {
      "case_id": "case_thermal_sensor",
      "case_title": "Thermal sensor mismatch masks accelerator hotspot",
      "correlation_id": "F-2026-003",
      "status": "closed",
      "event_count": 8,
      "duration_ms": 1000,
      "first_symptom": "Local VRM hotspot estimate exceeds derating threshold while inlet remains normal.",
      "closure": "Thermal soak passes with updated sensor map, inspection step, and derating policy.",
      "domains": [
        "BMC",
        "Thermal",
        "Validation",
        "Workload"
      ],
      "events": [
        {
          "ts_ms": 6000,
          "case_id": "case_thermal_sensor",
          "case_title": "Thermal sensor mismatch masks accelerator hotspot",
          "event_type": "workload_marker",
          "source": "host",
          "workload": "thermal_soak",
          "phase": "steady_state",
          "message": "Thermal soak reaches steady workload phase."
        },
        {
          "ts_ms": 6120,
          "case_id": "case_thermal_sensor",
          "case_title": "Thermal sensor mismatch masks accelerator hotspot",
          "event_type": "thermal_sample",
          "source": "bmc",
          "sensor_id": "GPU2_INLET",
          "temp_c": 31.4,
          "state": "nominal",
          "message": "Inlet temperature appears normal during thermal soak."
        },
        {
          "ts_ms": 6130,
          "case_id": "case_thermal_sensor",
          "case_title": "Thermal sensor mismatch masks accelerator hotspot",
          "event_type": "thermal_sample",
          "source": "bmc",
          "sensor_id": "GPU2_VRM_HOTSPOT",
          "temp_c": 91.8,
          "state": "degraded",
          "correlation_id": "F-2026-003",
          "message": "Local VRM hotspot estimate exceeds derating threshold while inlet remains normal."
        },
        {
          "ts_ms": 6134,
          "case_id": "case_thermal_sensor",
          "case_title": "Thermal sensor mismatch masks accelerator hotspot",
          "event_type": "bmc_event",
          "source": "bmc",
          "severity": "warning",
          "code": "THERM_GPU2_DERATE",
          "correlation_id": "F-2026-003",
          "message": "BMC derates GPU2 due to local hotspot policy."
        },
        {
          "ts_ms": 6150,
          "case_id": "case_thermal_sensor",
          "case_title": "Thermal sensor mismatch masks accelerator hotspot",
          "event_type": "validation_event",
          "source": "lab",
          "test": "sensor_map_review",
          "verdict": "fail",
          "closure": "open",
          "correlation_id": "F-2026-003",
          "message": "Review finds inlet sensor cannot explain VRM hotspot; local sensor placement and cold-plate contact require inspection."
        },
        {
          "ts_ms": 6420,
          "case_id": "case_thermal_sensor",
          "case_title": "Thermal sensor mismatch masks accelerator hotspot",
          "event_type": "validation_event",
          "source": "lab",
          "test": "mechanical_inspection",
          "verdict": "reviewed",
          "closure": "open",
          "correlation_id": "F-2026-003",
          "message": "TIM contact adjusted and VRM zone threshold added to service checklist."
        },
        {
          "ts_ms": 6900,
          "case_id": "case_thermal_sensor",
          "case_title": "Thermal sensor mismatch masks accelerator hotspot",
          "event_type": "thermal_sample",
          "source": "bmc",
          "sensor_id": "GPU2_VRM_HOTSPOT",
          "temp_c": 78.2,
          "state": "nominal",
          "correlation_id": "F-2026-003",
          "message": "Retest hotspot temperature remains below derating threshold."
        },
        {
          "ts_ms": 7000,
          "case_id": "case_thermal_sensor",
          "case_title": "Thermal sensor mismatch masks accelerator hotspot",
          "event_type": "validation_event",
          "source": "lab",
          "test": "thermal_soak_regression",
          "verdict": "pass",
          "closure": "closed",
          "correlation_id": "F-2026-003",
          "message": "Thermal soak passes with updated sensor map, inspection step, and derating policy."
        }
      ]
    }
  ],
  "timeline": [
    {
      "ts_ms": 0,
      "case_id": "case_power_pcie",
      "case_title": "GPU load-step rail droop causes PCIe downtrain",
      "event_type": "workload_marker",
      "source": "host",
      "workload": "boot",
      "phase": "start",
      "message": "System enters validation boot sequence."
    },
    {
      "ts_ms": 1180,
      "case_id": "case_power_pcie",
      "case_title": "GPU load-step rail droop causes PCIe downtrain",
      "event_type": "rail_sample",
      "source": "pmbus",
      "rail_id": "GPU0_CORE",
      "voltage_v": 0.842,
      "nominal_v": 0.85,
      "current_a": 168.4,
      "state": "nominal",
      "message": "GPU0 core rail is stable before workload step."
    },
    {
      "ts_ms": 1800,
      "case_id": "case_power_pcie",
      "case_title": "GPU load-step rail droop causes PCIe downtrain",
      "event_type": "workload_marker",
      "source": "host",
      "workload": "gpu_stress",
      "phase": "load_step",
      "message": "GPU stress workload transitions from idle to high-current phase."
    },
    {
      "ts_ms": 1806,
      "case_id": "case_power_pcie",
      "case_title": "GPU load-step rail droop causes PCIe downtrain",
      "event_type": "rail_sample",
      "source": "pmbus",
      "rail_id": "GPU0_CORE",
      "voltage_v": 0.781,
      "nominal_v": 0.85,
      "current_a": 312.8,
      "state": "degraded",
      "correlation_id": "F-2026-001",
      "message": "GPU0 core rail droop exceeds validation threshold during load step."
    },
    {
      "ts_ms": 1812,
      "case_id": "case_power_pcie",
      "case_title": "GPU load-step rail droop causes PCIe downtrain",
      "event_type": "bmc_event",
      "source": "bmc",
      "severity": "warning",
      "code": "PWR_GPU0_UV_WARN",
      "correlation_id": "F-2026-001",
      "message": "BMC records PMBus undervoltage warning on GPU0_CORE."
    },
    {
      "ts_ms": 1826,
      "case_id": "case_power_pcie",
      "case_title": "GPU load-step rail droop causes PCIe downtrain",
      "event_type": "pcie_event",
      "source": "host",
      "link_id": "CPU0-SW0-GPU0",
      "negotiated_speed": "Gen5",
      "negotiated_width": "x16",
      "state": "training_retry",
      "correlation_id": "F-2026-001",
      "message": "PCIe link enters training retry window after rail droop."
    },
    {
      "ts_ms": 1872,
      "case_id": "case_power_pcie",
      "case_title": "GPU load-step rail droop causes PCIe downtrain",
      "event_type": "pcie_event",
      "source": "host",
      "link_id": "CPU0-SW0-GPU0",
      "negotiated_speed": "Gen4",
      "negotiated_width": "x16",
      "state": "downtrained",
      "correlation_id": "F-2026-001",
      "message": "Link downtrains after repeated training retries."
    },
    {
      "ts_ms": 1930,
      "case_id": "case_power_pcie",
      "case_title": "GPU load-step rail droop causes PCIe downtrain",
      "event_type": "validation_event",
      "source": "lab",
      "test": "oscilloscope_capture",
      "verdict": "fail",
      "closure": "open",
      "correlation_id": "F-2026-001",
      "message": "Scope capture confirms rail droop overlaps PCIe training retry window."
    },
    {
      "ts_ms": 2200,
      "case_id": "case_power_pcie",
      "case_title": "GPU load-step rail droop causes PCIe downtrain",
      "event_type": "validation_event",
      "source": "lab",
      "test": "mitigation_plan",
      "verdict": "reviewed",
      "closure": "open",
      "correlation_id": "F-2026-001",
      "message": "Mitigation selected: add bulk capacitance, tune compensation, extend OCP blanking, and guard retimer reset release."
    },
    {
      "ts_ms": 2920,
      "case_id": "case_power_pcie",
      "case_title": "GPU load-step rail droop causes PCIe downtrain",
      "event_type": "rail_sample",
      "source": "pmbus",
      "rail_id": "GPU0_CORE",
      "voltage_v": 0.836,
      "nominal_v": 0.85,
      "current_a": 316.2,
      "state": "nominal",
      "correlation_id": "F-2026-001",
      "message": "Retest rail droop stays within target after mitigation."
    },
    {
      "ts_ms": 2965,
      "case_id": "case_power_pcie",
      "case_title": "GPU load-step rail droop causes PCIe downtrain",
      "event_type": "pcie_event",
      "source": "host",
      "link_id": "CPU0-SW0-GPU0",
      "negotiated_speed": "Gen5",
      "negotiated_width": "x16",
      "state": "stable",
      "correlation_id": "F-2026-001",
      "message": "PCIe link trains at expected speed and width after mitigation."
    },
    {
      "ts_ms": 3100,
      "case_id": "case_power_pcie",
      "case_title": "GPU load-step rail droop causes PCIe downtrain",
      "event_type": "validation_event",
      "source": "lab",
      "test": "regression_matrix",
      "verdict": "pass",
      "closure": "closed",
      "correlation_id": "F-2026-001",
      "message": "Regression passes cold boot, warm reboot, load-step, and thermal soak coverage."
    },
    {
      "ts_ms": 4000,
      "case_id": "case_retimer_refclk",
      "case_title": "Retimer reset race causes Gen5 training retries",
      "event_type": "workload_marker",
      "source": "host",
      "workload": "warm_reboot",
      "phase": "start",
      "message": "Warm reboot validation begins for fabric stability check."
    },
    {
      "ts_ms": 4060,
      "case_id": "case_retimer_refclk",
      "case_title": "Retimer reset race causes Gen5 training retries",
      "event_type": "bmc_event",
      "source": "bmc",
      "severity": "info",
      "code": "RST_RETIMER_RELEASE",
      "correlation_id": "F-2026-002",
      "message": "BMC releases retimer reset shortly after reference clock lock indication."
    },
    {
      "ts_ms": 4074,
      "case_id": "case_retimer_refclk",
      "case_title": "Retimer reset race causes Gen5 training retries",
      "event_type": "pcie_event",
      "source": "host",
      "link_id": "CPU0-SW1-GPU3",
      "negotiated_speed": "Gen5",
      "negotiated_width": "x16",
      "state": "training_retry",
      "correlation_id": "F-2026-002",
      "message": "PCIe link retries during Gen5 training on warm reboot."
    },
    {
      "ts_ms": 4108,
      "case_id": "case_retimer_refclk",
      "case_title": "Retimer reset race causes Gen5 training retries",
      "event_type": "pcie_event",
      "source": "host",
      "link_id": "CPU0-SW1-GPU3",
      "negotiated_speed": "Gen4",
      "negotiated_width": "x16",
      "state": "downtrained",
      "correlation_id": "F-2026-002",
      "message": "Link falls back to Gen4 after retry budget is exhausted."
    },
    {
      "ts_ms": 4140,
      "case_id": "case_retimer_refclk",
      "case_title": "Retimer reset race causes Gen5 training retries",
      "event_type": "validation_event",
      "source": "lab",
      "test": "refclk_reset_capture",
      "verdict": "fail",
      "closure": "open",
      "correlation_id": "F-2026-002",
      "message": "Capture shows retimer reset release is too close to refclk lock settle time."
    },
    {
      "ts_ms": 4300,
      "case_id": "case_retimer_refclk",
      "case_title": "Retimer reset race causes Gen5 training retries",
      "event_type": "validation_event",
      "source": "lab",
      "test": "mitigation_plan",
      "verdict": "reviewed",
      "closure": "open",
      "correlation_id": "F-2026-002",
      "message": "Mitigation selected: add 25 ms guard time after refclk lock before retimer reset release."
    },
    {
      "ts_ms": 4620,
      "case_id": "case_retimer_refclk",
      "case_title": "Retimer reset race causes Gen5 training retries",
      "event_type": "pcie_event",
      "source": "host",
      "link_id": "CPU0-SW1-GPU3",
      "negotiated_speed": "Gen5",
      "negotiated_width": "x16",
      "state": "stable",
      "correlation_id": "F-2026-002",
      "message": "PCIe link trains at Gen5 x16 after reset guard change."
    },
    {
      "ts_ms": 4700,
      "case_id": "case_retimer_refclk",
      "case_title": "Retimer reset race causes Gen5 training retries",
      "event_type": "validation_event",
      "source": "lab",
      "test": "warm_reboot_regression",
      "verdict": "pass",
      "closure": "closed",
      "correlation_id": "F-2026-002",
      "message": "Warm reboot regression closes with stable Gen5 training over repeated cycles."
    },
    {
      "ts_ms": 6000,
      "case_id": "case_thermal_sensor",
      "case_title": "Thermal sensor mismatch masks accelerator hotspot",
      "event_type": "workload_marker",
      "source": "host",
      "workload": "thermal_soak",
      "phase": "steady_state",
      "message": "Thermal soak reaches steady workload phase."
    },
    {
      "ts_ms": 6120,
      "case_id": "case_thermal_sensor",
      "case_title": "Thermal sensor mismatch masks accelerator hotspot",
      "event_type": "thermal_sample",
      "source": "bmc",
      "sensor_id": "GPU2_INLET",
      "temp_c": 31.4,
      "state": "nominal",
      "message": "Inlet temperature appears normal during thermal soak."
    },
    {
      "ts_ms": 6130,
      "case_id": "case_thermal_sensor",
      "case_title": "Thermal sensor mismatch masks accelerator hotspot",
      "event_type": "thermal_sample",
      "source": "bmc",
      "sensor_id": "GPU2_VRM_HOTSPOT",
      "temp_c": 91.8,
      "state": "degraded",
      "correlation_id": "F-2026-003",
      "message": "Local VRM hotspot estimate exceeds derating threshold while inlet remains normal."
    },
    {
      "ts_ms": 6134,
      "case_id": "case_thermal_sensor",
      "case_title": "Thermal sensor mismatch masks accelerator hotspot",
      "event_type": "bmc_event",
      "source": "bmc",
      "severity": "warning",
      "code": "THERM_GPU2_DERATE",
      "correlation_id": "F-2026-003",
      "message": "BMC derates GPU2 due to local hotspot policy."
    },
    {
      "ts_ms": 6150,
      "case_id": "case_thermal_sensor",
      "case_title": "Thermal sensor mismatch masks accelerator hotspot",
      "event_type": "validation_event",
      "source": "lab",
      "test": "sensor_map_review",
      "verdict": "fail",
      "closure": "open",
      "correlation_id": "F-2026-003",
      "message": "Review finds inlet sensor cannot explain VRM hotspot; local sensor placement and cold-plate contact require inspection."
    },
    {
      "ts_ms": 6420,
      "case_id": "case_thermal_sensor",
      "case_title": "Thermal sensor mismatch masks accelerator hotspot",
      "event_type": "validation_event",
      "source": "lab",
      "test": "mechanical_inspection",
      "verdict": "reviewed",
      "closure": "open",
      "correlation_id": "F-2026-003",
      "message": "TIM contact adjusted and VRM zone threshold added to service checklist."
    },
    {
      "ts_ms": 6900,
      "case_id": "case_thermal_sensor",
      "case_title": "Thermal sensor mismatch masks accelerator hotspot",
      "event_type": "thermal_sample",
      "source": "bmc",
      "sensor_id": "GPU2_VRM_HOTSPOT",
      "temp_c": 78.2,
      "state": "nominal",
      "correlation_id": "F-2026-003",
      "message": "Retest hotspot temperature remains below derating threshold."
    },
    {
      "ts_ms": 7000,
      "case_id": "case_thermal_sensor",
      "case_title": "Thermal sensor mismatch masks accelerator hotspot",
      "event_type": "validation_event",
      "source": "lab",
      "test": "thermal_soak_regression",
      "verdict": "pass",
      "closure": "closed",
      "correlation_id": "F-2026-003",
      "message": "Thermal soak passes with updated sensor map, inspection step, and derating policy."
    }
  ]
}