Update Inspur parsing and align release docs

This commit is contained in:
2026-02-15 23:13:47 +03:00
parent c13788132b
commit 514da76ddb
11 changed files with 1231 additions and 190 deletions

View File

@@ -8,9 +8,9 @@
## Принципы импорта ## Принципы импорта
1. **Snapshot данных** - JSON содержит состояние сервера на момент сбора, без исторической информации 1. **Snapshot данных** - JSON содержит состояние сервера на момент сбора и может включать историю изменений статуса компонентов
2. **Автоматическое определение LOT** - классификация компонентов определяется приложением на основе vendor/model/type 2. **Автоматическое определение LOT** - классификация компонентов определяется приложением на основе vendor/model/type
3. **Статус компонентов** - каждый компонент имеет статус работоспособности (OK, Warning, Critical, Unknown) 3. **Статус компонентов** - каждый компонент имеет статус работоспособности (OK, Warning, Critical, Unknown) и может передавать время проверки статуса
4. **Идемпотентность** - повторный импорт с тем же snapshot не создает дубликаты 4. **Идемпотентность** - повторный импорт с тем же snapshot не создает дубликаты
5. **Event-driven обновления** - импорт создает события в timeline (LOG_COLLECTED, INSTALLED, REMOVED, FIRMWARE_CHANGED) 5. **Event-driven обновления** - импорт создает события в timeline (LOG_COLLECTED, INSTALLED, REMOVED, FIRMWARE_CHANGED)
@@ -53,6 +53,33 @@
- `filename` (string, опционально) - идентификатор источника данных - `filename` (string, опционально) - идентификатор источника данных
- `hardware` (object, обязательно) - структура с аппаратными компонентами - `hardware` (object, обязательно) - структура с аппаратными компонентами
### Общее поле статуса для компонентных секций
Для секций `cpus`, `memory`, `storage`, `pcie_devices`, `power_supplies` поддерживается дополнительное поле:
- `status_checked_at` (string RFC3339, опционально) - дата/время, когда был проверен статус работоспособности компонента
- `status_changed_at` (string RFC3339, опционально) - дата/время последнего изменения статуса компонента
- `status_at_collection` (object, опционально) - зафиксированный статус на момент сбора логов:
- `status` (string) - статус в момент сбора (`OK`, `Warning`, `Critical`, `Unknown`, `Empty`)
- `at` (string RFC3339) - дата/время, к которому относится этот статус
- `status_history` (array, опционально) - история статусов компонента:
- `status` (string) - статус (`OK`, `Warning`, `Critical`, `Unknown`, `Empty`)
- `changed_at` (string RFC3339) - дата/время смены статуса
- `details` (string, опционально) - пояснение к переходу статуса
- `error_description` (string, опционально) - текст ошибки/диагностики для статуса компонента (например при `Warning`/`Critical`)
### Правила экспорта JSON для внешнего проекта
Используйте эти правила, если JSON формируется внешним сервисом/экспортером:
1. Всегда передавайте `status` как текущее состояние компонента в snapshot.
2. Если есть точное время последней смены, передавайте `status_changed_at` (RFC3339, UTC).
3. Если источник умеет фиксировать состояние именно на момент сбора, передавайте `status_at_collection` c полями `status` и `at`.
4. Если источник хранит историю (например Windows Event Log), передавайте `status_history` отсортированным по `changed_at` по возрастанию.
5. В `status_history` не отправляйте записи без `changed_at`; такие записи игнорируются.
6. Для совместимости допускается передавать только старые поля (`status`, `status_checked_at`) без истории.
7. Все даты/время в исторических полях должны быть RFC3339; рекомендуется использовать UTC (`Z`).
--- ---
## Секция hardware ## Секция hardware
@@ -99,7 +126,8 @@
"frequency_mhz": 2100, "frequency_mhz": 2100,
"max_frequency_mhz": 4000, "max_frequency_mhz": 4000,
"manufacturer": "Intel", "manufacturer": "Intel",
"status": "OK" "status": "OK",
"status_checked_at": "2026-02-10T15:28:00Z"
}, },
{ {
"socket": 1, "socket": 1,
@@ -109,7 +137,8 @@
"frequency_mhz": 2100, "frequency_mhz": 2100,
"max_frequency_mhz": 4000, "max_frequency_mhz": 4000,
"manufacturer": "Intel", "manufacturer": "Intel",
"status": "OK" "status": "OK",
"status_checked_at": "2026-02-10T15:28:00Z"
} }
] ]
} }
@@ -124,6 +153,7 @@
- `max_frequency_mhz` (int, опционально) - максимальная частота в МГц - `max_frequency_mhz` (int, опционально) - максимальная частота в МГц
- `manufacturer` (string, опционально) - производитель (Intel, AMD, etc.) - `manufacturer` (string, опционально) - производитель (Intel, AMD, etc.)
- `status` (string, опционально) - статус: `OK`, `Warning`, `Critical`, `Unknown` - `status` (string, опционально) - статус: `OK`, `Warning`, `Critical`, `Unknown`
- `status_checked_at` (string RFC3339, опционально) - дата/время проверки статуса
**Генерация serial_number:** **Генерация serial_number:**
- Формат: `{board_serial}-CPU-{socket}` - Формат: `{board_serial}-CPU-{socket}`
@@ -153,7 +183,8 @@
"manufacturer": "Hynix", "manufacturer": "Hynix",
"serial_number": "80AD032419E17CEEC1", "serial_number": "80AD032419E17CEEC1",
"part_number": "HMCG88AGBRA191N", "part_number": "HMCG88AGBRA191N",
"status": "OK" "status": "OK",
"status_checked_at": "2026-02-10T15:28:00Z"
}, },
{ {
"slot": "CPU0_C1D0", "slot": "CPU0_C1D0",
@@ -182,6 +213,7 @@
- `serial_number` (string, условно обязательно если present=true) - серийный номер - `serial_number` (string, условно обязательно если present=true) - серийный номер
- `part_number` (string, опционально) - партномер - `part_number` (string, опционально) - партномер
- `status` (string, опционально) - статус: `OK`, `Warning`, `Critical`, `Unknown`, `Empty` - `status` (string, опционально) - статус: `OK`, `Warning`, `Critical`, `Unknown`, `Empty`
- `status_checked_at` (string RFC3339, опционально) - дата/время проверки статуса
**Обработка:** **Обработка:**
- Если `present = false` или `status = "Empty"`, компонент не создается/не обновляется - Если `present = false` или `status = "Empty"`, компонент не создается/не обновляется
@@ -239,6 +271,7 @@
- `interface` (string, опционально) - интерфейс: `NVMe`, `SATA`, `SAS` - `interface` (string, опционально) - интерфейс: `NVMe`, `SATA`, `SAS`
- `present` (bool, обязательно) - наличие диска в слоте - `present` (bool, обязательно) - наличие диска в слоте
- `status` (string, опционально) - статус: `OK`, `Warning`, `Critical`, `Unknown` - `status` (string, опционально) - статус: `OK`, `Warning`, `Critical`, `Unknown`
- `status_checked_at` (string RFC3339, опционально) - дата/время проверки статуса
**Обработка firmware:** **Обработка firmware:**
- Если версия firmware изменилась относительно предыдущего observation - создается событие FIRMWARE_CHANGED - Если версия firmware изменилась относительно предыдущего observation - создается событие FIRMWARE_CHANGED
@@ -266,6 +299,7 @@
"part_number": "V0310C9000000000", "part_number": "V0310C9000000000",
"firmware": "00.03.05", "firmware": "00.03.05",
"status": "OK", "status": "OK",
"status_checked_at": "2026-02-10T15:28:00Z",
"input_type": "ACWideRange", "input_type": "ACWideRange",
"input_power_w": 137, "input_power_w": 137,
"output_power_w": 104, "output_power_w": 104,
@@ -285,6 +319,7 @@
- `part_number` (string, опционально) - партномер - `part_number` (string, опционально) - партномер
- `firmware` (string, опционально) - версия прошивки - `firmware` (string, опционально) - версия прошивки
- `status` (string, опционально) - статус: `OK`, `Warning`, `Critical`, `Unknown` - `status` (string, опционально) - статус: `OK`, `Warning`, `Critical`, `Unknown`
- `status_checked_at` (string RFC3339, опционально) - дата/время проверки статуса
- `input_type` (string, опционально) - тип входа - `input_type` (string, опционально) - тип входа
- `input_power_w` (int, опционально) - входная мощность (telemetry) - `input_power_w` (int, опционально) - входная мощность (telemetry)
- `output_power_w` (int, опционально) - выходная мощность (telemetry) - `output_power_w` (int, опционально) - выходная мощность (telemetry)
@@ -319,7 +354,8 @@
"max_link_speed": "Gen3", "max_link_speed": "Gen3",
"serial_number": "RAID-001-12345", "serial_number": "RAID-001-12345",
"firmware": "50.9.1-4296", "firmware": "50.9.1-4296",
"status": "OK" "status": "OK",
"status_checked_at": "2026-02-10T15:28:00Z"
}, },
{ {
"slot": "PCIeCard2", "slot": "PCIeCard2",
@@ -355,6 +391,7 @@
- `serial_number` (string, опционально) - серийный номер (если доступен, иначе генерируется) - `serial_number` (string, опционально) - серийный номер (если доступен, иначе генерируется)
- `firmware` (string, опционально) - версия прошивки - `firmware` (string, опционально) - версия прошивки
- `status` (string, опционально) - статус: `OK`, `Warning`, `Critical`, `Unknown` - `status` (string, опционально) - статус: `OK`, `Warning`, `Critical`, `Unknown`
- `status_checked_at` (string RFC3339, опционально) - дата/время проверки статуса
**Генерация serial_number (если отсутствует):** **Генерация serial_number (если отсутствует):**
- Формат: `{board_serial}-PCIE-{slot}` - Формат: `{board_serial}-PCIE-{slot}`
@@ -872,7 +909,7 @@ Content-Type: application/json
} }
``` ```
### Пример 2: Server с отказавшим диском ### Пример 2: Server с историей "сломан -> починен"
```json ```json
{ {
@@ -893,7 +930,24 @@ Content-Type: application/json
"firmware": "9CV10510", "firmware": "9CV10510",
"interface": "NVMe", "interface": "NVMe",
"present": true, "present": true,
"status": "Critical" "status": "OK",
"status_changed_at": "2026-02-10T15:22:00Z",
"status_at_collection": {
"status": "OK",
"at": "2026-02-10T15:30:00Z"
},
"status_history": [
{
"status": "Critical",
"changed_at": "2026-02-10T15:10:00Z",
"details": "I/O timeout on NVMe queue 3"
},
{
"status": "OK",
"changed_at": "2026-02-10T15:22:00Z",
"details": "Recovered after controller reset"
}
]
}, },
{ {
"slot": "Disk.Bay.1", "slot": "Disk.Bay.1",
@@ -911,9 +965,9 @@ Content-Type: application/json
``` ```
**Обработка:** **Обработка:**
- Disk.Bay.0 получит статус Critical - Disk.Bay.0 получит текущий статус `OK`
- Автоматически создастся failure_event для компонента S5GUNG0N123456 - История статусов сохранится в `observations.details.status_history`
- Timeline event COMPONENT_FAILED - Автоматический `failure_event` не создается, так как текущий статус snapshot не `Critical`
### Пример 3: Замена памяти ### Пример 3: Замена памяти

View File

@@ -4,6 +4,7 @@ import (
"fmt" "fmt"
"net/url" "net/url"
"regexp" "regexp"
"sort"
"strings" "strings"
"time" "time"
@@ -27,20 +28,22 @@ func ConvertToReanimator(result *models.AnalysisResult) (*ReanimatorExport, erro
// Determine target host (optional field) // Determine target host (optional field)
targetHost := inferTargetHost(result.TargetHost, result.Filename) targetHost := inferTargetHost(result.TargetHost, result.Filename)
collectedAt := formatRFC3339(result.CollectedAt)
export := &ReanimatorExport{ export := &ReanimatorExport{
Filename: result.Filename, Filename: result.Filename,
SourceType: normalizeSourceType(result.SourceType), SourceType: normalizeSourceType(result.SourceType),
Protocol: normalizeProtocol(result.Protocol), Protocol: normalizeProtocol(result.Protocol),
TargetHost: targetHost, TargetHost: targetHost,
CollectedAt: formatRFC3339(result.CollectedAt), CollectedAt: collectedAt,
Hardware: ReanimatorHardware{ Hardware: ReanimatorHardware{
Board: convertBoard(result.Hardware.BoardInfo), Board: convertBoard(result.Hardware.BoardInfo),
Firmware: convertFirmware(result.Hardware.Firmware), Firmware: dedupeFirmware(convertFirmware(result.Hardware.Firmware)),
CPUs: convertCPUs(result.Hardware.CPUs), CPUs: dedupeCPUs(convertCPUs(result.Hardware.CPUs, collectedAt)),
Memory: convertMemory(result.Hardware.Memory), Memory: dedupeMemory(convertMemory(result.Hardware.Memory, collectedAt)),
Storage: convertStorage(result.Hardware.Storage), Storage: dedupeStorage(convertStorage(result.Hardware.Storage, collectedAt)),
PCIeDevices: convertPCIeDevices(result.Hardware), PCIeDevices: dedupePCIe(convertPCIeDevices(result.Hardware, collectedAt)),
PowerSupplies: convertPowerSupplies(result.Hardware.PowerSupply), PowerSupplies: dedupePSUs(convertPowerSupplies(result.Hardware.PowerSupply, collectedAt)),
}, },
} }
@@ -83,7 +86,7 @@ func convertFirmware(firmware []models.FirmwareInfo) []ReanimatorFirmware {
} }
// convertCPUs converts CPU information to Reanimator format // convertCPUs converts CPU information to Reanimator format
func convertCPUs(cpus []models.CPU) []ReanimatorCPU { func convertCPUs(cpus []models.CPU, collectedAt string) []ReanimatorCPU {
if len(cpus) == 0 { if len(cpus) == 0 {
return nil return nil
} }
@@ -92,6 +95,20 @@ func convertCPUs(cpus []models.CPU) []ReanimatorCPU {
for _, cpu := range cpus { for _, cpu := range cpus {
manufacturer := inferCPUManufacturer(cpu.Model) manufacturer := inferCPUManufacturer(cpu.Model)
cpuStatus := normalizeStatus(cpu.Status, false)
if strings.TrimSpace(cpu.Status) == "" {
cpuStatus = "Unknown"
}
meta := buildStatusMeta(
cpuStatus,
cpu.StatusCheckedAt,
cpu.StatusChangedAt,
cpu.StatusAtCollect,
cpu.StatusHistory,
cpu.ErrorDescription,
collectedAt,
)
result = append(result, ReanimatorCPU{ result = append(result, ReanimatorCPU{
Socket: cpu.Socket, Socket: cpu.Socket,
Model: cpu.Model, Model: cpu.Model,
@@ -100,14 +117,19 @@ func convertCPUs(cpus []models.CPU) []ReanimatorCPU {
FrequencyMHz: cpu.FrequencyMHz, FrequencyMHz: cpu.FrequencyMHz,
MaxFrequencyMHz: cpu.MaxFreqMHz, MaxFrequencyMHz: cpu.MaxFreqMHz,
Manufacturer: manufacturer, Manufacturer: manufacturer,
Status: "Unknown", Status: cpuStatus,
StatusCheckedAt: meta.StatusCheckedAt,
StatusChangedAt: meta.StatusChangedAt,
StatusAtCollect: meta.StatusAtCollection,
StatusHistory: meta.StatusHistory,
ErrorDescription: meta.ErrorDescription,
}) })
} }
return result return result
} }
// convertMemory converts memory modules to Reanimator format // convertMemory converts memory modules to Reanimator format
func convertMemory(memory []models.MemoryDIMM) []ReanimatorMemory { func convertMemory(memory []models.MemoryDIMM, collectedAt string) []ReanimatorMemory {
if len(memory) == 0 { if len(memory) == 0 {
return nil return nil
} }
@@ -123,6 +145,16 @@ func convertMemory(memory []models.MemoryDIMM) []ReanimatorMemory {
} }
} }
meta := buildStatusMeta(
status,
mem.StatusCheckedAt,
mem.StatusChangedAt,
mem.StatusAtCollect,
mem.StatusHistory,
mem.ErrorDescription,
collectedAt,
)
result = append(result, ReanimatorMemory{ result = append(result, ReanimatorMemory{
Slot: mem.Slot, Slot: mem.Slot,
Location: mem.Location, Location: mem.Location,
@@ -135,13 +167,18 @@ func convertMemory(memory []models.MemoryDIMM) []ReanimatorMemory {
SerialNumber: mem.SerialNumber, SerialNumber: mem.SerialNumber,
PartNumber: mem.PartNumber, PartNumber: mem.PartNumber,
Status: status, Status: status,
StatusCheckedAt: meta.StatusCheckedAt,
StatusChangedAt: meta.StatusChangedAt,
StatusAtCollect: meta.StatusAtCollection,
StatusHistory: meta.StatusHistory,
ErrorDescription: meta.ErrorDescription,
}) })
} }
return result return result
} }
// convertStorage converts storage devices to Reanimator format // convertStorage converts storage devices to Reanimator format
func convertStorage(storage []models.Storage) []ReanimatorStorage { func convertStorage(storage []models.Storage, collectedAt string) []ReanimatorStorage {
if len(storage) == 0 { if len(storage) == 0 {
return nil return nil
} }
@@ -154,6 +191,18 @@ func convertStorage(storage []models.Storage) []ReanimatorStorage {
} }
status := inferStorageStatus(stor) status := inferStorageStatus(stor)
if strings.TrimSpace(stor.Status) != "" {
status = normalizeStatus(stor.Status, false)
}
meta := buildStatusMeta(
status,
stor.StatusCheckedAt,
stor.StatusChangedAt,
stor.StatusAtCollect,
stor.StatusHistory,
stor.ErrorDescription,
collectedAt,
)
result = append(result, ReanimatorStorage{ result = append(result, ReanimatorStorage{
Slot: stor.Slot, Slot: stor.Slot,
@@ -166,17 +215,36 @@ func convertStorage(storage []models.Storage) []ReanimatorStorage {
Interface: stor.Interface, Interface: stor.Interface,
Present: stor.Present, Present: stor.Present,
Status: status, Status: status,
StatusCheckedAt: meta.StatusCheckedAt,
StatusChangedAt: meta.StatusChangedAt,
StatusAtCollect: meta.StatusAtCollection,
StatusHistory: meta.StatusHistory,
ErrorDescription: meta.ErrorDescription,
}) })
} }
return result return result
} }
// convertPCIeDevices converts PCIe devices, GPUs, and network adapters to Reanimator format // convertPCIeDevices converts PCIe devices, GPUs, and network adapters to Reanimator format
func convertPCIeDevices(hw *models.HardwareConfig) []ReanimatorPCIe { func convertPCIeDevices(hw *models.HardwareConfig, collectedAt string) []ReanimatorPCIe {
result := make([]ReanimatorPCIe, 0) result := make([]ReanimatorPCIe, 0)
gpuSlots := make(map[string]struct{}, len(hw.GPUs))
for _, gpu := range hw.GPUs {
slot := strings.ToLower(strings.TrimSpace(gpu.Slot))
if slot != "" {
gpuSlots[slot] = struct{}{}
}
}
// Convert regular PCIe devices // Convert regular PCIe devices
for _, pcie := range hw.PCIeDevices { for _, pcie := range hw.PCIeDevices {
slot := strings.ToLower(strings.TrimSpace(pcie.Slot))
if _, isDedicatedGPU := gpuSlots[slot]; isDedicatedGPU || isDisplayClass(pcie.DeviceClass) {
// Skip GPU-like PCIe entries to avoid duplicates:
// dedicated GPUs are exported from hw.GPUs with richer metadata.
continue
}
serialNumber := normalizedSerial(pcie.SerialNumber) serialNumber := normalizedSerial(pcie.SerialNumber)
// Determine model (prefer PartNumber, fallback to DeviceClass) // Determine model (prefer PartNumber, fallback to DeviceClass)
@@ -185,6 +253,17 @@ func convertPCIeDevices(hw *models.HardwareConfig) []ReanimatorPCIe {
model = pcie.DeviceClass model = pcie.DeviceClass
} }
status := normalizeStatus(pcie.Status, false)
meta := buildStatusMeta(
status,
pcie.StatusCheckedAt,
pcie.StatusChangedAt,
pcie.StatusAtCollect,
pcie.StatusHistory,
pcie.ErrorDescription,
collectedAt,
)
result = append(result, ReanimatorPCIe{ result = append(result, ReanimatorPCIe{
Slot: pcie.Slot, Slot: pcie.Slot,
VendorID: pcie.VendorID, VendorID: pcie.VendorID,
@@ -199,7 +278,12 @@ func convertPCIeDevices(hw *models.HardwareConfig) []ReanimatorPCIe {
MaxLinkSpeed: pcie.MaxLinkSpeed, MaxLinkSpeed: pcie.MaxLinkSpeed,
SerialNumber: serialNumber, SerialNumber: serialNumber,
Firmware: "", // PCIeDevice doesn't have firmware in models Firmware: "", // PCIeDevice doesn't have firmware in models
Status: "Unknown", Status: status,
StatusCheckedAt: meta.StatusCheckedAt,
StatusChangedAt: meta.StatusChangedAt,
StatusAtCollect: meta.StatusAtCollection,
StatusHistory: meta.StatusHistory,
ErrorDescription: meta.ErrorDescription,
}) })
} }
@@ -210,6 +294,17 @@ func convertPCIeDevices(hw *models.HardwareConfig) []ReanimatorPCIe {
// Determine device class // Determine device class
deviceClass := "DisplayController" deviceClass := "DisplayController"
status := normalizeStatus(gpu.Status, false)
meta := buildStatusMeta(
status,
gpu.StatusCheckedAt,
gpu.StatusChangedAt,
gpu.StatusAtCollect,
gpu.StatusHistory,
gpu.ErrorDescription,
collectedAt,
)
result = append(result, ReanimatorPCIe{ result = append(result, ReanimatorPCIe{
Slot: gpu.Slot, Slot: gpu.Slot,
VendorID: gpu.VendorID, VendorID: gpu.VendorID,
@@ -224,7 +319,12 @@ func convertPCIeDevices(hw *models.HardwareConfig) []ReanimatorPCIe {
MaxLinkSpeed: gpu.MaxLinkSpeed, MaxLinkSpeed: gpu.MaxLinkSpeed,
SerialNumber: serialNumber, SerialNumber: serialNumber,
Firmware: gpu.Firmware, Firmware: gpu.Firmware,
Status: normalizeStatus(gpu.Status, false), Status: status,
StatusCheckedAt: meta.StatusCheckedAt,
StatusChangedAt: meta.StatusChangedAt,
StatusAtCollect: meta.StatusAtCollection,
StatusHistory: meta.StatusHistory,
ErrorDescription: meta.ErrorDescription,
}) })
} }
@@ -236,6 +336,17 @@ func convertPCIeDevices(hw *models.HardwareConfig) []ReanimatorPCIe {
serialNumber := normalizedSerial(nic.SerialNumber) serialNumber := normalizedSerial(nic.SerialNumber)
status := normalizeStatus(nic.Status, false)
meta := buildStatusMeta(
status,
nic.StatusCheckedAt,
nic.StatusChangedAt,
nic.StatusAtCollect,
nic.StatusHistory,
nic.ErrorDescription,
collectedAt,
)
result = append(result, ReanimatorPCIe{ result = append(result, ReanimatorPCIe{
Slot: nic.Slot, Slot: nic.Slot,
VendorID: nic.VendorID, VendorID: nic.VendorID,
@@ -250,15 +361,27 @@ func convertPCIeDevices(hw *models.HardwareConfig) []ReanimatorPCIe {
MaxLinkSpeed: "", MaxLinkSpeed: "",
SerialNumber: serialNumber, SerialNumber: serialNumber,
Firmware: nic.Firmware, Firmware: nic.Firmware,
Status: normalizeStatus(nic.Status, false), Status: status,
StatusCheckedAt: meta.StatusCheckedAt,
StatusChangedAt: meta.StatusChangedAt,
StatusAtCollect: meta.StatusAtCollection,
StatusHistory: meta.StatusHistory,
ErrorDescription: meta.ErrorDescription,
}) })
} }
return result return result
} }
func isDisplayClass(deviceClass string) bool {
class := strings.ToLower(strings.TrimSpace(deviceClass))
return strings.Contains(class, "display") ||
strings.Contains(class, "vga") ||
strings.Contains(class, "3d controller")
}
// convertPowerSupplies converts power supplies to Reanimator format // convertPowerSupplies converts power supplies to Reanimator format
func convertPowerSupplies(psus []models.PSU) []ReanimatorPSU { func convertPowerSupplies(psus []models.PSU, collectedAt string) []ReanimatorPSU {
if len(psus) == 0 { if len(psus) == 0 {
return nil return nil
} }
@@ -271,6 +394,15 @@ func convertPowerSupplies(psus []models.PSU) []ReanimatorPSU {
} }
status := normalizeStatus(psu.Status, false) status := normalizeStatus(psu.Status, false)
meta := buildStatusMeta(
status,
psu.StatusCheckedAt,
psu.StatusChangedAt,
psu.StatusAtCollect,
psu.StatusHistory,
psu.ErrorDescription,
collectedAt,
)
result = append(result, ReanimatorPSU{ result = append(result, ReanimatorPSU{
Slot: psu.Slot, Slot: psu.Slot,
@@ -286,11 +418,267 @@ func convertPowerSupplies(psus []models.PSU) []ReanimatorPSU {
InputPowerW: psu.InputPowerW, InputPowerW: psu.InputPowerW,
OutputPowerW: psu.OutputPowerW, OutputPowerW: psu.OutputPowerW,
InputVoltage: psu.InputVoltage, InputVoltage: psu.InputVoltage,
StatusCheckedAt: meta.StatusCheckedAt,
StatusChangedAt: meta.StatusChangedAt,
StatusAtCollect: meta.StatusAtCollection,
StatusHistory: meta.StatusHistory,
ErrorDescription: meta.ErrorDescription,
}) })
} }
return result return result
} }
type convertedStatusMeta struct {
StatusCheckedAt string
StatusChangedAt string
StatusAtCollection *ReanimatorStatusAtCollection
StatusHistory []ReanimatorStatusHistoryEntry
ErrorDescription string
}
func buildStatusMeta(
currentStatus string,
checkedAt time.Time,
changedAt time.Time,
statusAtCollection *models.StatusAtCollection,
history []models.StatusHistoryEntry,
errorDescription string,
collectedAt string,
) convertedStatusMeta {
meta := convertedStatusMeta{
StatusCheckedAt: formatOptionalRFC3339(checkedAt),
StatusChangedAt: formatOptionalRFC3339(changedAt),
ErrorDescription: strings.TrimSpace(errorDescription),
}
convertedHistory := make([]ReanimatorStatusHistoryEntry, 0, len(history))
for _, h := range history {
changed := formatOptionalRFC3339(h.ChangedAt)
if changed == "" {
continue
}
convertedHistory = append(convertedHistory, ReanimatorStatusHistoryEntry{
Status: normalizeStatus(h.Status, true),
ChangedAt: changed,
Details: strings.TrimSpace(h.Details),
})
}
sort.Slice(convertedHistory, func(i, j int) bool {
return convertedHistory[i].ChangedAt < convertedHistory[j].ChangedAt
})
if len(convertedHistory) > 0 {
meta.StatusHistory = convertedHistory
if meta.StatusChangedAt == "" {
meta.StatusChangedAt = convertedHistory[len(convertedHistory)-1].ChangedAt
}
}
if statusAtCollection != nil {
at := formatOptionalRFC3339(statusAtCollection.At)
if at != "" && strings.TrimSpace(statusAtCollection.Status) != "" {
meta.StatusAtCollection = &ReanimatorStatusAtCollection{
Status: normalizeStatus(statusAtCollection.Status, true),
At: at,
}
}
}
if meta.StatusAtCollection == nil && strings.TrimSpace(currentStatus) != "" && collectedAt != "" {
meta.StatusAtCollection = &ReanimatorStatusAtCollection{
Status: currentStatus,
At: collectedAt,
}
}
if meta.StatusCheckedAt == "" && len(meta.StatusHistory) > 0 {
meta.StatusCheckedAt = meta.StatusHistory[len(meta.StatusHistory)-1].ChangedAt
}
if meta.StatusCheckedAt == "" && strings.TrimSpace(currentStatus) != "" && collectedAt != "" {
meta.StatusCheckedAt = collectedAt
}
return meta
}
func formatOptionalRFC3339(t time.Time) string {
if t.IsZero() {
return ""
}
return t.UTC().Format(time.RFC3339)
}
func dedupeFirmware(items []ReanimatorFirmware) []ReanimatorFirmware {
if len(items) < 2 {
return items
}
seen := make(map[string]struct{}, len(items))
result := make([]ReanimatorFirmware, 0, len(items))
for _, item := range items {
key := strings.ToLower(strings.TrimSpace(item.DeviceName))
if key == "" {
key = strings.ToLower(strings.TrimSpace(item.Version))
}
if _, ok := seen[key]; ok {
continue
}
seen[key] = struct{}{}
result = append(result, item)
}
return result
}
func dedupeCPUs(items []ReanimatorCPU) []ReanimatorCPU {
if len(items) < 2 {
return items
}
seen := make(map[int]struct{}, len(items))
result := make([]ReanimatorCPU, 0, len(items))
for _, item := range items {
if _, ok := seen[item.Socket]; ok {
continue
}
seen[item.Socket] = struct{}{}
result = append(result, item)
}
return result
}
func dedupeMemory(items []ReanimatorMemory) []ReanimatorMemory {
if len(items) < 2 {
return items
}
seen := make(map[string]struct{}, len(items))
result := make([]ReanimatorMemory, 0, len(items))
for _, item := range items {
key := strings.ToLower(strings.TrimSpace(item.Slot))
if key == "" {
key = strings.ToLower(strings.TrimSpace(item.Location))
}
if _, ok := seen[key]; ok {
continue
}
seen[key] = struct{}{}
result = append(result, item)
}
return result
}
func dedupeStorage(items []ReanimatorStorage) []ReanimatorStorage {
if len(items) < 2 {
return items
}
seen := make(map[string]struct{}, len(items))
result := make([]ReanimatorStorage, 0, len(items))
for _, item := range items {
key := strings.ToLower(strings.TrimSpace(item.SerialNumber))
if key == "" {
key = "slot:" + strings.ToLower(strings.TrimSpace(item.Slot))
}
if _, ok := seen[key]; ok {
continue
}
seen[key] = struct{}{}
result = append(result, item)
}
return result
}
func dedupePSUs(items []ReanimatorPSU) []ReanimatorPSU {
if len(items) < 2 {
return items
}
seen := make(map[string]struct{}, len(items))
result := make([]ReanimatorPSU, 0, len(items))
for _, item := range items {
key := strings.ToLower(strings.TrimSpace(item.SerialNumber))
if key == "" {
key = "slot:" + strings.ToLower(strings.TrimSpace(item.Slot))
}
if _, ok := seen[key]; ok {
continue
}
seen[key] = struct{}{}
result = append(result, item)
}
return result
}
func dedupePCIe(items []ReanimatorPCIe) []ReanimatorPCIe {
if len(items) < 2 {
return items
}
type scored struct {
item ReanimatorPCIe
score int
idx int
}
byKey := make(map[string]scored, len(items))
order := make([]string, 0, len(items))
for i, item := range items {
key := pcieDedupKey(item)
curr := scored{item: item, score: pcieQualityScore(item), idx: i}
existing, ok := byKey[key]
if !ok {
byKey[key] = curr
order = append(order, key)
continue
}
if curr.score > existing.score {
byKey[key] = curr
}
}
result := make([]ReanimatorPCIe, 0, len(byKey))
for _, key := range order {
result = append(result, byKey[key].item)
}
return result
}
func pcieDedupKey(item ReanimatorPCIe) string {
slot := strings.ToLower(strings.TrimSpace(item.Slot))
serial := strings.ToLower(strings.TrimSpace(item.SerialNumber))
bdf := strings.ToLower(strings.TrimSpace(item.BDF))
if slot != "" {
return "slot:" + slot
}
if serial != "" {
return "sn:" + serial
}
if bdf != "" {
return "bdf:" + bdf
}
return strings.ToLower(strings.TrimSpace(item.DeviceClass)) + "|" + strings.ToLower(strings.TrimSpace(item.Model))
}
func pcieQualityScore(item ReanimatorPCIe) int {
score := 0
if strings.TrimSpace(item.SerialNumber) != "" {
score += 4
}
if strings.TrimSpace(item.Model) != "" && !isGenericPCIeModel(item.Model) {
score += 3
}
status := strings.ToLower(strings.TrimSpace(item.Status))
if status == "ok" || status == "warning" || status == "critical" {
score += 2
}
if strings.TrimSpace(item.BDF) != "" {
score++
}
if strings.EqualFold(strings.TrimSpace(item.DeviceClass), "DisplayController") {
score++
}
return score
}
func isGenericPCIeModel(model string) bool {
switch strings.ToLower(strings.TrimSpace(model)) {
case "", "unknown", "vga", "3d controller", "display controller":
return true
default:
return false
}
}
// inferCPUManufacturer determines CPU manufacturer from model string // inferCPUManufacturer determines CPU manufacturer from model string
func inferCPUManufacturer(model string) string { func inferCPUManufacturer(model string) string {
upper := strings.ToUpper(model) upper := strings.ToUpper(model)

View File

@@ -210,7 +210,7 @@ func TestConvertCPUs(t *testing.T) {
}, },
} }
result := convertCPUs(cpus) result := convertCPUs(cpus, "2026-02-10T15:30:00Z")
if len(result) != 2 { if len(result) != 2 {
t.Fatalf("expected 2 CPUs, got %d", len(result)) t.Fatalf("expected 2 CPUs, got %d", len(result))
@@ -245,7 +245,7 @@ func TestConvertMemory(t *testing.T) {
}, },
} }
result := convertMemory(memory) result := convertMemory(memory, "2026-02-10T15:30:00Z")
if len(result) != 2 { if len(result) != 2 {
t.Fatalf("expected 2 memory modules, got %d", len(result)) t.Fatalf("expected 2 memory modules, got %d", len(result))
@@ -278,7 +278,7 @@ func TestConvertStorage(t *testing.T) {
}, },
} }
result := convertStorage(storage) result := convertStorage(storage, "2026-02-10T15:30:00Z")
if len(result) != 1 { if len(result) != 1 {
t.Fatalf("expected 1 storage device (skipped one without serial), got %d", len(result)) t.Fatalf("expected 1 storage device (skipped one without serial), got %d", len(result))
@@ -329,7 +329,7 @@ func TestConvertPCIeDevices(t *testing.T) {
}, },
} }
result := convertPCIeDevices(hw) result := convertPCIeDevices(hw, "2026-02-10T15:30:00Z")
// Should have: 2 PCIe devices + 1 GPU + 1 NIC = 4 total // Should have: 2 PCIe devices + 1 GPU + 1 NIC = 4 total
if len(result) != 4 { if len(result) != 4 {
@@ -369,7 +369,7 @@ func TestConvertPCIeDevices_NVSwitchWithoutSerialRemainsEmpty(t *testing.T) {
}, },
} }
result := convertPCIeDevices(hw) result := convertPCIeDevices(hw, "2026-02-10T15:30:00Z")
if len(result) != 1 { if len(result) != 1 {
t.Fatalf("expected 1 PCIe device, got %d", len(result)) t.Fatalf("expected 1 PCIe device, got %d", len(result))
@@ -380,6 +380,74 @@ func TestConvertPCIeDevices_NVSwitchWithoutSerialRemainsEmpty(t *testing.T) {
} }
} }
func TestConvertPCIeDevices_SkipsDisplayControllerDuplicates(t *testing.T) {
hw := &models.HardwareConfig{
PCIeDevices: []models.PCIeDevice{
{
Slot: "#GPU0",
DeviceClass: "3D Controller",
},
},
GPUs: []models.GPU{
{
Slot: "#GPU0",
Model: "B200 180GB HBM3e",
Manufacturer: "NVIDIA",
SerialNumber: "1655024043371",
Status: "OK",
},
},
}
result := convertPCIeDevices(hw, "2026-02-10T15:30:00Z")
if len(result) != 1 {
t.Fatalf("expected only dedicated GPU record without duplicate display PCIe, got %d", len(result))
}
if result[0].DeviceClass != "DisplayController" {
t.Fatalf("expected GPU record with DisplayController class, got %q", result[0].DeviceClass)
}
if result[0].Status != "OK" {
t.Fatalf("expected GPU status OK, got %q", result[0].Status)
}
}
func TestConvertPCIeDevices_MapsGPUStatusHistory(t *testing.T) {
hw := &models.HardwareConfig{
GPUs: []models.GPU{
{
Slot: "#GPU6",
Model: "B200 180GB HBM3e",
Manufacturer: "NVIDIA",
SerialNumber: "1655024043204",
Status: "Critical",
StatusHistory: []models.StatusHistoryEntry{
{
Status: "Critical",
ChangedAt: time.Date(2026, 1, 12, 15, 5, 18, 0, time.UTC),
Details: "BIOS miss F_GPU6",
},
},
ErrorDescription: "BIOS miss F_GPU6",
},
},
}
result := convertPCIeDevices(hw, "2026-02-10T15:30:00Z")
if len(result) != 1 {
t.Fatalf("expected 1 converted GPU, got %d", len(result))
}
if len(result[0].StatusHistory) != 1 {
t.Fatalf("expected 1 history entry, got %d", len(result[0].StatusHistory))
}
if result[0].StatusHistory[0].ChangedAt != "2026-01-12T15:05:18Z" {
t.Fatalf("unexpected history changed_at: %q", result[0].StatusHistory[0].ChangedAt)
}
if result[0].StatusAtCollect == nil || result[0].StatusAtCollect.At != "2026-02-10T15:30:00Z" {
t.Fatalf("expected status_at_collection to be populated from collected_at")
}
}
func TestConvertPowerSupplies(t *testing.T) { func TestConvertPowerSupplies(t *testing.T) {
psus := []models.PSU{ psus := []models.PSU{
{ {
@@ -398,7 +466,7 @@ func TestConvertPowerSupplies(t *testing.T) {
}, },
} }
result := convertPowerSupplies(psus) result := convertPowerSupplies(psus, "2026-02-10T15:30:00Z")
if len(result) != 1 { if len(result) != 1 {
t.Fatalf("expected 1 PSU (skipped empty), got %d", len(result)) t.Fatalf("expected 1 PSU (skipped empty), got %d", len(result))
@@ -506,3 +574,75 @@ func TestInferTargetHost(t *testing.T) {
}) })
} }
} }
func TestConvertToReanimator_DeduplicatesAllSections(t *testing.T) {
input := &models.AnalysisResult{
Filename: "dup-test.json",
CollectedAt: time.Date(2026, 2, 10, 15, 30, 0, 0, time.UTC),
Hardware: &models.HardwareConfig{
BoardInfo: models.BoardInfo{SerialNumber: "BOARD-001"},
Firmware: []models.FirmwareInfo{
{DeviceName: "BMC", Version: "1.0"},
{DeviceName: "BMC", Version: "1.1"},
},
CPUs: []models.CPU{
{Socket: 0, Model: "CPU-A"},
{Socket: 0, Model: "CPU-A-DUP"},
},
Memory: []models.MemoryDIMM{
{Slot: "DIMM_A1", Present: true, SerialNumber: "MEM-1", Status: "OK"},
{Slot: "DIMM_A1", Present: true, SerialNumber: "MEM-1-DUP", Status: "OK"},
},
Storage: []models.Storage{
{Slot: "U.2-1", SerialNumber: "SSD-1", Model: "Disk1", Present: true},
{Slot: "U.2-2", SerialNumber: "SSD-1", Model: "Disk1-dup", Present: true},
},
PCIeDevices: []models.PCIeDevice{
{Slot: "#GPU0", DeviceClass: "3D Controller", BDF: "17:00.0"},
{Slot: "SLOT-NIC1", DeviceClass: "NetworkController", BDF: "18:00.0"},
{Slot: "SLOT-NIC1", DeviceClass: "NetworkController", BDF: "18:00.1"},
},
GPUs: []models.GPU{
{Slot: "#GPU0", Model: "B200 180GB HBM3e", SerialNumber: "GPU-1", Status: "OK"},
},
PowerSupply: []models.PSU{
{Slot: "0", Present: true, SerialNumber: "PSU-1", Status: "OK"},
{Slot: "1", Present: true, SerialNumber: "PSU-1", Status: "OK"},
},
},
}
out, err := ConvertToReanimator(input)
if err != nil {
t.Fatalf("ConvertToReanimator() failed: %v", err)
}
if len(out.Hardware.Firmware) != 1 {
t.Fatalf("expected deduped firmware len=1, got %d", len(out.Hardware.Firmware))
}
if len(out.Hardware.CPUs) != 1 {
t.Fatalf("expected deduped cpus len=1, got %d", len(out.Hardware.CPUs))
}
if len(out.Hardware.Memory) != 1 {
t.Fatalf("expected deduped memory len=1, got %d", len(out.Hardware.Memory))
}
if len(out.Hardware.Storage) != 1 {
t.Fatalf("expected deduped storage len=1, got %d", len(out.Hardware.Storage))
}
if len(out.Hardware.PowerSupplies) != 1 {
t.Fatalf("expected deduped psu len=1, got %d", len(out.Hardware.PowerSupplies))
}
if len(out.Hardware.PCIeDevices) != 2 {
t.Fatalf("expected deduped pcie len=2 (gpu+nic), got %d", len(out.Hardware.PCIeDevices))
}
gpuCount := 0
for _, dev := range out.Hardware.PCIeDevices {
if dev.Slot == "#GPU0" {
gpuCount++
}
}
if gpuCount != 1 {
t.Fatalf("expected single #GPU0 record, got %d", gpuCount)
}
}

View File

@@ -36,6 +36,17 @@ type ReanimatorFirmware struct {
Version string `json:"version"` Version string `json:"version"`
} }
type ReanimatorStatusAtCollection struct {
Status string `json:"status"`
At string `json:"at"`
}
type ReanimatorStatusHistoryEntry struct {
Status string `json:"status"`
ChangedAt string `json:"changed_at"`
Details string `json:"details,omitempty"`
}
// ReanimatorCPU represents processor information // ReanimatorCPU represents processor information
type ReanimatorCPU struct { type ReanimatorCPU struct {
Socket int `json:"socket"` Socket int `json:"socket"`
@@ -46,6 +57,11 @@ type ReanimatorCPU struct {
MaxFrequencyMHz int `json:"max_frequency_mhz,omitempty"` MaxFrequencyMHz int `json:"max_frequency_mhz,omitempty"`
Manufacturer string `json:"manufacturer,omitempty"` Manufacturer string `json:"manufacturer,omitempty"`
Status string `json:"status,omitempty"` Status string `json:"status,omitempty"`
StatusCheckedAt string `json:"status_checked_at,omitempty"`
StatusChangedAt string `json:"status_changed_at,omitempty"`
StatusAtCollect *ReanimatorStatusAtCollection `json:"status_at_collection,omitempty"`
StatusHistory []ReanimatorStatusHistoryEntry `json:"status_history,omitempty"`
ErrorDescription string `json:"error_description,omitempty"`
} }
// ReanimatorMemory represents a memory module (DIMM) // ReanimatorMemory represents a memory module (DIMM)
@@ -61,6 +77,11 @@ type ReanimatorMemory struct {
SerialNumber string `json:"serial_number,omitempty"` SerialNumber string `json:"serial_number,omitempty"`
PartNumber string `json:"part_number,omitempty"` PartNumber string `json:"part_number,omitempty"`
Status string `json:"status,omitempty"` Status string `json:"status,omitempty"`
StatusCheckedAt string `json:"status_checked_at,omitempty"`
StatusChangedAt string `json:"status_changed_at,omitempty"`
StatusAtCollect *ReanimatorStatusAtCollection `json:"status_at_collection,omitempty"`
StatusHistory []ReanimatorStatusHistoryEntry `json:"status_history,omitempty"`
ErrorDescription string `json:"error_description,omitempty"`
} }
// ReanimatorStorage represents a storage device // ReanimatorStorage represents a storage device
@@ -75,6 +96,11 @@ type ReanimatorStorage struct {
Interface string `json:"interface,omitempty"` Interface string `json:"interface,omitempty"`
Present bool `json:"present"` Present bool `json:"present"`
Status string `json:"status,omitempty"` Status string `json:"status,omitempty"`
StatusCheckedAt string `json:"status_checked_at,omitempty"`
StatusChangedAt string `json:"status_changed_at,omitempty"`
StatusAtCollect *ReanimatorStatusAtCollection `json:"status_at_collection,omitempty"`
StatusHistory []ReanimatorStatusHistoryEntry `json:"status_history,omitempty"`
ErrorDescription string `json:"error_description,omitempty"`
} }
// ReanimatorPCIe represents a PCIe device // ReanimatorPCIe represents a PCIe device
@@ -93,6 +119,11 @@ type ReanimatorPCIe struct {
SerialNumber string `json:"serial_number,omitempty"` SerialNumber string `json:"serial_number,omitempty"`
Firmware string `json:"firmware,omitempty"` Firmware string `json:"firmware,omitempty"`
Status string `json:"status,omitempty"` Status string `json:"status,omitempty"`
StatusCheckedAt string `json:"status_checked_at,omitempty"`
StatusChangedAt string `json:"status_changed_at,omitempty"`
StatusAtCollect *ReanimatorStatusAtCollection `json:"status_at_collection,omitempty"`
StatusHistory []ReanimatorStatusHistoryEntry `json:"status_history,omitempty"`
ErrorDescription string `json:"error_description,omitempty"`
} }
// ReanimatorPSU represents a power supply unit // ReanimatorPSU represents a power supply unit
@@ -110,4 +141,9 @@ type ReanimatorPSU struct {
InputPowerW int `json:"input_power_w,omitempty"` InputPowerW int `json:"input_power_w,omitempty"`
OutputPowerW int `json:"output_power_w,omitempty"` OutputPowerW int `json:"output_power_w,omitempty"`
InputVoltage float64 `json:"input_voltage,omitempty"` InputVoltage float64 `json:"input_voltage,omitempty"`
StatusCheckedAt string `json:"status_checked_at,omitempty"`
StatusChangedAt string `json:"status_changed_at,omitempty"`
StatusAtCollect *ReanimatorStatusAtCollection `json:"status_at_collection,omitempty"`
StatusHistory []ReanimatorStatusHistoryEntry `json:"status_history,omitempty"`
ErrorDescription string `json:"error_description,omitempty"`
} }

View File

@@ -43,6 +43,19 @@ const (
SeverityInfo Severity = "info" SeverityInfo Severity = "info"
) )
// StatusAtCollection captures component status at a specific timestamp.
type StatusAtCollection struct {
Status string `json:"status"`
At time.Time `json:"at"`
}
// StatusHistoryEntry represents a status transition point.
type StatusHistoryEntry struct {
Status string `json:"status"`
ChangedAt time.Time `json:"changed_at"`
Details string `json:"details,omitempty"`
}
// SensorReading represents a single sensor reading // SensorReading represents a single sensor reading
type SensorReading struct { type SensorReading struct {
Name string `json:"name"` Name string `json:"name"`
@@ -112,6 +125,13 @@ type CPU struct {
TDP int `json:"tdp_w,omitempty"` TDP int `json:"tdp_w,omitempty"`
PPIN string `json:"ppin,omitempty"` PPIN string `json:"ppin,omitempty"`
SerialNumber string `json:"serial_number,omitempty"` SerialNumber string `json:"serial_number,omitempty"`
Status string `json:"status,omitempty"`
StatusCheckedAt time.Time `json:"status_checked_at,omitempty"`
StatusChangedAt time.Time `json:"status_changed_at,omitempty"`
StatusAtCollect *StatusAtCollection `json:"status_at_collection,omitempty"`
StatusHistory []StatusHistoryEntry `json:"status_history,omitempty"`
ErrorDescription string `json:"error_description,omitempty"`
} }
// MemoryDIMM represents a memory module // MemoryDIMM represents a memory module
@@ -129,6 +149,12 @@ type MemoryDIMM struct {
PartNumber string `json:"part_number,omitempty"` PartNumber string `json:"part_number,omitempty"`
Status string `json:"status,omitempty"` Status string `json:"status,omitempty"`
Ranks int `json:"ranks,omitempty"` Ranks int `json:"ranks,omitempty"`
StatusCheckedAt time.Time `json:"status_checked_at,omitempty"`
StatusChangedAt time.Time `json:"status_changed_at,omitempty"`
StatusAtCollect *StatusAtCollection `json:"status_at_collection,omitempty"`
StatusHistory []StatusHistoryEntry `json:"status_history,omitempty"`
ErrorDescription string `json:"error_description,omitempty"`
} }
// Storage represents a storage device // Storage represents a storage device
@@ -144,6 +170,13 @@ type Storage struct {
Present bool `json:"present"` Present bool `json:"present"`
Location string `json:"location,omitempty"` // Front/Rear Location string `json:"location,omitempty"` // Front/Rear
BackplaneID int `json:"backplane_id,omitempty"` BackplaneID int `json:"backplane_id,omitempty"`
Status string `json:"status,omitempty"`
StatusCheckedAt time.Time `json:"status_checked_at,omitempty"`
StatusChangedAt time.Time `json:"status_changed_at,omitempty"`
StatusAtCollect *StatusAtCollection `json:"status_at_collection,omitempty"`
StatusHistory []StatusHistoryEntry `json:"status_history,omitempty"`
ErrorDescription string `json:"error_description,omitempty"`
} }
// PCIeDevice represents a PCIe device // PCIeDevice represents a PCIe device
@@ -161,6 +194,13 @@ type PCIeDevice struct {
PartNumber string `json:"part_number,omitempty"` PartNumber string `json:"part_number,omitempty"`
SerialNumber string `json:"serial_number,omitempty"` SerialNumber string `json:"serial_number,omitempty"`
MACAddresses []string `json:"mac_addresses,omitempty"` MACAddresses []string `json:"mac_addresses,omitempty"`
Status string `json:"status,omitempty"`
StatusCheckedAt time.Time `json:"status_checked_at,omitempty"`
StatusChangedAt time.Time `json:"status_changed_at,omitempty"`
StatusAtCollect *StatusAtCollection `json:"status_at_collection,omitempty"`
StatusHistory []StatusHistoryEntry `json:"status_history,omitempty"`
ErrorDescription string `json:"error_description,omitempty"`
} }
// NIC represents a network interface card // NIC represents a network interface card
@@ -189,6 +229,12 @@ type PSU struct {
InputVoltage float64 `json:"input_voltage,omitempty"` InputVoltage float64 `json:"input_voltage,omitempty"`
OutputVoltage float64 `json:"output_voltage,omitempty"` OutputVoltage float64 `json:"output_voltage,omitempty"`
TemperatureC int `json:"temperature_c,omitempty"` TemperatureC int `json:"temperature_c,omitempty"`
StatusCheckedAt time.Time `json:"status_checked_at,omitempty"`
StatusChangedAt time.Time `json:"status_changed_at,omitempty"`
StatusAtCollect *StatusAtCollection `json:"status_at_collection,omitempty"`
StatusHistory []StatusHistoryEntry `json:"status_history,omitempty"`
ErrorDescription string `json:"error_description,omitempty"`
} }
// GPU represents a graphics processing unit // GPU represents a graphics processing unit
@@ -220,6 +266,12 @@ type GPU struct {
CurrentLinkWidth int `json:"current_link_width,omitempty"` CurrentLinkWidth int `json:"current_link_width,omitempty"`
CurrentLinkSpeed string `json:"current_link_speed,omitempty"` CurrentLinkSpeed string `json:"current_link_speed,omitempty"`
Status string `json:"status,omitempty"` Status string `json:"status,omitempty"`
StatusCheckedAt time.Time `json:"status_checked_at,omitempty"`
StatusChangedAt time.Time `json:"status_changed_at,omitempty"`
StatusAtCollect *StatusAtCollection `json:"status_at_collection,omitempty"`
StatusHistory []StatusHistoryEntry `json:"status_history,omitempty"`
ErrorDescription string `json:"error_description,omitempty"`
} }
// NetworkAdapter represents a network adapter with detailed info // NetworkAdapter represents a network adapter with detailed info
@@ -238,4 +290,10 @@ type NetworkAdapter struct {
PortType string `json:"port_type,omitempty"` PortType string `json:"port_type,omitempty"`
MACAddresses []string `json:"mac_addresses,omitempty"` MACAddresses []string `json:"mac_addresses,omitempty"`
Status string `json:"status,omitempty"` Status string `json:"status,omitempty"`
StatusCheckedAt time.Time `json:"status_checked_at,omitempty"`
StatusChangedAt time.Time `json:"status_changed_at,omitempty"`
StatusAtCollect *StatusAtCollection `json:"status_at_collection,omitempty"`
StatusHistory []StatusHistoryEntry `json:"status_history,omitempty"`
ErrorDescription string `json:"error_description,omitempty"`
} }

View File

@@ -2,6 +2,7 @@ package inspur
import ( import (
"regexp" "regexp"
"sort"
"strconv" "strconv"
"strings" "strings"
@@ -15,13 +16,84 @@ func applyGPUStatusFromEvents(hw *models.HardwareConfig, events []models.Event)
return return
} }
faulty := make(map[int]bool) gpuByIndex := make(map[int]*models.GPU)
for _, e := range events { for i := range hw.GPUs {
if !isGPUFaultEvent(e) { gpu := &hw.GPUs[i]
idx, ok := extractLogicalGPUIndex(gpu.Slot)
if !ok {
continue continue
} }
gpuByIndex[idx] = gpu
gpu.StatusHistory = nil
gpu.ErrorDescription = ""
}
matches := reFaultGPU.FindAllStringSubmatch(e.Description, -1) relevantEvents := make([]models.Event, 0)
for _, e := range events {
if !isGPUFaultEvent(e) || len(extractFaultyGPUSet(e.Description)) == 0 {
continue
}
relevantEvents = append(relevantEvents, e)
}
if len(relevantEvents) == 0 {
for _, gpu := range gpuByIndex {
if strings.TrimSpace(gpu.Status) == "" {
gpu.Status = "OK"
}
}
return
}
sort.Slice(relevantEvents, func(i, j int) bool {
return relevantEvents[i].Timestamp.Before(relevantEvents[j].Timestamp)
})
currentStatus := make(map[int]string, len(gpuByIndex))
lastCriticalDetails := make(map[int]string, len(gpuByIndex))
for idx := range gpuByIndex {
currentStatus[idx] = "OK"
}
for _, e := range relevantEvents {
faultySet := extractFaultyGPUSet(e.Description)
for idx, gpu := range gpuByIndex {
newStatus := "OK"
if faultySet[idx] {
newStatus = "Critical"
lastCriticalDetails[idx] = strings.TrimSpace(e.Description)
}
if currentStatus[idx] != newStatus {
gpu.StatusHistory = append(gpu.StatusHistory, models.StatusHistoryEntry{
Status: newStatus,
ChangedAt: e.Timestamp,
Details: strings.TrimSpace(e.Description),
})
gpu.StatusChangedAt = e.Timestamp
currentStatus[idx] = newStatus
}
gpu.StatusCheckedAt = e.Timestamp
}
}
for idx, gpu := range gpuByIndex {
gpu.Status = currentStatus[idx]
if gpu.Status == "Critical" {
gpu.ErrorDescription = lastCriticalDetails[idx]
} else {
gpu.ErrorDescription = ""
}
if gpu.StatusCheckedAt.IsZero() && strings.TrimSpace(gpu.Status) == "" {
gpu.Status = "OK"
}
}
}
func extractFaultyGPUSet(description string) map[int]bool {
faulty := make(map[int]bool)
matches := reFaultGPU.FindAllStringSubmatch(description, -1)
for _, m := range matches { for _, m := range matches {
if len(m) < 2 { if len(m) < 2 {
continue continue
@@ -31,20 +103,7 @@ func applyGPUStatusFromEvents(hw *models.HardwareConfig, events []models.Event)
faulty[idx] = true faulty[idx] = true
} }
} }
} return faulty
for i := range hw.GPUs {
gpu := &hw.GPUs[i]
idx, ok := extractLogicalGPUIndex(gpu.Slot)
if ok && faulty[idx] {
gpu.Status = "Critical"
continue
}
if strings.TrimSpace(gpu.Status) == "" {
gpu.Status = "OK"
}
}
} }
func isGPUFaultEvent(e models.Event) bool { func isGPUFaultEvent(e models.Event) bool {

View File

@@ -0,0 +1,69 @@
package inspur
import (
"testing"
"git.mchus.pro/mchus/logpile/internal/models"
)
func TestAppendHGXFirmwareFromHWInfo_AppendsInventoryEntries(t *testing.T) {
hw := &models.HardwareConfig{
Firmware: []models.FirmwareInfo{
{DeviceName: "BIOS", Version: "1.0.0"},
},
}
content := []byte(`
{
"@odata.id": "/redfish/v1/UpdateService/FirmwareInventory/HGX_FW_BMC_0",
"Id": "HGX_FW_BMC_0",
"Oem": {
"Nvidia": {
"ActiveFirmwareSlot": {"Version": "25.05-A"},
"InactiveFirmwareSlot": {"Version": "25.04-B"}
}
},
"Version": "25.05-A",
"WriteProtected": false
}
{
"@odata.id": "/redfish/v1/UpdateService/FirmwareInventory/HGX_FW_GPU_SXM_1",
"Id": "HGX_FW_GPU_SXM_1",
"Version": "97.00.C5.00.0E",
"WriteProtected": false
}
{
"@odata.id": "/redfish/v1/UpdateService/FirmwareInventory/HGX_Driver_GPU_SXM_1",
"Id": "HGX_Driver_GPU_SXM_1",
"Version": "",
"WriteProtected": false
}
`)
appendHGXFirmwareFromHWInfo(content, hw)
if len(hw.Firmware) != 5 {
t.Fatalf("expected 5 firmware entries after append, got %d", len(hw.Firmware))
}
seen := make(map[string]string)
for _, fw := range hw.Firmware {
seen[fw.DeviceName] = fw.Version
}
if seen["HGX_FW_BMC_0"] != "25.05-A" {
t.Fatalf("expected HGX_FW_BMC_0 version 25.05-A, got %q", seen["HGX_FW_BMC_0"])
}
if seen["HGX_FW_BMC_0 Active Slot"] != "25.05-A" {
t.Fatalf("expected active slot version, got %q", seen["HGX_FW_BMC_0 Active Slot"])
}
if seen["HGX_FW_BMC_0 Inactive Slot"] != "25.04-B" {
t.Fatalf("expected inactive slot version, got %q", seen["HGX_FW_BMC_0 Inactive Slot"])
}
if seen["HGX_FW_GPU_SXM_1"] != "97.00.C5.00.0E" {
t.Fatalf("expected GPU FW entry, got %q", seen["HGX_FW_GPU_SXM_1"])
}
if _, ok := seen["HGX_Driver_GPU_SXM_1"]; ok {
t.Fatalf("did not expect empty version driver entry")
}
}

View File

@@ -24,6 +24,10 @@ func TestEnrichGPUsFromHGXHWInfo_UsesHGXLogicalMapping(t *testing.T) {
{"Name":"GPU Board Assembly","Model":"B200 180GB HBM3e","PartNumber":"PN3","SerialNumber":"SXM3SN"} {"Name":"GPU Board Assembly","Model":"B200 180GB HBM3e","PartNumber":"PN3","SerialNumber":"SXM3SN"}
# curl -X GET http://127.0.0.1/redfish/v1/Chassis/HGX_GPU_SXM_5/Assembly # curl -X GET http://127.0.0.1/redfish/v1/Chassis/HGX_GPU_SXM_5/Assembly
{"Name":"GPU Board Assembly","Model":"B200 180GB HBM3e","PartNumber":"PN5","SerialNumber":"SXM5SN"} {"Name":"GPU Board Assembly","Model":"B200 180GB HBM3e","PartNumber":"PN5","SerialNumber":"SXM5SN"}
{"Id":"HGX_FW_GPU_SXM_1","Version":"FW1"}
{"Id":"HGX_FW_GPU_SXM_3","Version":"FW3"}
{"Id":"HGX_FW_GPU_SXM_5","Version":"FW5"}
{"Id":"HGX_InfoROM_GPU_SXM_3","Version":"IR3"}
`) `)
enrichGPUsFromHGXHWInfo(content, hw) enrichGPUsFromHGXHWInfo(content, hw)
@@ -37,6 +41,15 @@ func TestEnrichGPUsFromHGXHWInfo_UsesHGXLogicalMapping(t *testing.T) {
if hw.GPUs[2].SerialNumber != "SXM5SN" { if hw.GPUs[2].SerialNumber != "SXM5SN" {
t.Fatalf("expected #GPU0 to map to SXM5 serial, got %q", hw.GPUs[2].SerialNumber) t.Fatalf("expected #GPU0 to map to SXM5 serial, got %q", hw.GPUs[2].SerialNumber)
} }
if hw.GPUs[0].Firmware != "FW3" {
t.Fatalf("expected #GPU6 firmware FW3, got %q", hw.GPUs[0].Firmware)
}
if hw.GPUs[0].VideoBIOS != "IR3" {
t.Fatalf("expected #GPU6 InfoROM in VideoBIOS IR3, got %q", hw.GPUs[0].VideoBIOS)
}
if hw.GPUs[2].Firmware != "FW5" {
t.Fatalf("expected #GPU0 firmware FW5, got %q", hw.GPUs[2].Firmware)
}
for _, g := range hw.GPUs { for _, g := range hw.GPUs {
if g.Slot == "#CPU0_PE1_E_BMC" { if g.Slot == "#CPU0_PE1_E_BMC" {
t.Fatalf("expected non-HGX BMC VGA entry to be filtered out") t.Fatalf("expected non-HGX BMC VGA entry to be filtered out")
@@ -104,6 +117,44 @@ func TestApplyGPUStatusFromEvents_MarksFaultedGPU(t *testing.T) {
} }
} }
func TestApplyGPUStatusFromEvents_UsesLatestEventAsCurrentStatusAndKeepsHistory(t *testing.T) {
hw := &models.HardwareConfig{
GPUs: []models.GPU{
{Slot: "#GPU1"},
{Slot: "#GPU3"},
{Slot: "#GPU6"},
},
}
events := []models.Event{
{
ID: "17FFB002",
Timestamp: time.Date(2026, 1, 12, 22, 51, 16, 0, time.FixedZone("UTC+8", 8*3600)),
Description: "PCIe Present mismatch BIOS miss F_GPU1 F_GPU3 F_GPU6",
},
{
ID: "17FFB002",
Timestamp: time.Date(2026, 1, 12, 23, 5, 18, 0, time.FixedZone("UTC+8", 8*3600)),
Description: "PCIe Present mismatch BIOS miss F_GPU6",
},
}
applyGPUStatusFromEvents(hw, events)
if hw.GPUs[0].Status != "OK" {
t.Fatalf("expected #GPU1 to recover to OK on latest event, got %q", hw.GPUs[0].Status)
}
if hw.GPUs[1].Status != "OK" {
t.Fatalf("expected #GPU3 to recover to OK on latest event, got %q", hw.GPUs[1].Status)
}
if hw.GPUs[2].Status != "Critical" {
t.Fatalf("expected #GPU6 to remain Critical, got %q", hw.GPUs[2].Status)
}
if len(hw.GPUs[0].StatusHistory) == 0 {
t.Fatalf("expected #GPU1 status history to be populated")
}
}
func TestParseIDLLog_ParsesStructuredJSONLine(t *testing.T) { func TestParseIDLLog_ParsesStructuredJSONLine(t *testing.T) {
content := []byte(`{ "MESSAGE": "|2026-01-12T23:05:18+08:00|PCIE|Assert|Critical|17FFB002|PCIe Present mismatch BIOS miss F_GPU6 - Assert|" }`) content := []byte(`{ "MESSAGE": "|2026-01-12T23:05:18+08:00|PCIE|Assert|Critical|17FFB002|PCIe Present mismatch BIOS miss F_GPU6 - Assert|" }`)

View File

@@ -15,6 +15,18 @@ type hgxGPUAssemblyInfo struct {
Serial string Serial string
} }
type hgxGPUFirmwareInfo struct {
Firmware string
InfoROM string
}
type hgxFirmwareInventoryEntry struct {
ID string
Version string
ActiveVersion string
InactiveVersion string
}
// Logical GPU index mapping used by HGX B200 UI ordering. // Logical GPU index mapping used by HGX B200 UI ordering.
// Example from real logs/UI: // Example from real logs/UI:
// GPU0->SXM5, GPU1->SXM7, GPU2->SXM6, GPU3->SXM8, GPU4->SXM2, GPU5->SXM4, GPU6->SXM3, GPU7->SXM1. // GPU0->SXM5, GPU1->SXM7, GPU2->SXM6, GPU3->SXM8, GPU4->SXM2, GPU5->SXM4, GPU6->SXM3, GPU7->SXM1.
@@ -31,6 +43,10 @@ var hgxLogicalToSXM = map[int]int{
var ( var (
reHGXGPUBlock = regexp.MustCompile(`(?s)/redfish/v1/Chassis/HGX_GPU_SXM_(\d+)/Assembly.*?"Name":\s*"GPU Board Assembly".*?"Model":\s*"([^"]+)".*?"PartNumber":\s*"([^"]+)".*?"SerialNumber":\s*"([^"]+)"`) reHGXGPUBlock = regexp.MustCompile(`(?s)/redfish/v1/Chassis/HGX_GPU_SXM_(\d+)/Assembly.*?"Name":\s*"GPU Board Assembly".*?"Model":\s*"([^"]+)".*?"PartNumber":\s*"([^"]+)".*?"SerialNumber":\s*"([^"]+)"`)
reHGXFWBlock = regexp.MustCompile(`(?s)"Id":\s*"HGX_FW_GPU_SXM_(\d+)".*?"Version":\s*"([^"]*)"`)
reHGXInfoROM = regexp.MustCompile(`(?s)"Id":\s*"HGX_InfoROM_GPU_SXM_(\d+)".*?"Version":\s*"([^"]*)"`)
reIDLine = regexp.MustCompile(`"Id":\s*"([^"]+)"`)
reVersion = regexp.MustCompile(`"Version":\s*"([^"]*)"`)
reSlotGPU = regexp.MustCompile(`(?i)gpu\s*#?\s*(\d+)`) reSlotGPU = regexp.MustCompile(`(?i)gpu\s*#?\s*(\d+)`)
) )
@@ -43,6 +59,7 @@ func enrichGPUsFromHGXHWInfo(content []byte, hw *models.HardwareConfig) {
if len(bySXM) == 0 { if len(bySXM) == 0 {
return return
} }
fwBySXM := parseHGXGPUFirmware(content)
normalizeHGXGPUInventory(hw, bySXM) normalizeHGXGPUInventory(hw, bySXM)
@@ -72,6 +89,57 @@ func enrichGPUsFromHGXHWInfo(content []byte, hw *models.HardwareConfig) {
if strings.TrimSpace(gpu.Manufacturer) == "" { if strings.TrimSpace(gpu.Manufacturer) == "" {
gpu.Manufacturer = "NVIDIA" gpu.Manufacturer = "NVIDIA"
} }
if fw, ok := fwBySXM[sxm]; ok {
if strings.TrimSpace(gpu.Firmware) == "" && strings.TrimSpace(fw.Firmware) != "" {
gpu.Firmware = fw.Firmware
}
if strings.TrimSpace(gpu.VideoBIOS) == "" && strings.TrimSpace(fw.InfoROM) != "" {
gpu.VideoBIOS = fw.InfoROM
}
}
}
}
func appendHGXFirmwareFromHWInfo(content []byte, hw *models.HardwareConfig) {
if hw == nil || len(content) == 0 {
return
}
entries := parseHGXFirmwareInventory(content)
if len(entries) == 0 {
return
}
existing := make(map[string]bool, len(hw.Firmware))
for _, fw := range hw.Firmware {
key := strings.ToLower(strings.TrimSpace(fw.DeviceName) + "|" + strings.TrimSpace(fw.Version))
existing[key] = true
}
appendFW := func(name, version string) {
name = strings.TrimSpace(name)
version = strings.TrimSpace(version)
if name == "" || version == "" {
return
}
key := strings.ToLower(name + "|" + version)
if existing[key] {
return
}
existing[key] = true
hw.Firmware = append(hw.Firmware, models.FirmwareInfo{
DeviceName: name,
Version: version,
})
}
for _, e := range entries {
appendFW(e.ID, e.Version)
if e.ActiveVersion != "" && e.InactiveVersion != "" && e.ActiveVersion != e.InactiveVersion {
appendFW(e.ID+" Active Slot", e.ActiveVersion)
appendFW(e.ID+" Inactive Slot", e.InactiveVersion)
}
} }
} }
@@ -97,6 +165,123 @@ func parseHGXGPUAssembly(content []byte) map[int]hgxGPUAssemblyInfo {
return result return result
} }
func parseHGXGPUFirmware(content []byte) map[int]hgxGPUFirmwareInfo {
result := make(map[int]hgxGPUFirmwareInfo)
matchesFW := reHGXFWBlock.FindAllSubmatch(content, -1)
for _, m := range matchesFW {
if len(m) != 3 {
continue
}
sxmIdx, err := strconv.Atoi(string(m[1]))
if err != nil || sxmIdx <= 0 {
continue
}
version := strings.TrimSpace(string(m[2]))
if version == "" {
continue
}
current := result[sxmIdx]
if current.Firmware == "" {
current.Firmware = version
}
result[sxmIdx] = current
}
matchesInfoROM := reHGXInfoROM.FindAllSubmatch(content, -1)
for _, m := range matchesInfoROM {
if len(m) != 3 {
continue
}
sxmIdx, err := strconv.Atoi(string(m[1]))
if err != nil || sxmIdx <= 0 {
continue
}
version := strings.TrimSpace(string(m[2]))
if version == "" {
continue
}
current := result[sxmIdx]
if current.InfoROM == "" {
current.InfoROM = version
}
result[sxmIdx] = current
}
return result
}
func parseHGXFirmwareInventory(content []byte) []hgxFirmwareInventoryEntry {
lines := strings.Split(string(content), "\n")
result := make([]hgxFirmwareInventoryEntry, 0)
var current *hgxFirmwareInventoryEntry
section := ""
flush := func() {
if current == nil {
return
}
if current.Version == "" && current.ActiveVersion == "" && current.InactiveVersion == "" {
current = nil
section = ""
return
}
result = append(result, *current)
current = nil
section = ""
}
for _, line := range lines {
if m := reIDLine.FindStringSubmatch(line); len(m) > 1 {
flush()
id := strings.TrimSpace(m[1])
if strings.HasPrefix(id, "HGX_") {
current = &hgxFirmwareInventoryEntry{ID: id}
}
continue
}
if current == nil {
continue
}
if strings.Contains(line, `"ActiveFirmwareSlot"`) {
section = "active"
}
if strings.Contains(line, `"InactiveFirmwareSlot"`) {
section = "inactive"
}
if m := reVersion.FindStringSubmatch(line); len(m) > 1 {
version := strings.TrimSpace(m[1])
if version == "" {
section = ""
continue
}
switch section {
case "active":
if current.ActiveVersion == "" {
current.ActiveVersion = version
}
case "inactive":
if current.InactiveVersion == "" {
current.InactiveVersion = version
}
default:
// Keep top-level version from the last seen plain "Version" in current entry.
current.Version = version
}
section = ""
}
}
flush()
return result
}
func extractLogicalGPUIndex(slot string) (int, bool) { func extractLogicalGPUIndex(slot string) (int, bool) {
m := reSlotGPU.FindStringSubmatch(slot) m := reSlotGPU.FindStringSubmatch(slot)
if len(m) < 2 { if len(m) < 2 {

View File

@@ -161,6 +161,7 @@ func (p *Parser) Parse(files []parser.ExtractedFile) (*models.AnalysisResult, er
// Enrich GPU inventory from HGX Redfish snapshot (serial/model/part mapping). // Enrich GPU inventory from HGX Redfish snapshot (serial/model/part mapping).
if f := parser.FindFileByName(files, "HGX_HWInfo_FWVersion.log"); f != nil && result.Hardware != nil { if f := parser.FindFileByName(files, "HGX_HWInfo_FWVersion.log"); f != nil && result.Hardware != nil {
enrichGPUsFromHGXHWInfo(f.Content, result.Hardware) enrichGPUsFromHGXHWInfo(f.Content, result.Hardware)
appendHGXFirmwareFromHWInfo(f.Content, result.Hardware)
} }
// Mark problematic GPUs from IDL errors like "BIOS miss F_GPU6". // Mark problematic GPUs from IDL errors like "BIOS miss F_GPU6".