v1.3.0: Add multiple vendor parsers and enhanced hardware detection

New parsers:
- NVIDIA Field Diagnostics parser with dmidecode output support
- NVIDIA Bug Report parser with comprehensive hardware extraction
- Supermicro crashdump (CDump.txt) parser
- Generic fallback parser for unrecognized text files

Enhanced GPU parsing (nvidia-bug-report):
- Model and manufacturer detection (NVIDIA H100 80GB HBM3)
- UUID, Video BIOS version, IRQ information
- Bus location (BDF), DMA size/mask, device minor
- PCIe bus type details

New hardware detection (nvidia-bug-report):
- System Information: server S/N, UUID, manufacturer, product name
- CPU: model, S/N, cores, threads, frequencies from dmidecode
- Memory: P/N, S/N, manufacturer, speed for all DIMMs
- Power Supplies: manufacturer, model, S/N, wattage, status
- Network Adapters: Ethernet/InfiniBand controllers with VPD data
  - Model, P/N, S/N from lspci Vital Product Data
  - Port count/type detection (QSFP56, OSFP, etc.)
  - Support for ConnectX-6/7 adapters

Archive handling improvements:
- Plain .gz file support (not just tar.gz)
- Increased size limit for plain gzip files (50MB)
- Better error handling for mixed archive formats

Web interface enhancements:
- Display parser name and filename badges
- Improved file info section with visual indicators

Co-Authored-By: Claude (qwen3-coder:480b) <noreply@anthropic.com>
This commit is contained in:
Mikhail Chusavitin
2026-01-30 17:19:47 +03:00
parent 21f4e5a67e
commit 70cd541d9e
24 changed files with 2930 additions and 12 deletions

View File

@@ -0,0 +1,275 @@
# NVIDIA Bug Report Parser
Парсер для файлов nvidia-bug-report, генерируемых скриптом `nvidia-bug-report.sh`.
## Назначение
Этот парсер обрабатывает диагностические логи NVIDIA драйверов и извлекает:
- Информацию о модулях памяти (из dmidecode)
- Информацию о GPU устройствах
- Версию NVIDIA драйвера
## Формат файла
- Имя файла: `nvidia-bug-report-*.log.gz`
- Формат: Gzip-сжатый текстовый файл
- Генерируется: `nvidia-bug-report.sh` скриптом
## Confidence Score
**85** - высокий приоритет для файлов nvidia-bug-report
## Извлекаемые данные
### 1. System Information (из dmidecode)
Информация о сервере:
- **Serial Number**: Серийный номер сервера (например, 2KD501412)
- **UUID**: Уникальный идентификатор системы (например, 2e4054bc-1dd2-11b2-0284-6b0a21737950)
- **Manufacturer**: Производитель сервера
- **Product Name**: Модель сервера
- **Version**: Версия системы
### 2. CPU Information (из dmidecode)
Для каждого процессора извлекается:
- **Model**: Модель процессора (например, Intel(R) Xeon(R) Platinum 8480+)
- **Serial Number**: Серийный номер (например, 5DB0D6C0DD30ABD8)
- **Core Count**: Количество ядер (например, 56)
- **Thread Count**: Количество потоков (например, 112)
- **Max Speed**: Максимальная частота (например, 3800 MHz)
- **Current Speed**: Текущая частота (например, 2000 MHz)
Пример:
```
Socket 0: Intel(R) Xeon(R) Platinum 8480+
Serial Number: 5DB0D6C0DD30ABD8
Cores: 56, Threads: 112
Frequency: 2000 MHz (Max: 3800 MHz)
```
### 3. Memory Modules (из dmidecode)
Для каждого модуля памяти извлекается:
- **Slot/Location**: Например, CPU0_C0D0
- **Size**: Размер в GB (например, 64 GB)
- **Type**: Тип памяти (DDR5, DDR4, etc.)
- **Manufacturer**: Производитель (Hynix, Samsung, Micron, etc.)
- **Part Number**: P/N модуля (например, HMCG94AGBRA179N)
- **Serial Number**: S/N модуля (например, 80AD0224322B3834E6)
- **Speed**: Max/Current скорость (например, 5600/4400 MHz)
- **Ranks**: Количество рангов
Пример:
```
Slot: CPU0_C0D0
Size: 64 GB
Type: DDR5
Manufacturer: Hynix
Part Number: HMCG94AGBRA179N
Serial Number: 80AD0224322B3834E6
Speed: 5600 MT/s (configured: 4400 MT/s)
Ranks: 2
```
### 4. Power Supplies (из dmidecode)
Для каждого блока питания извлекается:
- **Location**: Позиция (например, PSU0, PSU1)
- **Manufacturer**: Производитель (например, DELTA, Great Wall)
- **Model Part Number**: Модель БП (например, V0310DT000000000)
- **Serial Number**: Серийный номер (например, DGPLV251500LZ)
- **Max Power Capacity**: Максимальная мощность (например, 2700 W)
- **Revision**: Версия прошивки (например, 00.01.04)
- **Status**: Статус (например, Present, OK)
Пример:
```
PSU0: V0310DT000000000 (DELTA)
Serial Number: DGPLV251500LZ
Power: 2700 W, Revision: 00.01.04
Status: Present, OK
```
### 5. Network Adapters (из lspci)
Для каждого сетевого адаптера (Ethernet, Network, InfiniBand) извлекается:
- **Model**: Полное название модели из VPD (например, "NVIDIA ConnectX-7 HHHL Adapter card, 400GbE / NDR IB (default mode), Single-port OSFP, PCIe 5.0 x16")
- **Location**: PCI BDF адрес (например, 0000:0e:00.0)
- **Slot**: Физический слот (например, 108)
- **Part Number**: P/N адаптера (например, MCX75310AAS-NEAT)
- **Serial Number**: S/N адаптера (например, MT2430600249)
- **Vendor**: Производитель (Mellanox, NVIDIA)
- **Vendor ID / Device ID**: PCI идентификаторы (например, 15b3:1021)
- **Port Count**: Количество портов (определяется из модели: Dual-port = 2, Single-port = 1)
- **Port Type**: Тип портов (QSFP56, OSFP, SFP+)
Пример:
```
0000:0e:00.0: NVIDIA ConnectX-7 HHHL Adapter card, 400GbE / NDR IB (default mode), Single-port OSFP
Slot: 108
P/N: MCX75310AAS-NEAT
S/N: MT2430600249
Ports: 1 x OSFP
```
### 6. GPU Devices
Для каждого GPU извлекается:
- **Model**: Модель GPU (например, NVIDIA H100 80GB HBM3)
- **BDF (Bus:Device.Function)**: PCI адрес (например, 0000:0f:00.0)
- **UUID**: Уникальный идентификатор GPU (например, GPU-64674e47-e036-c12a-3e8d-55a2a9ac8db3)
- **Video BIOS**: Версия BIOS видеокарты (например, 96.00.99.00.01)
- **IRQ**: Прерывание (например, 17)
- **Bus Type**: Тип шины (PCIe)
- **DMA Size**: Размер DMA (например, 52 bits)
- **DMA Mask**: Маска DMA (например, 0xfffffffffffff)
- **Device Minor**: Номер устройства (например, 0)
- **Manufacturer**: NVIDIA
Пример:
```
0000:0f:00.0: NVIDIA H100 80GB HBM3
UUID: GPU-64674e47-e036-c12a-3e8d-55a2a9ac8db3
Video BIOS: 96.00.99.00.01
IRQ: 17
```
### 7. Events
- **Memory Configuration**: Сводка по модулям памяти (количество, производители, общий размер)
- **GPU Detection**: Обнаруженные GPU устройства
- **Driver Version**: Версия NVIDIA драйвера
## Пример использования
```bash
# Запуск с nvidia-bug-report файлом
./logpile --file nvidia-bug-report-2KD501412.log.gz
# Веб-интерфейс будет доступен на http://localhost:8082
```
## Пример вывода
```
✓ Detected vendor: NVIDIA Bug Report Parser
✓ CPUs: 2
✓ Memory: 32 modules
✓ Power Supplies: 8
✓ GPUs: 8
✓ Network Adapters: 12
System Information:
Serial Number: 2KD501412
UUID: 2e4054bc-1dd2-11b2-0284-6b0a21737950
Version: 0
CPU Information:
Socket 0: Intel(R) Xeon(R) Platinum 8480+
S/N: 5DB0D6C0DD30ABD8, Cores: 56, Threads: 112
Socket 1: Intel(R) Xeon(R) Platinum 8480+
S/N: 5DB017C05685B3ED, Cores: 56, Threads: 112
Power Supplies:
PSU0: V0310DT000000000 (DELTA)
S/N: DGPLV251500LZ
Power: 2700 W, Revision: 00.01.04
Status: Present, OK
PSU1: V0310DT000000000 (DELTA)
S/N: DGPLV251500GY
Power: 2700 W, Revision: 00.01.04
Status: Present, OK
[... 6 more PSUs ...]
Memory Modules:
CPU0_C0D0: 64 GB, Hynix
P/N: HMCG94AGBRA179N, S/N: 80AD0224322B3834E6
Type: DDR5, Speed: 4400/5600 MHz
[... 31 more modules ...]
Network Adapters: 12 devices
0000:0e:00.0: NVIDIA ConnectX-7 HHHL Adapter card, 400GbE / NDR IB (default mode), Single-port OSFP
Slot: 108
P/N: MCX75310AAS-NEAT
S/N: MT2430600249
Ports: 1 x OSFP
0000:1f:00.0: ConnectX-6 Dx EN adapter card, 100GbE, Dual-port QSFP56
Slot: 12
P/N: MCX623106AN-CDAT
S/N: MT2434J00PCD
Ports: 2 x QSFP56
[... 10 more adapters ...]
GPUs: 8 devices
0000:0f:00.0: NVIDIA H100 80GB HBM3
UUID: GPU-64674e47-e036-c12a-3e8d-55a2a9ac8db3
Video BIOS: 96.00.99.00.01
IRQ: 17
0000:34:00.0: NVIDIA H100 80GB HBM3
UUID: GPU-fa796345-c23a-54aa-1b67-709ac2542852
Video BIOS: 96.00.99.00.01
IRQ: 16
[... 6 more GPUs ...]
```
## Версионирование
**Текущая версия парсера:** 1.0.0
### История версий
- **1.0.0** - Первоначальная версия с парсингом System Info, CPU, Memory, PSU, GPU, Network Adapters и Driver
## Структура данных
Парсер использует следующие секции в bug report:
1. **dmidecode output (System Information)** - для извлечения информации о сервере
2. **dmidecode output (Processor Information)** - для извлечения информации о CPU
3. **dmidecode output (Memory Device)** - для извлечения информации о памяти
4. **dmidecode output (System Power Supply)** - для извлечения информации о блоках питания
5. **lspci -vvv output (Ethernet/Network/Infiniband controller)** - для извлечения информации о сетевых адаптерах
6. **lspci VPD (Vital Product Data)** - для извлечения P/N, S/N и модели сетевых адаптеров
7. **/proc/driver/nvidia/gpus/.../information** - для детальной информации о GPU
8. **NVRM version** - для версии драйвера
## Известные ограничения
1. Ошибки и предупреждения из логов пока не извлекаются
2. Некоторые специфичные характеристики GPU (температура, утилизация) не парсятся
3. Информация о производительности и метрики GPU требуют парсинга других секций
## Расширение
Для добавления новых возможностей:
1. **Ошибки драйвера**: Парсить секции с ошибками NVIDIA драйвера
2. **nvidia-smi output**: Извлекать детальную информацию из вывода nvidia-smi (температура, утилизация)
3. **GPU производительность**: Парсить метрики производительности и использования памяти GPU
4. **PCIe информация**: Извлекать детали о PCIe конфигурации (скорость линка, ширина)
## Пример структуры файла
```
Start of NVIDIA bug report log file
nvidia-bug-report.sh Version: 34275561
Date: Thu Jul 17 18:18:18 EDT 2025
[... system info ...]
Memory Device
Data Width: 64 bits
Size: 64 GB
Form Factor: DIMM
Locator: CPU0_C0D0
Type: DDR5
Speed: 5600 MT/s
Manufacturer: Hynix
Serial Number: 80AD0224322B3834E6
Part Number: HMCG94AGBRA179N
[... more memory modules ...]
*** /proc/driver/nvidia/./gpus/0000:0f:00.0/power
[... GPU info ...]
```

View File

@@ -0,0 +1,140 @@
package nvidia_bug_report
import (
"bufio"
"strconv"
"strings"
"git.mchus.pro/mchus/logpile/internal/models"
)
// parseCPUInfo extracts CPU information from dmidecode output
func parseCPUInfo(content string, result *models.AnalysisResult) {
scanner := bufio.NewScanner(strings.NewReader(content))
var currentCPU *models.CPU
inProcessorInfo := false
cpuSocket := 0
for scanner.Scan() {
line := scanner.Text()
trimmed := strings.TrimSpace(line)
// Start of Processor Information section
if strings.Contains(trimmed, "Processor Information") {
inProcessorInfo = true
currentCPU = &models.CPU{
Socket: cpuSocket,
}
cpuSocket++
continue
}
// End of current section (empty line or new section with Handle)
if inProcessorInfo && (trimmed == "" || strings.HasPrefix(trimmed, "Handle ")) {
// Save CPU if it has valid data
if currentCPU != nil && currentCPU.Model != "" {
result.Hardware.CPUs = append(result.Hardware.CPUs, *currentCPU)
}
inProcessorInfo = false
currentCPU = nil
continue
}
// Parse fields within Processor Information section
if inProcessorInfo && currentCPU != nil && strings.Contains(line, ":") {
parts := strings.SplitN(trimmed, ":", 2)
if len(parts) != 2 {
continue
}
field := strings.TrimSpace(parts[0])
value := strings.TrimSpace(parts[1])
if value == "" || value == "Not Specified" || value == "Unknown" || value == "UNKNOWN" || value == "<OUT OF SPEC>" {
continue
}
switch field {
case "Version":
// CPU model name
currentCPU.Model = value
case "Serial Number":
currentCPU.SerialNumber = value
case "Part Number":
// Store part number if available
// Could be stored in a custom field if needed
case "Core Count":
if cores, err := strconv.Atoi(value); err == nil {
currentCPU.Cores = cores
}
case "Core Enabled":
// Could store this if needed
case "Thread Count":
if threads, err := strconv.Atoi(value); err == nil {
currentCPU.Threads = threads
}
case "Max Speed":
// Parse speed like "3800 MHz"
if speed := parseCPUSpeed(value); speed > 0 {
currentCPU.MaxFreqMHz = speed
}
case "Current Speed":
// Parse current speed like "2000 MHz"
if speed := parseCPUSpeed(value); speed > 0 {
currentCPU.FrequencyMHz = speed
}
case "Voltage":
// Could parse voltage if needed (e.g., "1.6 V")
case "Status":
// Status like "Populated, Enabled"
// Check if CPU is enabled
if !strings.Contains(value, "Populated") {
// Skip unpopulated CPUs
currentCPU = nil
inProcessorInfo = false
}
}
}
}
// Save last CPU if exists
if currentCPU != nil && currentCPU.Model != "" {
result.Hardware.CPUs = append(result.Hardware.CPUs, *currentCPU)
}
}
// parseCPUSpeed parses CPU speed strings like "3800 MHz" or "2.0 GHz"
func parseCPUSpeed(speedStr string) int {
parts := strings.Fields(speedStr)
if len(parts) < 2 {
return 0
}
// Try to parse the number (may be int or float)
speedStr = parts[0]
var speed float64
var err error
if strings.Contains(speedStr, ".") {
speed, err = strconv.ParseFloat(speedStr, 64)
} else {
var speedInt int
speedInt, err = strconv.Atoi(speedStr)
speed = float64(speedInt)
}
if err != nil {
return 0
}
unit := strings.ToUpper(parts[1])
switch unit {
case "MHZ":
return int(speed)
case "GHZ":
return int(speed * 1000)
default:
return 0
}
}

View File

@@ -0,0 +1,170 @@
package nvidia_bug_report
import (
"bufio"
"regexp"
"strconv"
"strings"
"time"
"git.mchus.pro/mchus/logpile/internal/models"
)
// parseGPUInfo extracts GPU information from the bug report
func parseGPUInfo(content string, result *models.AnalysisResult) {
scanner := bufio.NewScanner(strings.NewReader(content))
var currentGPU *models.GPU
inGPUInfo := false
for scanner.Scan() {
line := scanner.Text()
// Look for GPU information section markers (but skip ls listings)
if strings.Contains(line, "/proc/driver/nvidia") && strings.Contains(line, "/gpus/") &&
strings.Contains(line, "/information") && !strings.Contains(line, "ls:") {
// Extract PCI address
re := regexp.MustCompile(`/gpus/([\da-f]{4}:[\da-f]{2}:[\da-f]{2}\.[\da-f])`)
matches := re.FindStringSubmatch(line)
if len(matches) > 1 {
pciAddr := matches[1]
// Save previous GPU if exists
if currentGPU != nil {
result.Hardware.GPUs = append(result.Hardware.GPUs, *currentGPU)
}
// Start new GPU entry
currentGPU = &models.GPU{
BDF: pciAddr,
Manufacturer: "NVIDIA",
}
inGPUInfo = true
continue
}
}
// End of GPU info section (separator line or new section, but not ls lines)
if inGPUInfo && (strings.HasPrefix(line, "___") || (strings.HasPrefix(line, "***") && !strings.Contains(line, "ls:"))) {
inGPUInfo = false
continue
}
// Parse GPU fields within information section
if inGPUInfo && currentGPU != nil && strings.Contains(line, ":") {
// Split on first colon and trim whitespace/tabs
parts := strings.SplitN(line, ":", 2)
if len(parts) != 2 {
continue
}
field := strings.TrimSpace(parts[0])
value := strings.TrimSpace(parts[1])
if value == "" {
continue
}
switch field {
case "Model":
currentGPU.Model = value
case "IRQ":
if irq, err := strconv.Atoi(value); err == nil {
currentGPU.IRQ = irq
}
case "GPU UUID":
currentGPU.UUID = value
case "Video BIOS":
currentGPU.VideoBIOS = value
case "Bus Type":
currentGPU.BusType = value
case "DMA Size":
currentGPU.DMASize = value
case "DMA Mask":
currentGPU.DMAMask = value
case "Bus Location":
// BDF already set from path, but verify consistency
if currentGPU.BDF != value {
// Use the value from the information section as it's more explicit
currentGPU.BDF = value
}
case "Device Minor":
if minor, err := strconv.Atoi(value); err == nil {
currentGPU.DeviceMinor = minor
}
case "GPU Excluded":
// Store as status if "Yes"
if strings.ToLower(value) == "yes" {
currentGPU.Status = "Excluded"
}
}
}
}
// Save last GPU if exists
if currentGPU != nil {
result.Hardware.GPUs = append(result.Hardware.GPUs, *currentGPU)
}
// Create event for GPU summary
if len(result.Hardware.GPUs) > 0 {
result.Events = append(result.Events, models.Event{
Timestamp: time.Now(),
Source: "NVIDIA Driver",
EventType: "GPU Detection",
Description: "NVIDIA GPUs detected",
Severity: models.SeverityInfo,
RawData: formatGPUSummary(result.Hardware.GPUs),
})
}
}
// parseDriverVersion extracts NVIDIA driver version
func parseDriverVersion(content string, result *models.AnalysisResult) {
scanner := bufio.NewScanner(strings.NewReader(content))
for scanner.Scan() {
line := scanner.Text()
// Look for NVRM version line
if strings.Contains(line, "NVRM version:") {
// Extract version info
parts := strings.Split(line, "NVRM version:")
if len(parts) > 1 {
version := strings.TrimSpace(parts[1])
result.Events = append(result.Events, models.Event{
Timestamp: time.Now(),
Source: "NVIDIA Driver",
EventType: "Driver Version",
Description: "NVIDIA driver version detected",
Severity: models.SeverityInfo,
RawData: version,
})
break
}
}
}
}
// formatGPUSummary creates a summary string for GPUs
func formatGPUSummary(gpus []models.GPU) string {
if len(gpus) == 0 {
return ""
}
var summary strings.Builder
for i, gpu := range gpus {
if i > 0 {
summary.WriteString("; ")
}
summary.WriteString(gpu.BDF)
if gpu.Model != "" {
summary.WriteString(" (")
summary.WriteString(gpu.Model)
summary.WriteString(")")
}
}
return summary.String()
}

View File

@@ -0,0 +1,183 @@
package nvidia_bug_report
import (
"bufio"
"strconv"
"strings"
"time"
"git.mchus.pro/mchus/logpile/internal/models"
)
// parseMemoryModules extracts memory module information from dmidecode output
func parseMemoryModules(content string, result *models.AnalysisResult) {
scanner := bufio.NewScanner(strings.NewReader(content))
var currentModule *models.MemoryDIMM
inMemoryDevice := false
for scanner.Scan() {
line := scanner.Text()
trimmed := strings.TrimSpace(line)
// Start of Memory Device section
if strings.Contains(trimmed, "Memory Device") && !strings.Contains(trimmed, "Array") {
inMemoryDevice = true
currentModule = &models.MemoryDIMM{
Present: true,
}
continue
}
// End of current section (empty line or new section)
if inMemoryDevice && (trimmed == "" || strings.HasPrefix(trimmed, "Handle ")) {
// Save module if it has valid data
if currentModule != nil && currentModule.Slot != "" && currentModule.SizeMB > 0 {
result.Hardware.Memory = append(result.Hardware.Memory, *currentModule)
}
inMemoryDevice = false
currentModule = nil
continue
}
// Parse fields within Memory Device section
if inMemoryDevice && currentModule != nil && strings.Contains(line, ":") {
parts := strings.SplitN(trimmed, ":", 2)
if len(parts) != 2 {
continue
}
field := strings.TrimSpace(parts[0])
value := strings.TrimSpace(parts[1])
if value == "" || value == "Not Specified" || value == "Unknown" || value == "NO DIMM" {
continue
}
switch field {
case "Size":
// Parse size like "64 GB" or "32768 MB"
currentModule.SizeMB = parseMemorySize(value)
case "Locator":
currentModule.Slot = value
currentModule.Location = value
case "Bank Locator":
// Store in location if slot is empty
if currentModule.Location == "" {
currentModule.Location = value
}
case "Type":
currentModule.Type = value
case "Type Detail":
currentModule.Technology = value
case "Speed":
// Parse speed like "5600 MT/s"
currentModule.MaxSpeedMHz = parseMemorySpeed(value)
case "Configured Memory Speed":
currentModule.CurrentSpeedMHz = parseMemorySpeed(value)
case "Manufacturer":
currentModule.Manufacturer = value
case "Serial Number":
currentModule.SerialNumber = value
case "Part Number":
currentModule.PartNumber = strings.TrimSpace(value)
case "Rank":
// Parse rank
if rank, err := strconv.Atoi(value); err == nil {
currentModule.Ranks = rank
}
}
}
}
// Save last module if exists
if currentModule != nil && currentModule.Slot != "" && currentModule.SizeMB > 0 {
result.Hardware.Memory = append(result.Hardware.Memory, *currentModule)
}
// Create event for memory summary
if len(result.Hardware.Memory) > 0 {
totalMemoryGB := 0
for _, mem := range result.Hardware.Memory {
totalMemoryGB += mem.SizeMB / 1024
}
result.Events = append(result.Events, models.Event{
Timestamp: time.Now(),
Source: "DMI",
EventType: "Memory Configuration",
Description: "Memory modules detected",
Severity: models.SeverityInfo,
RawData: formatMemorySummary(result.Hardware.Memory, totalMemoryGB),
})
}
}
// parseMemorySize parses memory size strings like "64 GB" or "32768 MB"
func parseMemorySize(sizeStr string) int {
parts := strings.Fields(sizeStr)
if len(parts) < 2 {
return 0
}
size, err := strconv.Atoi(parts[0])
if err != nil {
return 0
}
unit := strings.ToUpper(parts[1])
switch unit {
case "GB":
return size * 1024
case "MB":
return size
case "TB":
return size * 1024 * 1024
default:
return 0
}
}
// parseMemorySpeed parses speed strings like "5600 MT/s" or "4400 MHz"
func parseMemorySpeed(speedStr string) int {
parts := strings.Fields(speedStr)
if len(parts) < 1 {
return 0
}
speed, err := strconv.Atoi(parts[0])
if err != nil {
return 0
}
return speed
}
// formatMemorySummary creates a summary string for memory modules
func formatMemorySummary(modules []models.MemoryDIMM, totalGB int) string {
if len(modules) == 0 {
return ""
}
// Group by manufacturer
manufacturerCount := make(map[string]int)
for _, mem := range modules {
if mem.Manufacturer != "" {
manufacturerCount[mem.Manufacturer]++
}
}
summary := ""
for mfr, count := range manufacturerCount {
if summary != "" {
summary += ", "
}
summary += mfr + ": " + strconv.Itoa(count) + " modules"
}
if summary == "" {
summary = strconv.Itoa(len(modules)) + " modules"
}
return summary + ", Total: " + strconv.Itoa(totalGB) + " GB"
}

View File

@@ -0,0 +1,160 @@
package nvidia_bug_report
import (
"bufio"
"regexp"
"strconv"
"strings"
"git.mchus.pro/mchus/logpile/internal/models"
)
// parseNetworkAdapters extracts network adapter information from lspci output
func parseNetworkAdapters(content string, result *models.AnalysisResult) {
scanner := bufio.NewScanner(strings.NewReader(content))
var currentAdapter *models.NetworkAdapter
inVPD := false
currentBDF := ""
for scanner.Scan() {
line := scanner.Text()
trimmed := strings.TrimSpace(line)
// Check if this is a new PCI device line
re := regexp.MustCompile(`^([\da-f]{4}:[\da-f]{2}:[\da-f]{2}\.[\da-f])\s+`)
matches := re.FindStringSubmatch(line)
if len(matches) > 0 {
// Save previous adapter if exists before processing new device
if currentAdapter != nil && currentAdapter.Model != "" {
result.Hardware.NetworkAdapters = append(result.Hardware.NetworkAdapters, *currentAdapter)
}
currentAdapter = nil
inVPD = false
}
// Match PCI device line: "0000:1f:00.0 Ethernet controller [0200]: Mellanox Technologies..."
if strings.Contains(line, "Ethernet controller") || strings.Contains(line, "Network controller") || strings.Contains(line, "Infiniband controller") {
// Extract BDF (Bus:Device.Function)
if len(matches) > 1 {
currentBDF = matches[1]
currentAdapter = &models.NetworkAdapter{
Location: currentBDF,
Present: true,
}
// Extract vendor and device info
// Format: "Vendor description [DeviceClass]: Vendor Name Device Name [VendorID:DeviceID]"
re2 := regexp.MustCompile(`:\s+(.+?)\s+\[([0-9a-f]{4}):([0-9a-f]{4})\]`)
matches2 := re2.FindStringSubmatch(line)
if len(matches2) > 3 {
// Parse vendor name from description
vendorDesc := matches2[1]
if idx := strings.Index(vendorDesc, " "); idx > 0 {
currentAdapter.Vendor = strings.Split(vendorDesc, " ")[0]
}
// Parse vendor ID and device ID
if vendorID, err := strconv.ParseInt(matches2[2], 16, 32); err == nil {
currentAdapter.VendorID = int(vendorID)
}
if deviceID, err := strconv.ParseInt(matches2[3], 16, 32); err == nil {
currentAdapter.DeviceID = int(deviceID)
}
}
continue
}
}
// Skip if not processing an adapter
if currentAdapter == nil {
continue
}
// Parse Physical Slot
if strings.HasPrefix(trimmed, "Physical Slot:") {
slotStr := strings.TrimPrefix(trimmed, "Physical Slot:")
currentAdapter.Slot = strings.TrimSpace(slotStr)
continue
}
// Start of Vital Product Data section
if strings.Contains(trimmed, "Vital Product Data") {
inVPD = true
continue
}
// End of VPD section
if inVPD && (trimmed == "End" || strings.HasPrefix(trimmed, "Capabilities:")) {
if trimmed == "End" {
inVPD = false
}
continue
}
// Parse Product Name in VPD
if inVPD && strings.HasPrefix(trimmed, "Product Name:") {
productName := strings.TrimPrefix(trimmed, "Product Name:")
currentAdapter.Model = strings.TrimSpace(productName)
// Extract port count from model name
if strings.Contains(currentAdapter.Model, "Dual-port") {
currentAdapter.PortCount = 2
} else if strings.Contains(currentAdapter.Model, "Single-port") {
currentAdapter.PortCount = 1
} else if strings.Contains(currentAdapter.Model, "Quad-port") {
currentAdapter.PortCount = 4
}
// Extract port type from model name
if strings.Contains(currentAdapter.Model, "QSFP56") {
currentAdapter.PortType = "QSFP56"
} else if strings.Contains(currentAdapter.Model, "QSFP28") {
currentAdapter.PortType = "QSFP28"
} else if strings.Contains(currentAdapter.Model, "OSFP") {
currentAdapter.PortType = "OSFP"
} else if strings.Contains(currentAdapter.Model, "SFP") {
currentAdapter.PortType = "SFP+"
}
continue
}
// Parse VPD fields
if inVPD && strings.HasPrefix(trimmed, "[") {
// Match pattern: [TAG] Description: Value
re := regexp.MustCompile(`^\[([A-Z0-9]+)\]\s+([^:]+):\s+(.+)`)
matches := re.FindStringSubmatch(trimmed)
if len(matches) > 3 {
tag := matches[1]
value := strings.TrimSpace(matches[3])
switch tag {
case "PN":
// Part number
currentAdapter.PartNumber = value
case "SN":
// Serial number
currentAdapter.SerialNumber = value
case "EC":
// Engineering changes - could be stored as firmware/revision
if currentAdapter.Firmware == "" {
currentAdapter.Firmware = value
}
}
}
continue
}
// End of current device section (empty line followed by hex dump or new device)
if currentAdapter != nil && trimmed == "" {
// Check if next lines are hex dump (config space)
continue
}
}
// Save last adapter if exists
if currentAdapter != nil && currentAdapter.Model != "" {
result.Hardware.NetworkAdapters = append(result.Hardware.NetworkAdapters, *currentAdapter)
}
}

View File

@@ -0,0 +1,107 @@
// Package nvidia_bug_report provides parser for NVIDIA bug report files
// Generated by nvidia-bug-report.sh script
package nvidia_bug_report
import (
"strings"
"git.mchus.pro/mchus/logpile/internal/models"
"git.mchus.pro/mchus/logpile/internal/parser"
)
// parserVersion - version of this parser module
const parserVersion = "1.0.0"
func init() {
parser.Register(&Parser{})
}
// Parser implements VendorParser for NVIDIA bug reports
type Parser struct{}
// Name returns human-readable parser name
func (p *Parser) Name() string {
return "NVIDIA Bug Report Parser"
}
// Vendor returns vendor identifier
func (p *Parser) Vendor() string {
return "nvidia_bug_report"
}
// Version returns parser version
func (p *Parser) Version() string {
return parserVersion
}
// Detect checks if this is an NVIDIA bug report
// Returns confidence 0-100
func (p *Parser) Detect(files []parser.ExtractedFile) int {
// Only detect if there's exactly one file
if len(files) != 1 {
return 0
}
file := files[0]
// Check filename
if !strings.Contains(strings.ToLower(file.Path), "nvidia-bug-report") {
return 0
}
// Check content markers
content := string(file.Content)
if !strings.Contains(content, "nvidia-bug-report.sh") ||
!strings.Contains(content, "NVIDIA bug report log file") {
return 0
}
// High confidence for nvidia-bug-report files
return 85
}
// Parse parses NVIDIA bug report file
func (p *Parser) Parse(files []parser.ExtractedFile) (*models.AnalysisResult, error) {
result := &models.AnalysisResult{
Events: make([]models.Event, 0),
FRU: make([]models.FRUInfo, 0),
Sensors: make([]models.SensorReading, 0),
}
// Initialize hardware config
result.Hardware = &models.HardwareConfig{
CPUs: make([]models.CPU, 0),
Memory: make([]models.MemoryDIMM, 0),
GPUs: make([]models.GPU, 0),
PowerSupply: make([]models.PSU, 0),
}
if len(files) == 0 {
return result, nil
}
content := string(files[0].Content)
// Parse system information
parseSystemInfo(content, result)
// Parse CPU information
parseCPUInfo(content, result)
// Parse memory modules
parseMemoryModules(content, result)
// Parse power supplies
parsePSUInfo(content, result)
// Parse GPU information
parseGPUInfo(content, result)
// Parse network adapters
parseNetworkAdapters(content, result)
// Parse driver version
parseDriverVersion(content, result)
return result, nil
}

View File

@@ -0,0 +1,116 @@
package nvidia_bug_report
import (
"bufio"
"strconv"
"strings"
"git.mchus.pro/mchus/logpile/internal/models"
)
// parsePSUInfo extracts Power Supply information from dmidecode output
func parsePSUInfo(content string, result *models.AnalysisResult) {
scanner := bufio.NewScanner(strings.NewReader(content))
var currentPSU *models.PSU
inPowerSupply := false
for scanner.Scan() {
line := scanner.Text()
trimmed := strings.TrimSpace(line)
// Start of System Power Supply section
if strings.Contains(trimmed, "System Power Supply") {
inPowerSupply = true
currentPSU = &models.PSU{}
continue
}
// End of current section (empty line or new section with Handle)
if inPowerSupply && (trimmed == "" || strings.HasPrefix(trimmed, "Handle ")) {
// Save PSU if it has valid data
if currentPSU != nil && currentPSU.Slot != "" {
// Only add if PSU is present
if strings.Contains(strings.ToLower(currentPSU.Status), "present") {
result.Hardware.PowerSupply = append(result.Hardware.PowerSupply, *currentPSU)
}
}
inPowerSupply = false
currentPSU = nil
continue
}
// Parse fields within System Power Supply section
if inPowerSupply && currentPSU != nil && strings.Contains(line, ":") {
parts := strings.SplitN(trimmed, ":", 2)
if len(parts) != 2 {
continue
}
field := strings.TrimSpace(parts[0])
value := strings.TrimSpace(parts[1])
if value == "" || value == "Not Specified" || value == "Unknown" || value == "UNKNOWN" {
continue
}
switch field {
case "Location":
currentPSU.Slot = value
case "Name":
// Use Name as Model if Model is not set later
if currentPSU.Model == "" {
currentPSU.Model = value
}
case "Manufacturer":
currentPSU.Vendor = value
case "Serial Number":
currentPSU.SerialNumber = value
case "Model Part Number":
// Use Model Part Number as the primary model identifier
currentPSU.Model = value
case "Revision":
currentPSU.Firmware = value
case "Max Power Capacity":
// Parse wattage like "2700 W"
if wattage := parsePowerWattage(value); wattage > 0 {
currentPSU.WattageW = wattage
}
case "Status":
currentPSU.Status = value
case "Type":
// Could store PSU type if needed (e.g., "Switching")
case "Plugged":
// Could track if PSU is plugged
case "Hot Replaceable":
// Could track if hot-swappable
}
}
}
// Save last PSU if exists
if currentPSU != nil && currentPSU.Slot != "" {
if strings.Contains(strings.ToLower(currentPSU.Status), "present") {
result.Hardware.PowerSupply = append(result.Hardware.PowerSupply, *currentPSU)
}
}
}
// parsePowerWattage parses power capacity strings like "2700 W" or "1200 Watts"
func parsePowerWattage(powerStr string) int {
parts := strings.Fields(powerStr)
if len(parts) < 1 {
return 0
}
// Try to parse the number
wattageStr := parts[0]
wattage, err := strconv.Atoi(wattageStr)
if err != nil {
return 0
}
// Check if unit is specified (W, Watts, etc.) and convert if needed
// For now, assume it's always in Watts
return wattage
}

View File

@@ -0,0 +1,61 @@
package nvidia_bug_report
import (
"bufio"
"strings"
"git.mchus.pro/mchus/logpile/internal/models"
)
// parseSystemInfo extracts System Information from dmidecode output
func parseSystemInfo(content string, result *models.AnalysisResult) {
scanner := bufio.NewScanner(strings.NewReader(content))
inSystemInfo := false
for scanner.Scan() {
line := scanner.Text()
trimmed := strings.TrimSpace(line)
// Start of System Information section
if trimmed == "System Information" {
inSystemInfo = true
continue
}
// End of section (empty line or new Handle)
if inSystemInfo && (trimmed == "" || strings.HasPrefix(trimmed, "Handle ")) {
inSystemInfo = false
continue
}
// Parse fields within System Information section
if inSystemInfo && strings.Contains(line, ":") {
parts := strings.SplitN(trimmed, ":", 2)
if len(parts) != 2 {
continue
}
field := strings.TrimSpace(parts[0])
value := strings.TrimSpace(parts[1])
// Skip empty, NULL, or "Not specified" values
if value == "" || value == "NULL" || value == "Not specified" || value == "Not Specified" {
continue
}
switch field {
case "Manufacturer":
result.Hardware.BoardInfo.Manufacturer = value
case "Product Name":
result.Hardware.BoardInfo.ProductName = value
case "Version":
result.Hardware.BoardInfo.Version = value
case "Serial Number":
result.Hardware.BoardInfo.SerialNumber = value
case "UUID":
result.Hardware.BoardInfo.UUID = value
}
}
}
}