use the lastest image.

Author: DF-wu

.env

@@ -45,11 +45,17 @@ OPERATION_MODE=auto
 # 自動模式下的溫度檢查間隔時間 (秒)
 CHECK_INTERVAL=60
 
+
 # =============================================================================
 # GPU 溫度監控設定 (可選)
 # =============================================================================
 # 是否啟用 GPU 溫度監控
-# 設為 true 時，使用決策溫度 = max(磁碟溫度, GPU溫度-20)
+# 設為 true 時，使用決策溫度 = max(磁碟溫度, GPU溫度 - GPU_TEMP_OFFSET)
 # 設為 false 時，僅使用磁碟溫度
 # 注意：啟用 GPU 監控需要使用 --gpus all 運行容器
-WITH_GPU_TEMP=false
+WITH_GPU_TEMP=false
+
+# GPU 溫度偏移量 (攝氏度)
+# 用於調整 GPU 溫度對風扇控制的影響程度
+# 預設值為 15°C，即 GPU 溫度會減去此值後與磁碟溫度比較
+GPU_TEMP_OFFSET=15

.env.example

@@ -41,14 +41,22 @@ CHECK_INTERVAL=60
 # ===== GPU 溫度監控設定 / GPU Temperature Monitoring =====
 # GPU 溫度控制開關
 # 設定為 'true' 以啟用 GPU 溫度監控進行決策溫度計算
-# 決策溫度 = max(磁碟溫度, GPU溫度 - 20°C)
+# 決策溫度 = max(磁碟溫度, GPU溫度 - GPU_TEMP_OFFSET)
 # 預設值：false (僅使用磁碟溫度)
 # GPU temperature control flag
 # Set to 'true' to enable GPU temperature monitoring for decision temperature calculation
-# Decision Temperature = max(Disk Temperature, GPU Temperature - 20°C)
+# Decision Temperature = max(Disk Temperature, GPU Temperature - GPU_TEMP_OFFSET)
 # Default: false (only use disk temperature)
 WITH_GPU_TEMP=false
 
+# GPU 溫度偏移量 (攝氏度)
+# 用於調整 GPU 溫度對風扇控制的影響程度
+# 預設值為 15°C，即 GPU 溫度會減去此值後與磁碟溫度比較
+# GPU temperature offset (Celsius)
+# Used to adjust GPU temperature's impact on fan control
+# Default is 15°C, meaning GPU temperature minus this value will be compared with disk temperature
+GPU_TEMP_OFFSET=15
+
 # ===== ESXi 主機設定 / ESXi Host Configuration =====
 # ESXi 主機連線設定 - 用於取得磁碟溫度
 # ESXi host connection settings - Used for retrieving disk temperature

Dockerfile

@@ -1,7 +1,7 @@
 # Dockerfile for iDRAC Fan Speed Control with GPU Support
-# 基於 NVIDIA 的 Ubuntu 基礎映像檔以支援 GPU 溫度監控
+# 基於 NVIDIA 的 Ubuntu 24.04 基礎映像檔以支援 GPU 溫度監控
 # 如果不需要 GPU 支援，可以改用 alpine:latest
-FROM nvidia/cuda:11.8-base-ubuntu22.04
+FROM nvidia/cuda:12.9.0-runtime-ubuntu24.04
 
 # 設定維護者資訊
 LABEL maintainer="iDRAC Fan Control Service"

README.md

@@ -12,7 +12,7 @@
 - 🎮 **GPU Temperature Support**: Optional GPU temperature monitoring with NVIDIA Container Toolkit
 - ⚙️ **Environment Variable Control**: Complete configuration through `.env` files
 - 🐳 **Docker Deployment**: One-click deployment with no manual environment setup
-- 🔥 **Decision Temperature Algorithm**: `max(disk_temp, gpu_temp - 20°C)` ensures optimal cooling
+- 🔥 **Decision Temperature Algorithm**: `max(disk_temp, gpu_temp - GPU_TEMP_OFFSET)` ensures optimal cooling
 - 📊 **Configurable Thresholds**: Customize temperature and fan speed settings
 - 🔄 **Auto/Manual Modes**: Flexible operation modes for different use cases
 
@@ -105,22 +105,24 @@ CHECK_INTERVAL=60     # Check interval in seconds (auto mode only)
 When GPU temperature monitoring is enabled, the system uses this algorithm to calculate the decision temperature:
 
 ```text
-Decision Temperature = max(Disk Temperature, GPU Temperature - 20°C)
+Decision Temperature = max(Disk Temperature, GPU Temperature - GPU_TEMP_OFFSET)
 ```
 
 This algorithm ensures:
 
 - **Proactive GPU cooling**: High GPU temperatures trigger increased fan speeds
-- **Temperature offset compensation**: Accounts for thermal differences between GPU and system
+- **Configurable offset compensation**: Adjustable offset (`GPU_TEMP_OFFSET`) accounts for thermal differences between GPU and system
 - **Disk temperature baseline**: Disk temperature always serves as the minimum baseline
 
 ### Algorithm Examples
 
-| Disk Temp | GPU Temp | GPU-20 | Decision Temp | Reasoning |
+With default `GPU_TEMP_OFFSET=15°C`:
+
+| Disk Temp | GPU Temp | GPU-15 | Decision Temp | Reasoning |
 |-----------|----------|--------|---------------|-----------|
-| 65°C | 70°C | 50°C | **65°C** | Disk temperature is higher |
-| 65°C | 90°C | 70°C | **70°C** | GPU-20 is higher, use adjusted GPU temp |
-| 75°C | 80°C | 60°C | **75°C** | Disk temperature remains baseline |
+| 65°C | 70°C | 55°C | **65°C** | Disk temperature is higher |
+| 65°C | 85°C | 70°C | **70°C** | GPU-15 is higher, use adjusted GPU temp |
+| 75°C | 80°C | 65°C | **75°C** | Disk temperature remains baseline |
 
 ## 🔧 Troubleshooting
 
@@ -133,7 +135,7 @@ This algorithm ensures:
 nvidia-smi
 
 # Verify Docker GPU support
-docker run --rm --gpus all nvidia/cuda:11.0-base nvidia-smi
+docker run --rm --gpus all nvidia/cuda:12.9.0-runtime-ubuntu24.04 nvidia-smi
 ```
 
 #### Cannot Connect to ESXi Host
@@ -184,6 +186,7 @@ docker exec idrac-fan-control tail -f /var/log/fan-control/fan_control.log
 | `OPERATION_MODE` | auto | Operation mode (auto/manual) |
 | `CHECK_INTERVAL` | 60 | Check interval in seconds |
 | `WITH_GPU_TEMP` | false | Enable GPU temperature monitoring |
+| `GPU_TEMP_OFFSET` | 15 | GPU temperature offset for decision algorithm (°C) |
 
 ## 🐳 Docker Compose Examples
 

USAGE_GUIDE.md

@@ -307,7 +307,7 @@ ssh root@your_esxi_host
 
 ```bash
 # Verify NVIDIA Container Toolkit installation
-docker run --rm --gpus all nvidia/cuda:11.0-base nvidia-smi
+docker run --rm --gpus all nvidia/cuda:12.9.0-runtime-ubuntu24.04 nvidia-smi
 
 # Check if GPU is accessible in container
 docker exec idrac-fan-control nvidia-smi

src/FanControlWithEsxiSmart.sh

@@ -34,6 +34,8 @@ CHECK_INTERVAL=${CHECK_INTERVAL:-60}
 # 預設為 false，只有當使用者明確設定 WITH_GPU_TEMP=true 時才啟用 GPU 溫度監控
 WITH_GPU_TEMP=${WITH_GPU_TEMP:-"false"}
 
+GPU_TEMP_OFFSET=${GPU_TEMP_OFFSET:-15}
+
 # ESXi variables
 ESXI_HOST=${ESXI_HOST:-"REPLACE_TO_YOUR_ESXI_HOST"}
 ESXI_USERNAME=${ESXI_USERNAME:-"REPLACE_TO_YOUR_ESXI_USERNAME"}
@@ -66,7 +68,7 @@ get_nvidia_temp() {
 }
 
 # Function to calculate decision temperature
-# 決策溫度 = max(磁碟溫度, GPU溫度-20)
+# 決策溫度 = max(磁碟溫度, GPU溫度 - GPU_TEMP_OFFSET)
 # 這個邏輯確保風扇轉速基於較高的溫度來源進行調節
 get_decision_temp() {
     local disk_temp=$(get_drive_temp)
@@ -75,7 +77,7 @@ get_decision_temp() {
     # 只有在啟用 GPU 溫度監控時才考慮 GPU 溫度
     if [[ "$WITH_GPU_TEMP" == "true" ]]; then
         local gpu_temp=$(get_nvidia_temp)
-        local gpu_adjusted_temp=$((gpu_temp - 20))
+        local gpu_adjusted_temp=$((gpu_temp - GPU_TEMP_OFFSET))
 
         # 記錄原始溫度用於除錯
         echo "$(date '+%Y-%m-%d %H:%M:%S') - Debug: Disk=${disk_temp}°C, GPU=${gpu_temp}°C, GPU_Adjusted=${gpu_adjusted_temp}°C" >&2
@@ -85,7 +87,7 @@ get_decision_temp() {
             decision_temp=$gpu_adjusted_temp
         fi
 
-        echo "$(date '+%Y-%m-%d %H:%M:%S') - Decision temperature: ${decision_temp}°C (Disk: ${disk_temp}°C, GPU-20: ${gpu_adjusted_temp}°C)" >&2
+        echo "$(date '+%Y-%m-%d %H:%M:%S') - Decision temperature: ${decision_temp}°C (Disk: ${disk_temp}°C, GPU-${GPU_TEMP_OFFSET}: ${gpu_adjusted_temp}°C)" >&2
     else
         echo "$(date '+%Y-%m-%d %H:%M:%S') - Decision temperature: ${decision_temp}°C (Disk only mode)" >&2
     fi
@@ -159,10 +161,9 @@ manual_mode() {
 
 # Function to run in automatic mode
 auto_mode() {
-    echo "Automatic fan speed control mode"
-    if [[ "$WITH_GPU_TEMP" == "true" ]]; then
+    echo "Automatic fan speed control mode"    if [[ "$WITH_GPU_TEMP" == "true" ]]; then
         echo "GPU temperature monitoring enabled - using decision temperature logic"
-        echo "Decision Temperature = max(Disk Temperature, GPU Temperature - 20°C)"
+        echo "Decision Temperature = max(Disk Temperature, GPU Temperature - ${GPU_TEMP_OFFSET}°C)"
     else
         echo "GPU temperature monitoring disabled - using disk temperature only"
     fi