feat: add check_events.sh, check_resources.sh, check_all.sh for diagnostics

cluster/check_all.sh

@@ -0,0 +1,40 @@
+#!/usr/bin/env bash
+# 用途：一鍵執行所有檢查（診斷問題時使用）
+
+set -euo pipefail
+
+# 顏色定義
+BLUE='\033[0;34m'
+NC='\033[0m'
+
+echo -e "${BLUE}"
+cat << "EOF"
+╔════════════════════════════════════════════════╗
+║                                            ║
+║           K3s 叢集全面診斷                  ║
+║                                            ║
+╚════════════════════════════════════════════════╝
+EOF
+echo -e "${NC}"
+echo ""
+
+SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
+
+# 執行各項檢查
+for script in check_nodes.sh check_storage.sh check_access.sh check_events.sh; do
+  if [ -f "$SCRIPT_DIR/$script" ]; then
+    echo "▶️  執行: $script"
+    echo "=========================================="
+    bash "$SCRIPT_DIR/$script"
+    echo ""
+    echo ""
+  fi
+done
+
+echo "=========================================="
+echo "診斷完成！"
+echo ""
+echo "💡 提示："
+echo "   - 如果有問題，請把整個輸出貼給 Claude"
+echo "   - 或只貼有紅色/黃色警告的部分"
+echo "=========================================="

cluster/check_events.sh

@@ -0,0 +1,36 @@
+#!/usr/bin/env bash
+# 用途：檢查叢集事件（診斷問題最重要！）
+
+set -euo pipefail
+
+# 顏色定義
+RED='\033[0;31m'
+GREEN='\033[0;32m'
+YELLOW='\033[1;33m'
+NC='\033[0m'
+
+echo "=========================================="
+echo "        K3s 叢集事件 (最近1小時)"
+echo "=========================================="
+echo ""
+
+# 顯示最近 1 小時的 Warning 和 Error 事件
+echo "⚠️  Warning / Error 事件"
+echo "------------------------------------------"
+kubectl get events -A --sort-by='.lastTimestamp' --field-selector type!=Normal | \
+  grep -E "(Warning|Error)" | tail -20 || echo "無異常事件"
+
+echo ""
+echo "📋 所有最近事件 (最近20筆)"
+echo "------------------------------------------"
+kubectl get events -A --sort-by='.lastTimestamp' | tail -20
+
+echo ""
+echo "🔄 計畫失敗的 Pod"
+echo "------------------------------------------"
+kubectl get pods -A | grep -E "(Failed|Error|CrashLoopBackOff|ImagePullBackOff)" || echo -e "${GREEN}無異常 Pod${NC}"
+
+echo ""
+echo "=========================================="
+echo "提示: 事件太多可用 kubectl get events -A --field-selector type=Warning"
+echo "=========================================="

cluster/check_resources.sh

@@ -0,0 +1,40 @@
+#!/usr/bin/env bash
+# 用途：檢查資源使用率（CPU/Memory）
+
+set -euo pipefail
+
+# 顏色定義
+RED='\033[0;31m'
+GREEN='\033[0;32m'
+YELLOW='\033[1;33m'
+NC='\033[0m'
+
+echo "=========================================="
+echo "        K3s 資源使用率"
+echo "=========================================="
+echo ""
+
+# 節點資源
+echo "🖥️  節點資源"
+echo "------------------------------------------"
+kubectl top nodes 2>/dev/null || echo "需要 metrics-server，請稍後再試"
+
+echo ""
+echo "📦 Pod 資源使用 (Top 10)"
+echo "------------------------------------------"
+kubectl top pods -A --sort-by=cpu 2>/dev/null | head -11 || echo "需要 metrics-server"
+
+echo ""
+echo "🔋 高資源使用 Pod (>80% CPU 或 >1G Memory)"
+echo "------------------------------------------"
+kubectl top pods -A 2>/dev/null | awk -v OFS='\t' 'NR>1 && ($3 > 80 || $4 > 1000) {print $1, $2, $3"%", $4"Mi"}' || echo "無法計算"
+
+echo ""
+echo "📊 資源請求 vs 限制"
+echo "------------------------------------------"
+echo "查看 Pod 是否設定了資源限制："
+kubectl get pods -A -o jsonpath='{range .items[*]}{.metadata.namespace}{"\t"}{.metadata.name}{"\t"}{.spec.containers[*].resources.requests.cpu}{"\t"}{.spec.containers[*].resources.limits.cpu}{"\n"}{end}' | \
+  awk '{if ($3 != "<none>" && $4 == "<none>") print $1"/"$2 " ❌ 只有 request 無 limit"; else if ($3 == "<none>") print $1"/"$2 " ⚠️  無資料設定"}' | head -10 || echo -e "${GREEN}所有 Pod 都有設定${NC}"
+
+echo ""
+echo "=========================================="