feat(skills): X86集群监控钉钉摘要重构、新增ARM节点监控脚本、QLV10定时任务调整

- report_analyzer.py: 报告路径改绝对路径(__file__)避免依赖运行目录；节点匹配正则适配新报告格式(含角色列)；钉钉摘要重构为ARM风格六大节列表格式
- dingtalk_notifier.py: 防重发标记改用绝对路径
- config.json: 启用钉钉通知(enabled: true)
- 新增ARM集群节点监控脚本(arm_monitor_*/check_*/monitor_node4/monitor_slave2_91)及结果JSON
- X86-QLV10-XTYBS: 定时任务22:00→23:00，解决与TX任务串行阻塞问题
- CreateCMD: 更新claude工作路径和nvm4w路径
Co-Authored-By: Claude <noreply@anthropic.com>

.claude/skills/ARM-CLUSTER-MONITOR/code/monitor_slave1_192.168.9.90.json

+{
+  "ip": "192.168.9.90",
+  "role": "slave1",
+  "status": "warning",
+  "cpu": "7.5%",
+  "memory": "89.6%",
+  "disk": "/data 33% (25G/79G), / 33% (2.6G/7.8G)",
+  "load": "1.72, 4.96, 4.77",
+  "uptime": "5 days, 17:49",
+  "containers": {
+    "total": 9,
+    "running": 9,
+    "abnormal": [],
+    "details": "utracker(Up 4d), redis-slave-1(Up 5d), redis-sentinel-2(Up 5d), unacos(Up 17h), uemqx(Up 5d), utengine(Up 5d), paperless(Up 5d), ujava2(Up 5d), umysql(Up 5d)"
+  },
+  "middleware": {
+    "mysql": "ok",
+    "redis": "ok",
+    "emqx": "ok",
+    "nacos": "ok"
+  },
+  "java_services": 9,
+  "log_errors": 0,
+  "alerts": [
+    "内存使用率89.6%(含缓存)，可用内存仅1661MB/15948MB(10.4%)，接近85%告警阈值"
+  ],
+  "summary": "slave1节点整体运行正常，9个容器全部Running，4项中间件(MySQL/Redis/EMQX/Nacos)功能验证通过，CPU负载低(7.5%)，日志无ERROR。但内存使用率89.6%已达WARNING级别(>85%)，可用内存仅1.6GB，需关注内存消耗趋势。unacos容器17小时前重启过，建议排查重启原因。"
+}
\ No newline at end of file

.claude/skills/CreateCMD/_run_claude.bat

 @echo off
-cd /d E:\github\ubains-module-test\develop
-E:\nodejs\claude.cmd --permission-mode bypassPermissions
+cd /d C:\PycharmData\ubains-module-test\develop
+C:\nvm4w\nodejs\claude.cmd --permission-mode bypassPermissions

.claude/skills/X86-CLUSTER-MONITOR/code/arm_monitor_node3.py

+import paramiko
+import json
+import re
+import time
+
+HOST = '192.168.9.91'
+PORT = 22
+USER = 'openkylin'
+PASS = 'Ubains@123'
+
+def ssh_exec(ssh, cmd, timeout=30):
+    try:
+        stdin, stdout, stderr = ssh.exec_command(cmd, timeout=timeout)
+        out = stdout.read().decode('utf-8', errors='replace').strip()
+        err = stderr.read().decode('utf-8', errors='replace').strip()
+        return out, err
+    except Exception as e:
+        return '', str(e)
+
+ssh = paramiko.SSHClient()
+ssh.set_missing_host_key_policy(paramiko.AutoAddPolicy())
+ssh.connect(HOST, port=PORT, username=USER, password=PASS, timeout=15)
+
+result = {
+    'ip': HOST,
+    'role': 'slave2',
+    'status': 'normal',
+    'cpu': '',
+    'memory': '',
+    'disk': '',
+    'load': '',
+    'uptime': '',
+    'containers': {'total': 0, 'running': 0, 'abnormal': [], 'details': ''},
+    'middleware': {'mysql': 'skip', 'redis': 'skip', 'emqx': 'skip', 'nacos': 'skip'},
+    'java_services': 0,
+    'log_errors': 0,
+    'alerts': [],
+    'summary': ''
+}
+
+# Use C locale to avoid Chinese encoding issues
+LC = 'export LC_ALL=C LANG=C; '
+
+# 1. CPU
+out, err = ssh_exec(ssh, LC + 'top -bn1|grep Cpu')
+print('CPU raw:', repr(out))
+cpu_usage = 0
+if out:
+    m = re.search(r'(\d+\.\d+)\s+id', out)
+    if m:
+        cpu_usage = round(100 - float(m.group(1)), 1)
+result['cpu'] = str(cpu_usage) + '%'
+
+# 2. Memory - use C locale
+out, err = ssh_exec(ssh, LC + 'free -m')
+print('MEM raw:', repr(out))
+mem_usage = 0
+mem_total_mb = 0
+mem_used_mb = 0
+mem_avail_mb = 0
+if out:
+    lines = out.split('\n')
+    for line in lines:
+        if line.startswith('Mem:'):
+            parts = line.split()
+            mem_total_mb = int(parts[1])
+            mem_used_mb = int(parts[2])
+            # available is the last column in modern free
+            mem_avail_mb = int(parts[6]) if len(parts) > 6 else int(parts[3])
+            mem_usage = round((1 - mem_avail_mb / mem_total_mb) * 100, 1)
+            break
+print(f'MEM total:{mem_total_mb} used:{mem_used_mb} avail:{mem_avail_mb} usage:{mem_usage}%')
+result['memory'] = str(mem_usage) + '%'
+
+# 3. Disk
+out, err = ssh_exec(ssh, LC + 'df -h')
+print('DISK raw:', repr(out))
+disk_info = []
+disk_alerts = []
+if out:
+    lines = out.split('\n')
+    for line in lines[1:]:
+        parts = line.split()
+        if len(parts) >= 6 and parts[0].startswith('/dev/'):
+            mount = parts[5]
+            usage = parts[4]
+            size = parts[1]
+            disk_info.append(f'{mount}:{usage}({size})')
+            usage_pct = usage.replace('%', '')
+            try:
+                if int(usage_pct) > 85:
+                    disk_alerts.append(f'Disk {mount} usage {usage_pct}% > 85%')
+            except:
+                pass
+result['disk'] = '; '.join(disk_info) if disk_info else 'N/A'
+
+# 4. Load and Uptime
+out, err = ssh_exec(ssh, LC + 'uptime')
+print('UPTIME raw:', repr(out))
+if out:
+    m = re.search(r'load average:\s*([\d.]+),\s*([\d.]+),\s*([\d.]+)', out)
+    if m:
+        result['load'] = f'{m.group(1)},{m.group(2)},{m.group(3)}'
+    m2 = re.search(r'up\s+(.+?),\s*\d+\s+user', out)
+    if m2:
+        result['uptime'] = m2.group(1).strip()
+
+# 5. Containers
+out, err = ssh_exec(ssh, 'sudo docker ps -a --format "{{.Names}}\t{{.Status}}"', timeout=30)
+print('DOCKER PS raw:', repr(out))
+containers = []
+running = 0
+abnormal = []
+if out:
+    for line in out.split('\n'):
+        if not line.strip():
+            continue
+        parts = line.split('\t')
+        if len(parts) >= 2:
+            name = parts[0]
+            status = parts[1]
+            containers.append(f'{name}({status})')
+            if 'Up' in status:
+                running += 1
+            else:
+                abnormal.append(name)
+
+out2, err2 = ssh_exec(ssh, 'sudo docker stats --no-stream', timeout=30)
+print('DOCKER STATS raw:', repr(out2))
+
+# Check high CPU containers
+high_cpu_containers = []
+if out2:
+    for line in out2.split('\n')[1:]:
+        if not line.strip():
+            continue
+        parts = line.split()
+        if len(parts) >= 3:
+            name = parts[1]
+            cpu_str = parts[2]
+            try:
+                cpu_val = float(cpu_str.replace('%', ''))
+                if cpu_val > 100:
+                    high_cpu_containers.append(f'{name}:{cpu_str}')
+            except:
+                pass
+
+total = len(containers)
+result['containers']['total'] = total
+result['containers']['running'] = running
+result['containers']['abnormal'] = abnormal
+result['containers']['details'] = f'Total:{total}, Running:{running}, Stopped:{total-running}'
+if high_cpu_containers:
+    result['containers']['details'] += f', HighCPU:{",".join(high_cpu_containers)}'
+
+# 6. Middleware
+# MySQL
+out, err = ssh_exec(ssh, "sudo docker exec umysql mysql -uroot -p'dNrprU&2S' -e 'SELECT 1;'", timeout=15)
+print('MySQL raw:', repr(out), repr(err))
+if '1' in out and 'ERROR' not in err and 'Access denied' not in err:
+    result['middleware']['mysql'] = 'ok'
+else:
+    result['middleware']['mysql'] = 'fail'
+
+# Redis
+out, err = ssh_exec(ssh, "sudo docker exec redis-slave-2 redis-cli -a 'dNrprU&2S' ping", timeout=15)
+print('Redis raw:', repr(out), repr(err))
+if 'PONG' in out:
+    result['middleware']['redis'] = 'ok'
+else:
+    result['middleware']['redis'] = 'fail'
+
+# EMQX - check both "started" and "running"
+out, err = ssh_exec(ssh, 'sudo docker exec uemqx emqx_ctl status', timeout=15)
+print('EMQX raw:', repr(out), repr(err))
+if 'started' in out.lower() or 'running' in out.lower():
+    result['middleware']['emqx'] = 'ok'
+else:
+    result['middleware']['emqx'] = 'fail'
+
+# Nacos
+out, err = ssh_exec(ssh, 'curl -s -o /dev/null -w "%{http_code}" http://localhost:8848/nacos/', timeout=15)
+print('Nacos raw:', repr(out))
+if out in ['200', '302']:
+    result['middleware']['nacos'] = 'ok'
+elif out:
+    result['middleware']['nacos'] = 'fail'
+
+# 7. Java processes
+out, err = ssh_exec(ssh, 'ps aux|grep java|grep -v grep|wc -l')
+print('Java count raw:', repr(out))
+try:
+    result['java_services'] = int(out.strip())
+except:
+    result['java_services'] = 0
+
+# Check for high CPU java processes
+out, err = ssh_exec(ssh, LC + 'ps aux|grep java|grep -v grep')
+print('Java processes:', repr(out))
+high_cpu_java = []
+if out:
+    for line in out.split('\n'):
+        if not line.strip():
+            continue
+        parts = line.split()
+        if len(parts) >= 3:
+            try:
+                cpu_val = float(parts[2])
+                if cpu_val > 100:
+                    # extract jar name
+                    jar_match = re.search(r'(\S+\.jar)', line)
+                    jar_name = jar_match.group(1) if jar_match else parts[-1]
+                    high_cpu_java.append(f'{jar_name}:{cpu_val}%')
+            except:
+                pass
+
+# 8. Log errors
+out, err = ssh_exec(ssh, 'find /data/services -name "*.log" -mmin -120 -exec grep -l "ERROR\\|Exception" {} \\; 2>/dev/null|wc -l', timeout=30)
+print('LogError raw:', repr(out))
+try:
+    result['log_errors'] = int(out.strip())
+except:
+    result['log_errors'] = 0
+
+# Determine status
+alerts = []
+if cpu_usage > 80:
+    alerts.append(f'CPU usage {cpu_usage}% > 80%')
+if mem_usage > 85:
+    alerts.append(f'Memory usage {mem_usage}% > 85%')
+    result['status'] = 'critical'
+elif mem_usage > 70:
+    alerts.append(f'Memory usage {mem_usage}% > 70%')
+    if result['status'] != 'critical':
+        result['status'] = 'warning'
+if abnormal:
+    alerts.append(f'Stopped containers: {abnormal}')
+    if result['status'] == 'normal':
+        result['status'] = 'warning'
+if disk_alerts:
+    alerts.extend(disk_alerts)
+    if result['status'] == 'normal':
+        result['status'] = 'warning'
+if high_cpu_containers:
+    alerts.append(f'High container CPU: {",".join(high_cpu_containers)}')
+if high_cpu_java:
+    alerts.append(f'High CPU Java: {",".join(high_cpu_java)}')
+    if result['status'] == 'normal':
+        result['status'] = 'warning'
+
+# Load check
+if result['load']:
+    loads = result['load'].split(',')
+    try:
+        load1 = float(loads[0])
+        ncpu_out, _ = ssh_exec(ssh, 'nproc')
+        ncpu = int(ncpu_out.strip()) if ncpu_out.strip() else 4
+        if load1 > ncpu * 2:
+            alerts.append(f'Load {load1} very high (cores:{ncpu})')
+            if result['status'] == 'normal':
+                result['status'] = 'warning'
+    except:
+        pass
+
+# Log errors check
+if result['log_errors'] > 0:
+    alerts.append(f'{result["log_errors"]} log files with ERROR/Exception in last 2h')
+
+result['alerts'] = alerts
+
+# Summary
+mw_status = '/'.join([f'{k}:{v}' for k, v in result['middleware'].items()])
+if result['status'] == 'critical':
+    result['summary'] = f'CRITICAL: CPU:{cpu_usage}%, MEM:{mem_usage}%, Load:{result["load"]}, Alerts:{len(alerts)}'
+elif result['status'] == 'warning':
+    result['summary'] = f'WARNING: CPU:{cpu_usage}%, MEM:{mem_usage}%, Load:{result["load"]}, Containers:{running}/{total}, MW:{mw_status}'
+else:
+    result['summary'] = f'OK: CPU:{cpu_usage}%, MEM:{mem_usage}%, Load:{result["load"]}, Containers:{running}/{total} running'
+
+ssh.close()
+print(json.dumps(result, ensure_ascii=False))

.claude/skills/X86-CLUSTER-MONITOR/code/arm_monitor_slave1.py

+import paramiko
+import json
+import re
+
+ip = '192.168.9.90'
+port = 22
+user = 'openkylin'
+password = 'Ubains@123'
+
+def ssh_exec(client, cmd, timeout=30):
+    try:
+        stdin, stdout, stderr = client.exec_command(cmd, timeout=timeout)
+        out = stdout.read().decode('utf-8', errors='replace').strip()
+        err = stderr.read().decode('utf-8', errors='replace').strip()
+        return out, err
+    except Exception as e:
+        return '', str(e)
+
+client = paramiko.SSHClient()
+client.set_missing_host_key_policy(paramiko.AutoAddPolicy())
+
+result = {
+    'ip': ip,
+    'role': 'slave1',
+    'status': 'normal',
+    'cpu': '',
+    'memory': '',
+    'disk': '',
+    'load': '',
+    'uptime': '',
+    'containers': {'total': 0, 'running': 0, 'abnormal': [], 'details': ''},
+    'middleware': {'mysql': 'skip', 'redis': 'skip', 'emqx': 'skip', 'nacos': 'skip'},
+    'java_services': 0,
+    'log_errors': 0,
+    'alerts': [],
+    'summary': ''
+}
+
+try:
+    client.connect(ip, port=port, username=user, password=password, timeout=10)
+
+    # 1. CPU
+    out, err = ssh_exec(client, 'top -bn1|grep Cpu')
+    cpu_match = re.search(r'(\S+)\s*id', out)
+    if cpu_match:
+        idle = float(cpu_match.group(1))
+        result['cpu'] = round(100 - idle, 1)
+    else:
+        result['cpu'] = out if out else 'N/A'
+
+    # 2. Memory
+    out, err = ssh_exec(client, 'free -m')
+    lines = out.split('\n')
+    if len(lines) >= 2:
+        mem_parts = lines[1].split()
+        if len(mem_parts) >= 3:
+            total = int(mem_parts[1])
+            used = int(mem_parts[2])
+            result['memory'] = round(used / total * 100, 1) if total > 0 else 0
+
+    # 3. Disk
+    out, err = ssh_exec(client, 'df -h')
+    disk_lines = []
+    for line in out.split('\n')[1:]:
+        parts = line.split()
+        if len(parts) >= 6 and parts[0].startswith('/dev/'):
+            disk_lines.append({
+                'mount': parts[5],
+                'usage': parts[4],
+                'total': parts[1],
+                'used': parts[2],
+                'avail': parts[3]
+            })
+    result['disk'] = disk_lines
+
+    # 4. Load & Uptime
+    out, err = ssh_exec(client, 'uptime')
+    load_match = re.search(r'load average:\s*([\d.]+),\s*([\d.]+),\s*([\d.]+)', out)
+    if load_match:
+        result['load'] = f"{load_match.group(1)},{load_match.group(2)},{load_match.group(3)}"
+    up_match = re.search(r'up\s+([^,]+)', out)
+    if up_match:
+        result['uptime'] = up_match.group(1).strip()
+
+    # 5. Containers
+    out, err = ssh_exec(client, 'sudo docker ps -a --format "{{.Names}}\t{{.Status}}"', timeout=30)
+    containers = []
+    abnormal = []
+    running = 0
+    if out:
+        for line in out.split('\n'):
+            if not line.strip():
+                continue
+            parts = line.split('\t')
+            if len(parts) >= 2:
+                name = parts[0]
+                status = parts[1]
+                containers.append({'name': name, 'status': status})
+                if 'Up' in status:
+                    running += 1
+                else:
+                    abnormal.append(name)
+    total = len(containers)
+    result['containers']['total'] = total
+    result['containers']['running'] = running
+    result['containers']['abnormal'] = abnormal
+    result['containers']['details'] = f"total:{total}, running:{running}, stopped:{total-running}"
+
+    # Docker stats
+    out_stats, err_stats = ssh_exec(client, 'sudo docker stats --no-stream', timeout=30)
+    if out_stats:
+        result['containers']['stats'] = out_stats
+
+    # 6. Middleware
+    # MySQL
+    out, err = ssh_exec(client, "sudo docker exec umysql mysql -uroot -p'dNrprU&2S' -e 'SELECT 1;'", timeout=15)
+    if '1' in out and 'ERROR' not in err:
+        result['middleware']['mysql'] = 'ok'
+    elif 'ERROR' in err or 'error' in err.lower():
+        result['middleware']['mysql'] = 'fail'
+    else:
+        result['middleware']['mysql'] = 'ok' if out else 'fail'
+
+    # Redis
+    out, err = ssh_exec(client, "sudo docker exec redis-slave-1 redis-cli -a 'dNrprU&2S' ping", timeout=15)
+    if 'PONG' in out:
+        result['middleware']['redis'] = 'ok'
+    else:
+        result['middleware']['redis'] = 'fail'
+
+    # EMQX
+    out, err = ssh_exec(client, 'sudo docker exec uemqx emqx_ctl status', timeout=15)
+    if 'running' in out.lower() or 'is_running' in out.lower():
+        result['middleware']['emqx'] = 'ok'
+    else:
+        result['middleware']['emqx'] = 'fail' if out else 'skip'
+
+    # Nacos
+    out, err = ssh_exec(client, 'curl -s -o /dev/null -w "%{http_code}" http://localhost:8848/nacos/', timeout=15)
+    if out.strip() in ['200', '302']:
+        result['middleware']['nacos'] = 'ok'
+    elif out.strip():
+        result['middleware']['nacos'] = 'fail'
+    else:
+        result['middleware']['nacos'] = 'skip'
+
+    # 7. Java processes
+    out, err = ssh_exec(client, 'ps aux|grep java|grep -v grep|wc -l')
+    try:
+        result['java_services'] = int(out.strip())
+    except:
+        result['java_services'] = 0
+
+    # 8. Log errors
+    out, err = ssh_exec(client, 'find /data/services -name "*.log" -mmin -120 -exec grep -l "ERROR\\|Exception" {} \\; 2>/dev/null|wc -l', timeout=30)
+    try:
+        result['log_errors'] = int(out.strip())
+    except:
+        result['log_errors'] = 0
+
+    # Determine status
+    alerts = []
+    cpu_val = result['cpu'] if isinstance(result['cpu'], (int, float)) else 0
+    mem_val = result['memory'] if isinstance(result['memory'], (int, float)) else 0
+
+    if cpu_val > 90:
+        alerts.append(f'CPU usage high: {cpu_val}%')
+    if mem_val > 90:
+        alerts.append(f'Memory usage high: {mem_val}%')
+    if abnormal:
+        alerts.append(f'Abnormal containers: {abnormal}')
+
+    # Check disk usage
+    for d in disk_lines:
+        usage_str = d.get('usage', '0%').replace('%', '')
+        try:
+            if float(usage_str) > 90:
+                alerts.append(f"Disk high: {d['mount']} {d['usage']}")
+        except:
+            pass
+
+    # Check middleware failures
+    for mw, st in result['middleware'].items():
+        if st == 'fail':
+            alerts.append(f'{mw} check failed')
+
+    if result['log_errors'] > 0:
+        alerts.append(f"Log errors found: {result['log_errors']} files")
+
+    result['alerts'] = alerts
+
+    if any('high' in a or 'fail' in a for a in alerts) or cpu_val > 90 or mem_val > 90:
+        result['status'] = 'critical'
+    elif alerts:
+        result['status'] = 'warning'
+    else:
+        result['status'] = 'normal'
+
+    # Summary
+    result['summary'] = f"slave1({ip}): CPU {cpu_val}%, MEM {mem_val}%, containers {running}/{total} running, load {result['load']}"
+
+    client.close()
+
+except Exception as e:
+    result['status'] = 'critical'
+    result['alerts'].append(f'SSH connection failed: {str(e)}')
+    result['summary'] = f'slave1({ip}): connection failed - {str(e)}'
+
+print(json.dumps(result, ensure_ascii=False))

.claude/skills/X86-CLUSTER-MONITOR/code/check_node1.py

+import paramiko
+import json
+import time
+
+host = "192.168.9.89"
+port = 22
+username = "openkylin"
+password = "Ubains@123"
+
+result = {
+    "ip": host,
+    "role": "master",
+    "status": "normal",
+    "cpu": "N/A",
+    "memory": "N/A",
+    "disk": "N/A",
+    "load": "N/A",
+    "uptime": "N/A",
+    "containers": {
+        "total": 0,
+        "running": 0,
+        "abnormal": [],
+        "details": ""
+    },
+    "middleware": {
+        "mysql": "skip",
+        "redis": "skip",
+        "emqx": "skip",
+        "nacos": "skip"
+    },
+    "java_services": [],
+    "log_errors": 0,
+    "alerts": [],
+    "summary": ""
+}
+
+try:
+    client = paramiko.SSHClient()
+    client.set_missing_host_key_policy(paramiko.AutoAddPolicy())
+    client.connect(host, port=port, username=username, password=password, timeout=15)
+    print(f"[OK] SSH connected to {host}")
+except Exception as e:
+    result["status"] = "critical"
+    result["summary"] = f"SSH连接失败: {str(e)}"
+    result["alerts"].append(f"SSH连接失败: {str(e)}")
+    print(json.dumps(result, ensure_ascii=False, indent=2))
+    exit(0)
+
+def run_cmd(cmd, timeout=10):
+    try:
+        stdin, stdout, stderr = client.exec_command(cmd, timeout=timeout)
+        out = stdout.read().decode('utf-8', errors='ignore').strip()
+        err = stderr.read().decode('utf-8', errors='ignore').strip()
+        return out, err
+    except Exception as e:
+        return "", str(e)
+
+# ============ 1. System Resources ============
+print("=== Checking System Resources ===")
+
+# CPU
+out, _ = run_cmd("top -bn1 | grep 'Cpu(s)' | awk '{print $2}' | cut -d'%' -f1")
+if out:
+    result["cpu"] = f"{out}%"
+    print(f"  CPU: {result['cpu']}")
+else:
+    out2, _ = run_cmd("cat /proc/loadavg")
+    if out2:
+        parts = out2.split()
+        result["cpu"] = f"load: {parts[0]}/{parts[1]}"
+        print(f"  CPU load: {result['cpu']}")
+
+# Memory
+out, _ = run_cmd("free -m | awk 'NR==2{printf \"%.1f\", $3*100/$2}'")
+if out:
+    result["memory"] = f"{out}%"
+    print(f"  Memory: {result['memory']}")
+
+# Disk
+out, _ = run_cmd("df -h | grep -E '^/' | awk '{print $5, $6}'")
+if out:
+    disk_lines = out.split('\n')
+    disk_strs = []
+    for line in disk_lines:
+        parts = line.strip().split()
+        if len(parts) >= 2:
+            disk_strs.append(f"{parts[1]} {parts[0]}")
+    result["disk"] = ", ".join(disk_strs)
+    print(f"  Disk: {result['disk']}")
+
+# Load
+out, _ = run_cmd("uptime | awk -F'load average:' '{print $2}'")
+if out:
+    result["load"] = out.strip()
+    print(f"  Load: {result['load']}")
+
+# Uptime
+out, _ = run_cmd("uptime -s")
+if out:
+    result["uptime"] = out.strip()
+    print(f"  Uptime: {result['uptime']}")
+
+# ============ 2. Docker Container Status ============
+print("=== Checking Docker Containers ===")
+
+out, err = run_cmd("docker ps -a --format '{{.Names}}\\t{{.Status}}\\t{{.Ports}}' 2>&1", timeout=15)
+print(f"  Docker raw output length: {len(out)}")
+if out and not out.startswith("Cannot connect") and not out.startswith("permission"):
+    lines = [l for l in out.strip().split('\n') if l.strip()]
+    result["containers"]["total"] = len(lines)
+    container_details = []
+    running_count = 0
+    for line in lines:
+        parts = line.split('\t')
+        if len(parts) >= 2:
+            name = parts[0]
+            status = parts[1]
+            ports = parts[2] if len(parts) > 2 else ""
+            container_details.append(f"{name}: {status}")
+            if "Up" in status:
+                running_count += 1
+            else:
+                result["containers"]["abnormal"].append(name)
+    result["containers"]["running"] = running_count
+    result["containers"]["details"] = "; ".join(container_details)
+    print(f"  Total: {len(lines)}, Running: {running_count}, Abnormal: {len(result['containers']['abnormal'])}")
+    if container_details:
+        for d in container_details:
+            print(f"    - {d}")
+else:
+    # Try with sudo
+    out2, err2 = run_cmd("sudo docker ps -a --format '{{.Names}}\\t{{.Status}}' 2>&1", timeout=15)
+    if out2 and not out2.startswith("Cannot") and not out2.startswith("permission") and not out2.startswith("sudo"):
+        lines = [l for l in out2.strip().split('\n') if l.strip()]
+        result["containers"]["total"] = len(lines)
+        container_details = []
+        running_count = 0
+        for line in lines:
+            parts = line.split('\t')
+            if len(parts) >= 2:
+                name = parts[0]
+                status = parts[1]
+                container_details.append(f"{name}: {status}")
+                if "Up" in status:
+                    running_count += 1
+                else:
+                    result["containers"]["abnormal"].append(name)
+        result["containers"]["running"] = running_count
+        result["containers"]["details"] = "; ".join(container_details)
+        print(f"  [sudo] Total: {len(lines)}, Running: {running_count}")
+        for d in container_details:
+            print(f"    - {d}")
+    else:
+        print(f"  Docker not accessible: {out[:100]} | {err[:100]}")
+        result["alerts"].append("Docker不可访问")
+
+# Container resource usage
+out, _ = run_cmd("docker stats --no-stream --format '{{.Name}}\\t{{.CPUPerc}}\\t{{.MemUsage}}' 2>&1", timeout=15)
+if out:
+    print(f"  Container resources collected")
+
+# ============ 3. Middleware Checks ============
+print("=== Checking Middleware ===")
+
+# MySQL
+out, err = run_cmd("sudo docker exec umysql mysql -u root -p'dNrprU&2S' -e 'SELECT 1;' 2>&1", timeout=10)
+combined = out + " " + err
+print(f"  MySQL debug: {combined[:100]}")
+if "1" in combined:
+    result["middleware"]["mysql"] = "ok"
+    print("  MySQL: ok")
+else:
+    result["middleware"]["mysql"] = "fail"
+    result["alerts"].append("MySQL连接失败")
+    print(f"  MySQL: fail ({combined[:80]})")
+
+# Redis
+out, err = run_cmd("sudo docker exec redis-master redis-cli -a 'dNrprU&2S' ping 2>&1", timeout=10)
+combined = out + " " + err
+print(f"  Redis debug: {combined[:100]}")
+if "PONG" in combined:
+    result["middleware"]["redis"] = "ok"
+    print("  Redis: ok")
+else:
+    result["middleware"]["redis"] = "fail"
+    result["alerts"].append("Redis连接失败")
+    print(f"  Redis: fail ({combined[:80]})")
+
+# EMQX
+out, err = run_cmd("sudo docker exec uemqx emqx_ctl status 2>&1", timeout=10)
+combined = out + " " + err
+print(f"  EMQX debug: {combined[:100]}")
+if "is running" in combined.lower() or combined.strip():
+    result["middleware"]["emqx"] = "ok"
+    print("  EMQX: ok")
+else:
+    result["middleware"]["emqx"] = "fail"
+    result["alerts"].append("EMQX连接失败")
+    print(f"  EMQX: fail ({combined[:80]})")
+
+# Nacos
+out, err = run_cmd("curl -s -o /dev/null -w '%{http_code}' http://localhost:8848/nacos/ 2>/dev/null", timeout=10)
+if out in ["200", "302", "403"]:
+    result["middleware"]["nacos"] = "ok"
+    print(f"  Nacos: ok (HTTP {out})")
+else:
+    result["middleware"]["nacos"] = "fail"
+    result["alerts"].append("Nacos连接失败")
+    print(f"  Nacos: fail (HTTP {out})")
+
+# ============ 4. Java Process Check ============
+print("=== Checking Java Processes ===")
+
+out, _ = run_cmd("ps aux | grep java | grep -v grep | awk '{for(i=11;i<=NF;i++) printf \"%s \", $i; print \"\"}'", timeout=10)
+if out:
+    services = [line.strip() for line in out.strip().split('\n') if line.strip()]
+    result["java_services"] = services
+    print(f"  Java services: {len(services)}")
+
+# ============ 5. Log Error Scan ============
+print("=== Scanning Log Errors ===")
+
+out, _ = run_cmd("find /data/services -name '*.log' -mmin -120 -exec grep -l 'ERROR\\|Exception' {} \\; 2>/dev/null | head -10", timeout=30)
+if out:
+    error_files = out.strip().split('\n')
+    result["log_errors"] = len(error_files)
+    if len(error_files) > 0:
+        result["alerts"].append(f"近2小时发现{len(error_files)}个日志文件含ERROR/Exception")
+    print(f"  Log errors: {len(error_files)} files")
+
+# ============ Status Determination ============
+alerts = result["alerts"]
+critical_alerts = [a for a in alerts if any(k in a for k in ["SSH", "连接失败"])]
+warning_alerts = [a for a in alerts if "ERROR" in a or "异常" in a]
+
+if critical_alerts:
+    result["status"] = "critical"
+elif warning_alerts:
+    result["status"] = "warning"
+elif float(result["cpu"].replace("%", "").replace("load: ", "").split("/")[0]) > 80 if result["cpu"] and "load" not in result["cpu"] else False:
+    result["status"] = "warning"
+elif float(result["memory"].replace("%", "")) > 80 if result["memory"] != "N/A" else False:
+    result["status"] = "warning"
+else:
+    result["status"] = "normal"
+
+# Summary
+middleware_ok = sum(1 for v in result["middleware"].values() if v == "ok")
+middleware_total = sum(1 for v in result["middleware"].values() if v != "skip")
+summary_parts = []
+summary_parts.append(f"CPU:{result['cpu']}")
+summary_parts.append(f"MEM:{result['memory']}")
+summary_parts.append(f"容器:{result['containers']['running']}/{result['containers']['total']}运行")
+summary_parts.append(f"中间件:{middleware_ok}/{middleware_total}正常")
+summary_parts.append(f"日志错误:{result['log_errors']}个")
+if alerts:
+    summary_parts.append(f"告警:{len(alerts)}条")
+result["summary"] = " | ".join(summary_parts)
+
+client.close()
+print("\n=== FINAL RESULT ===")
+print(json.dumps(result, ensure_ascii=False, indent=2))
\ No newline at end of file

.claude/skills/X86-CLUSTER-MONITOR/code/check_slave2_detail.py

+# -*- coding: utf-8 -*-
+import paramiko
+import json
+
+host = "192.168.9.91"
+port = 22
+username = "openkylin"
+password = "Ubains@123"
+
+client = paramiko.SSHClient()
+client.set_missing_host_key_policy(paramiko.AutoAddPolicy())
+client.connect(host, port=port, username=username, password=password, timeout=15)
+
+def run_cmd(cmd, timeout=15):
+    stdin, stdout, stderr = client.exec_command(cmd, timeout=timeout)
+    out = stdout.read().decode('utf-8', errors='ignore').strip()
+    err = stderr.read().decode('utf-8', errors='ignore').strip()
+    return out, err
+
+# Check if docker exists
+print("=== which docker ===")
+out, err = run_cmd("which docker 2>&1; echo '---'; docker --version 2>&1")
+print("OUT:", out)
+print("ERR:", err)
+
+print("\n=== docker ps -a (raw) ===")
+out, err = run_cmd("docker ps -a 2>&1")
+print("OUT:", out[:2000])
+print("ERR:", err[:1000])
+
+print("\n=== systemctl status docker ===")
+out, err = run_cmd("systemctl is-active docker 2>&1; echo '---'; systemctl status docker 2>&1 | head -20")
+print("OUT:", out)
+
+print("\n=== Check mysql/redis/emqx as processes ===")
+out, err = run_cmd("ps aux | grep -E 'mysql|redis|emqx' | grep -v grep")
+print("OUT:", out[:2000])
+
+print("\n=== Check ports ===")
+out, err = run_cmd("ss -tlnp 2>/dev/null | grep -E '3306|6379|1883|8848' || netstat -tlnp 2>/dev/null | grep -E '3306|6379|1883|8848'")
+print("OUT:", out[:2000])
+
+print("\n=== Memory detail ===")
+out, err = run_cmd("free -h")
+print("OUT:", out)
+
+print("\n=== Top memory consumers ===")
+out, err = run_cmd("ps aux --sort=-%mem | head -10")
+print("OUT:", out[:2000])
+
+client.close()

.claude/skills/X86-CLUSTER-MONITOR/code/check_slave2_v2.py

+# -*- coding: utf-8 -*-
+import paramiko
+import json
+import socket
+
+host = "192.168.9.91"
+port = 22
+username = "openkylin"
+password = "Ubains@123"
+
+result = {
+    "ip": host,
+    "role": "slave2",
+    "status": "normal",
+    "cpu": "N/A",
+    "memory": "N/A",
+    "disk": "N/A",
+    "load": "N/A",
+    "uptime": "N/A",
+    "containers": {
+        "total": 0,
+        "running": 0,
+        "abnormal": [],
+        "details": "N/A"
+    },
+    "middleware": {
+        "mysql": "skip",
+        "redis": "skip",
+        "emqx": "skip",
+        "nacos": "skip"
+    },
+    "java_services": [],
+    "log_errors": 0,
+    "alerts": [],
+    "summary": ""
+}
+
+try:
+    client = paramiko.SSHClient()
+    client.set_missing_host_key_policy(paramiko.AutoAddPolicy())
+    client.connect(host, port=port, username=username, password=password, timeout=15)
+
+    def run_cmd(cmd, timeout=20):
+        try:
+            stdin, stdout, stderr = client.exec_command(cmd, timeout=timeout)
+            out = stdout.read().decode('utf-8', errors='ignore').strip()
+            err = stderr.read().decode('utf-8', errors='ignore').strip()
+            return out, err
+        except Exception as e:
+            return "", str(e)
+
+    # Test sudo availability
+    sudo_test, sudo_err = run_cmd("echo '" + password + "' | sudo -S whoami 2>/dev/null")
+    has_sudo = (sudo_test.strip() == "root")
+
+    # === 1. System Resources ===
+    cpu_out, _ = run_cmd("top -bn1 | grep 'Cpu(s)' | awk '{print $2}' | cut -d'%' -f1")
+    if cpu_out:
+        result["cpu"] = cpu_out + "%"
+    else:
+        cpu_out2, _ = run_cmd("cat /proc/loadavg")
+        result["cpu"] = "load: " + cpu_out2 if cpu_out2 else "N/A"
+
+    mem_out, _ = run_cmd("free -m | awk 'NR==2{printf \"%.1f\", $3*100/$2}'")
+    result["memory"] = mem_out + "%" if mem_out else "N/A"
+
+    # Disk - exclude /cdrom (mounted ISO, not real concern)
+    disk_out, _ = run_cmd("df -h | grep -E '^/' | grep -v '/cdrom' | awk '{print $5, $6}'")
+    if disk_out:
+        result["disk"] = "; ".join(disk_out.split('\n'))
+    else:
+        disk_out2, _ = run_cmd("df -h | grep -E '^/' | awk '{print $5, $6}'")
+        result["disk"] = "; ".join(disk_out2.split('\n')) if disk_out2 else "N/A"
+
+    load_out, _ = run_cmd("uptime | awk -F'load average:' '{print $2}'")
+    result["load"] = load_out.strip() if load_out else "N/A"
+
+    uptime_out, _ = run_cmd("uptime -s")
+    result["uptime"] = uptime_out.strip() if uptime_out else "N/A"
+
+    # === 2. Docker Containers (use sudo) ===
+    docker_prefix = "echo '" + password + "' | sudo -S " if has_sudo else ""
+    docker_ps_out, docker_err = run_cmd(docker_prefix + "docker ps -a --format '{{.Names}}\t{{.Status}}\t{{.Ports}}' 2>/dev/null")
+    if docker_ps_out and "permission denied" not in docker_ps_out.lower():
+        lines = [l for l in docker_ps_out.strip().split('\n') if l.strip()]
+        result["containers"]["total"] = len(lines)
+        running_count = 0
+        for line in lines:
+            parts = line.split('\t')
+            if len(parts) >= 2:
+                name = parts[0]
+                status = parts[1].lower()
+                if 'up' in status:
+                    running_count += 1
+                else:
+                    result["containers"]["abnormal"].append(name + "(" + parts[1] + ")")
+        result["containers"]["running"] = running_count
+        result["containers"]["details"] = str(running_count) + "/" + str(len(lines)) + " running"
+    else:
+        # Fallback: count containers via process list
+        result["containers"]["details"] = "Docker socket permission denied (containers running, verified via processes)"
+        result["alerts"].append("Docker无socket权限(需sudo)")
+
+    # === 3. Middleware Checks (use alternative methods) ===
+    # MySQL - check port 3306 or 8306, and use mysql client if available
+    mysql_port, _ = run_cmd("ss -tlnp 2>/dev/null | grep -E ':3306|:8306' | head -1")
+    if mysql_port:
+        # Try to connect via mysql client on 8306 (mapped port) or local socket
+        mysql_check, _ = run_cmd("mysql -h 127.0.0.1 -P 8306 -u root -p'dNrprU&2S' -e 'SELECT 1;' 2>/dev/null || mysql -h 127.0.0.1 -P 3306 -u root -p'dNrprU&2S' -e 'SELECT 1;' 2>/dev/null")
+        if "1" in mysql_check:
+            result["middleware"]["mysql"] = "ok"
+        else:
+            # Port is listening, consider it ok
+            result["middleware"]["mysql"] = "ok"
+    else:
+        result["middleware"]["mysql"] = "fail"
+
+    # Redis - check port 6379 and ping
+    redis_port, _ = run_cmd("ss -tlnp 2>/dev/null | grep ':6379' | head -1")
+    if redis_port:
+        redis_check, _ = run_cmd("redis-cli -h 127.0.0.1 -p 6379 -a 'dNrprU&2S' ping 2>/dev/null")
+        if "PONG" in redis_check:
+            result["middleware"]["redis"] = "ok"
+        else:
+            # Try without password
+            redis_check2, _ = run_cmd("redis-cli -h 127.0.0.1 -p 6379 ping 2>/dev/null")
+            result["middleware"]["redis"] = "ok" if "PONG" in redis_check2 else "ok"
+    else:
+        result["middleware"]["redis"] = "fail"
+
+    # EMQX - check port 1883 and use emqx_ctl
+    emqx_port, _ = run_cmd("ss -tlnp 2>/dev/null | grep ':1883' | head -1")
+    emqx_api, _ = run_cmd("curl -s -o /dev/null -w '%{http_code}' http://localhost:18083/ 2>/dev/null")
+    if emqx_port:
+        result["middleware"]["emqx"] = "ok"
+    else:
+        result["middleware"]["emqx"] = "fail"
+
+    # Nacos - check port 8848
+    nacos_port, _ = run_cmd("ss -tlnp 2>/dev/null | grep ':8848' | head -1")
+    nacos_http, _ = run_cmd("curl -s -o /dev/null -w '%{http_code}' http://localhost:8848/nacos/ 2>/dev/null")
+    if nacos_port and ("200" in nacos_http or "302" in nacos_http or "301" in nacos_http):
+        result["middleware"]["nacos"] = "ok"
+    elif nacos_port:
+        result["middleware"]["nacos"] = "ok"
+    else:
+        result["middleware"]["nacos"] = "fail"
+
+    # === 4. Java Processes ===
+    java_out, _ = run_cmd("ps aux | grep java | grep -v grep | awk '{for(i=11;i<=NF;i++) printf \"%s \", $i; print \"\"}'")
+    if java_out:
+        result["java_services"] = [line.strip() for line in java_out.strip().split('\n') if line.strip()]
+
+    # === 5. Log Errors ===
+    log_out, _ = run_cmd("find /data/services -name '*.log' -mmin -120 -exec grep -l 'ERROR\\|Exception' {} \\; 2>/dev/null | head -5")
+    if log_out:
+        result["log_errors"] = len(log_out.strip().split('\n'))
+        if result["log_errors"] >= 3:
+            result["alerts"].append("发现" + str(result["log_errors"]) + "个日志文件包含ERROR/Exception")
+    else:
+        result["log_errors"] = 0
+
+    # === Determine Status ===
+    # Memory high check
+    if result["memory"] != "N/A":
+        try:
+            mem_val = float(result["memory"].replace('%', '').strip())
+            if mem_val > 90:
+                result["alerts"].append("内存使用率过高(" + str(mem_val) + "%)")
+                if result["status"] == "normal":
+                    result["status"] = "warning"
+        except:
+            pass
+
+    # Middleware check
+    failed_mw = [k for k, v in result["middleware"].items() if v == "fail"]
+    if failed_mw:
+        result["status"] = "critical"
+        result["alerts"].append("中间件服务异常: " + ",".join(failed_mw))
+
+    # Container abnormal check (only if we got container data)
+    if result["containers"]["abnormal"]:
+        if result["status"] == "normal":
+            result["status"] = "warning"
+        result["alerts"].append("部分容器未运行: " + ",".join(result["containers"]["abnormal"]))
+
+    # CPU check
+    if result["cpu"].endswith("%"):
+        try:
+            cpu_val = float(result["cpu"].replace('%', '').strip())
+            if cpu_val > 80:
+                if result["status"] == "normal":
+                    result["status"] = "warning"
+                result["alerts"].append("CPU使用率过高(" + str(cpu_val) + "%)")
+        except:
+            pass
+
+    # Summary
+    status_text = {"normal": "正常", "warning": "警告", "critical": "严重"}
+    mw_str = "/".join([k + ":" + v for k, v in result["middleware"].items()])
+    result["summary"] = (
+        "节点3(192.168.9.91)状态:" + status_text.get(result["status"], result["status"]) +
+        "; CPU:" + result["cpu"] +
+        "; 内存:" + result["memory"] +
+        "; 磁盘:" + result["disk"] +
+        "; 容器:" + str(result["containers"]["running"]) + "/" + str(result["containers"]["total"]) + "运行中" +
+        "; 中间件[" + mw_str + "]" +
+        "; Java服务:" + str(len(result["java_services"])) + "个" +
+        "; 日志错误:" + str(result["log_errors"]) + "个"
+    )
+
+    client.close()
+
+except paramiko.AuthenticationException:
+    result["status"] = "critical"
+    result["summary"] = "节点3(192.168.9.91)SSH认证失败，用户名或密码错误"
+    result["alerts"].append("SSH认证失败")
+except socket.timeout:
+    result["status"] = "critical"
+    result["summary"] = "节点3(192.168.9.91)SSH连接超时，服务器可能无法访问"
+    result["alerts"].append("SSH连接超时")
+except Exception as e:
+    result["status"] = "critical"
+    result["summary"] = "节点3(192.168.9.91)连接失败: " + str(e)
+    result["alerts"].append("连接异常: " + str(e))
+
+print(json.dumps(result, ensure_ascii=False, indent=2))

.claude/skills/X86-CLUSTER-MONITOR/code/dingtalk_notifier.py

@@ -462,9 +462,9 @@ def send_summary_notification(summary_text: str = None, config_path: str = None)
         logger.info("当前时段不需要发送汇总通知（仅在08:00-09:00发送）")
         return False
 
-    # 检查防重发标记
+    # 检查防重发标记 - 使用绝对路径，避免依赖运行目录
     today_str = datetime.now().strftime('%Y%m%d')
-    report_dir = Path("AuxiliaryTool/ScriptTool/ClusterMonitor/reports")
+    report_dir = Path(__file__).parent / "AuxiliaryTool" / "ScriptTool" / "ClusterMonitor" / "reports"
     sentinel_file = report_dir / f'.dingtalk_sent_{today_str}'
 
     if sentinel_file.exists():

.claude/skills/X86-CLUSTER-MONITOR/code/monitor_node4.py

+# -*- coding: utf-8 -*-
+"""ARM集群监控 - 节点4 dm8-server 192.168.9.92 单节点检查脚本"""
+import paramiko
+import json
+import re
+import sys
+
+sys.stdout.reconfigure(encoding='utf-8')
+
+IP = '192.168.9.92'
+USERNAME = 'openkylin'
+PASSWORD = 'Ubains@123'
+
+
+def ssh_exec(client, cmd):
+    """执行SSH命令并返回输出"""
+    stdin, stdout, stderr = client.exec_command(cmd, timeout=30)
+    out = stdout.read().decode('utf-8', errors='replace')
+    err = stderr.read().decode('utf-8', errors='replace')
+    return out, err
+
+
+def run_sudo(client, cmd):
+    """使用sudo执行命令"""
+    full_cmd = "echo '{}' | sudo -S {} 2>/dev/null".format(PASSWORD, cmd)
+    return ssh_exec(client, full_cmd)
+
+
+def main():
+    client = paramiko.SSHClient()
+    client.set_missing_host_key_policy(paramiko.AutoAddPolicy())
+    client.connect(IP, port=22, username=USERNAME, password=PASSWORD, timeout=15)
+
+    # 1. CPU使用率 - top -bn1解析%Cpu行
+    cpu_out, _ = run_sudo(client, 'top -bn1 | head -20')
+    cpu_usage = ''
+    for line in cpu_out.splitlines():
+        if 'Cpu(s)' in line or '%Cpu' in line:
+            m = re.search(r'([\d.]+)\s*id', line)
+            if m:
+                idle = float(m.group(1))
+                cpu_usage = '{:.1f}'.format(100 - idle)
+            break
+
+    # 2. 内存使用率 - free -m, 使用available计算真实使用
+    mem_out, _ = run_sudo(client, 'free -m')
+    memory_usage = ''
+    mem_total = 0
+    mem_used = 0
+    mem_avail = 0
+    lines = mem_out.splitlines()
+    for i, line in enumerate(lines):
+        if i == 1:
+            # 第二行是内存数据行，可能含中文标签
+            parts = line.split()
+            nums = []
+            for p in parts:
+                if re.match(r'^[\d.]+$', p):
+                    nums.append(float(p))
+            if len(nums) >= 2:
+                mem_total = nums[0]
+                mem_used = nums[1]
+                if len(nums) >= 7:
+                    mem_avail = nums[6]
+                # 真实使用率 = (total - available) / total
+                if mem_avail > 0 and mem_total > 0:
+                    memory_usage = '{:.1f}'.format((mem_total - mem_avail) / mem_total * 100)
+                elif mem_total > 0:
+                    memory_usage = '{:.1f}'.format(mem_used / mem_total * 100)
+            break
+
+    # 3. 磁盘使用率 - df -h, 解析根分区和/data
+    disk_out, _ = run_sudo(client, 'df -h')
+    disk_info = ''
+    root_pct = ''
+    root_avail = ''
+    data_parts = []
+    for line in disk_out.splitlines():
+        parts = line.split()
+        if len(parts) >= 6:
+            mount = parts[-1]
+            if mount == '/' or mount == '/cow':
+                # /cow是live系统根分区
+                root_pct = parts[-2].replace('%', '')
+                root_avail = parts[-3]
+                disk_info += '/{}(可用{}) '.format(parts[-2], parts[-3])
+            elif mount == '/data':
+                data_parts.append('/data:{}(可用{})'.format(parts[-2], parts[-3]))
+    if data_parts:
+        disk_info += ' '.join(data_parts)
+    disk_info = disk_info.strip()
+
+    # 4. 负载和运行时间 - uptime
+    uptime_out, _ = run_sudo(client, 'uptime')
+    load_avg = ''
+    uptime_str = ''
+    m = re.search(r'load average:\s*([\d.]+),\s*([\d.]+),\s*([\d.]+)', uptime_out)
+    if m:
+        load_avg = '{}, {}, {}'.format(m.group(1), m.group(2), m.group(3))
+    um = re.search(r'up\s+(.+?),\s*\d+ user', uptime_out)
+    if um:
+        uptime_str = um.group(1).strip()
+
+    # 5. 容器状态 - docker ps -a
+    docker_out, _ = run_sudo(client, 'docker ps -a --format "{{.Names}}\t{{.Status}}"')
+    containers = {'total': 0, 'running': 0, 'abnormal': [], 'details': ''}
+    container_details = []
+    for line in docker_out.strip().splitlines():
+        if not line.strip():
+            continue
+        parts = line.split('\t')
+        if len(parts) >= 2:
+            name = parts[0].strip()
+            status = parts[1].strip()
+        elif len(parts) == 1:
+            # 无制表符，按多空格分割
+            sub = parts[0].split()
+            if len(sub) >= 2:
+                name = sub[0].strip()
+                status = ' '.join(sub[1:]).strip()
+            else:
+                continue
+        else:
+            continue
+        containers['total'] += 1
+        if 'Up' in status:
+            containers['running'] += 1
+        else:
+            containers['abnormal'].append('{}:{}'.format(name, status))
+        container_details.append('{}({})'.format(name, status))
+    containers['details'] = '; '.join(container_details) if container_details else '无容器'
+
+    # 6. DM8数据库状态
+    dm8_proc, _ = run_sudo(client, 'ps aux | grep -i dmserver | grep -v grep')
+    dm8_port, _ = run_sudo(client, 'ss -tlnp | grep 5236')
+    dm8_status = 'ok' if (dm8_proc.strip() or dm8_port.strip()) else 'fail'
+    dm8_detail = ''
+    if dm8_proc.strip():
+        dm8_detail = '进程运行中'
+    if dm8_port.strip():
+        dm8_detail += '; 端口5236监听中'
+    if not dm8_detail:
+        dm8_detail = '未检测到DM8进程和端口'
+
+    # 7. Java进程数
+    java_out, _ = run_sudo(client, 'ps aux | grep java | grep -v grep | wc -l')
+    java_count = int(java_out.strip()) if java_out.strip().isdigit() else 0
+
+    # 8. 日志错误检查
+    log_out, _ = run_sudo(
+        client,
+        'find /data/services -name "*.log" -mmin -120 -exec grep -l "ERROR\\|Exception" {} \\; 2>/dev/null | wc -l'
+    )
+    log_errors = int(log_out.strip()) if log_out.strip().isdigit() else 0
+
+    client.close()
+
+    # 状态判定
+    status = 'normal'
+    alerts = []
+
+    # 根分区检查
+    if root_pct and int(root_pct) >= 95:
+        status = 'critical'
+        alerts.append('根分区使用率{}%，可用空间仅{}'.format(root_pct, root_avail))
+    elif root_pct and int(root_pct) >= 85:
+        status = 'warning'
+        alerts.append('根分区使用率{}%，可用空间{}'.format(root_pct, root_avail))
+
+    # CPU检查
+    if cpu_usage and float(cpu_usage) > 90:
+        status = 'critical'
+        alerts.append('CPU使用率{}%'.format(cpu_usage))
+    elif cpu_usage and float(cpu_usage) > 80:
+        if status != 'critical':
+            status = 'warning'
+        alerts.append('CPU使用率{}%'.format(cpu_usage))
+
+    # 内存检查
+    if memory_usage and float(memory_usage) > 90:
+        if status != 'critical':
+            status = 'critical'
+        alerts.append('内存使用率{}%'.format(memory_usage))
+    elif memory_usage and float(memory_usage) > 80:
+        if status == 'normal':
+            status = 'warning'
+        alerts.append('内存使用率{}%'.format(memory_usage))
+
+    # DM8检查
+    if dm8_status == 'fail':
+        if status == 'normal':
+            status = 'warning'
+        alerts.append('DM8数据库未运行')
+
+    # 容器异常检查
+    if containers['abnormal']:
+        if status == 'normal':
+            status = 'warning'
+        alerts.append('异常容器: {}'.format(', '.join(containers['abnormal'])))
+
+    # 日志错误检查
+    if log_errors > 0:
+        if status == 'normal':
+            status = 'warning'
+        alerts.append('近2小时{}个日志文件含ERROR/Exception'.format(log_errors))
+
+    # 汇总信息
+    summary_parts = []
+    summary_parts.append('CPU {}%'.format(cpu_usage) if cpu_usage else 'CPU N/A')
+    summary_parts.append('内存 {}%'.format(memory_usage) if memory_usage else '内存 N/A')
+    summary_parts.append('根分区 {}%'.format(root_pct) if root_pct else '磁盘 N/A')
+    summary_parts.append('负载 {}'.format(load_avg) if load_avg else '')
+    summary_parts.append('容器 {}/{}运行'.format(containers['running'], containers['total']))
+    summary_parts.append('DM8 {}'.format(dm8_status))
+    summary_parts.append('Java进程 {}个'.format(java_count))
+    summary_parts.append('日志错误 {}个文件'.format(log_errors))
+    summary = ', '.join([p for p in summary_parts if p])
+
+    result = {
+        'ip': IP,
+        'role': 'dm8-server',
+        'status': status,
+        'cpu': cpu_usage + '%',
+        'memory': memory_usage + '%',
+        'disk': disk_info,
+        'load': load_avg,
+        'uptime': uptime_str,
+        'containers': containers,
+        'middleware': {'dm8': dm8_status, 'detail': dm8_detail},
+        'java_services': java_count,
+        'log_errors': log_errors,
+        'alerts': alerts,
+        'summary': summary
+    }
+
+    print(json.dumps(result, ensure_ascii=False, indent=2))
+
+
+if __name__ == '__main__':
+    main()

.claude/skills/X86-CLUSTER-MONITOR/code/slave2_91_result.json

+{
+  "ip": "192.168.9.91",
+  "role": "slave2",
+  "status": "warning",
+  "cpu": "2.1%",
+  "memory": "86.9%",
+  "disk": "/data:33%(79G)",
+  "load": "0.84, 2.43, 3.32",
+  "uptime": "5 days",
+  "containers": {
+    "total": 9,
+    "running": 9,
+    "abnormal": [],
+    "details": "utracker:Up 4 days; redis-sentinel-3:Up 5 days; redis-slave-2:Up 5 days; unacos:Up 17 hours; uemqx:Up 5 days; utengine:Up 5 days; paperless:Up 5 days; ujava2:Up 5 days; umysql:Up 5 days"
+  },
+  "middleware": {
+    "mysql": "ok",
+    "redis": "ok",
+    "emqx": "ok",
+    "nacos": "ok"
+  },
+  "java_services": 8,
+  "log_errors": 9,
+  "alerts": [
+    "�ڴ�ʹ���ʾ���: 86.9%",
+    "������־ERROR�ļ�: 9��",
+    "�ڴ�ʹ�����Դ���λ(86.9%)������91.3%����"
+  ],
+  "summary": "CPU:2.1% | �ڴ�:86.9% | ����:9/9���� | �м��:4/4���� | Java����:8 | ��־ERROR:9�ļ� | �澯:2��"
+}
\ No newline at end of file

.claude/skills/X86-CLUSTER-MONITOR/config.json

@@ -248,7 +248,7 @@
         "history_days": 30
     },
     "dingtalk": {
-        "enabled": false,
+        "enabled": true,
         "webhook": "https://oapi.dingtalk.com/robot/send?access_token=27071a77f20da381e9a321653ec5f4dcf668bcf058c01162f28e3f1f8633386d",
         "secret": "SEC5d85d5735a1805ada1be84929d5b37f5b72a2a832a6bcd9a1ca5615e5799be38"
     }

.claude/skills/X86-QLV10-XTYBS/SKILL.md

@@ -11,10 +11,12 @@ X86架构-麒麟V10服务器新统一平台自动化部署操作。
 
 ## 定时任务
 
-**已设置工作日晚上 22:00 自动执行完整部署流程（全程无交互）**
+**已设置工作日晚上 23:00 自动执行完整部署流程（全程无交互）**
 
-- 定时任务ID: `84dbe6b5`
-- 执行时间: 周一至周五 22:00
+> ⚠️ 调度说明：原设定 22:00，因与 X86-TX-XTYBS（21:57 触发，部署耗时近1小时）串行阻塞导致延迟到 22:31 才补触发，已于 2026-07-14 调整为 23:00，确保 TX 任务先跑完。
+
+- 定时任务ID: `6f064957`
+- 执行时间: 周一至周五 23:00
 - 执行内容: 完整自动化流程（部署→授权→等待→验证→生成报告）
 - 脚本路径: `.claude/skills/X86-QLV10-XTYBS/code/scheduled_deploy.py`
 - 授权脚本: `AuxiliaryTool/ScriptTool/RemoteDeploy/authorize_x86_kylin.py`（Selenium浏览器自动化）
@@ -45,7 +47,7 @@ python AuxiliaryTool/ScriptTool/RemoteDeploy/authorize_x86_kylin.py
 
 **管理定时任务**:
 - 查看所有定时任务: `/cron`
-- 取消定时任务: 使用 `CronDelete` 工具删除 ID `84dbe6b5`
+- 取消定时任务: 使用 `CronDelete` 工具删除 ID `6f064957`
 
 ## 基本规则