refactor: limpeza de testes, reestruturacao multi-tenant e auto-cura scoreodonto
This commit is contained in:
Rui
@@ -0,0 +1,344 @@
|
||||
#!/usr/bin/env python3
|
||||
import sys
|
||||
import os
|
||||
import time
|
||||
import subprocess
|
||||
import json
|
||||
import syslog
|
||||
import socket
|
||||
from datetime import datetime, timedelta
|
||||
|
||||
CONFIG_PATH = "/etc/ssh/ssh_alerts_config.json"
|
||||
STATE_FILE = "/tmp/system_resources_state.json"
|
||||
ALERT_THRESHOLD = 85.0
|
||||
COOLDOWN_SECONDS = 3600 # Remind every 1 hour if resource stays above threshold
|
||||
|
||||
def log_to_syslog(message, is_error=False):
|
||||
level = syslog.LOG_ERR if is_error else syslog.LOG_INFO
|
||||
syslog.openlog(ident="system_resources_monitor", logoption=syslog.LOG_PID, facility=syslog.LOG_DAEMON)
|
||||
syslog.syslog(level, message)
|
||||
syslog.closelog()
|
||||
|
||||
def get_vps_identity():
|
||||
# Attempt to auto-detect WireGuard IP
|
||||
try:
|
||||
output = subprocess.check_output(
|
||||
['ip', '-o', '-4', 'addr', 'show', 'dev', 'wg0'],
|
||||
stderr=subprocess.DEVNULL
|
||||
).decode('utf-8')
|
||||
for line in output.split('\n'):
|
||||
parts = line.split()
|
||||
if len(parts) >= 4:
|
||||
return f"VPS 3 ({parts[3].split('/')[0]})"
|
||||
except Exception:
|
||||
pass
|
||||
|
||||
# Fallback to general IP/hostname
|
||||
try:
|
||||
return f"VPS ({socket.gethostname()})"
|
||||
except Exception:
|
||||
return "VPS (10.99.0.3)"
|
||||
|
||||
def get_cpu_usage():
|
||||
def read_cpu_times():
|
||||
with open('/proc/stat', 'r') as f:
|
||||
first_line = f.readline()
|
||||
parts = first_line.split()
|
||||
times = [float(x) for x in parts[1:9]]
|
||||
idle = times[3] + times[4] # idle + iowait
|
||||
total = sum(times)
|
||||
return idle, total
|
||||
|
||||
try:
|
||||
idle1, total1 = read_cpu_times()
|
||||
time.sleep(1.0)
|
||||
idle2, total2 = read_cpu_times()
|
||||
|
||||
idle_delta = idle2 - idle1
|
||||
total_delta = total2 - total1
|
||||
|
||||
if total_delta > 0:
|
||||
return (1.0 - idle_delta / total_delta) * 100.0
|
||||
except Exception as e:
|
||||
log_to_syslog(f"Erro ao obter uso de CPU: {e}", is_error=True)
|
||||
return 0.0
|
||||
|
||||
def get_memory_usage():
|
||||
try:
|
||||
mem_info = {}
|
||||
with open('/proc/meminfo', 'r') as f:
|
||||
for line in f:
|
||||
parts = line.split(':')
|
||||
if len(parts) == 2:
|
||||
key = parts[0].strip()
|
||||
val = parts[1].strip().split()[0]
|
||||
mem_info[key] = int(val)
|
||||
|
||||
total = mem_info.get('MemTotal', 0)
|
||||
available = mem_info.get('MemAvailable', 0)
|
||||
if total > 0:
|
||||
used = total - available
|
||||
return (used / total) * 100.0
|
||||
except Exception as e:
|
||||
log_to_syslog(f"Erro ao obter uso de memoria: {e}", is_error=True)
|
||||
return 0.0
|
||||
|
||||
def get_disk_usage():
|
||||
try:
|
||||
output = subprocess.check_output(['df', '-Pl', '/']).decode('utf-8')
|
||||
lines = output.strip().split('\n')
|
||||
if len(lines) >= 2:
|
||||
parts = lines[-1].split()
|
||||
if len(parts) >= 5:
|
||||
pct_str = parts[4]
|
||||
if pct_str.endswith('%'):
|
||||
return float(pct_str[:-1])
|
||||
except Exception as e:
|
||||
log_to_syslog(f"Erro ao obter uso de disco: {e}", is_error=True)
|
||||
return 0.0
|
||||
|
||||
def get_top_processes():
|
||||
try:
|
||||
ps_output = subprocess.check_output(
|
||||
['ps', '-eo', 'pid,%mem,%cpu,cmd', '--sort=-%mem'],
|
||||
stderr=subprocess.DEVNULL
|
||||
).decode('utf-8', errors='ignore').strip().split('\n')
|
||||
return ps_output[:6] # Header + top 5 processes
|
||||
except Exception as e:
|
||||
log_to_syslog(f"Erro ao obter processos ofensores: {e}", is_error=True)
|
||||
return []
|
||||
|
||||
def load_state():
|
||||
if os.path.exists(STATE_FILE):
|
||||
try:
|
||||
with open(STATE_FILE, "r") as f:
|
||||
return json.load(f)
|
||||
except Exception:
|
||||
pass
|
||||
return {}
|
||||
|
||||
def save_state(state):
|
||||
try:
|
||||
with open(STATE_FILE, "w") as f:
|
||||
json.dump(state, f)
|
||||
except Exception as e:
|
||||
log_to_syslog(f"Erro ao salvar estado de alertas: {e}", is_error=True)
|
||||
|
||||
def send_telegram_alert(token, chat_id, vps_identity, alerts_active, top_processes):
|
||||
try:
|
||||
url = f"https://api.telegram.org/bot{token}/sendMessage"
|
||||
|
||||
alerts_text = ""
|
||||
for resource, val in alerts_active.items():
|
||||
alerts_text += f"• *{resource}:* `{val:.1f}%` (Limite: {ALERT_THRESHOLD:.1f}%)\n"
|
||||
|
||||
proc_text = ""
|
||||
if top_processes:
|
||||
proc_text = "\n🔥 *Top 5 Processos Consumindo mais RAM:*\n```\n"
|
||||
proc_text += "\n".join(top_processes)
|
||||
proc_text += "\n```"
|
||||
|
||||
text = (
|
||||
f"🚨 *ALERTA DE USO DE RECURSOS CRÍTICOS*\n\n"
|
||||
f"• *Servidor:* `{vps_identity}`\n"
|
||||
f"{alerts_text}"
|
||||
f"{proc_text}\n"
|
||||
f"📅 _Registrado automaticamente pelo monitor do sistema._"
|
||||
)
|
||||
|
||||
payload = json.dumps({
|
||||
"chat_id": chat_id,
|
||||
"text": text,
|
||||
"parse_mode": "Markdown"
|
||||
}).encode("utf-8")
|
||||
|
||||
req = urllib.request.Request(
|
||||
url,
|
||||
data=payload,
|
||||
headers={
|
||||
'Content-Type': 'application/json',
|
||||
'User-Agent': 'Mozilla/5.0'
|
||||
}
|
||||
)
|
||||
import urllib.request
|
||||
with urllib.request.urlopen(req, timeout=5) as r:
|
||||
res = json.loads(r.read().decode())
|
||||
return res.get("ok", False)
|
||||
except Exception as e:
|
||||
log_to_syslog(f"Erro ao enviar alerta para o Telegram: {e}", is_error=True)
|
||||
return False
|
||||
|
||||
def send_telegram_resolved(token, chat_id, vps_identity, resolved_resources):
|
||||
try:
|
||||
url = f"https://api.telegram.org/bot{token}/sendMessage"
|
||||
|
||||
resolved_text = ""
|
||||
for resource, val in resolved_resources.items():
|
||||
resolved_text += f"• *{resource}:* `{val:.1f}%` (Normalizado)\n"
|
||||
|
||||
text = (
|
||||
f"✅ *RECURSOS NORMALIZADOS*\n\n"
|
||||
f"• *Servidor:* `{vps_identity}`\n"
|
||||
f"{resolved_text}\n"
|
||||
f"📅 _O servidor voltou a operar dentro dos limites de segurança._"
|
||||
)
|
||||
|
||||
payload = json.dumps({
|
||||
"chat_id": chat_id,
|
||||
"text": text,
|
||||
"parse_mode": "Markdown"
|
||||
}).encode("utf-8")
|
||||
|
||||
req = urllib.request.Request(
|
||||
url,
|
||||
data=payload,
|
||||
headers={
|
||||
'Content-Type': 'application/json',
|
||||
'User-Agent': 'Mozilla/5.0'
|
||||
}
|
||||
)
|
||||
import urllib.request
|
||||
with urllib.request.urlopen(req, timeout=5) as r:
|
||||
res = json.loads(r.read().decode())
|
||||
return res.get("ok", False)
|
||||
except Exception as e:
|
||||
log_to_syslog(f"Erro ao enviar resolucao para o Telegram: {e}", is_error=True)
|
||||
return False
|
||||
|
||||
def create_calendar_event(vps_identity, alerts_active, top_processes, calendar_id):
|
||||
try:
|
||||
from google.oauth2 import service_account
|
||||
from googleapiclient.discovery import build
|
||||
|
||||
creds = service_account.Credentials.from_service_account_file(
|
||||
'/etc/ssh/gcal_key.json',
|
||||
scopes=['https://www.googleapis.com/auth/calendar']
|
||||
)
|
||||
|
||||
service = build('calendar', 'v3', credentials=creds)
|
||||
|
||||
now = datetime.utcnow()
|
||||
end_time = now + timedelta(minutes=30)
|
||||
|
||||
start_iso = now.isoformat() + 'Z'
|
||||
end_iso = end_time.isoformat() + 'Z'
|
||||
|
||||
resources_str = ", ".join(alerts_active.keys())
|
||||
summary = f"{vps_identity}: 🚨 Estresse de Sistema - {resources_str}"
|
||||
|
||||
description = (
|
||||
f"=== 🖥️ ALERTA DE RECURSOS DO SISTEMA ===\n\n"
|
||||
f"• Servidor: {vps_identity}\n"
|
||||
f"• Recursos acima do limite (>={ALERT_THRESHOLD}%):\n"
|
||||
)
|
||||
for res, val in alerts_active.items():
|
||||
description += f" - {res}: {val:.1f}%\n"
|
||||
|
||||
if top_processes:
|
||||
description += f"\n--- TOP PROCESSOS POR MEMÓRIA ---\n"
|
||||
description += "\n".join(top_processes)
|
||||
|
||||
description += f"\n\nRegistrado automaticamente em {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}."
|
||||
|
||||
event_body = {
|
||||
'summary': summary,
|
||||
'description': description,
|
||||
'start': {'dateTime': start_iso, 'timeZone': 'UTC'},
|
||||
'end': {'dateTime': end_iso, 'timeZone': 'UTC'},
|
||||
'colorId': "11", # Vermelho Bold
|
||||
'reminders': {
|
||||
'useDefault': False,
|
||||
'overrides': [{'method': 'popup', 'minutes': 0}]
|
||||
}
|
||||
}
|
||||
|
||||
event = service.events().insert(
|
||||
calendarId=calendar_id,
|
||||
body=event_body
|
||||
).execute()
|
||||
return True
|
||||
except Exception as e:
|
||||
log_to_syslog(f"Erro ao criar evento de estresse no Google Calendar: {e}", is_error=True)
|
||||
return False
|
||||
|
||||
def main():
|
||||
# Load config
|
||||
if not os.path.exists(CONFIG_PATH):
|
||||
log_to_syslog(f"Arquivo de configuracao nao encontrado: {CONFIG_PATH}", is_error=True)
|
||||
return
|
||||
|
||||
try:
|
||||
with open(CONFIG_PATH, "r") as f:
|
||||
config = json.load(f)
|
||||
except Exception as e:
|
||||
log_to_syslog(f"Erro ao carregar configuracoes: {e}", is_error=True)
|
||||
return
|
||||
|
||||
telegram_token = config.get("telegram_token")
|
||||
telegram_chat_id = config.get("telegram_chat_id")
|
||||
calendar_id = config.get("calendar_id")
|
||||
|
||||
if not telegram_token or not telegram_chat_id:
|
||||
log_to_syslog("Telegram credentials missing in config.", is_error=True)
|
||||
return
|
||||
|
||||
# Check metrics
|
||||
cpu = get_cpu_usage()
|
||||
mem = get_memory_usage()
|
||||
disk = get_disk_usage()
|
||||
|
||||
metrics = {
|
||||
"CPU": cpu,
|
||||
"Memoria": mem,
|
||||
"Disco (/)": disk
|
||||
}
|
||||
|
||||
vps_identity = get_vps_identity()
|
||||
state = load_state()
|
||||
now = time.time()
|
||||
|
||||
alerts_to_send = {}
|
||||
resolved_to_send = {}
|
||||
|
||||
state_changed = False
|
||||
|
||||
for resource, val in metrics.items():
|
||||
is_above = val >= ALERT_THRESHOLD
|
||||
prev_alert_time = state.get(resource, 0)
|
||||
|
||||
if is_above:
|
||||
# Check if we should notify: either first time or after cooldown
|
||||
if prev_alert_time == 0 or (now - prev_alert_time >= COOLDOWN_SECONDS):
|
||||
alerts_to_send[resource] = val
|
||||
state[resource] = now
|
||||
state_changed = True
|
||||
else:
|
||||
# If it was previously in alert state, send resolution message
|
||||
if prev_alert_time > 0:
|
||||
resolved_to_send[resource] = val
|
||||
state[resource] = 0
|
||||
state_changed = True
|
||||
|
||||
if state_changed:
|
||||
save_state(state)
|
||||
|
||||
# Process resolutions
|
||||
if resolved_to_send:
|
||||
log_to_syslog(f"Métricas normalizadas para: {', '.join(resolved_to_send.keys())}")
|
||||
send_telegram_resolved(telegram_token, telegram_chat_id, vps_identity, resolved_to_send)
|
||||
|
||||
# Process alerts
|
||||
if alerts_to_send:
|
||||
log_to_syslog(f"ALERTA: Métricas acima de {ALERT_THRESHOLD}%: " + ", ".join([f"{k} ({v:.1f}%)" for k, v in alerts_to_send.items()]))
|
||||
|
||||
top_processes = get_top_processes()
|
||||
|
||||
# Send to Telegram
|
||||
send_telegram_alert(telegram_token, telegram_chat_id, vps_identity, alerts_to_send, top_processes)
|
||||
|
||||
# Log to Calendar
|
||||
if calendar_id:
|
||||
create_calendar_event(vps_identity, alerts_to_send, top_processes, calendar_id)
|
||||
|
||||
if __name__ == "__main__":
|
||||
main()
|
||||
Reference in New Issue
Block a user