Монитор ресурсов через WEB, или мониторим свою VPS с помощью Grafana с самописными метриками

nano /etc/nginx/sites-available/domain.com

    # Настройки для графаны
    location /grafs {
        proxy_pass http://localhost:3000;
        rewrite ^/grafs/(.*) /$1 break;
        proxy_set_header Host $http_host;
        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
        proxy_set_header X-Real-IP $remote_addr;
    }

sudo nano /etc/grafana/grafana.ini

[server]
domain = mydomain.com
root_url = %(protocol)s://%(domain)s:%(http_port)s/grafs
serve_from_sub_path = true

def do_alarm(t_alarmtext):
    headers = {"Content-type": "application/json"}
    payload = {"text": f"{t_alarmtext}", "chat_id": f"{admin_id}"}
    requests.post(url=webhook_url, data=json.dumps(payload), headers=headers)

def get_cpu_procent():
    #print(f'psutil.cpu_percent(): {psutil.cpu_percent()}')
    return psutil.cpu_percent()


def get_virtual_memory_procent_usage():
    #print(f'psutil.virtual_memory().percent: {psutil.virtual_memory().percent}')
    return psutil.virtual_memory().percent


def get_disc_usage():
    #print(f'psutil.disk_usage.used.precent:{psutil.disk_usage("/").percent}')
    return psutil.disk_usage("/").percent

class CustomCollector(object):
    def __init__(self):
        pass

    def collect(self):
        cpu_usage = GaugeMetricFamily('system_cpu_percent_usage', '% использования cpu')
        cpu_usage_metric = get_cpu_procent()
        cpu_usage.add_metric([], cpu_usage_metric)

        ram_usage = GaugeMetricFamily('system_ram_percent_usage', '% использования оперативы')
        ram_usage_metric = get_virtual_memory_procent_usage()
        ram_usage.add_metric([], ram_usage_metric)

        disk_usage = GaugeMetricFamily('system_disk_usage', '% занятого места на диске')
        disk_usage_metric = get_disc_usage()
        disk_usage.add_metric([], disk_usage_metric)

        yield cpu_usage
        yield ram_usage
        yield disk_usage

if __name__ == '__main__':
    try:
        start_http_server(8000)
        REGISTRY.register(CustomCollector())
        while True:
            time.sleep(1)
    except KeyboardInterrupt:
        logger.info('Program stopped')
    except Exception as e:
        t_alarmtext = f'prometheus-custom-collector (app.py):\n {str(e)}'
        do_alarm(t_alarmtext)
        logger.error(f'Other except error Exception', exc_info=True)

nano /etc/prometheus/prometheus.yml

  - job_name: 'prometheus-custom'
    scrape_interval: 5s
    static_configs:
      - targets: ['localhost:8000']