Hello @vivecentreon  ,

There is 2 main possibility for this to append :

• Some of the gorgone module are not loaded correclty :

Check with this api call the module proxy, node and register are correctly loaded

curl --request GET "http://127.0.0.1:8085/api/internal/information" | jq .

• a poller is configured in the central but don’t respond :

please check every poller configured in the web app is correctly responding

you can list all poller gorgone see with this request on the centreon database :

SELECT id, name, localhost, ns_ip_address, gorgone_port, remote_id, remote_server_use_as_proxy, gorgone_communication_type
FROM nagios_server
WHERE ns_activate = '1'

curl http://127.0.0.1:8085/api/internal/constatus

Hello

Thank you for your answer.

I don’t see something strange. All the Pollers are Green in the Gui and work normally (one host only at the moment).

Here are the result of the Queries:
 

sroot@ ~]# curl --request GET "http://127.0.0.1:8085/api/internal/information" | jq .data.modules
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
100  2642  100  2642    0     0   645k      0 --:--:-- --:--:-- --:--:--  645k
{
  "engine": "gorgone::modules::centreon::engine::hooks",
  "nodes": "gorgone::modules::centreon::nodes::hooks",
  "cron": "gorgone::modules::core::cron::hooks",
  "legacycmd": "gorgone::modules::centreon::legacycmd::hooks",
  "dbcleaner": "gorgone::modules::core::dbcleaner::hooks",
  "httpserver": "gorgone::modules::core::httpserver::hooks",
  "statistics": "gorgone::modules::centreon::statistics::hooks",
  "register": "gorgone::modules::core::register::hooks",
  "autodiscovery": "gorgone::modules::centreon::autodiscovery::hooks",
  "action": "gorgone::modules::core::action::hooks",
  "audit": "gorgone::modules::centreon::audit::hooks",
  "proxy": "gorgone::modules::core::proxy::hooks"
}

MariaDB icentreon]> SELECT id, name, localhost, ns_ip_address, gorgone_port, remote_id, remote_server_use_as_proxy, gorgone_communication_type FROM nagios_server WHERE ns_activate = '1';
+----+-----------------+-----------+----------------+--------------+-----------+----------------------------+----------------------------+
| id | name            | localhost | ns_ip_address  | gorgone_port | remote_id | remote_server_use_as_proxy | gorgone_communication_type |
+----+-----------------+-----------+----------------+--------------+-----------+----------------------------+----------------------------+
|  1 | Central         | 1         | 127.0.0.1      |         5556 |      NULL | 1                          | 1                          |
|  2 | 1-VM-POLLER24   | 0         | Public_ip_1    |         5556 |      NULL | 1                          | 1                          |
|  3 | 2-VM-POLLER24   | 0         | Public_ip_2    |         5556 |      NULL | 1                          | 1                          |
|  4 | 3-VM-POLLER24   | 0         | Public_ip_3    |         5556 |      NULL | 1                          | 1                          |
|  5 | 3-VM-POLLER24   | 0         | Public_ip_4    |         5556 |      NULL | 1                          | 1                          |
|  6 | 4-VM-POLLER24   | 0         | Public_ip_5    |         5556 |      NULL | 1                          | 1                          |
+----+-----------------+-----------+----------------+--------------+-----------+----------------------------+----------------------------+
6 rows in set (0.000 sec)

MariaDB icentreon]>

aroot@~]# curl http://127.0.0.1:8085/api/internal/constatus | jq .
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
100  1102  100  1102    0     0   269k      0 --:--:-- --:--:-- --:--:--  269k
{
  "action": "constatus",
  "message": "ok",
  "data": {
    "6": {
      "type": "pull",
      "ping_timeout": 0,
      "next_ping": 1719235508,
      "in_progress_ping_pull": 1719235448,
      "nodes": {},
      "last_ping_sent": 1719235448,
      "in_progress_ping": 0,
      "ping_failed": 1,
      "ping_ok": 7,
      "last_ping_recv": 1719235448
    },
    "2": {
      "last_ping_sent": 1719235408,
      "in_progress_ping": 0,
      "ping_failed": 1,
      "ping_ok": 7,
      "last_ping_recv": 1719235448,
      "type": "pull",
      "in_progress_ping_pull": 1719235408,
      "next_ping": 1719235468,
      "ping_timeout": 0,
      "nodes": {}
    },
    "4": {
      "last_ping_recv": 1719235448,
      "ping_ok": 6,
      "ping_failed": 1,
      "last_ping_sent": 1719235408,
      "in_progress_ping": 0,
      "nodes": {},
      "in_progress_ping_pull": 1719235408,
      "ping_timeout": 0,
      "next_ping": 1719235468,
      "type": "pull"
    },
    "5": {
      "in_progress_ping_pull": 1719235408,
      "ping_timeout": 0,
      "next_ping": 1719235468,
      "type": "pull",
      "nodes": {},
      "last_ping_sent": 1719235408,
      "in_progress_ping": 0,
      "last_ping_recv": 1719235448,
      "ping_ok": 6,
      "ping_failed": 1
    },
    "3": {
      "ping_timeout": 0,
      "next_ping": 1719235508,
      "in_progress_ping_pull": 1719235448,
      "type": "pull",
      "nodes": {},
      "last_ping_sent": 1719235448,
      "in_progress_ping": 0,
      "last_ping_recv": 1719235448,
      "ping_failed": 1,
      "ping_ok": 7
    }
  }
}

Pierre

Hello
Here are the CPU & Load graphs of the Central server:

 

When I have disabled all Pollers with the Gui, we can see that the CPU and Load goes down to almost zero. When the Pollers are enabled, the CPU & Load goes High.

I have update Central and Poller to lastest version.

Does anyboy else have the same trouble and was abble to fix it ?

Regards

Pierre
 

Hello everybody
 

I have postponed my migration from 21 to latest 24 version due to the gogrgoned-proxy issue that I have.

I have 1 Central with 6 remote pollers (reverse flow).
When I enable the pollers (with only 1 host monitored each poller, in fact the loller itseld), the CPU goes high after one hour on the Central.

Am I the only one to have this problem ?

Thank you in advance.

Pierre

Hi @vivecentreon,

I just saw your issue, and I had this problem I think when updating my test platform.

I don’t remember exactly what I did but can you check that your Central’s Gorgone is using the register module as such :

    - name: register
      package: "gorgone::modules::core::register::hooks"
      enable: true
      config_file: /etc/centreon-gorgone/nodes-register-override.yml

(I put it before nodes and proxy modules)

And that you have all your Pollers in the config file like :

nodes:
  - id: 7
    type: pull
    prevail: 1

  - id: 8
    type: pull
    prevail: 1

And that this config file is readable by Gorgone like 640/centreon-gorgone.

If you set logging level to debug and restart gorgoned, you should have logs like:

2024-08-19 10:53:19 - INFO - Fcore] Module 'register' is loading
2024-08-19 10:53:19 - INFO - Fcore] Module 'register' is loaded
2024-08-19 10:53:20 - DEBUG - Uregister] internal message: aREGISTERNODES] D] ]] {"nodes":d{"prevail":"1","type":"pull","id":"7"}]}
2024-08-19 10:53:20 - DEBUG - Unodes] internal message: aREGISTERNODES] D] ]] {"nodes":d{"id":"7","port":5556,"type":"push_zmq","address":"10.2.3.4"}]}
2024-08-19 10:53:20 - DEBUG - Ucore] Message received internal - aREGISTERNODES] D] ]] {"nodes":d{"prevail":"1","type":"pull","id":"7"}]}
2024-08-19 10:53:20 - INFO - Fproxy] Node '7' is registered

Logs will show that nodes module try to register Poller but register module configuration will prevail.

And you will see this log every minute:

2024-08-19 11:01:15 - DEBUG - Ucore] Message received external - aREGISTERNODES] D] ]] {"nodes":d{"identity":"gorgone-7-a23434674e3e90da6d75806d","type":"pull","id":"7"}]}
2024-08-19 11:01:15 - INFO - Fproxy] cannot override node '7' registration: prevails!!!

As I said I don’t remember exactly what was the problem but I think the proxy module was trying to access Gorgone’s Poller, which was not possible, because of misconfiguration or right issue.

Hello Stewart

Thank you for the answer.
I have try what you said but everything was already like that.

However I have also noticed in the log this kinf of line (every 10 minutes for each poller)

[nodes] internal message: [registernodes] [] [] {"nodes":[{"id":"7","port":5556,"type":"push_zmq","address":"xx.xx.xx.xx"}

It seems that the nodes module of the Central Server try to connect the my remote pollers with public IP (which is impossible).

Finally, I have disabled the nodes module on the Central Server and since then, everything seems to run smoothly (CPU around 1%).

I will check on the next days when activating other remote pollers to see if it changes.

Best regards

Pierre