Solved

Trying to connect poller to centreon serveur central : ERROR - Couldn't connect to https

  • 9 September 2022
  • 6 replies
  • 85 views

Badge +2

Hi everyone,

I’ve recently installed a centreon central server (using À partir des paquets | Centreon Documentation) and a poller (using À partir des paquets | Centreon Documentation), I’ve also done the secure part with Sécurisez votre plateforme | Centreon Documentation.

After that I’ve been trying to connect poller to the centreon central server without any success with /usr/share/centreon/bin/registerServerTopology.sh -u admin -t poller -h x.x.x.x -n poller-1.

 

At first I just had the typicall non helping message (at least to me) “ ERROR - Couldn't connect to https:x.x.x.x:443”. So i checked if port 443 was opened on central server and it was.

 

I tried looking through different stuff using Centreon Troubleshooting Series | Episode 1 : Help! Mon Poller ne fonctionne pas.

 

-I’ve noticed that :

  • the gorgoned service was not started on the poller and that the 31-centreon-api.yaml had no information in it (which is probably normal ?)
  • that the systemctl status gorgoned didn’t have gorgone-dbclearner engine or action (again I have no idea if it’s normal or not)
gorgone:
tpapi:
- name: centreonv2
base_url: ""
username: "@GORGONE_USER@"
password: "@GORGONE_PASSWORD@"
- name: clapi
username: "@GORGONE_USER@"
password: "@GORGONE_PASSWORD@"
  • the 40-gorgoned.yaml file is not present but I guess it will be after the link is done.

I’ve also tried to look at logs:

  • the poller is trying to ping himself on port 5669 and it isn’t working (with or without firewall) which is probably not helping since 5669 is supposed to be from poller to central server / remote server ?

[2022-09-09T06:51:12.339-04:00] [tcp] [error] TCP: Connection refused to :5669
[2022-09-09T06:51:12.339-04:00] [core] [error] failover: global error: failover: 'central-module-master-output' cannot connect endpoint.

  • Same thing is happening on centreon server central with port 5670 (?)
  • I also have this error in the gorgoned.log on the central server which is probably not linked at all to this problem and is a few days old
2022-09-04 03:33:51 - ERROR - [autodiscovery] -class- host discovery - cannot get platform versions - http request error
2022-09-04 03:34:51 - ERROR - [autodiscovery] Receiving DIE: SSL peer certificate or SSH remote key was not OK
2022-09-04 03:34:51 - ERROR - [autodiscovery] Receiving DIE: SSL peer certificate or SSH remote key was not OK
2022-09-04 03:34:51 - ERROR - [autodiscovery] Receiving DIE: SSL peer certificate or SSH remote key was not OK

 

 

I’ve also tried adding the poller through the central server instead but I couldn’t manage to get the poller to be running.

 

The install is “fresh” and i’ve followed every step in the three urls.

By this point I have no clue what is wrong.

 

It may certainly be a stupid mistake but I have no idea what it is.

 

If you have any idea of where I should look / what I should change, thanks for your help !

Regards

Arnaud

icon

Best answer by ponchoh 9 September 2022, 15:44

View original

6 replies

Badge +2

I’m trying to enter log messages but i get the error something’s gone wrong.
 

Badge +10

👋

 

@ArnaudT is it a self signed?

 

/usr/share/centreon/bin/registerServerTopology.sh -u admin -t poller -h x.x.x.x -n poller-1 --help

 

 --insecure <optional>                allow self-signed certificate

Badge +2

The certificate on the central server is self signed indeed, I didn’t think that it would be the reason for that. 

Thanks!

Badge +2

@hmorales since we are on the subjects of certifcates do you know why i get this for the autodiscovery ?

The tasks for autodiscovery seems to no start and I guess its linked to this (in gorgoned.log)

Is there an insecure option or something to activate somwhere about autodiscovery?

I’m trying to do that on Windows Server for now so, snmp agents (2c), I had it working once (different server) but can’t remember what I did… 

2022-09-12 05:45:52 - ERROR - [autodiscovery] Receiving DIE: SSL peer certificate or SSH remote key was not OK
2022-09-12 05:45:52 - ERROR - [autodiscovery] Receiving DIE: SSL peer certificate or SSH remote key was not OK
2022-09-12 05:45:52 - ERROR - [autodiscovery] Receiving DIE: SSL peer certificate or SSH remote key was not OK
2022-09-12 05:45:52 - ERROR - [autodiscovery] Receiving DIE: SSL peer certificate or SSH remote key was not OK
2022-09-12 05:45:52 - ERROR - [autodiscovery] Receiving DIE: SSL peer certificate or SSH remote key was not OK
2022-09-12 05:45:52 - ERROR - [autodiscovery] Receiving DIE: SSL peer certificate or SSH remote key was not OK
2022-09-12 05:45:52 - ERROR - [autodiscovery] Receiving DIE: SSL peer certificate or SSH remote key was not OK
2022-09-12 05:45:52 - ERROR - [autodiscovery] Receiving DIE: SSL peer certificate or SSH remote key was not OK
2022-09-12 05:45:52 - ERROR - curl perform error : SSL peer certificate or SSH remote key was not OK

 

Badge +10

@ArnaudT looks like it all comes down to the selfsigned cert and the ip/fqdn is part of your alt_names

 

https://docs.centreon.com/docs/administration/secure-platform/#creating-a-self-signed-certificate

 

ie

 

 

Badge +2

I’ll try to check and recreate the certificates, is that normal that the task stays on programmed though ?

 

 

Reply