Logs problem: Failed auth

[thomas789]

INFO:

WAPT Server version: 2.6.0.16552
WAPT Agent version: 2.6.0
WAPT Console version: 2.6.0.16767
My admin PC version: Ubuntu 24.04

Hello, I'm taking the liberty of bumping an old topic (viewtopic.php?t=3683We're experiencing the same problem on ours. What troubleshooting steps could we take? Thanks in advance.

Code for our case that spams the logs every minute:

Code: Select all

[waptserver] CRITICAL host_tasks_status failed EWaptHostUnreachable('Host not connected, Websocket sid not in database') 2025-03-12T16:08 NOM_DU_SERVEUR waptserver[881]: 2025-03-12 16:08:17 
[waptauth] WARNING Failed auth for None from IP_DU_POSTE path /add_host

[thomas789]

Good morning,

To add context, we recently changed the WAPT server name, and upon closer inspection of the configuration, I noticed that the Nginx configuration still has the old server name for the following line:

Code: Select all

ssl_client_certificate "/opt/wapt/conf/ca-ANCIEN_NOM_DE_SERVEUR.crt";

As well as the crt + pem keys, present in /opt/wapt/conf/, still have this old name.

This might explain these errors. Furthermore, could you please tell me the procedure to follow to correct this?

Thanks in advance,

Sincerely.

[dcardon]

Hello Thomas,

this certificate is used to sign and validate client certificates for WAPT agents. In practice, the only impact of the server name is the name displayed when you double-click the .crt file on the agent. So, having the old name here isn't a problem.

If you want the new server name, you'll need to create a new .crt + .pem file, but all agents will have to re-register with the WAPT server. Therefore, the Kerberos configuration must be correct.

Regarding the initial problem, it seems that one agent isn't properly registered with the server; we'll need to investigate why.

Best regards,

Denis