Services status unknown after update?

I’ve had the same problem since the update.

Yeah trying to figure out what it is.

I’ve made some changes to the engine configuration and have had some success. I’m still getting timeouts but not nearly as much as several hundred services going into status unknown.

Timeouts still happen from time to time. No point going back to an older version now so I’ll try to live with this and see if I can improve it.

Might also be a resource problem since our VM environment started crapping out before christmas. Currently waiting for memory repair on one of two hosts.

So basically the environment our centreon is hosted on is a bit tight on resources and I’m currently on 2500 services and 500 hosts on 4 vCPU 16 GB ram with 1 central poller so might be that we are hitting some limit.

I have the same problem on two different systems/pollers after upgrading the pollers centreon-engine and centreon-connector from 24.10.13 to 24.10.14.

Only the plugins/commands without perl connector was going all the time. After a restart is working for a while until next time, when suddenly all checks go to (Process Timeout)

I can see this error in engine log:

[commands] [error] [967680] unknown query type: 0
[commands] [error] [967680] recv query failed: command_id(2233) not found into queries
[commands] [error] [967680] time out of id=1480, command_id=1480 start_time=Tue Dec 23 10:28:55 2025: , end_time=Tue Dec 23 10:29:55 2025: , exit_code=3, exit_status=timeout, output='(Process Timeout)'

After a downgrade to 24.10.13 the problem has gone away again. I would like to add that other pollers with the 24.10.13 engine on the same system never had this problem.


 

The issue occurs only with services running AWS, MongoDB, and Kubernetes plugins hosted across different cloud environments, using the Perl connector. All other services operate without any problems.

• After restarting the centengine service, checks start working correctly.
• Over time, services begin to accumulate in (Process Timeout) state, reaching approximately 570 affected services.
• When commands are executed manually from the shell, they always work without errors.
• Cloud connections are fast, and the configured timeout values are high (300 seconds).

System Resources:

• 12 vCPU
• 32 GB RAM
• 1 central poller
• 755 hosts
• 7,100 services

Seems like the problem is with the perl connector as I’m also using that.

There is an update on the perl-library. Maybe I’ll try to update it to see if it helps in some way.

centreon-perl-libs/bookworm 24.10.16-1+deb12u1 all [upgradable from: 24.10.15-1+deb12u1]

Based on the update I did last time these are what has changed:

 centreon-perl-libs:amd64 (24.10.14-1+deb12u1, 24.10.15-1+deb12u1)
 centreon-connector-perl:amd64 (24.10.13-1+deb12u1, 24.10.14-1+deb12u1)