Hosts and services status for faulting poller

Hi @FredericGerard,

Maybe there has been a regression that I am not aware of, but the central broker service already behaves as you described. Once a poller’s link to the central has been cut, after some time (I’m not sure whether it is 10 or 20 minutes) the central broker changes the status of all hosts monitored by the poller to UNREACHABLE and all service to UNKNOWN.

If you do not meet this behaviour, please let me know 🙏

Hi @omercier ,

We’ve just replay the scenario to be sure : shutdown of a poller and check the “resources status” panel. After more than an hour, last status of hosts and services are still displayed :

Ok, I thought you were asking for a change when a poller is lost, but what you describe here seems to be a clean stop of the centengine service. And actually there’s a recent regression with this part. 

Normally when engine stops, it sends a particular message to the central broker and all the hosts and services disappear from the monitoring. Currently, in some cases, the message is not sent and the resources remain and the poller still seems to be running.

We are working on a fix!

Declining this idea since this is more of a bug than an idea and it’s being worked on.

It’s ok for Centreon admins who should be quickly aware of the shutdown or the lost of a poller.

For other Centreon users (with no knowledge of the centreon topology) or even in case of panels displayed on screens, an “unreachable” status would be more explicit and accurate than no status at all which would be more puzzling. Especially if the rest is working properly.

Regards.