There have been a couple instances where the VM hosting centreon has gone offline either due to glitch with esxi or by some other outside interference. When this happens centreon begins to register all the hosts as being down since it is failing to communicate with them. When the server comes back online the email alerts to notify us these hosts are down (which in this case they weren't) have been stuck in the queue and begin to process now that connectivity is restored. At the same time new alerts are being generated and sent out to notify us that the hosts have recovered... I have some ideas on how to potentially script a fail-safe when something like this happens but I'm curious if there is some built in options within centreon that address this problem that I'm missing? Or if someone has already developed their own solution to this problem that they'd be willing to share?