There is some "bug" in events processing by Nagios.
New event results with calling system php process.
Ins some environments it results in massive process calls, thousand of process are are called on OS level - and it kills system.
Above behaviour is possible when Nagios checks passively thousand of devices and check freshness process is invoked within Nagios.
The enough solution is to queuing events internally (in database) before invoking OS to process events.
Queuing of event processing
Re: Queuing of event processing
First, I am moving this out of Nagios Ideas since this is an issue with Core, not a suggestion.
Second, can you provide some more detail? What Core version is this for? What do you mean by "New event results with calling system php process" exactly?
Second, can you provide some more detail? What Core version is this for? What do you mean by "New event results with calling system php process" exactly?
Former Nagios employee