Windows services state

[Mommo333]

Hello,

I asked the NSClient++ team for help about a "CheckServiceState" command but they sayd that the value came from Nagios and not from the NSCLient++ agent that sending the check command.
So my request/question: Actually when a "Automatic" service is stopped, then Nagios alert me that the service named XYZ is critical. If a second service is in the same state it would be added to the first one without setting a new alert. That's my problem. The for me searched notification is that every stopped service is notified again. So if i acknowledge one and a new one came up the first acknowledged notification came up in critical state again.
What is the easiest way to do this?

Thanks in advance for your answer.

Michael

[slansing]

Could you please clarify what you are trying to accomplish? It seems as though you would like to be able to acknowledge a critical service and be warned again if another service goes into a critical state?

[Mommo333]

Could you please clarify what you are trying to accomplish? It seems as though you would like to be able to acknowledge a critical service and be warned again if another service goes into a critical state?

That's right, that's what i search.
Actually while the acknowledged service a new Windows service is stopped then the acknowledged service doesn't change of state or wouldn't disacknowledge automatically.

[Mommo333]

slansling, do you have an idea ?

thanks in advance

[jsmurphy]

Ok so just to clarify what you want:

You have an Automatic service check, a problem happens with 1 service and you acknowledge it. This consequently stops you from being notified about any subsequent service failure. So what you want is a way for that service to become unacknowledged when the description changes?

Unfortunately there is no native way to do this in Nagios, the only way you could make this happen is if you wrote an 'event handler' script that compared the current state to the last received state and then use the Nagios command pipe to unacknowledge the service.

To be honest this feels like more of a process problem... are you excluding services that are automatic but don't run all the time? Do failed services really occur that often and sit around in your queue that long?

[Mommo333]

Hello jsmurphy,

You're right. That's exactly what i search!

I'm excluding some specific services yes, it's a customer request that we set the windows services check as you wrote. I totally agree with you that if a service is stopped the customer must check and resolve the problem...
I asked to know if something like this exist or not.

Thanks for your answer, you helped me!

[jsmurphy]

Glad I could help