Nagios Support Forum

Hello all! I installed nagios and check_nrpe to monitor the hosts and services associated with them in my cluster with 3 nodes. I have considered one of them as the monitoring server. Now, I want to monitor two of the services in one remotehost once in 2hrs and keep all others have the default check interval. I referred the documentation and found that I can change the pooling interval (normal_check_interval) to attain this. In the templates.cfg I can see that service named local-service sets the normal_check_interval to 5. With this, I can make the associated plugin execute once in 2hrs by adding a line normal_check_interval=2h in the corresponding service definition (Please correct me if I am wrong).
But with this, I can just make the service be checked once in 2hrs but cannot reduce the size of status.dat file. Even if the service is not being checked every 5 mins, the status.dat file always contains the entry for that service structure with old 'status information', 'perfdata' and other values. My intention is to reduce the unnecessary data transfer between the remote host and the server. But here, I think the old data is transfered for 2hrs every 5 mins. How to avoid this?

Your remote host has a status.dat file? How large is your current status.dat?

Yes. My remotehosts also have status.dat files. Status.dat file is not very large. It contains totally 9 services.
And not all these services in remotehost status.dat files get copied onto the server status.dat file. Only the services specified in NRPE (the services executed on the remotehost by the server nrpe) are documented in the server status.dat file.
So what I doubt is that, all the services (remote host services) specified in nrpe get checked everytime nrpe is executed by the server (even though the normal_check_interval is specified for those services in the remote hosts).

Just want to clear up one thing, you have 3 nagios installations checking each other? Just keep in mind it is not required of you to install Nagios on systems you intend to monitor, you can have one installation running checks to the others.. though this is how you may want your network set up. In that case, the status.dat file will most likely not look the same on each system, it is a pooling of status information for checks being sent and returned from that specific Nagios server. In the Nagios world the term "remote host" is generally used when talking about a remote server being monitored by the Nagios server, not an additional Nagios server.. so that may be causing a bit of confusion.