Freshness check chg my service checks fr passive to active?

[charmainequek]

hi all, i am pretty new to Nagios XI, hope to get some help and advice on this issue i am facing..
Currently we have a software that sends check results to Nagios XI using the NRDP agent running in Nagios. These are passive checks. However i noticed after some servers having some down time, the results showing in Nagios XI becomes "UNKNOWN" for the longest time and does not recover.
Whan i checked the results from a nrdp dump, the services are seen to be ok and running. I looked into the services' advanced configuration and notice they have changed from "passive, soft" to "active, hard" checks. Would like to have some advise on how to reverse the situation and change the service back to "passive, soft"?
Thanks.

[cdienger]

This sounds very odd. What version of XI is this?

Does the configuration change as well if you edit the service and look at it's "Check Settings" tab? The running config is stored in /usr/local/nagios/var/status.dat(or /var/nagiosramdisk if you have a ramdisk) - do you see the changes to the service in there as well ?

My initial thought is that someone must be modifying it - is there anyone with access to the machine besides you? The audit log can be checked under Admin > System Information > Audit Log. Modifing and applying changes to a service would generate a few lines like:

Service modified: 192.168.0.31
Config file deleted: 192.168.0.31.cfg
User applied a new monitoring configuration

[charmainequek]

hi cdienger,
thank you for your reply.
actually i am not sure if freshness checks changed it. it happened after these servers went through a maintenance window where they were shut down. i read from the guides that freshness checks (active checks) by default will kick in if the last check hits the freshness threshold and is considered stale by Nagios.

After the maintenance period, the service status seen in Nagios XI is "UNKNOWN: No results received"
Nagios XI version : 5.4.12

I checked the settings in services, advance options and compared with the servers that are working and saw only the "active" "passive "hard" "soft" as the only difference.
the checks are set up by the software thus in settings i do not see much configurations. I will look and see if i can find more info.

Please help.. thanks..

[cdienger]

Thanks for clarifying. The passive check does run an active check if the freshness threshold is exceeded and this will change the check type as seen on the Advanced tab. The service will still accept passive checks though and display 'passive' again on the Advanced tab when passive results come in. You can get it to switch back by using the "submit passive check result' link found on the Advanced tab.

[charmainequek]

hi cdienger,

i did as per your advice by clicking "submit passive check results" and entered a random string such as "test" in the "check output" field. it did return the check to "passive" but the status remains as "test" and does not change anymore. now seems like scheduled passive checks are broken?

please help.. thanks..

[cdienger]

It sounds like the checks may not be making it to the XI server. What nrdp agent are you using to send data to XI ? Can you confirm that that agent/machine can access http(s)://nagiosxi/nrdp ? You can change the hostname and servicename info on this screen to match your service's info and submit it to XI as a test. Activity should be seen in /usr/local/nagios/var/nagios.log when you do this as well.

[charmainequek]

hi cdienger,

we are using the NRDP agent together with NRPE and NSClient++. I can access http://nagiosxi/nrdp from the NagiosXI server and i was able to send a command to disable notifications on a host from there. So this will mean passive checks are working fine?
some queries :
1. Can there be more than 1 type of agents in the environment?
2. Can there be a mixture of passive and active checks in the environment?

Appreciate your advise.

[cdienger]

Yes, it sounds like it is working as expected on the XI side and you can have a mixture of agents, passive, and active checks.

[charmainequek]

hi cdienger,

is there anything that needs to be installed or configured on the host machines for NRDP passive monitoring? i realized that after maintenance (where servers are being power cycled) the information transfer from the hosts to Nagios XI are dropped / stuck / not transferred somehow. The last check time shown in Nagios XI also stops at the time and status of the last check before the servers were shut down. may i know how to check or troubleshoot the chain of Nagios XI accepting information and status from NRDP?

Thanks.

[cdienger]

What agent are you using on the clients to send the NRDP traffic? I'd start by reviewing the configuration on the agent. And is the agent starting automatically when the server it is on is rebooted?