Nagios Report has little data

[goldmund84]

Hello,

At the beginning of the month, I'm trying to create the Availability Report, as I usually do for the Last Month. But the data is showing me almost 100% Availability (for both Hosts and Services), that is very unusual. Seems that the data is partly unavailable or lost. Please, advice how I can check this.

Thanks.

[npolovenko]

Hello, @goldmund84. Can you try to generate other kinds of reports to see if any data is missing? For example, you may run a state history report for the previous month and see if the entries look right. Also, you may double check that you didn't have any wrong filters enabled when you ran the report for the first time.

[goldmund84]

Hi,
Other reports are generated normally (checked State history, Event Log, Notifications - all have a lot of data for the Last Month period). Alternatively, I've set manually Custom time period from Dec 1 to Dec 31. But the result is the same.

I have just one filter - Hostgroup. When I change the time period from Last Month to This Month, the report is Ok (the Availability value correlates with the Average I see each month). So, it is an issue with the older data.

I have a service that was down for several months, so in the reports it was always stated 100% Unavailability. But for the Last Month, I see, that this service was 96% Available. I conclude, that 96% of all the Availability data is lost/unavailable.

[npolovenko]

@goldmund84 , Please try to rerun the report with the following options:

Code: Select all

Assume Initial States: No
Assume State Retention: No
Assume States During Program Downtime: No
First Assumed Host State: Unspecified
First Assumed Service State: Unspecified

Does the report look more "normal" now?

[goldmund84]

Hi
It seems that the option "First Assumed Service State" caused the problem.
But why it caused the issue now? I've never changed that option.

[npolovenko]

@goldmund84, Perhaps the host that has 100% unavailability was already in the hard state when the month started, and probably it was in a downtime. By default, Nagios won't assume critical state for hosts in downtime when writing a report.