Werk #56: New concept of service periods in availability reporting
| Component | Reporting & availability | ||||||||||||||||||
| Title | New concept of service periods in availability reporting | ||||||||||||||||||
| Date | Nov 7, 2013 | ||||||||||||||||||
| Level | Prominent Change | ||||||||||||||||||
| Class | New Feature | ||||||||||||||||||
| Compatibility | Compatible - no manual interaction needed | ||||||||||||||||||
| Checkmk versions & editions |
|
There is now a pair of new WATO rules called Service period for hosts and Service period for services. With these you can define a timeperiod for a host and/or service within that object is expected to be available in terms of availability reporting. An object that is out of its service period is still being monitored and also notified. But in the reporting there is now a new option box Service Time for handling this new information. You have either:
- Base report only on service times - All phase out of the service period of the object in question is not taken into account when computing percentages. It is still visible in the timeline but grayed out. This enabled you to compute the availability of on object just within a define time frame.
- Base report only on non-service times - This does quite the opposite: It just shows the times where the object is out of its service time.
- Include both service and non-service times - This simply ignores the service time
Please note, that any change to a timeperiod definition always affects future times, never the past. But: it is possible to select an alternative timeperiod for a host and object an rerun the reporting. If that timeperiod aready had existed during the reporting time range then everything works as expected.
Note: The service period make use of new Livestatus columns. You thus need to update all of your monitoring nodes in order to use the availability reporting - even if you do not use this new feature.