Skip to:

  1. Skip to Jira Navigation
  2. Skip to Side Navigation
  3. Skip to Main Content

Outages ignoring incoming Scheduled Outages

Description

During a scheduled outage an existing outage will not be checked so the end of the outage cannot be detected during the scheduled outage and/or the duration is calculated wrong.

Example:

5:00 Outage begins
6:00 Scheduled Outage begins
7:00 Outage ends (service will NOT be polled so the end of outage can NOT be detected)
8:00 Scheduled Outage ends (service will be polled again and end of outage is detected)

This results into an outage with a duration of 3 hours instead of an outage of 1 hour (or at least 2 hours). The right answer should be 1 hour in my opinion.

Another example:

5:00 Outage begins
6:00 Scheduled Outage begins
7:00 Scheduled Outage ends
8:00 Outage ends

This results into an outage with a duration of 3 hours instead of 2 hours (1 hour scheduled).

Environment

Debian, Windows 2008, Windows 7

Acceptance / Success Criteria

None

Lucidchart Diagrams

Activity

Loopo June 21, 2012 at 5:46 AM

Well after rethinking the two solutions, the "simple" one would be the preferred one maybe. Because polling during a scheduled outage may not be wished.

Loopo February 7, 2012 at 10:54 AM

A simple solution would be, that an outage ends, if a scheduled outage starts. If the scheduled outage ends and the outage still exists, a new outage raises.

Maybe a better solution would be a hibernate state, if a scheduled outage starts. If the outages ends during a scheduled outage, it resolves backdated at the beginning of the scheduled outage. If it resolves after one (or more) scheduled outages, the duration is calculated by taking the scheduled outages into account. This duration could be stored into a new database field in the outages table.

Details

Assignee

Reporter

Labels

Components

Affects versions

Priority

PagerDuty

Created February 7, 2012 at 10:27 AM
Updated September 21, 2021 at 6:23 PM

Flag notifications