System Statuses and Outages
Effective scheduling and management of system outages are essential for meeting your CDR compliance obligations. Publishing these outages ensures ADRs are informed of any Data Sharing disruptions and both enables them to support their customers and avoid calling data from you during these periods.
The HaaS Manager console can be used to log and manage planned outages, which ensures that the planned downtime is accurately recorded and excluded from availability metrics. Proper outage management helps maintain transparency, minimises disruption and also supports accurate and compliant reporting aligned with regulatory expectations.
It is your responsibility to use HaaS Manager to schedule and manage system outages for your HaaS instance. Instructions for doing so can be found here. Please be aware that unscheduled outages affect your availability metric published in the getMetrics CDR endpoint.
Unscheduled Outagesβ
Unscheduled outages occur when your HaaS Instance becomes unavailable without prior notice. These outages are included in your availability metrics. This logic ensures unexpected downtime is accurately reflected in the figures reported to the ACCC. Unlike planned outages, unplanned outages canβt be excluded from reporting, so timely resolution and investigation are essential to maintaining compliance and overall system performance.
Metrics Calculationβ
The following formula is used to calculate the availability metrics where unplanned outages must be taken into account. It represents the percentage of time a HaaS instance is was available during a given period, excluding any unplanned outages.
availability = (totalSecondsInPeriod - totalUnplannedOutageSecondsInPeriod) / totalSecondsInPeriod
Period Definitionsβ
The period used in the calculation depends on the time range requested:
- Current Month: From UTC midnight on the 1st day of the current month to the time the ACCC calls the metrics endpoint.
- example: If the ACCC requests metrics at
2025-05-19T20:01:07 UTC, then the current month period is2025-05-01T00:00:00 UTC to 2025-05-19T20:01:07 UTC.
- example: If the ACCC requests metrics at
- Previous Month: Each full month is defined from UTC midnight on the 1st day of the month to UTC midnight on the 1st day of the next month.
- example: For April 2025, the period is 2025-04-01T00:00:00 UTC ~ 2025-05-01T00:00:00 UTC.
If an unplanned outage occurred is April and lasted for 1 hour and 15 minutes (4,500 seconds), the availability for that month would be calculated as:
(30 * 24 * 3600 - 75 * 60) / (30 * 24 * 3600) = 99.826388888%
Please note that the same values will be reported for aggregate, authenticated and unauthenticated calls.