diff --git a/communicate/status-pages/maintenance-windows.mdx b/communicate/status-pages/maintenance-windows.mdx index 7d1c1653..f03b0106 100644 --- a/communicate/status-pages/maintenance-windows.mdx +++ b/communicate/status-pages/maintenance-windows.mdx @@ -69,6 +69,7 @@ With the setting off: - The maintenance card and its detail page list no components, and no component on the page is marked as under maintenance. - Subscriber notifications leave out the affected components. +- Incident downtime during the maintenance lowers uptime as usual. ## Posting updates to your status page @@ -110,4 +111,4 @@ Turn on **Don't open incidents** in the status page visibility settings. While t This option applies only to the components covered by the maintenance on the pages it is shown on. -Maintenance doesn't change your status page **uptime**: incident downtime that overlaps a maintenance counts the same as any other downtime. +Incident downtime on an affected component while the maintenance is active does not lower the component's **uptime**. diff --git a/communicate/status-pages/overview.mdx b/communicate/status-pages/overview.mdx index 258409a8..fdfdce72 100644 --- a/communicate/status-pages/overview.mdx +++ b/communicate/status-pages/overview.mdx @@ -62,6 +62,8 @@ Uptime is calculated over the **last 90 days**, per component, from incident imp Uptime = ((Total time - Weighted downtime) / Total time) × 100 ``` +Downtime during a [maintenance window](/communicate/status-pages/maintenance-windows) that covers the component is not counted. + A group shows the average uptime of its child components. Uptime is driven by incidents, not directly by check results: without an automation rule or a manually created incident, a failing check does not lower uptime. ## Monitoring as code