All Systems Operational

Our API experienced elevated response times across all regions for approximately 18 minutes due to a database connection pool saturation during a scheduled index rebuild. Users in Europe and North America reported intermittent timeouts on document saves. All endpoints recovered automatically after the rebuild completed. No data was lost.

10:42 UTC — 11:00 UTC · Duration: 18 minutes · Affected regions: EU-West, US-East

Outbound email notifications (invoices, receipts, password resets) were delayed by up to 12 minutes for users in the Asia-Pacific region due to a transient issue with our email provider's Singapore relay. All queued emails were delivered successfully. No emails were lost.

14:15 UTC — 14:27 UTC · Duration: 12 minutes · Affected regions: AP-Southeast

A subset of PDF exports (invoices, reports) returned timeout errors for approximately 25 minutes affecting users across all regions. Root cause was a memory limit on the PDF rendering service during peak business hours in the Middle East and Africa. The limit was increased and the service was restarted. All failed exports were automatically retried.

09:03 UTC — 09:28 UTC · Duration: 25 minutes · Affected regions: All

Planned maintenance window for database schema migration across all regions. The platform was in read-only mode for 8 minutes while the migration ran. Scheduled during low-traffic hours (02:00 UTC) to minimize impact. All write operations queued during this window were processed immediately after completion.

02:00 UTC — 02:08 UTC · Duration: 8 minutes · Affected regions: All (planned)

The main dashboard experienced slow load times (5–8 seconds vs normal <1 second) for approximately 35 minutes, primarily affecting users in North America and Europe during business hours. Root cause was a spike in concurrent report generation jobs saturating the read replica. Job concurrency limits were tightened to prevent recurrence.

16:10 UTC — 16:45 UTC · Duration: 35 minutes · Affected regions: US-East, EU-West