Issue with API

This incident is resolved. Between approximately 06:03 and 06:13 UTC on 20 October, most requests to the Qonversion API returned errors and the dashboard was briefly unavailable, due to an infrastructure availability event with our cloud provider. From 06:13 UTC the API was served through a fallback path: requests were accepted, while remote configuration delivery was degraded and purchase and user-data updates were applied with a delay until the backlog was processed. Full service on the primary path has been restored since 09:43 UTC, and we have confirmed no recurrence since. We're sorry for the disruption.

Recovery is complete. API traffic has been returned to the primary path and all services are operating normally. The backlog of requests received during the incident has been fully processed; purchases, user data and remote configuration updates from that period have been applied. We will continue to monitor closely and post a final update within 60 minutes.

Update 07:50 UTC — The API continues to operate on a fallback path and is responding normally.
We have started processing the backlog of requests received during the incident; purchase and user-data updates from that period are being applied progressively and may appear with a delay. Remote configuration delivery remains degraded until the recovery completes. The original impact window was approximately 06:03–06:13 UTC, when most API requests returned errors. Next update within 60 minutes.

Between approximately 06:03 and 06:13 UTC, most requests to the Qonversion API returned errors, and the dashboard was briefly unavailable. We have identified the cause as an infrastructure availability event with our cloud provider and have switched API traffic to a fallback path. The API is responding again; some functionality, including remote configuration delivery, remains degraded and some purchase events may be processed with a delay. We are working on full recovery and will post the next update within 30 minutes.

We're experiencing an elevated level of API errors and are currently looking into the issue.