Summary
On November 20, 2026, between 13:18 and 13:33 UTC, the Adapty Dashboard and several of our APIs returned errors for about 15 minutes. The Dashboard was fully unavailable for about 2 minutes (13:18–13:21 UTC). The errors happened while we applied an urgent security update to the infrastructure that routes traffic to our services. The update addressed a recently disclosed vulnerability in a widely used web server. The SDK API kept serving profile, paywall and purchase requests.
Impact
-
Dashboard: unavailable 13:18–13:21 UTC, with intermittent errors until about 13:33 UTC.
-
Ads Manager and some other non-SDK APIs: intermittent errors between 13:18 and 13:33 UTC.
-
SDK API: almost no errors. Some apps may have seen brief request delays around 13:16–13:18 UTC; the SDK retried those requests automatically.
- Purchases, renewals, profiles and paywall delivery were not affected.
Timeline (UTC)
-
13:12 — Security update to the traffic-routing layer begins.
-
13:16 — Traffic-routing components start restarting with the patched version.
-
13:18 — Dashboard and several APIs begin returning errors; our monitoring alerts the on-call engineer.
-
13:21 — Dashboard available again.
-
13:33 — Errors stop on all affected APIs.
-
14:55 — Update fully completed across our infrastructure.
Root cause and contributing factors
The security update replaced the components that route incoming traffic to our services. Connections that were open while a component restarted were dropped instead of finishing, and the affected requests failed. Contributing factors:
- The update was applied quickly because of the severity of the vulnerability, so it could not wait for our regular maintenance window.
- Traffic-routing components were replaced without first draining open connections.
What went well
- The vulnerability was mitigated as soon as it was disclosed, before the update was applied.
- The SDK API served traffic throughout.
- External monitoring detected the Dashboard errors within seconds.
What we will do
-
Prevent: drain open connections before traffic-routing components restart, and update them in smaller groups.
-
Detect: alert on error rates for each public API, not only the Dashboard and the SDK.
-
Mitigate: announce urgent infrastructure updates on this status page before we apply them.
We apologize for the disruption. If you noticed anything unexpected in your data for this window, please contact your Adapty account manager and mention the 2026-11-20 incident.