Summary

  • RIPE NCC reported some requests to the RIPE Database web application failing with 502 responses at 09:46 CEST on 7 October. At 09:47 it said a fix was in place and under monitoring; the incident was marked resolved at 10:21 CEST on 8 October.
  • The 24h34 interval runs from the monitoring update to the resolved notice. It does not show errors persisted for 24 hours, how many requests failed, or whether another query method was available.

A status chronology is not a failure duration

The public record is compact: an initial report, a monitoring update one minute later, and a resolved notice the following morning. “Some requests” is the only scope RIPE NCC gives. The page names no endpoint or operation and identifies no cause. The minute between updates is a difference between posted timestamps, not evidence that RIPE found and repaired the problem in one minute.

From 09:47 CEST on 7 October to 10:21 CEST on 8 October is 24 hours and 34 minutes. That is how long the public incident remained between “monitoring” and “resolved.” It is not a measured outage window: the status page does not state when the first request failed, when the last one failed, or whether errors continued after the fix.

RIPE Database documentation places the web query form within the Database web interface, which also includes Webupdates. The incident notice does not say which requests or functions were affected, so it cannot support a claim that either search or updates failed specifically. RIPE documents RESTful queries separately. That distinction describes the service architecture; it does not prove that REST or command-line access worked during this event.

For operators, the practical question is which path returned which result. A failed browser request is not evidence that a registration record changed, disappeared, or became corrupt. Nor does a status notice about some web requests establish a database-wide outage. The public record reports no data loss or compromise, but it also offers no client-level impact measure.

The disciplined reading is narrow: RIPE NCC reported some 502 failures in the web application, announced a fix under monitoring, and later marked the incident resolved. The current status dashboard lists the Database as operational, but that is a present-state check, not a reconstruction of 7 October. The status page lets users follow the service state; it leaves the affected request window and scale unknown.

Sources