Outage
New connections failed, old ones worked: the failover that sent writes to a read-only database
DEV · By Kashif Manzer ·
The Brief
- On 28 October 2026 the RIPE Database began failing some updates while others succeeded.
- Failed requests were refused, and the failure was not visible from outside the service.
- The cause as described is a failover that sent writes to a read-only database.
Why it matters
Teams relying on database writes during failovers can see intermittent, hard-to-predict failures when some connections are routed to a read-only node while others are not.
Who’s affected
- RIPE Database users submitting updates
- #failover
- #database
- #read-only
- #incident
- #ripe
On Wednesday 28 October 2026, the RIPE Database started failing updates. Not all of them. Some went through fine. Others were refused. The same service, the same hour, the same kind of request, and nobody could tell from the outside which ones would fail. That is the worst kind…