Fluxer outage on October 29

Issues with platform connectivity

1. Resolved · 1:24 AM

   On 30 October at 00:36 UTC, a short network blip in our cluster set off a mass reconnect. The presence service stalled under the load, so heartbeats timed out and clients kept reconnecting. We patched the gateway live and rebuilt presence data. Service is fully restored.

2. Resolved · 1:03 AM

   This incident has been resolved.

3. Identified · 12:53 AM

   We're still working on stabilising everything after a brief network outage at our hosting provider during our peak hours requires manual intervention to restore connections to all users without saturating our backend resources. Hang tight!

4. Identified · 12:46 AM

   At 00:36 UTC a short network disruption inside our server network cut connections between our web servers and our backend for about 3 minutes. We're still working on recovering all stuck connections. Bear with us!

5. Investigating · 12:38 AM

   We are currently investigating this incident.