Realtime latency

Timeline

1. 22:59 UTC Supabase status page · Reported Realtime latency
2. 22:59 UTC Supabase status page · Investigating We are experiencing increased latency with Realtime and are actively investigating.
3. 00:02 UTC Supabase status page · Monitoring We have identified the cause and taken actions to remediate the high latency issues. We will continue to monitor.
4. 06:38 UTC Supabase status page · Monitoring We have taken preventative steps to expand our realtime cluster capacity and are continuing to monitor our system stability for full recovery.
5. 13:49 UTC Supabase status page · Identified We’re seeing increased Realtime latency and instability affecting the presence feature, which may impact join rates. This recurrence comes as we continue work on the remaining preventative measures. We are actively investigating and taking steps to mitigate the impact. We’ll share another update as soon as we have more information.
6. 17:34 UTC Supabase status page · Identified We are continuing to investigate increased Realtime latency and instability affecting the presence feature, which may impact join rates. Our team is actively working on mitigation and we will share another update as soon as more information becomes available.
7. 21:15 UTC Supabase status page · Identified We’ve added extra capacity to the system and are currently testing it. We expect to begin stabilizing the affected clusters shortly. We’ll continue to share further updates as more information becomes available.
8. 23:27 UTC Supabase status page · Identified We’re seeing recovery for a large percentage of affected customers, though some issues remain in certain regions. We’re continuing to work on resolving the remaining issues and will share further updates as more information becomes available.
9. 03:00 UTC Supabase status page · Monitoring We have stabilised impacted users and enabled improvements to our Realtime infrastructure. We will continue to monitor for full recovery.
10. 20:25 UTC Supabase status page · Resolved Realtime has fully recovered.