Emergency maintenance: Inference gateway
resolved- Started
- Aug 13, 4:00 AM EDT
- Resolved
- Aug 13, 5:00 AM EDT
- Duration
- 1h
Cerebrium's updates (2)
- update · Aug 13, 4:00 AM EDT
We are performing emergency maintenance on our gateway infrastructure to improve reliability and capacity headroom. During the window, active connections will be dropped and in-flight requests may be terminated. New requests may fail for up to 30 seconds while the gateway cycles. Clients with standard retry logic should recover automatically. What to expect: - Brief connection drops during the window - Possible request failures for up to 30 seconds - No action needed if your integration retries failed requests What we recommend: - Avoid starting long-running requests just before or during the window - Confirm your client retries on connection errors - We'll post updates here as the maintenance progresses and confirm once complete.
- update · Aug 13, 5:00 AM EDT
Maintenance ended
What our checks recorded
We checked Cerebrium 12 times from Cerebrium's start time to its resolution, once every five minutes. Every one of them reached Cerebrium's site, so this incident did not show up as a failed page load. Our check loads the public site; it cannot see an API error or a broken feature. Median response time was 182 ms, against 145 ms over the 24 hours before. Cerebrium posted the incident at Aug 13, 4:00 AM EDT.
Updates are Cerebrium's own words from its status page (status: resolved); we never add an inferred cause.