Traces
Service Error Ratio
Updated Nov 3, 2026 · 1d ago
Watches the share of each service's own requests that fail, and tells you when that share rises well above its normal level.
Example alert
Exact wording varies, because the detector generates titles from the anomaly it finds. This is representative of what an alert looks like when it fires.
How it works
Counts failed server spans per service and per main operation, using the span error status, so a service that returns HTTP 200 on a failed call is still counted. It compares each 5-minute window over the last 30 minutes with a budget learned from that service's clean hours, and needs at least 5 failed requests and a 2% failure share before it fires. Late spans still count, because long failed calls can arrive many minutes after they start. It sends a warning and never pages on its own.
Availability
Runs on these tiers:
Want to see this detector firing in the live demo?
Open alerts in the sandbox →Related detectors
- Trace Latency Break
Catches a service whose request latency regressed against its own past, with the whole distribution shifted slower.
- Trace Tail Shift
Catches partial degradation: most requests are fine but a slice falls off a cliff. The tail blows up while the median holds.
- Trace Error Origin
Walks the span graph to find the origin service of an error cascade: where the failure started, rather than the services that merely reported it.
- Trace Throughput Drop
Catches a service whose request volume collapsed (crashed, deadlocked, or off the load balancer) when no slow or error spans give it away.
- Trace New Operation
Flags a service serving an operation it never served before: an unannounced deploy, a new endpoint, or a client calling a new path.