bentoml.api_server.request.total
Total API server requestsDimensions:None
Available on:
OpenTelemetry (1)
Interface Metrics (1)
Technical Annotations (4)
Configuration Parameters (1)
traffic.max_concurrencyrecommended: set based on worker resource capacityError Signatures (1)
503http statusTechnical References (2)
output data driftconceptpredicted distributionsconceptRelated Insights (3)
Output data drift detected through prediction distribution changeswarning
▸
Uneven request distribution across workers causes load imbalancewarning
▸
MaxConcurrencyMiddleware returns 503 under loadwarning
▸