BentoML insights

Open SourceVersions: []22 metrics

OpenTelemetry·

Metrics endpoint recursion prevented by skipping self-instrumentation

24d ago▸

Service readiness determined by startup lifecycle completion

24d ago▸

Custom command health check polling timeoutwarning

24d ago▸

Histogram bucket configuration impacts cardinality and accuracy

bentoml.api_server.request.duration bentoml.runner.request.duration

24d ago▸

Tracing disabled when sample rate is zero

24d ago▸

gRPC trace context extraction from metadatainfo

24d ago▸

OpenTelemetry resource attributes derived from server contextinfo

24d ago▸

Layered configuration loading order affects final valueswarning

24d ago▸

Thread pool exhaustion prevents synchronous API method executioncritical

bentoml.api_server.request.in_progress http.server.duration

24d ago▸

Request timeout configuration prevents long-running inferencewarning

http.server.duration

24d ago▸

CORS preflight failures block cross-origin API accesswarning

24d ago▸

MaxConcurrencyMiddleware returns 503 under loadwarning

bentoml.api_server.request.in_progress bentoml.api_server.request.total

24d ago▸

Generic exception details leak in development modeinfo

24d ago▸

Circus socket file descriptor sharing enables zero-downtime reloadinfo

24d ago▸

Health check endpoints bypass middleware to avoid metric pollutioninfo

24d ago▸

Adaptive batching queue depth causes latency spikeswarning

bentoml.runner.adaptive_batch.wait_duration bentoml.runner.adaptive_batch.size

24d ago▸

Generation stalls cause multi-second inter-token delays degrading user experiencecritical

bentoml.runner.processing_latency

2mo ago▸

Aggregated request-level metrics mask micro-stalls in token generationwarning

bentoml.api_server.request.duration bentoml.runner.processing_latency

2mo ago▸

Static latency thresholds fail under variable request length causing false positiveswarning

bentoml.runner.processing_latency

2mo ago▸

Minute-level monitoring lag loses transient anomaly context before diagnosiswarning

2mo ago▸

Prefill stage latency varies wildly with KV-cache layout making baseline modeling noisy

2mo ago▸

Python-level logs and GPU hardware metrics operate on divergent timebases causing misalignmentwarning

2mo ago▸

Observer effect from indiscriminate tracing competes for resources masking true bottleneckswarning

bentoml.system.cpu.usage

2mo ago▸

BentoML 1.4.31+ containerization fails with uv directory resolution error

3mo ago▸

BentoML containerize fails with NotImplementedError due to Click 8.3.0 incompatibilitycritical

6mo ago▸

Duplicate --quiet parameter warning in BentoML CLI with Click 8.3.0warning

6mo ago▸

Long cold start times delay development iterationwarning

1y ago▸

GPU over-provisioning drives up infrastructure costswarning

bentoml.system.cpu.usage

1y ago▸

Manual infrastructure setup delays AI model deploymentwarning

1y ago▸

Locked runtime versions prevent using newer AI frameworkswarning

1y ago▸