Metrics
Prometheus
To enable the Prometheus /metrics endpoint, set the PROMETHEUS_ENABLED environment variable to true.
When enabled, Flagsmith serves the /metrics endpoint on a standalone HTTP server on port 9100, separate from the main API server. This design ensures that metrics collection remains available even when the main API is under heavy load.
The metrics provided by Flagsmith are described below.
flagsmith_build_info
Gauge.
Flagsmith version and build information.
Labels:
ci_commit_shaversion
flagsmith_environment_document_cache_queries
Counter.
Results of cache retrieval for environment document. result label is either hit or miss.
Labels:
result
flagsmith_http_server_request_duration_seconds
Histogram.
HTTP request duration in seconds.
Labels:
routemethodresponse_status
flagsmith_http_server_requests
Counter.
Total number of HTTP requests.
Labels:
routemethodresponse_status
flagsmith_http_server_response_size_bytes
Histogram.
HTTP response size in bytes.
Labels:
routemethodresponse_status
flagsmith_task_processor_enqueued_tasks
Counter.
Total number of enqueued tasks.
Labels:
task_identifier
flagsmith_task_processor_finished_tasks
Counter.
Total number of finished tasks. Only collected by Task Processor. task_type label is either recurring or standard.
Labels:
task_identifiertask_typeresult
flagsmith_task_processor_task_duration_seconds
Histogram.
Task processor task duration in seconds. Only collected by Task Processor. task_type label is either recurring or standard.
Labels:
task_identifiertask_typeresult