System Metrics
This document describes the System Level metrics captured by the DynamoAI API. These metrics provide insights into API health, performance, and reliability at the HTTP level.
Note: These metrics are collected by the opentelemetry-collector deployment and sent to OpenTelemetry, from which they can be exported to any desired backend (Prometheus in the DynamoAI package).
Currently, only DynamoAI's API level metrics are supported.
| Type | Otel Receivers used | Enabled in | Description |
|---|---|---|---|
| API Metrics | otlp receiver | opentelemetry-collector deployment | HTTP level metrics covering the http response time, the error instances that happened, health of the API. API pushes these metrics to the opentelemetry-collector deployment. |
API
| Metric Name | Description |
|---|---|
http_client_request_error_count_total | Total number of client error requests |
http_server_duration_milliseconds_bucket | Cumulative counters for the observation buckets (The duration of the inbound HTTP request) |
http_server_duration_milliseconds_count | Count of events that have been observed for the histogram metric (The duration of the inbound HTTP request) |
http_server_duration_milliseconds_sum | Total sum of all observed values for the histogram metric (The duration of the inbound HTTP request) |
http_server_request_count_requests_total | Total number of HTTP requests |
http_server_request_size_bytes_bucket | Cumulative counters for the observation buckets (Size of incoming bytes) |
http_server_request_size_bytes_count | Count of events that have been observed for the histogram metric (Size of incoming bytes) |
http_server_request_size_bytes_sum | Total sum of all observed values for the histogram metric (Size of incoming bytes) |
http_server_response_count_responses_total | Total number of HTTP responses |
http_server_response_error_count_total | Total number of all response errors |
http_server_response_size_bytes_bucket | Cumulative counters for the observation buckets (Size of outgoing bytes) |
http_server_response_size_bytes_count | Count of events that have been observed for the histogram metric (Size of outgoing bytes) |
http_server_response_size_bytes_sum | Total sum of all observed values for the histogram metric (Size of outgoing bytes) |
http_server_response_success_count_responses_total | Total number of all successful responses |
httpcheck_duration_milliseconds | Measures the duration of the HTTP check. |
httpcheck_status | 1 if the check resulted in status_code matching the status_class, otherwise 0. |