Documentation Grafana Cloud Knowledge graph Advanced configuration Enable metrics collection Application frameworks Flask

Configure Flask to generate Prometheus metrics

Flask is a python module that you can enable for Prometheus metrics using Flask Exporter for Prometheus.

To configure Flask to generate Prometheus metrics, perform the following steps:

1. Add the following export code in the Python module.

   Copy

   from prometheus_flask_exporter import PrometheusMetrics

2. To enable custom metrics, follow the steps provided in prometheus_flask_exporter.

3. Verify the presence of the following metrics in Prometheus:

   • flask_http_request_duration_seconds
   • flask_http_request_total

Metrics

histogram_quantile (0.99,sum(rate(flask_http_request_duration_seconds_bucket[1m])) by (le))

Alerts

Service KPI dashboard

This dashboard has the following KPIs that include resource and requests metrics:

• Request Rate
• Latency Average
• Latency P99
• Error Rate
• CPU %
• CPU Cores Used
• CPU Throttle
• Memory %
• Memory Bytes
• Disk Usage
• Network Usage

You can include any custom dashboard from the community. For example, this dashboard shows Flask specific metrics directly as a custom dashboard for your service.