How was your monitoring organized? Prometheus, Grafana, logs?

Monitoring in projects is usually organized comprehensively using several tools:

• Prometheus — collects metrics from applications and infrastructure, stores them, and allows query building for analysis.
• Grafana — visualizes metrics from Prometheus and other sources, creating dashboards for convenient system monitoring.
• Logs — collected using systems like ELK (Elasticsearch, Logstash, Kibana) or Loki, enabling event and error analysis.

For example, in one project, we configured metric export from a Python application via the Prometheus client library, collected them in a Prometheus server, and created dashboards in Grafana with key metrics (latency, errors, load). Application logs were sent to Elasticsearch, where we could quickly search and analyze errors.

This approach allows for prompt problem detection and performance analysis.