Monitoring & Alerts Overview

The monitoring system in frugally.app tracks signals across your AWS environment, establishes baselines for normal behaviour, detects anomalies, and raises alerts when something needs attention.

---

How monitoring differs from notifications

Concept	Purpose
Monitoring	Detection — continuously analyses your environment to identify issues
Notifications	Delivery — routes detected issues to the right people via the right channels

Monitoring produces alerts; notifications deliver them. You configure what to detect here and how to deliver in the Notifications section.

---

Key concepts

Concept	Definition
Signal	A metric tracked over time — cost, resource usage, resource count, etc.
Baseline	The expected normal range for a signal, established from historical data
Anomaly	A signal value that deviates significantly from its baseline
Alert	A notification triggered when an anomaly or threshold condition is met

---

What is monitored

Signal category	Examples
Cost	Daily spend, per-service spend, per-account spend
Usage	EC2 CPU utilisation, RDS connections, ECS task count
Resource count	Number of running instances, active services, provisioned endpoints
Execution health	Execution success rate, failure frequency, partial failure rate

---

In this section

• Signals and Baselines — What is tracked and how baselines are established
• Anomaly Detection — How anomalies are identified and configured
• Alert Management — Creating alert rules, severity levels, and managing alert lifecycle