Home› Docs› Monitoring & Alerting

Monitoring & Alerting

Built-in metrics

Every deployed service automatically emits the following metrics: request rate, error rate, p50/p95/p99 latency, CPU, memory, and network I/O.

Custom metrics

import { metrics } from '@nexatech/monitor';

metrics.increment('orders.created', 1, { region: 'us-east', plan: 'pro' });
metrics.histogram('order.value_usd', 49.99);
metrics.gauge('queue.depth', queue.length);

Alert policies

Alerts consist of a condition, evaluation window, and notification channel. Example:

alert: high-error-rate
condition: error_rate > 1%
window: 5m
notify:
  - pagerduty: P-XXXXXXX
  - slack: "#incidents"
runbook: https://runbooks.internal/high-error-rate

Dashboards

Dashboards are defined as code (JSON or YAML) and version-controlled alongside your application. Import our starter dashboard templates from the gallery.