Error Tracking with Sentry

Logging vs Error Tracking

Logs tell you what happened. Error tracking tells you what broke, for which users, how often, with the exact stack trace pointing to the line of code. They complement each other.

Sentry Setup

npm install @sentry/node @sentry/profiling-node

// sentry.ts — initialize BEFORE everything else
import * as Sentry from "@sentry/node";
Sentry.init({
  dsn: process.env.SENTRY_DSN,
  environment: process.env.NODE_ENV,
  release: process.env.GIT_SHA,         // Tag errors to specific deploys
  tracesSampleRate: 0.1,
});

// Express: add as FIRST middleware
Sentry.setupExpressErrorHandler(app);

// Add user context after auth middleware
app.use((req, res, next) => {
  if (req.user) Sentry.setUser({ id: req.user.id, email: req.user.email });
  next();
});

// Capture manually with extra context
try {
  await riskyOperation();
} catch (err) {
  Sentry.captureException(err, { extra: { orderId: order.id } });
  res.status(500).json({ error: 'Something went wrong' });
}

Source Maps

# Upload source maps in CI/CD so Sentry shows original code:
- name: Upload Source Maps to Sentry
  env:
    SENTRY_AUTH_TOKEN: FROM_SECRETS
    SENTRY_ORG: your-org
    SENTRY_PROJECT: your-project
  run: |
    npx @sentry/cli releases new GIT_SHA_VAR
    npx @sentry/cli releases files GIT_SHA_VAR upload-sourcemaps ./dist
    npx @sentry/cli releases finalize GIT_SHA_VAR

Alert Rules

In Sentry → Alerts → Create Alert Rule:

• New error type → immediate Slack notification
• Error rate spike (10x normal in 5 min) → PagerDuty
• Specific error > 100/hour → escalation