Monitoring

Work-in-progress: This content is incomplete; you can help by expanding it

• Ensure you have some cloud-based logging in place that will allow you to debug failures. You must be informed when something fails, be able to figure out why, and take actionable steps to resolve for the user.
• Set up monitoring alerts - your dev team should be receiving some kind of notification (likely email or Slack) if your application has unexpected behavior
  • Example alerts include repeated HTTP request or async job run errors, dramatic spikes in traffic, memory or CPU usage exceeding a threshold
• Set up cost tracking - you should be receiving notifications (likely email or Slack) if your cloud usage exceeds a certain threshold that would result in over-budget issues.