One of the biggest challenges in developing and host applications in cloud is how to monitor and debugging issues. This post presents the basic concept of observability and AWS’s services.
Challenges
The challenges in application in cloud (especially in the serverless).
- More services and devices
- Independent and quick updates of each component
- Short-lived resources
- Fast-release cycle
Monitoring
Monitoring means watching the layers of your applications and resources in the infrastructure.
- Watch all layers of your applications and resources.
- Identify failures
- Provide metrics
Observability
Observability is a general term, which means the visibility into your system. It is a broader concept than “monitoring”.
- Check if your system works as expected
- Inspect the metrics and gain insight to
- Customer usage
- Trend
- Root cause of the issues
Components
- Logs
- timestamped records of events
- representation of the status: success, failure, or change
- Traces
- the flow of your application
- Metrics
- numeric data measured at intervals
- Service level: error rate, success rate, …
- Business level: the number of orders, the number of downloads, …
AWS Services
You need to be familiar with the following services:
Scriptorium 