| Configuration & Lifecycle | - Open to which package manager used
- Multiple interfaces
| - Installable Helm v3 chart
- Rolling Upgrades
- Has liveness probes
- Has readiness probes
- No naked pods
- Valid Helm Chart
- No NodePort usage
|
| Observability/Monitoring | - Not specified yet (looks like it will be part of Chapter 3.2)
| - Uses Fluentd for logging
- Uses Jaegar for tracing
- Uses Prometheus for time series metrics
|
| Installable/Upgradeable | - A package manager must be used, but not specific as to which
| - Supports rolling upgrade update/downgrade
- Stored in a Git repository
- Published to public repository
|
| Statelessness | - Persistent storage availability is optional
| - Continues to function during chaos events (i.e. Litmus)
- Data not lost during restart
|
| Microservice |
| - Max number of processes in a container
- Container image size
- Startup time
|
| Scalability |
| - Can autoscale
- Can scale manually
|
| Security | - No hard coded passwords
- Must use TLS for network access
- CIS benchmark
- Vulnerability scans
- Signed images
| - No container using privileged mode
- No shells
|
