Severity Levels
Understanding degraded and outage severity levels for status page issues.
Issues drive component status on your status page. Each issue has a severity level that determines how the component's health is displayed. Understanding severity levels helps you interpret your status page at a glance.
Severity levels
Sazabi uses two severity levels for status page issues:
Degraded
Degraded indicates partial issues that affect service quality but do not represent complete failure. Examples include:
- Elevated error rates (5-20% of requests failing)
- Increased latency (response times above normal thresholds)
- Reduced capacity (some instances unhealthy)
- Intermittent failures (sporadic errors)
When a component has a degraded issue:
- The status page shows the component with a yellow indicator
- The component is functional but not performing optimally
- Users may experience slower performance or occasional errors
Outage
Outage indicates complete failure where the service is unavailable or unusable. Examples include:
- 100% error rate (all requests failing)
- Service unreachable (connection timeouts)
- Critical dependency down (database unavailable)
- Complete data loss or corruption
When a component has an outage issue:
- The status page shows the component with a red indicator
- The component is considered non-functional
- Users cannot use the affected feature or service
Status determination
A component's displayed status is determined by the worst active severity among all its open issues.
Worst severity wins
If a component has multiple active issues:
| Active Issues | Displayed Status |
|---|---|
| None | Operational |
| Degraded only | Degraded |
| Outage only | Outage |
| Degraded + Outage | Outage |
The outage takes precedence because it represents the more severe condition. Even if some aspects of the service are merely degraded, an outage in any area means the overall component status is "Outage."
Operational status
A component shows Operational status when:
- No active issues exist for the component
- All previous issues have been resolved
Operational is the default state and indicates the service is functioning normally with no known problems.
How severity is assigned
When Chat creates an issue, it determines severity based on the evidence from its investigation:
- Degraded: Partial failures, elevated errors, performance problems
- Outage: Complete failures, service unavailable, critical errors
You can also manually specify severity when creating issues through alerts. See Alerts & Status Pages for details on severity mapping.