Demo
How it works
Incidents represent production issues caused by a deployment. They track the full lifecycle from detection to resolution, linking back to the deployment that caused the issue and the deployment that fixed it.Key fields
| Field | Description |
|---|---|
| Cause deployment | The deployment that introduced the issue (required). |
| Fix deployment | The deployment that resolved the issue. |
| Detected at | When the incident was detected. |
| Resolved at | When the incident was resolved. |
| Team | The team responsible for the incident. |
| Leader | The person leading incident response. |
| Postmortem URL | Link to the postmortem document. |
Creating incidents
Incidents can be created in two ways:- Manually: Create an incident from the UI, linking it to the deployment that caused it.
- Automatically: Use the Incident Detection automation to detect incidents from rollbacks, hotfixes, and reverts.
Impact on metrics
Incidents directly feed into two DORA metrics:- Change Failure Rate: The percentage of deployments that caused an incident.
- Mean Time to Recovery: The average time between incident detection and resolution.