Incidents API

Use the Incidents API to review, acknowledge, retry, resolve, and reopen operational incidents.

All endpoints require ACCESS_BPM_ADMIN.

Operations

Method	Path	Summary
GET	/incidents	List incidents
GET	/incidents/{id}	Get incident by ID
POST	/incidents/{id}/acknowledge	Acknowledge incident
GET	/incidents/{id}/events	Get incident timeline
POST	/incidents/{id}/reopen	Reopen incident
POST	/incidents/{id}/resolve	Resolve incident
POST	/incidents/{id}/retry	Retry incident
GET	/incidents/process-instances/{processInstanceId}	Get incidents for process instance
GET	/incidents/summary	Get incident summary

GET /incidents

List incidents.

Query parameters:

Name	Type	Description
status	string	Optional. OPEN, ACKNOWLEDGED, or RESOLVED.
source	string	Optional. PROCESS_ENGINE, WORKER, CODE_TASK, AI_TASK, or MESSAGE.
processInstanceId	integer	Optional process instance filter.
page, size, sort	pageable	Standard Spring pageable parameters.

Example:

curl -X GET "http://localhost:8080/incidents?status=OPEN&source=WORKER&page=0&size=20" \
  -H "Authorization: Bearer $TOKEN"

GET /incidents/summary

Returns dashboard counts.

Example response:

{
  "openIncidents": 3,
  "criticalIncidents": 1,
  "acknowledgedIncidents": 2,
  "incidentsCreatedToday": 4
}

GET /incidents/{id}

Returns a single incident.

Example response:

{
  "id": 25,
  "processInstanceId": 456,
  "nodeId": "sync-crm",
  "status": "OPEN",
  "severity": "HIGH",
  "source": "WORKER",
  "message": "API task 'sync-crm' timed out after 2 minutes without completion",
  "technicalDetails": "Process instance 456 failed at node 'sync-crm'",
  "externalReferenceId": "worker_request:91",
  "occurrenceCount": 1,
  "lastOccurredAt": "2026-06-23T10:15:00",
  "createdAt": "2026-06-23T10:15:00",
  "updatedAt": "2026-06-23T10:15:00",
  "acknowledgedAt": null,
  "acknowledgedBy": null,
  "resolvedAt": null,
  "resolvedBy": null,
  "resolutionNote": null,
  "resolutionAction": null
}

GET /incidents/{id}/events

Returns the incident lifecycle timeline.

Example response:

[
  {
    "id": 100,
    "incidentId": 25,
    "eventType": "CREATED",
    "message": "Incident created from WORKER.",
    "actor": null,
    "createdAt": "2026-06-23T10:15:00"
  }
]

POST /incidents/{id}/acknowledge

Mark an open incident as acknowledged.

Request:

{
  "acknowledgedBy": "admin"
}

POST /incidents/{id}/resolve

Resolve an incident.

Request:

{
  "resolvedBy": "admin",
  "resolutionNote": "CRM endpoint recovered and retry completed.",
  "resolutionAction": "RETRIED_SUCCESSFULLY"
}

POST /incidents/{id}/reopen

Move a resolved incident back to OPEN.

curl -X POST "http://localhost:8080/incidents/25/reopen" \
  -H "Authorization: Bearer $TOKEN"

POST /incidents/{id}/retry

Retry requires ACCESS_BPM_ADMIN. In addition to recoverable worker/API incidents, SLA incidents retry the failed SLA action subject to task, process, and action eligibility. SLA retries may return 409 when the action is unavailable. For WORKER incidents, The backend restores the instance to the failed node, resets the related worker request, and republishes the service task request.

Request:

{
  "requestedBy": "admin"
}

If the incident is not retryable, the API returns 400.

SLA escalation failures use incident source SLA.