Get Incident Details
Returns detailed incident data, including a timeline of failed/recovery checks and alert events.
GET /api/incidents/:incidentPublicId
Authentication
Requires a valid session.
- Header:
Authorization: Bearer <token>
Parameters
incidentPublicId(Path, required): Incident public UUID.
Example (cURL)
BASE_URL="https://uptimeify.io"
TOKEN="<your-api-token>"
curl -X GET "$BASE_URL/api/incidents/6bfec6f6-245a-47ce-843b-157d97d56f88" \
-H "Authorization: Bearer $TOKEN" \
-H "Accept: application/json"Note: evidenceCheck can be null. screenshotUrl is only set when hasScreenshot is true.
Timeline resolution
timeline.outageChecks lists every check between the confirmation (confirmationAt, exclusive) and the end of the incident, failed and successful alike, ascending and capped at 200; outageChecksTotal is the full count. Read status per entry: an aggregated bucket is failure as soon as one location failed in it.
Every entry in outageStartedCheck, failedChecks, outageChecks and recoveryChecks carries a resolution:
raw: a single check at a single location.location,statusCodeanderrorMessageare the ones that location reported.1min: one minute of the monitoring aggregate, used once an incident is older than the raw check retention window (24 hours).location,statusCodeanderrorMessagearenull,responseTimeMsis the average over the minute, and two extra fields say how many locations were behind it:totalLocationsandfailedLocations.15minand1day: one 15-minute or one daily bucket, used once the minute aggregate no longer reaches back (7 days for15min, 30 days for1day). Samenullfields as1min, but the counts are single location checks, not locations:totalChecksandfailedChecks.- Server and DNS monitors (TCP, SSH, FTP, SMTP, IMAP/POP, ICMP, DNS) have no 15-minute tier:
1minreaches back 30 days, older incidents get1dayentries withtotalChecksandfailedChecks.
The id of an aggregated entry is synthetic (<resolution>-<timestamp>, e.g. 1min-2026-09-01T10:00:00.000Z) and cannot be looked up as a check. The four lists are resolved independently, so an incident that spans a boundary can mix resolutions.
Example Response (excerpt)
{
"incident": {
"id": 123,
"websiteId": 101,
"type": "downtime",
"status": "open",
"startedAt": "2026-02-26T12:10:00.000Z",
"resolvedAt": null,
"statusCode": null,
"errorMessage": "Timeout",
"responseTimeMs": null,
"falseAlarmAt": null,
"falseAlarmReason": null,
"publicTitle": null,
"publicStatus": "open",
"publicResolvedAt": null
},
"timeline": {
"outageStartedAt": "2026-02-26T12:08:00.000Z",
"outageStartedCheck": {
"id": "chk_01H...",
"checkedAt": "2026-02-26T12:08:00.000Z",
"status": "failure",
"statusCode": 503,
"errorMessage": "Timeout",
"responseTimeMs": null,
"location": { "id": 7, "code": "de-nbg", "name": "Nuremberg (DE)" },
"resolution": "raw"
},
"confirmationAt": "2026-02-26T12:10:00.000Z",
"failedChecks": [
{
"id": "1min-2026-02-26T12:09:00.000Z",
"checkedAt": "2026-02-26T12:09:00.000Z",
"status": "failure",
"statusCode": null,
"errorMessage": null,
"responseTimeMs": 246,
"location": null,
"resolution": "1min",
"totalLocations": 3,
"failedLocations": 2
}
],
"failedChecksTotal": 2,
"alertEvents": [
{
"id": 987,
"sentAt": "2026-02-26T12:11:00.000Z",
"type": "email",
"status": "sent",
"channelName": "Ops Email",
"errorMessage": null
}
],
"recoveryChecks": [],
"recoveryChecksTotal": 0,
"outageChecks": [],
"outageChecksTotal": 0
},
"evidenceCheck": {
"id": "chk_01H...",
"checkedAt": "2026-02-26T12:08:00.000Z",
"diagnostics": null,
"hasScreenshot": false,
"screenshotUrl": null
}
}False-alarm mark
An incident can be marked as a false alarm (POST /api/incidents/{id}/false-alarm). Two read-only
fields on incident report that mark:
| Field | Type | Meaning |
|---|---|---|
falseAlarmAt | string | null | When the mark was set, ISO 8601. null means the incident is not marked. |
falseAlarmReason | string | null | The free-text reason given when marking, or null. |
The mark is reversible, so falseAlarmAt can return to null. It does not change status: an
incident marked as a false alarm still went out as an alert, which is the point the mark records.
The user who set the mark is not part of the response.
Setting or removing the mark follows the same write rule as every other change to the incident: within its customer scope, every linked status page must allow it (organization-wide admin or editor; customer-bound callers, also with the role readonly, on self_service pages, or on managed pages with the customer's canEditManaged exception, otherwise 403 managed_by_organization). An incident linked to no page needs admin or editor.
Status page links
statusPageLinks (since 27 October 2026) lists the status pages, and optionally components, the incident is linked to with Create Manual Incident or Update Incident: statusPageId, statusPagePublicId, statusPageName, componentId (null = page-wide), componentPublicId, componentName and impact. It is empty for an incident that only appears through its customer or monitor. A manual incident without a customer is readable by organization-wide roles only; a customer-scoped login or token gets 404.
publicTitle, publicStatus and publicResolvedAt (since 28 October 2026) describe the incident as status pages show it. publicTitle is the title set for the pages with Update Incident, or null (the pages then show title). publicStatus is open or resolved; for a monitor incident timeline updates set it independently of status, otherwise it follows status. publicResolvedAt is when it was resolved on the pages (ISO 8601).
postmortem (since 28 October 2026) is the public postmortem of the incident, or null when none was written: body (Markdown), publishedAt (null = draft, shown only in the status pages module) and updatedAt. It is written with Save Postmortem and published with Publish Postmortem.
Common Errors
400 Incident public ID (UUID) requiredif:incidentPublicIdis invalid401 Unauthorizedif you are not authenticated403 Forbiddenif the incident exists but you do not have access404 Incident not foundif the incident does not exist