REST API Problems

🔴 REST API: Problems

The complete usage guide for the Problem Management module of the REST API — every URL, method, parameter and response shape. Mirrors the interactive documentation at System → API → Documentation (with its live "Try it" tester).

Note

New to the API? The basics table on REST API: Tickets covers base URL, authentication, the response envelope, error codes, rate limits and PATCH semantics — identical across all modules.

What's distinctive about Problems:

🏢 Company-scoped	Problems belong to a company, like tickets. A key's company scope is enforced on every list, read and write — and linking an incident requires the ticket and problem to belong to the same company. Invisible on single-company installs.
📒 Append-only journal	Problem notes are immutable — you can add and read them, never edit or delete (same as the UI).
🗑️ Delete is permanent	Unlike tickets (trash + restore), deleting a problem is final and takes its links, notes and history with it — exactly like the UI's delete.

---

🚀 Quick start

Raise a problem from a monitoring correlation, link the incidents, and record the RCA:

B="https://your-server/api/v1"; K="Authorization: Bearer fitsm_…"

# 1. Create the problem
PRB=$(curl -s -X POST "$B/problems" -H "$K" -H "Content-Type: application/json" -d '{
  "title": "Recurring VPN drops on London office link",
  "description": "Multiple users disconnected since 28 June.",
  "priority": "High"}' | jq -r '.data.id')

# 2. Link the incidents behind it
curl -s -X POST "$B/problems/$PRB/tickets" -H "$K" -H "Content-Type: application/json" -d '{"ticket_id": 140}'

# 3. Later: record the root cause and flag the known error
curl -s -X PATCH "$B/problems/$PRB" -H "$K" -H "Content-Type: application/json" -d '{
  "status": "Root Cause Identified",
  "root_cause": "Faulty SFP on the primary switch uplink.",
  "workaround": "Failover to the secondary link.",
  "is_known_error": true}'

Every one of those steps writes the same audit entries the UI writes. 🧾

---

🔴 Problems

🟢 GET /problems — list / search   🔑 problems.read

Results are limited to the key's company scope.

Parameter	Description
state	open | closed | all (default all)
status / status_id	Filter by problem status name (e.g. Investigating) or id
priority / priority_id	Filter by problem priority name or id
assigned_analyst_id	Filter by assignee
is_known_error	true / false — the known-error database filter
company_id	One company (must be in the key's scope)
q	Search title and problem number (PRB-…)
created_since / updated_since	ISO 8601 date-time bounds
sort	created_at (default, desc), updated_at, closed_at, id, title, problem_number, priority, status — prefix - for descending
page / per_page	Pagination — default 1 / 25, max per_page 100

The problem shape:

{ "id": 5, "problem_number": "PRB-00005",
  "title": "Recurring VPN drops on London office link",
  "description": "Multiple users disconnected since 28 June.",
  "status":   { "id": 3, "name": "Root Cause Identified", "is_closed": false },
  "priority": { "id": 3, "name": "High" },
  "assigned_analyst": { "id": 1, "name": "Administrator" },
  "is_known_error": true,
  "root_cause": "Faulty SFP on the primary switch uplink.",
  "workaround": "Failover to the secondary link.",
  "company": { "id": 1, "name": "Default" },
  "created_by": { "id": 1, "name": "Administrator" },
  "linked_tickets_count": 3,
  "created_at": "2026-07-02T23:14:34Z", "updated_at": "2026-07-02T23:14:54Z", "closed_at": null }

🔵 POST /problems — create   🔑 problems.create

Creates a problem exactly like the UI: the PRB-##### number is stamped automatically, the module's default status applies if none is given, and a created entry is written to the audit trail. Files under the key's default company unless company_id says otherwise. Returns 201.

Field	Required	Description
title	✅	Problem title
description		What's the problem?
status / status_id		By name or id (default: the module's default status, usually New)
priority / priority_id		By name or id
assigned_analyst_id		Assign on creation
root_cause / workaround		RCA fields
is_known_error		Boolean
company_id		File under a specific company (must be in the key's scope)

🟢 GET /problems/{id} — get one   🔑 problems.read

The full problem plus its linked incidents and changes inline:

"linked_tickets": [ { "id": 140, "ticket_number": "MJI-729-47540",
                      "subject": "VPN drop — J.Smith", "status": "Open",
                      "linked_at": "2026-07-02T23:15:18Z" } ],
"linked_changes": [ { "id": 42, "title": "Replace SFP on core switch",
                      "status": "Scheduled", "relation_type": "fixes" } ]

🟠 PATCH /problems/{id} — update   🔑 problems.update

Send only what changes; null clears nullable fields; identical PATCHes are idempotent. Every change is written to the audit trail with the same field keys the UI writes (title, status, priority, assigned_to, root_cause, workaround, known_error as Yes/No) using display names for lookups. Moving into a closed-type status sets closed_at; moving out clears it.

Field	Notes
title	Cannot be blanked
description	null clears
status / status_id	Drives closed_at via the status's is_closed flag
priority / priority_id	null clears
assigned_analyst_id	null unassigns
root_cause / workaround	RCA fields, null clears
is_known_error	Boolean

🔴 DELETE /problems/{id} — delete   🔑 problems.delete

Warning

Permanent. Deletes the problem and cascades its links, journal and audit trail — the same as the UI's delete. There is no trash and no restore.

---

📒 Journal & history

🟢 GET /problems/{id}/notes   🔑 problem_notes.read

The append-only journal, newest first.

🔵 POST /problems/{id}/notes   🔑 problem_notes.create

Field	Required	Description
note	✅	The note text

Attributed to the analyst the key acts as. Immutable once written — no edit, no delete (parity with the UI). Returns 201.

🟢 GET /problems/{id}/audit   🔑 problem_audit.read

The system-generated change trail (created / modified rows), newest first:

{ "data": [ { "id": 18, "action": "modified", "field": "status",
              "old_value": "New", "new_value": "Root Cause Identified",
              "analyst": { "id": 1, "name": "Administrator" },
              "created_at": "2026-07-02T23:14:54Z" } ] }

---

🔗 Linking incidents & changes

Problems sit behind incidents (the tickets reporting the symptom) and in front of changes (the fix).

🔵 POST /problems/{id}/tickets — link an incident   🔑 problem_links.create

Body: {"ticket_id": 140}. Rules, all mirroring the UI:

• The ticket must be visible to the key (its company scope) — otherwise 404.
• Same-company rule: on multi-company installs the ticket and problem must belong to the same company — otherwise 422 "That incident belongs to a different company than this problem."
• Already linked → 409.
• Writes a linked_incident audit entry carrying the ticket number.

🔴 DELETE /problems/{id}/tickets/{ticket_id} — unlink   🔑 problem_links.delete

Removes the link (no audit entry — parity with the UI).

🔵 POST /problems/{id}/changes — link a change   🔑 problem_links.create

Body: {"change_id": 42}. Links via the shared change-relations mechanism (relation type fixes) and writes a linked_change audit entry. Unknown change → 422; already linked → 409; installs without Change Management get a clean 422.

🔴 DELETE /problems/{id}/changes/{change_id} — unlink   🔑 problem_links.delete

Removes the change link.

---

📚 Reference data

Under the shared 🔑 reference.read permission:

Endpoint	Returns
🟢 GET /problem-statuses	Problem statuses with is_closed, is_default, colour, is_active (seeded: New, Investigating, Root Cause Identified, Known Error, Resolved, Closed)
🟢 GET /problem-priorities	Problem priorities (seeded: Low, Medium, High, Critical)

---

🧪 Worked example: known-error database export

A read-only key (problems: read) feeds a status page or client report:

B="https://your-server/api/v1"; K="Authorization: Bearer fitsm_…"

# All current known errors with their workarounds
curl -s -H "$K" "$B/problems?is_known_error=true&state=open&per_page=100" \
  | jq -r '.data[] | "\(.problem_number)  \(.title)\n  Workaround: \(.workaround // "none documented")\n"'

And the monitoring direction — auto-linking a new alert ticket to its open problem:

# Find the open problem for this symptom, then link the fresh incident to it
PRB=$(curl -s -H "$K" "$B/problems?q=VPN+drops&state=open" | jq -r '.data[0].id')
curl -s -X POST "$B/problems/$PRB/tickets" -H "$K" -H "Content-Type: application/json" \
  -d "{\"ticket_id\": $NEW_TICKET_ID}"

---

Under the hood: REST API — How It Works · Other modules: Tickets · Assets · Keys & permissions: System → API (System module) · Module docs: Problem Management.