> ## Documentation Index
> Fetch the complete documentation index at: https://chatcli.edilsonfreitas.com/llms.txt
> Use this file to discover all available pages before exploring further.
# Resolver Incidente
> Resolve manualmente um incidente — essencial quando o sistema escala para ação humana ou quando um operador confirma que o problema foi corrigido
Nome único do incidente (ex.: `INC-20260319-001`)
Namespace Kubernetes do incidente
Descrição de como o incidente foi resolvido. Se omitido, assume o valor padrão `"Manually resolved via dashboard"`.
Boa prática: sempre forneça uma mensagem de resolução significativa para postmortem e trilha de auditoria.
## Quando Usar Este Endpoint
Este endpoint é crítico nos seguintes cenários:
### 1. Incidentes Escalados (Ação Humana Necessária)
Quando a plataforma AIOps esgota todas as tentativas de remediação automática (padrão: 3 tentativas), o incidente transiciona para o estado **Escalated**. Neste ponto:
* O sistema de notificação alerta a equipe de plantão pelos canais configurados (Slack, PagerDuty, etc.)
* A política de escalação progride através dos níveis (L1 → L2 → L3) com base nos tempos limite
* **Nenhuma remediação automática adicional será tentada**
* O incidente permanece no estado `Escalated` até que um humano o resolva por este endpoint, pelo painel web ou pela CLI
### 2. Verificação Manual Após Remediação Automatica
Mesmo quando a remediação automática e bem-sucedida, operadores podem querer verificar manualmente e fechar o incidente com contexto adicional.
### 3. Falsos Positivos
Quando um incidente e detectado mas o operador determina que e um falso positivo, use este endpoint com uma resolução como `"False positive — metric spike caused by scheduled batch job"`.
## Fluxo de Resolução para Incidentes Escalados
```
Incidente Detectado
↓
Análise IA → Remediação Runbook/Agentica
↓ ↓
Sucesso → Resolvido Falha (retry até 3x)
↓
Todas tentativas esgotadas
↓
ESCALADO ← Você está aqui
↓
┌─────────────────┼──────────────────┐
↓ ↓ ↓
API Resolver Botao Dashboard Comando CLI
POST /resolve Botao "Resolver" (versão futura)
↓ ↓ ↓
└─────────────────┼──────────────────┘
↓
Estado: Resolvido
PostMortem gerado
Cache de dedup limpo
Evento de auditoria registrado
```
## O Que Acontece Após a Resolução
1. **Mudança de Estado**: O incidente transiciona de `Escalated` (ou qualquer estado ativo) para `Resolved`
2. **Anotações Adicionadas**:
* `aiops.chatcli.io/resolved-by` — o papel/usuário autenticado
* `aiops.chatcli.io/resolved-at` — timestamp da resolução
* `aiops.chatcli.io/manual-resolution` — definido como `"true"`
3. **Cache de Dedup Limpo**: A entrada de deduplicação do recurso é invalidada, permitindo que novas anomalias sejam detectadas para o mesmo recurso
4. **PostMortem Gerado**: Um postmortem automático é criado com os detalhes da resolução
5. **Evento de Auditoria**: Um evento de auditoria imutável é registrado para conformidade
```json Response 200 theme={"system"}
{
"apiVersion": "v1",
"kind": "Incident",
"metadata": {
"name": "INC-20260319-001",
"namespace": "production",
"annotations": {
"aiops.chatcli.io/resolved-by": "operator",
"aiops.chatcli.io/resolved-at": "2026-03-19T16:45:00Z",
"aiops.chatcli.io/manual-resolution": "true"
}
},
"spec": {
"resourceRef": {
"kind": "Deployment",
"name": "payment-service",
"namespace": "production"
},
"severity": "critical",
"signalType": "oom_kill"
},
"status": {
"state": "Resolved",
"resolution": "Memory leak fixed in payment-service v2.4.1 hotfix deployed manually. Verified stable for 15 minutes.",
"resolvedAt": "2026-03-19T16:45:00Z",
"detectedAt": "2026-03-19T14:30:00Z",
"remediationAttempts": 3,
"escalatedAt": "2026-03-19T15:15:00Z"
}
}
```
```json Response 404 theme={"system"}
{
"apiVersion": "v1",
"kind": "Error",
"error": {
"code": 404,
"message": "Incident not found",
"details": "No incident named 'INC-20260319-999' found in namespace 'production'"
}
}
```
```json Response 409 theme={"system"}
{
"apiVersion": "v1",
"kind": "Error",
"error": {
"code": 409,
"message": "Incident already resolved",
"details": "The incident 'INC-20260319-001' was already resolved at 2026-03-19T16:45:00Z"
}
}
```
## OpenAPI
````yaml POST /incidents/{name}/resolve
openapi: 3.1.0
info:
title: ChatCLI AIOps Platform API
description: >-
REST API for the ChatCLI AIOps Platform — incidents, SLOs, runbooks,
approvals, postmortems, analytics, audit, federation and health.
The playground below uses **editable server variables** so you can point
requests at your own self-hosted instance. Adjust `host`, `port` and
`basePath` to match your deployment, then click **Try it**.
version: 1.0.0
contact:
name: ChatCLI
url: https://github.com/diillson/chatcli
license:
name: Apache-2.0
identifier: Apache-2.0
servers:
- url: http://{host}:{port}/{basePath}
description: >-
Self-hosted ChatCLI AIOps Platform — adjust host, port and basePath to
match your deployment.
variables:
host:
default: localhost
description: >-
Hostname or IP of your ChatCLI operator. For Kubernetes, use the
Service DNS (e.g. chatcli-api.chatcli-system.svc) or your Ingress
host.
port:
default: '8090'
description: >-
Port the API listens on. Default is 8090; configurable via Helm value
`apiPort` or env var CHATCLI_API_PORT.
basePath:
default: api/v1
description: >-
API base path. Always `api/v1` unless you have a custom proxy rewrite
in front of the operator.
security:
- ApiKeyAuth: []
tags:
- name: Incidents
description: >-
Incident lifecycle: list, fetch, acknowledge, resolve, snooze, timeline
and remediation views.
- name: Runbooks
description: Remediation runbooks (list, create, update, delete).
- name: SLOs
description: Service Level Objectives and error budgets.
- name: Approvals
description: Approval workflow for risky remediations.
- name: Postmortems
description: Auto-generated postmortems with review, feedback and close.
- name: Remediations
description: Remediation plans and agentic execution history.
- name: AI Insights
description: AI root-cause analysis and recommendations per incident.
- name: Analytics
description: 'Operational analytics: MTTD, MTTR, trends, capacity, compliance.'
- name: Audit
description: Immutable audit log and exports.
- name: Federation
description: 'Multi-cluster federation: clusters, status and cross-cluster correlations.'
- name: Health
description: Liveness and readiness probes.
paths:
/incidents/{name}/resolve:
post:
tags:
- Incidents
summary: Resolve Incident
description: >-
Manually resolves an incident — used when the system escalates to human
action or an operator confirms the issue is fixed.
operationId: resolveIncident
parameters:
- name: name
in: path
required: true
description: Unique incident name.
schema:
type: string
example: INC-20260319-001
- name: namespace
in: query
required: false
description: Kubernetes namespace of the incident.
schema:
type: string
default: default
requestBody:
required: false
content:
application/json:
schema:
type: object
properties:
resolution:
type: string
description: >-
Description of how the incident was resolved. Defaults to
`"Manually resolved via dashboard"` when omitted.
example:
resolution: >-
Memory leak fixed in payment-service v2.4.1 hotfix deployed
manually. Verified stable for 15 minutes.
responses:
'200':
description: Incident resolved
content:
application/json:
schema:
$ref: '#/components/schemas/Incident'
'401':
$ref: '#/components/responses/Unauthorized'
'404':
$ref: '#/components/responses/NotFound'
'409':
$ref: '#/components/responses/Conflict'
'429':
$ref: '#/components/responses/RateLimited'
components:
schemas:
Incident:
type: object
properties:
apiVersion:
type: string
example: v1
kind:
type: string
example: Incident
metadata:
type: object
properties:
name:
type: string
namespace:
type: string
createdAt:
type: string
format: date-time
annotations:
type: object
additionalProperties:
type: string
labels:
type: object
additionalProperties:
type: string
spec:
type: object
properties:
resource:
type: string
resourceRef:
type: object
properties:
kind:
type: string
name:
type: string
namespace:
type: string
severity:
type: string
description:
type: string
detectionSource:
type: string
correlationId:
type: string
runbook:
type: string
escalationPolicy:
type: string
signalType:
type: string
status:
type: object
properties:
state:
type: string
enum:
- Detected
- Analyzing
- Remediating
- Contained
- Resolved
- Escalated
- Failed
description: >-
Lifecycle state. Contained (GAP-03 fix) indicates the workload
was silenced via a containment action and human intervention is
required to restore service.
riskScore:
type: integer
analysisResult:
type: object
properties:
rootCause:
type: string
recommendation:
type: string
confidence:
type: number
relatedEvents:
type: array
items:
type: string
acknowledged:
type: boolean
acknowledgedBy:
type: string
acknowledgedAt:
type: string
format: date-time
message:
type: string
snoozed:
type: boolean
snoozedBy:
type: string
snoozedAt:
type: string
format: date-time
snoozeUntil:
type: string
format: date-time
snoozeReason:
type: string
assignedTo:
type: string
detectedAt:
type: string
format: date-time
resolvedAt:
type: string
format: date-time
escalatedAt:
type: string
format: date-time
lastUpdated:
type: string
format: date-time
timelineCount:
type: integer
remediationAttempts:
type: integer
resolution:
type: string
requiresHumanAction:
type: boolean
description: >-
True when the Issue is in Contained state OR carries the
RequiresHumanAction condition. GAP-07 fix promoted this to a
typed status field on the Issue CRD itself so `kubectl get issue
-o jsonpath='{.status.requiresHumanAction}'` works without
parsing conditions.
requiredAction:
type: string
description: >-
Concrete follow-up action the operator must perform when
requiresHumanAction=true (e.g., 'restore the deployment's
replicas to the desired count after fixing the root cause').
GAP-07 fix.
chaosInduced:
type: boolean
description: >-
True when the Issue was correlated with an active
ChaosExperiment at creation time (label
platform.chatcli.io/source=chaos-experiment). GAP-04 fix.
chaosExperiment:
type: string
description: >-
Name of the correlated ChaosExperiment CR when
chaosInduced=true. GAP-04 fix.
Error:
type: object
required:
- apiVersion
- kind
- error
properties:
apiVersion:
type: string
example: v1
kind:
type: string
example: Error
error:
oneOf:
- type: object
required:
- code
- message
properties:
code:
type: integer
example: 404
message:
type: string
example: Resource not found
details:
type: string
- type: string
code:
type: integer
description: Top-level code (only used by some endpoints).
example: 400
responses:
Unauthorized:
description: Missing or invalid API token
content:
application/json:
schema:
$ref: '#/components/schemas/Error'
NotFound:
description: Resource not found
content:
application/json:
schema:
$ref: '#/components/schemas/Error'
Conflict:
description: Conflict with current state of the resource
content:
application/json:
schema:
$ref: '#/components/schemas/Error'
RateLimited:
description: Rate limit exceeded for this token's role
content:
application/json:
schema:
$ref: '#/components/schemas/Error'
securitySchemes:
ApiKeyAuth:
type: apiKey
in: header
name: Authorization
description: >-
Bearer token issued by the operator. Format: `Authorization: Bearer
`.
````