fix(soc): trazer configs do Grafana/Promtail para o git e corrigir volumes

- Adiciona soc/promtail/config.yml (antes fora do git, em /opt/soc/) - Adiciona soc/grafana/provisioning/alerting/alerts.yml (antes fora do git) - Adiciona soc/grafana/provisioning/datasources/datasources.yml (antes fora do git) - Corrige docker-compose.yml: volumes do Promtail e Grafana agora apontam para paths dentro do repo (/home/deploy/stack/soc/) ao invés de /opt/soc/ - Corrige mount dos containers Docker: host path mapeia para /var/lib/docker/containers (path padrão que o Promtail espera dentro do container) - Corrige .gitignore: ignora apenas dados binários (grafana-data, loki-data, prom-data), não mais todos os subdiretórios do soc/ - Alertas Loki usam count_over_time() para retornar métrica numérica compatível
2026-05-12 05:16:21 +02:00
parent 7e1591303a
commit e6a1ab9af5
5 changed files with 208 additions and 5 deletions
@@ -22,12 +22,18 @@ ssh_alerts_config.json
 *.tar.gz
 *.zip

-# Monitoramento e Dados de Estado (SOC)
-**/prom-data/
+# Monitoramento e Dados de Estado (SOC) - apenas dados binários/estado, NÃO configs
+soc/prom-data/
+soc/loki-data/
+soc/grafana-data/
 **/wal/
 **/chunks/
 **/tombstones/
 **/positions.yaml
+# Configs SOC rastreados pelo git (NÃO ignorar):
+# soc/promtail/config.yml
+# soc/grafana/provisioning/**
+# soc/docker-compose.yml

 # Sessões e Mídia locais
 **/sessions/
@@ -7,7 +7,7 @@ services:
      - "10.99.0.1:3000:3000"
    volumes:
      - "/home/deploy/stack/soc/grafana-data:/var/lib/grafana"
-      - "/opt/soc/grafana/provisioning:/etc/grafana/provisioning"
+      - "/home/deploy/stack/soc/grafana/provisioning:/etc/grafana/provisioning"
      - "/opt/soc/grafana/dashboards:/var/lib/grafana/dashboards"
    environment:
      - GF_SECURITY_ADMIN_PASSWORD=wMjFvkQ3GRWCvJbd
@@ -24,9 +24,9 @@ services:
    restart: always
    volumes:
      - /var/log:/var/log:ro
-      - /opt/soc/promtail:/etc/promtail:ro
+      - /home/deploy/stack/soc/promtail:/etc/promtail:ro
      - /run/user/1000/docker.sock:/var/run/docker.sock:ro
-      - /home/deploy/.local/share/docker/containers:/home/deploy/.local/share/docker/containers:ro
+      - /home/deploy/.local/share/docker/containers:/var/lib/docker/containers:ro
    command: -config.file=/etc/promtail/config.yml
    networks:
      - soc
@@ -0,0 +1,145 @@
+apiVersion: 1
+groups:
+  # ──────────────────────────────────────────────────────────────────
+  # Alertas de infraestrutura (baseados em métricas do Prometheus)
+  # ──────────────────────────────────────────────────────────────────
+  - name: SOC Alerts
+    folder: SOC
+    interval: 30s
+    rules:
+      - uid: cpu_high
+        title: CPU ALTA
+        condition: A
+        data:
+          - refId: A
+            datasourceUid: PBFA97CFB590B2093
+            relativeTimeRange:
+              from: 300
+              to: 0
+            model:
+              expr: 100 - (avg(rate(node_cpu_seconds_total{mode="idle"}[5m])) * 100)
+        noDataState: OK
+        execErrState: OK
+        for: 2m
+        annotations:
+          description: "CPU acima de 80%"
+
+      - uid: falco_critical
+        title: FALCO CRITICAL
+        condition: A
+        data:
+          - refId: A
+            datasourceUid: P8E80F9AEF21F6940
+            relativeTimeRange:
+              from: 300
+              to: 0
+            model:
+              expr: '{job="falco"} |= "CRITICAL"'
+        for: 1m
+        annotations:
+          description: "Evento crítico detectado pelo Falco"
+
+      - uid: crypto_attack
+        title: POSSÍVEL CRYPTOJACKING
+        condition: A
+        data:
+          - refId: A
+            datasourceUid: PBFA97CFB590B2093
+            relativeTimeRange:
+              from: 300
+              to: 0
+            model:
+              expr: 100 - (avg(rate(node_cpu_seconds_total{mode="idle"}[5m])) * 100)
+          - refId: B
+            datasourceUid: P8E80F9AEF21F6940
+            relativeTimeRange:
+              from: 300
+              to: 0
+            model:
+              expr: '{job="falco"} |= "xmrig"'
+        for: 1m
+        annotations:
+          description: "CPU alta + minerador detectado"
+
+  # ──────────────────────────────────────────────────────────────────
+  # Alertas de aplicação (baseados em logs dos containers via Loki)
+  # Datasource: Loki (uid: P8E80F9AEF21F6940)
+  # Container monitorado: newwhats-backend-prod
+  # ──────────────────────────────────────────────────────────────────
+  - name: App Alerts
+    folder: SOC
+    interval: 60s
+    rules:
+      - uid: backend_unhandled_exception
+        title: BACKEND - Exceção Não Tratada
+        condition: A
+        data:
+          - refId: A
+            datasourceUid: P8E80F9AEF21F6940
+            relativeTimeRange:
+              from: 300
+              to: 0
+            model:
+              expr: |
+                count_over_time(
+                  {container="newwhats-backend-prod"}
+                  |~ "(?i)(unhandledRejection|uncaughtException|TypeError: Cannot|ReferenceError:|Cannot read properties)"
+                  [5m]
+                ) > 0
+        noDataState: OK
+        execErrState: OK
+        for: 0s
+        annotations:
+          summary: "Exceção não tratada detectada no backend WhatsApp"
+          description: "O container newwhats-backend-prod emitiu uma exceção crítica não tratada. Verifique os logs e o código-fonte."
+        labels:
+          severity: critical
+
+      - uid: backend_prisma_p2002
+        title: BACKEND - Colisão Prisma P2002
+        condition: A
+        data:
+          - refId: A
+            datasourceUid: P8E80F9AEF21F6940
+            relativeTimeRange:
+              from: 300
+              to: 0
+            model:
+              expr: |
+                count_over_time(
+                  {container="newwhats-backend-prod"} |= "P2002"
+                  [5m]
+                ) > 0
+        noDataState: OK
+        execErrState: OK
+        for: 2m
+        annotations:
+          summary: "Violação de unicidade P2002 detectada no banco de dados"
+          description: "Prisma está relatando violações de constraint única (P2002). Pode indicar race condition em mensagens concorrentes."
+        labels:
+          severity: warning
+
+      - uid: backend_http_5xx
+        title: BACKEND - Erros HTTP 5xx
+        condition: A
+        data:
+          - refId: A
+            datasourceUid: P8E80F9AEF21F6940
+            relativeTimeRange:
+              from: 300
+              to: 0
+            model:
+              expr: |
+                count_over_time(
+                  {container="newwhats-backend-prod"}
+                  |~ "statusCode.*5[0-9]{2}|HTTP.*5[0-9]{2}|\"status\":5[0-9]{2}"
+                  [5m]
+                ) > 0
+        noDataState: OK
+        execErrState: OK
+        for: 2m
+        annotations:
+          summary: "Erros HTTP 5xx detectados no backend"
+          description: "O backend está retornando erros 5xx. Pode indicar falha interna, banco de dados indisponível ou bug de runtime."
+        labels:
+          severity: warning
@@ -0,0 +1,11 @@
+apiVersion: 1
+datasources:
+  - name: Prometheus
+    type: prometheus
+    access: proxy
+    url: http://prometheus:9090
+
+  - name: Loki
+    type: loki
+    access: proxy
+    url: http://loki:3100
@@ -0,0 +1,41 @@
+server:
+  http_listen_port: 9080
+positions:
+  filename: /tmp/positions.yaml
+clients:
+  - url: http://loki:3100/loki/api/v1/push
+scrape_configs:
+  - job_name: falco
+    static_configs:
+      - targets: [localhost]
+        labels:
+          job: falco
+          __path__: /var/log/falco.log
+
+  - job_name: varlogs
+    static_configs:
+      - targets: [localhost]
+        labels:
+          job: varlogs
+          __path__: /var/log/syslog
+
+  # Coleta logs de todos os containers Docker via service discovery
+  - job_name: docker_containers
+    docker_sd_configs:
+      - host: unix:///var/run/docker.sock
+        refresh_interval: 5s
+    relabel_configs:
+      # Extrai o nome do container (remove a barra inicial)
+      - source_labels: ['__meta_docker_container_name']
+        regex: '/?(.+)'
+        target_label: container
+      # Stream (stdout/stderr)
+      - source_labels: ['__meta_docker_container_log_stream']
+        target_label: stream
+      # Nome do serviço do docker-compose
+      - source_labels: ['__meta_docker_container_label_com_docker_compose_service']
+        target_label: service
+      # Caminho do log dentro do container mountado
+      - source_labels: ['__meta_docker_container_id']
+        target_label: __path__
+        replacement: /var/lib/docker/containers/$1/$1-json.log