feat(ops): 添加生产监控告警基线

- 新增 Prometheus/Grafana/Loki/Promtail 监控 profile\n- 覆盖 DB、磁盘、慢 API、慢 Prisma、任务失败和小宝摘要 stale 告警\n- 补充 postgres-exporter 自定义查询、Dashboard、部署文档和校验\n\nCo-Authored-By: GPT-5 Codex <codex@openai.com>

deploy/monitoring/prometheus/alert-rules.yml

@@ -0,0 +1,62 @@
+groups:
+  - name: ftb-production-alerts
+    rules:
+      - alert: FtbPostgresDown
+        expr: pg_up == 0
+        for: 2m
+        labels:
+          severity: critical
+        annotations:
+          summary: PostgreSQL exporter cannot reach the FTB database.
+          runbook: docs/runbooks/migration-rollback.md
+
+      - alert: FtbDiskPressure
+        expr: |
+          (
+            node_filesystem_avail_bytes{mountpoint="/",fstype!~"tmpfs|overlay"}
+            /
+            node_filesystem_size_bytes{mountpoint="/",fstype!~"tmpfs|overlay"}
+          ) < 0.15
+        for: 10m
+        labels:
+          severity: warning
+        annotations:
+          summary: Production host root filesystem has less than 15% free space.
+          runbook: docs/deployment.md
+
+      - alert: FtbSlowApiLogBurst
+        expr: increase(ftb_slow_api_log_total[10m]) > 5
+        for: 2m
+        labels:
+          severity: warning
+        annotations:
+          summary: Slow API request log volume crossed the V2.8 baseline threshold.
+          runbook: docs/deployment.md
+
+      - alert: FtbSlowPrismaLogBurst
+        expr: increase(ftb_slow_prisma_log_total[10m]) > 3
+        for: 2m
+        labels:
+          severity: warning
+        annotations:
+          summary: Slow Prisma query log volume crossed the V2.8 baseline threshold.
+          runbook: docs/deployment.md
+
+      - alert: FtbJobFailureLogBurst
+        expr: increase(ftb_job_failure_log_total[10m]) > 0
+        for: 1m
+        labels:
+          severity: warning
+        annotations:
+          summary: Background or compatibility job failure logs were detected.
+          runbook: docs/runbooks/xiaobao-background-jobs.md
+
+      - alert: FtbXiaobaoSummaryStale
+        expr: ftb_xiaobao_stale_summary_count > 0
+        for: 15m
+        labels:
+          severity: warning
+        annotations:
+          summary: Xiaobao warning summaries are dirty or stale.
+          runbook: docs/runbooks/xiaobao-background-jobs.md
+

deploy/monitoring/prometheus/prometheus.yml

@@ -0,0 +1,45 @@
+global:
+  scrape_interval: 15s
+  evaluation_interval: 15s
+
+rule_files:
+  - /etc/prometheus/alert-rules.yml
+
+scrape_configs:
+  - job_name: prometheus
+    static_configs:
+      - targets: ['prometheus:9090']
+
+  - job_name: postgres-exporter
+    static_configs:
+      - targets: ['postgres-exporter:9187']
+
+  - job_name: node-exporter
+    static_configs:
+      - targets: ['node-exporter:9100']
+
+  - job_name: cadvisor
+    static_configs:
+      - targets: ['cadvisor:8080']
+
+  - job_name: promtail
+    static_configs:
+      - targets: ['promtail:9080']
+
+  - job_name: blackbox-http
+    metrics_path: /probe
+    params:
+      module: [http_2xx]
+    static_configs:
+      - targets:
+          - http://nginx/api/v1/health/version
+          - http://nginx/api/v1/config/ai
+          - http://nginx/api/v1/v2.2/requirements?productId=__smoke__&limit=1
+    relabel_configs:
+      - source_labels: [__address__]
+        target_label: __param_target
+      - source_labels: [__param_target]
+        target_label: instance
+      - target_label: __address__
+        replacement: blackbox-exporter:9115
+