7837a809ca
- 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>
63 lines
1.9 KiB
YAML
63 lines
1.9 KiB
YAML
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
|
|
|