Files
homelab-infra/platform/observability/metrics/rules/data-services.yaml
T
2026-09-25 19:35:38 +00:00

149 lines
7.1 KiB
YAML

apiVersion: monitoring.coreos.com/v1
kind: PrometheusRule
metadata:
name: data-services
namespace: monitoring
spec:
groups:
- name: data-services
interval: 30s
rules:
- alert: DataServiceMetricsUnavailable
expr: up{job="cnpg"} == 0 or absent(up{job="cnpg"})
for: 5m
labels:
severity: critical
job: cnpg
annotations:
summary: 数据服务 {{ $labels.job }} 指标不可用
description: 检查监控对象、Pod 与 exporter;目标消失同样触发。
runbook_url: https://git.ddupan.top/panxiao81/homelab-wiki/src/branch/main/guides/monitoring-data-services.md
- alert: DataServiceMetricsUnavailable
expr: up{job="seaweedfs-master"} == 0 or absent(up{job="seaweedfs-master"})
for: 5m
labels:
severity: critical
job: seaweedfs-master
annotations:
summary: 数据服务 {{ $labels.job }} 指标不可用
description: 检查监控对象、Pod 与 exporter;目标消失同样触发。
runbook_url: https://git.ddupan.top/panxiao81/homelab-wiki/src/branch/main/guides/monitoring-data-services.md
- alert: DataServiceMetricsUnavailable
expr: up{job="seaweedfs-volume"} == 0 or absent(up{job="seaweedfs-volume"})
for: 5m
labels:
severity: critical
job: seaweedfs-volume
annotations:
summary: 数据服务 {{ $labels.job }} 指标不可用
description: 检查监控对象、Pod 与 exporter;目标消失同样触发。
runbook_url: https://git.ddupan.top/panxiao81/homelab-wiki/src/branch/main/guides/monitoring-data-services.md
- alert: DataServiceMetricsUnavailable
expr: up{job="seaweedfs-filer"} == 0 or absent(up{job="seaweedfs-filer"})
for: 5m
labels:
severity: critical
job: seaweedfs-filer
annotations:
summary: 数据服务 {{ $labels.job }} 指标不可用
description: 检查监控对象、Pod 与 exporter;目标消失同样触发。
runbook_url: https://git.ddupan.top/panxiao81/homelab-wiki/src/branch/main/guides/monitoring-data-services.md
- alert: CnpgPostgresDown
expr: cnpg_collector_up{job="cnpg"} == 0
for: 2m
labels:
severity: critical
annotations:
summary: CNPG {{ $labels.pod }} PostgreSQL 不可用
description: 检查数据库 Pod 状态、实例日志和存储;exporter up 不代表数据库可用。
runbook_url: https://git.ddupan.top/panxiao81/homelab-wiki/src/branch/main/guides/monitoring-data-services.md
- alert: CnpgMetricsCollectionFailed
expr: cnpg_collector_last_collection_error{job="cnpg"} == 1
for: 5m
labels:
severity: warning
annotations:
summary: CNPG {{ $labels.pod }} SQL 指标采集失败
description: 检查内置 exporter 日志、数据库连接与查询兼容性。
runbook_url: https://git.ddupan.top/panxiao81/homelab-wiki/src/branch/main/guides/monitoring-data-services.md
- alert: CnpgConnectionsHigh
expr: sum by (job,namespace,pod) (cnpg_backends_total{job="cnpg"}) / on(job,namespace,pod)
max by(job,namespace,pod) (cnpg_pg_settings_setting{job="cnpg",name="max_connections"})
> 0.8
for: 10m
labels:
severity: warning
annotations:
summary: CNPG {{ $labels.pod }} 连接使用率超过 80%
description: 检查连接池、空闲连接和连接泄漏,不直接提高 max_connections。
runbook_url: https://git.ddupan.top/panxiao81/homelab-wiki/src/branch/main/guides/monitoring-data-services.md
- alert: CnpgLongTransaction
expr: cnpg_backends_max_tx_duration_seconds{job="cnpg",state!="idle"} > 300
for: 5m
labels:
severity: warning
annotations:
summary: CNPG {{ $labels.pod }} 存在长事务
description: 事务持续超过五分钟并维持五分钟,检查阻塞与 idle in transaction;先确认业务再处理。
runbook_url: https://git.ddupan.top/panxiao81/homelab-wiki/src/branch/main/guides/monitoring-data-services.md
- alert: NatsJetStreamStorageHigh
expr: nats_varz_jetstream_stats_storage{job="nats/nats"} / nats_varz_jetstream_config_max_storage{job="nats/nats"}
> 0.8
for: 10m
labels:
severity: warning
annotations:
summary: NATS JetStream 文件容量超过 80%
description: 检查服务端配额与消息保留策略;该指标不代表单个 Account 或 stream 配额。
runbook_url: https://git.ddupan.top/panxiao81/homelab-wiki/src/branch/main/guides/monitoring-data-services.md
- alert: NatsJetStreamMemoryHigh
expr: nats_varz_jetstream_stats_memory{job="nats/nats"} / nats_varz_jetstream_config_max_memory{job="nats/nats"}
> 0.8
for: 10m
labels:
severity: warning
annotations:
summary: NATS JetStream 内存容量超过 80%
description: 检查 memory store 消息保留与服务端内存配额。
runbook_url: https://git.ddupan.top/panxiao81/homelab-wiki/src/branch/main/guides/monitoring-data-services.md
- alert: NatsSlowConsumers
expr: increase(nats_varz_slow_consumers{job="nats/nats"}[10m]) > 0
for: 5m
labels:
severity: warning
annotations:
summary: NATS 检测到慢消费者
description: 检查消费者处理速率与网络;此规则不能代替 JetStream consumer 积压监控。
runbook_url: https://git.ddupan.top/panxiao81/homelab-wiki/src/branch/main/guides/monitoring-data-services.md
- alert: SeaweedVolumeDiskLow
expr: SeaweedFS_volumeServer_resource{job="seaweedfs-volume",type="avail"} /
ignoring(type) SeaweedFS_volumeServer_resource{job="seaweedfs-volume",type="all"}
< 0.1
for: 10m
labels:
severity: warning
annotations:
summary: SeaweedFS {{ $labels.name }} 可用空间不足 10%
description: 检查 volume 文件系统容量、保留策略与底层 ZFS;不要直接删除卷文件。
runbook_url: https://git.ddupan.top/panxiao81/homelab-wiki/src/branch/main/guides/monitoring-data-services.md
- alert: SeaweedVolumeDiskReadOnly
expr: SeaweedFS_volumeServer_read_only_volumes{job="seaweedfs-volume",type="isDiskSpaceLow"}
> 0
for: 5m
labels:
severity: critical
annotations:
summary: SeaweedFS 因磁盘空间不足限制卷写入
description: 检查磁盘容量和 volume 日志;主动只读或满卷切换不按此故障处理。
runbook_url: https://git.ddupan.top/panxiao81/homelab-wiki/src/branch/main/guides/monitoring-data-services.md
- alert: SeaweedVolumeWriteFailures
expr: increase(SeaweedFS_volumeServer_file_write_failures{job="seaweedfs-volume"}[5m])
> 0
for: 2m
labels:
severity: warning
annotations:
summary: SeaweedFS 卷写入失败
description: 检查 volume 日志、磁盘状态及上游请求;不代表已经验证 S3 端到端可用。
runbook_url: https://git.ddupan.top/panxiao81/homelab-wiki/src/branch/main/guides/monitoring-data-services.md