diff --git a/platform/observability/metrics/exporters/kube-state-metrics-helmrelease.yaml b/platform/observability/metrics/exporters/kube-state-metrics-helmrelease.yaml index af906ec..17bd746 100644 --- a/platform/observability/metrics/exporters/kube-state-metrics-helmrelease.yaml +++ b/platform/observability/metrics/exporters/kube-state-metrics-helmrelease.yaml @@ -198,6 +198,9 @@ spec: - conditions - '[type=Ready]' - status + repository_type: + - spec + - type - groupVersionKind: group: source.toolkit.fluxcd.io version: v1 diff --git a/platform/observability/metrics/rules/platform-controllers.yaml b/platform/observability/metrics/rules/platform-controllers.yaml index 1985ba5..f5394ba 100644 --- a/platform/observability/metrics/rules/platform-controllers.yaml +++ b/platform/observability/metrics/rules/platform-controllers.yaml @@ -140,7 +140,7 @@ spec: description: 检查 OpenBao 可达性与 ESO 身份鉴权;可能影响多个命名空间。 runbook_url: https://git.ddupan.top/panxiao81/homelab-wiki/src/branch/main/guides/monitoring-controllers.md - alert: FluxResourceNotReady - expr: gotk_resource_info{job="kube-state-metrics",ready!="True",suspended!="true"} + expr: gotk_resource_info{job="kube-state-metrics",ready!="True",suspended!="true",repository_type!="oci"} == 1 for: 15m labels: diff --git a/platform/observability/metrics/tests/platform.test.yaml b/platform/observability/metrics/tests/platform.test.yaml index 31b32f4..20f59d5 100644 --- a/platform/observability/metrics/tests/platform.test.yaml +++ b/platform/observability/metrics/tests/platform.test.yaml @@ -174,3 +174,12 @@ tests: summary: Flux 对象状态指标缺失 description: 检查 kube-state-metrics 自定义资源配置、RBAC 与采集;不能把空结果当作健康。 runbook_url: https://git.ddupan.top/panxiao81/homelab-wiki/src/branch/main/guides/monitoring-controllers.md +- name: OCI HelmRepository 没有 Ready 条件仍不误报 + interval: 1m + input_series: + - series: gotk_resource_info{job="kube-state-metrics",customresource_kind="HelmRepository",repository_type="oci",name="envoy-gateway"} + values: 1x30 + alert_rule_test: + - eval_time: 20m + alertname: FluxResourceNotReady + exp_alerts: []