diff --git a/flux/apps/kube-system/kube-ovn/observability/alerts.yml b/flux/apps/kube-system/kube-ovn/observability/alerts.yml index 76f7087e..3abd7f3a 100644 --- a/flux/apps/kube-system/kube-ovn/observability/alerts.yml +++ b/flux/apps/kube-system/kube-ovn/observability/alerts.yml @@ -337,6 +337,42 @@ spec: description: "{{ $value }} inconsistent port bindings between OVS and OVN-SB on node {{ $labels.node_name }}" + - name: kube-ovn-datapath + interval: 30s + rules: + - alert: HighDatapathFlowLookupMissRate + expr: rate(kube_ovn_dp_flows_lookup_missed[5m]) / rate(kube_ovn_dp_flows_lookup_hit[5m]) > 0.5 + for: 10m + labels: + team: devops + severity: warning + component: ovs-datapath + annotations: + summary: "High datapath flow lookup miss rate" + description: "Datapath {{ $labels.datapath }} on {{ $labels.hostname }} has high flow lookup miss rate: {{ $value | humanizePercentage }}" + + - alert: DatapathFlowsLost + expr: rate(kube_ovn_dp_flows_lookup_lost[5m]) > 1 + for: 5m + labels: + team: devops + severity: warning + component: ovs-datapath + annotations: + summary: "Datapath flows being lost" + description: "Datapath {{ $labels.datapath }} on {{ $labels.hostname }} is losing {{ $value }} flows per second" + + - alert: LowDatapathMaskHitRatio + expr: kube_ovn_dp_masks_hit_ratio < 1 + for: 10m + labels: + team: devops + severity: info + component: ovs-datapath + annotations: + summary: "Low datapath mask hit ratio" + description: "Datapath {{ $labels.datapath }} on {{ $labels.hostname }} has low mask hit ratio: {{ $value }}" + - name: kube-ovn-performance interval: 30s rules: