59 lines
2.5 KiB
YAML
59 lines
2.5 KiB
YAML
groups:
|
|
- name: EKS-Cilium-Alerts
|
|
rules:
|
|
- alert: CiliumClusterMeshFailureAlert
|
|
expr: sum(cilium_clustermesh_remote_cluster_failures) by (cluster, source_cluster, target_cluster) > 0
|
|
for: 5m
|
|
labels:
|
|
severity: critical
|
|
bu: infra
|
|
team: devops
|
|
service: eks-cilium
|
|
env: prd
|
|
priority: p0
|
|
annotations:
|
|
summary: "Cilium clustermesh failure from {{ $labels.source_cluster }} to {{ $labels.target_cluster }}"
|
|
description: "Cilium clustermesh failure VALUE = {{ $value }}\n LABELS: {{ $labels }}"
|
|
|
|
- alert: DroppedEgressPackets
|
|
expr: sum(rate(cilium_drop_count_total{direction="EGRESS", k8s_app="cilium", reason=~"No mapping for NAT masquerade|FIB lookup failed"}[1m])) by (kubernetes_pod_name, instance, reason, cluster) >= 5
|
|
for: 5m
|
|
labels:
|
|
severity: critical
|
|
bu: infra
|
|
team: devops
|
|
service: eks-cilium
|
|
env: prd
|
|
priority: p0
|
|
annotations:
|
|
summary: "Cilium Dropped Egress Packets has crossed 5 Ops per Second for Reason - {{ $labels.reason }}, Pod - {{ $labels.kubernetes_pod_name }}, Instance - {{ $labels.instance }}, cluster - {{ $labels.cluster }} and the value is {{ $value }}"
|
|
description: "Cilium Dropped Egress Packets has crossed 5 Ops per Second for Reason - {{ $labels.reason }}, Pod - {{ $labels.kubernetes_pod_name }}, Instance - {{ $labels.instance }}, cluster - {{ $labels.cluster }} and the value is {{ $value }}\n Labels - {{ $labels }}"
|
|
|
|
# - alert: SubnetOutofIpsByCilium
|
|
# expr: sum(cilium_operator_ipam_available_ips_per_subnet) by (cluster, subnet_id) < 300
|
|
# for: 1m
|
|
# labels:
|
|
# severity: warning
|
|
# bu: infra
|
|
# team: devops
|
|
# service: eks-cilium
|
|
# env: prd
|
|
# priority: p0
|
|
# annotations:
|
|
# summary: "Subnet {{ $labels.subnet_id }} has ip address less than 300. Please check"
|
|
# description: "Subnet {{ $labels.subnet_id }} is left with ip address of VALUE = {{ $value }} which is less than 500.\n LABELS: {{ $labels }}"
|
|
|
|
# - alert: SubnetOutofIpsByCilium_100
|
|
# expr: sum(cilium_operator_ipam_available_ips_per_subnet) by (cluster, subnet_id) < 100
|
|
# for: 1m
|
|
# labels:
|
|
# severity: critical
|
|
# bu: infra
|
|
# team: devops
|
|
# service: eks-cilium
|
|
# env: prd
|
|
# priority: p0
|
|
# annotations:
|
|
# summary: "Subnet {{ $labels.subnet_id }} has ip address less than 100. Please check"
|
|
# description: "Subnet {{ $labels.subnet_id }} is left with ip address of VALUE = {{ $value }} which is less than 100.\n LABELS: {{ $labels }}"
|