Files
homeprod/kubernetes/infrastructure/observability/vmstack-values.yaml
T

200 lines
9.3 KiB
YAML

# victoria-metrics-k8s-stack Helm values
# The chart installs the VictoriaMetrics operator + CRDs and creates VMSingle,
# VMAgent, VMAlert, VMAlertmanager CRs.
# ---------------------------------------------------------------------------
# VictoriaMetrics single-binary (the metrics database)
# ---------------------------------------------------------------------------
vmsingle:
enabled: true
spec:
retentionPeriod: 7d
storage:
storageClassName: longhorn
accessModes:
- ReadWriteOnce
resources:
requests:
storage: 20Gi
resources:
requests:
cpu: 100m
memory: 256Mi
limits:
cpu: "1"
memory: 1Gi
# ---------------------------------------------------------------------------
# vmagent — scrapes node-exporter, kube-state-metrics, kubelet, etc.
# The docker host runs its own vmagent that remote_writes through
# vm-internal.lan (the Traefik Ingress → this vmsingle).
# ---------------------------------------------------------------------------
vmagent:
enabled: true
spec:
selectAllByDefault: true
scrapeInterval: 20s
tolerations:
- key: ENC[AES256_GCM,data:SyutB0y9QtqQ,iv:EKRXNGtUovQYsESHFQTGygP09O2UWtLuBEQQcYZm3Oc=,tag:zh7CYOZleSgbbQhYZvcT+w==,type:str]
value: ENC[AES256_GCM,data:So+8xmHYSxs=,iv:7sfUBAaMZd7imjVx14lusVn58cJQ4HpTkaPy5Eal2dY=,tag:hRBjT6wuEfG3Z+ZiPJhwQQ==,type:str]
operator: Equal
effect: NoSchedule
- key: ENC[AES256_GCM,data:lj9OurynU8pRt4/MhgyevjmV0b7Zq+WG/9+zRRvPeA+RYZsV+A==,iv:LjuhErDT0bDlLcpo/AlNykz8Xu89B+nx1ja+HkPNXwo=,tag:5xM6XHrCdEejE7ituQ8W4Q==,type:str]
operator: Exists
effect: NoSchedule
# ---------------------------------------------------------------------------
# kube-state-metrics + node-exporter (metrics sources)
# ---------------------------------------------------------------------------
kube-state-metrics:
enabled: true
prometheus-node-exporter:
enabled: true
# ---------------------------------------------------------------------------
# Grafana (UI) — served behind ingress at grafana.lan.
# ---------------------------------------------------------------------------
grafana:
enabled: true
persistence:
enabled: true
storageClassName: longhorn
accessModes:
- ReadWriteOnce
size: 5Gi
adminUser: admin
# SOPS encrypts this when the file is processed.
adminPassword: ENC[AES256_GCM,data:vpPvJA/PktMnsg/IovYfsD4lDBndUw==,iv:b4eLMEcUMJj4BwyeYw1kw2sZuieyr2xL9et+R7ss0BU=,tag:u0RcTp6mkQNAGkblG4/xww==,type:str]
# Provision a Loki datasource alongside the chart's default VictoriaMetrics
# datasource so metric/log correlation works in one UI.
datasources:
datasources.yaml:
apiVersion: ENC[AES256_GCM,data:Uw==,iv:KYTMBV53yMbojMaqNinYpm/Uj5ylV86jcdJTdxuy+G0=,tag:IbIpiSBQFw7YSTQg0WRGqg==,type:int]
datasources:
- name: ENC[AES256_GCM,data:BPfR8Q==,iv:MkMhH3IJfBZzbIcqQw7y6bfJB/4Ew4feZNugjovk5O0=,tag:jaa43tzIXkE8kWZp2R/kcg==,type:str]
type: ENC[AES256_GCM,data:WWncww==,iv:hPqJE0KCidSBnf11wxFH+Je+YKztWmeToBqi39Qf4ZQ=,tag:c9enJ2VgQqDJVbasvgG+Dw==,type:str]
url: ENC[AES256_GCM,data:YkS66tpWXpW62Dtful0KNCwoZ3rG3nQtDBUsNQUwdQ+eqtG38JTrAudQqnsRD51q,iv:Mu3Sc6T2xypdot9dA1DHc8nD7UI8372nqlveLITR82E=,tag:1VmmIg+xR9qjWcvPKVYsJQ==,type:str]
access: ENC[AES256_GCM,data:af3GpjE=,iv:MAYKfklF+wCC/XN+4snebjPAnW4KpYzaMMGYds7rG18=,tag:X+oFiDAiCb1f7cfREqFT7A==,type:str]
isDefault: ENC[AES256_GCM,data:Zgu4Hn8=,iv:nnVB/in75nPAQos0Dfl6abBrHwT8tCt5X7lcCYBZbLc=,tag:u1UvkSj6SjcMm06NzKMGDA==,type:bool]
jsonData:
maxLines: ENC[AES256_GCM,data:88JJqA==,iv:lKsZa024I6223UEu0SHuZ3N9LZrzw0WBTfC+L2/3ljI=,tag:YvqieXK36xMmC/sSJ11NIg==,type:int]
ingress:
enabled: true
ingressClassName: traefik
annotations:
external-dns.alpha.kubernetes.io/enabled: "true"
hosts:
- grafana.lan
path: /
pathType: Prefix
tls: []
# ---------------------------------------------------------------------------
# vmalert — evaluates VMRule CRs against VictoriaMetrics, forwards firing
# alerts to Alertmanager. selectAllByDefault picks up all VMRules in the
# namespace (including our vmalert-rules.yaml VMRule CR).
# ---------------------------------------------------------------------------
vmalert:
enabled: true
spec:
selectAllByDefault: true
evaluationInterval: 20s
notifiers:
- url: http://vm-victoria-metrics-k8s-stack-alertmanager.observability.svc.cluster.local:9093
resources:
requests:
cpu: 50m
memory: 64Mi
limits:
cpu: 200m
memory: 128Mi
# ---------------------------------------------------------------------------
# Alertmanager — 1 replica; inline config with a single n8n webhook receiver.
# n8n runs on the docker host and fans out to email/Telegram/whatever.
# ---------------------------------------------------------------------------
alertmanager:
enabled: true
spec:
replicaCount: 1
port: "9093"
selectAllByDefault: true
storage:
volumeClaimTemplate:
spec:
storageClassName: longhorn
accessModes:
- ReadWriteOnce
resources:
requests:
storage: 2Gi
resources:
requests:
cpu: 50m
memory: 64Mi
limits:
cpu: 200m
memory: 128Mi
config:
route:
receiver: n8n-webhook
group_wait: 30s
group_interval: 5m
repeat_interval: 4h
receivers:
- name: n8n-webhook
webhook_configs:
- url: http://n8n.lan/webhook/observability-alert
send_resolved: true
extraRules:
node-alerts:
groups:
- name: node
rules:
- alert: HighNodeCPU
expr: 100 - (avg by (instance) (rate(node_cpu_seconds_total{mode="idle"}[5m])) * 100) > 80
for: 10m
labels:
severity: warning
annotations:
summary: High CPU on {{ $labels.instance }}
description: CPU usage above 80% for 10 minutes.
- alert: HighNodeRAM
expr: 100 - ((node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes) * 100) > 85
for: 10m
labels:
severity: warning
annotations:
summary: High RAM on {{ $labels.instance }}
description: RAM usage above 85% for 10 minutes.
- alert: LowDiskSpace
expr: |
100 - ((node_filesystem_avail_bytes{mountpoint!~"/run.*|/var/lib/docker.*"} /
node_filesystem_size_bytes{mountpoint!~"/run.*|/var/lib/docker.*"}) * 100) > 85
for: 10m
labels:
severity: warning
annotations:
summary: Low disk space on {{ $labels.instance }} {{ $labels.mountpoint }}
description: Disk usage above 85% for 10 minutes.
sops:
lastmodified: "2026-07-28T16:39:09Z"
mac: ENC[AES256_GCM,data:Ymxmmkui0Nuv35Yq6v4BXM4xGhvjAufexW7fIm9fUIdTTXZ3e10PitNCZXKiaDwwCJfdEkNZri7jPAYzhTzKMIdcTKnOMImsMjdiJZwRwwbKcYVay0A6KSLJ8eFjDRRYhlCXbsAIRlQHpFh4UkssKdu40mIiudU0qIVm04ppbno=,iv:R8kpITPYSbQ1yZw5IK0fArssxUtIjNmvzknOrLJQcuk=,tag:/CRcr1VZAIKIPbcag05uTw==,type:str]
pgp:
- created_at: "2026-07-28T16:39:08Z"
enc: |-
-----BEGIN PGP MESSAGE-----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=
=pmSq
-----END PGP MESSAGE-----
fp: DC6910268E657FF70BA7EC289974494E76938DDC
encrypted_regex: ^(password|value|ssh-key|api-key|user|username|privateKey|clientSecret|clientId|apiKey|extraArgs.*|.*Secret.*|extraEnvVars|.*SECRET.*|.*secret.*|key|.*Password|.*\.ya?ml)$
version: 3.10.2