mirror of
https://github.com/Alkatrazz24/Funk-lab.git
synced 2026-07-08 05:34:43 +02:00
649c6fde8e
Sondes actives ICMP/HTTP/DNS qui complètent le monitoring passif
(node_exporter + up{}) : joignabilité hôtes/passerelle, accès internet,
santé DNS *.lab.local + forwarding, latence LAN, services de bout en
bout via Traefik.
- blackbox_exporter (chart 11.13.0, ArgoCD, ns monitoring) — ICMP via
NET_RAW, DNS du pod = dnsmasq (chemin réel lab → Traefik)
- 3 jobs blackbox-icmp/-http/-dns (additionalScrapeConfigs)
- alerts-network.yaml : internet, DNS, passerelle, hôtes, services,
latence, exporter (→ pipeline AlertManager existant)
- Asa : outil network_status + contexte présélectionnable reseau
- doc admin/infra/reseau.md + CLAUDE.md + progress
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Rc3dLKfqEPHexvFCCx5ZEF
237 lines
6.1 KiB
YAML
237 lines
6.1 KiB
YAML
# kube-prometheus-stack — cluster Funk
|
|
# Ressources limitées pour compute-02/03 (8 GB RAM)
|
|
|
|
grafana:
|
|
initChownData:
|
|
enabled: false
|
|
grafana.ini:
|
|
database:
|
|
type: postgres
|
|
host: 192.168.10.1:5432
|
|
name: grafana
|
|
user: grafana
|
|
envFromSecrets:
|
|
- name: grafana-db-secret
|
|
ingress:
|
|
enabled: true
|
|
ingressClassName: traefik
|
|
hosts:
|
|
- grafana.lab.local
|
|
adminPassword: "funk-grafana"
|
|
resources:
|
|
requests:
|
|
cpu: 100m
|
|
memory: 256Mi
|
|
limits:
|
|
memory: 1Gi
|
|
persistence:
|
|
enabled: true
|
|
storageClassName: nfs
|
|
size: 2Gi
|
|
|
|
prometheus:
|
|
ingress:
|
|
enabled: true
|
|
ingressClassName: traefik
|
|
hosts:
|
|
- prometheus.lab.local
|
|
paths:
|
|
- /
|
|
pathType: Prefix
|
|
|
|
prometheusSpec:
|
|
retention: 15d
|
|
retentionSize: "8GB"
|
|
resources:
|
|
requests:
|
|
cpu: 200m
|
|
memory: 512Mi
|
|
limits:
|
|
memory: 1Gi
|
|
storageSpec:
|
|
volumeClaimTemplate:
|
|
spec:
|
|
storageClassName: nfs
|
|
accessModes: ["ReadWriteOnce"]
|
|
resources:
|
|
requests:
|
|
storage: 20Gi
|
|
|
|
# Scrape des nœuds hors cluster (storage-01, gpu-01)
|
|
additionalScrapeConfigs:
|
|
- job_name: storage-01
|
|
static_configs:
|
|
- targets: ['192.168.10.1:9100']
|
|
labels:
|
|
instance: storage-01
|
|
role: gateway
|
|
|
|
- job_name: gpu-01-node
|
|
static_configs:
|
|
- targets: ['192.168.10.20:9100']
|
|
labels:
|
|
instance: gpu-01
|
|
role: gpu
|
|
|
|
- job_name: llama-server-gpu
|
|
static_configs:
|
|
- targets: ['192.168.10.20:1234']
|
|
labels:
|
|
instance: gpu-01
|
|
model: qwen3-8b
|
|
type: gpu
|
|
|
|
# --- Sondes réseau actives (blackbox_exporter) ---
|
|
# Le relabeling envoie la cible réelle en ?target=… et fait pointer le
|
|
# scrape vers le service blackbox. Cf. admin/infra/reseau.md.
|
|
- job_name: blackbox-icmp
|
|
metrics_path: /probe
|
|
params:
|
|
module: [icmp]
|
|
static_configs:
|
|
- targets:
|
|
- 192.168.10.1 # storage-01 — passerelle / NAT / DNS
|
|
- 192.168.10.11 # compute-01 (control-plane)
|
|
- 192.168.10.12 # compute-02
|
|
- 192.168.10.13 # compute-03
|
|
- 192.168.10.20 # gpu-01
|
|
labels:
|
|
scope: lan
|
|
- targets:
|
|
- 1.1.1.1 # Cloudflare — joignabilité internet
|
|
- 9.9.9.9 # Quad9 — joignabilité internet
|
|
labels:
|
|
scope: internet
|
|
relabel_configs:
|
|
- source_labels: [__address__]
|
|
target_label: __param_target
|
|
- source_labels: [__param_target]
|
|
target_label: instance
|
|
- target_label: __address__
|
|
replacement: blackbox-exporter.monitoring.svc.cluster.local:9115
|
|
|
|
- job_name: blackbox-http
|
|
metrics_path: /probe
|
|
params:
|
|
module: [http_2xx]
|
|
static_configs:
|
|
- targets:
|
|
- http://grafana.lab.local
|
|
- http://prometheus.lab.local
|
|
- http://alertmanager.lab.local
|
|
- http://argocd.lab.local
|
|
- http://openwebui.lab.local
|
|
- http://n8n.lab.local
|
|
- http://ghostfolio.lab.local
|
|
- http://searxng.lab.local
|
|
- http://stt.lab.local
|
|
labels:
|
|
scope: ingress
|
|
relabel_configs:
|
|
- source_labels: [__address__]
|
|
target_label: __param_target
|
|
- source_labels: [__param_target]
|
|
target_label: instance
|
|
- target_label: __address__
|
|
replacement: blackbox-exporter.monitoring.svc.cluster.local:9115
|
|
|
|
# DNS : module choisi par cible (label module → __param_module), serveur
|
|
# interrogé = dnsmasq (192.168.10.1:53). dns_lab teste le wildcard
|
|
# *.lab.local, dns_external teste le forwarding vers l'upstream.
|
|
- job_name: blackbox-dns
|
|
metrics_path: /probe
|
|
static_configs:
|
|
- targets: ['192.168.10.1:53']
|
|
labels:
|
|
module: dns_lab
|
|
instance: 'dnsmasq:grafana.lab.local'
|
|
- targets: ['192.168.10.1:53']
|
|
labels:
|
|
module: dns_external
|
|
instance: 'dnsmasq:one.one.one.one'
|
|
relabel_configs:
|
|
- source_labels: [module]
|
|
target_label: __param_module
|
|
- source_labels: [__address__]
|
|
target_label: __param_target
|
|
- target_label: __address__
|
|
replacement: blackbox-exporter.monitoring.svc.cluster.local:9115
|
|
|
|
alertmanager:
|
|
ingress:
|
|
enabled: true
|
|
ingressClassName: traefik
|
|
hosts:
|
|
- alertmanager.lab.local
|
|
paths:
|
|
- /
|
|
pathType: Prefix
|
|
|
|
alertmanagerSpec:
|
|
resources:
|
|
requests:
|
|
cpu: 50m
|
|
memory: 64Mi
|
|
limits:
|
|
memory: 128Mi
|
|
config:
|
|
global:
|
|
resolve_timeout: 5m
|
|
route:
|
|
group_by: ['alertname', 'instance']
|
|
group_wait: 30s
|
|
group_interval: 5m
|
|
repeat_interval: 4h
|
|
receiver: hermes-monitor
|
|
routes:
|
|
# Watchdog (heartbeat interne kube-prometheus-stack) — silencé
|
|
- matchers:
|
|
- alertname = "Watchdog"
|
|
receiver: "null"
|
|
repeat_interval: 24h
|
|
receivers:
|
|
- name: "null"
|
|
- name: hermes-monitor
|
|
webhook_configs:
|
|
- url: 'http://192.168.10.1:9093/webhook'
|
|
send_resolved: true
|
|
- url: 'http://n8n.lab.local/webhook/alerts'
|
|
send_resolved: true
|
|
|
|
prometheusOperator:
|
|
resources:
|
|
requests:
|
|
cpu: 100m
|
|
memory: 128Mi
|
|
limits:
|
|
memory: 256Mi
|
|
|
|
kube-state-metrics:
|
|
resources:
|
|
requests:
|
|
cpu: 50m
|
|
memory: 64Mi
|
|
limits:
|
|
memory: 128Mi
|
|
|
|
nodeExporter:
|
|
resources:
|
|
requests:
|
|
cpu: 50m
|
|
memory: 32Mi
|
|
limits:
|
|
memory: 64Mi
|
|
|
|
prometheus-node-exporter:
|
|
extraArgs:
|
|
- --collector.rapl
|
|
|
|
# Désactiver etcd/scheduler monitoring (Talos — non accessible)
|
|
kubeEtcd:
|
|
enabled: false
|
|
kubeScheduler:
|
|
enabled: false
|
|
kubeControllerManager:
|
|
enabled: false
|
|
kubeProxy:
|
|
enabled: false
|