titan-iac/services/monitoring/dcgm-exporter.yaml

80 lines
1.9 KiB
YAML
Raw Normal View History

# services/monitoring/dcgm-exporter.yaml
apiVersion: apps/v1
kind: DaemonSet
metadata:
name: dcgm-exporter
namespace: monitoring
labels:
app: dcgm-exporter
spec:
selector:
matchLabels:
app: dcgm-exporter
updateStrategy:
rollingUpdate:
maxUnavailable: 2
template:
metadata:
labels:
app: dcgm-exporter
annotations:
prometheus.io/scrape: "true"
prometheus.io/port: "9400"
spec:
serviceAccountName: default
2026-01-14 10:07:31 -03:00
imagePullSecrets:
- name: harbor-regcred
runtimeClassName: nvidia
affinity:
nodeAffinity:
requiredDuringSchedulingIgnoredDuringExecution:
nodeSelectorTerms:
- matchExpressions:
- key: kubernetes.io/arch
operator: In
values:
- amd64
- key: jetson
operator: NotIn
values:
- "true"
2026-06-09 00:46:46 -03:00
- key: veles.bstein.dev/node-pool
operator: NotIn
values:
- oceanus
- key: node-role.kubernetes.io/accelerator
operator: Exists
tolerations:
- operator: Exists
containers:
- name: dcgm-exporter
2025-12-16 00:08:11 -03:00
image: registry.bstein.dev/monitoring/dcgm-exporter:4.4.2-4.7.0-ubuntu22.04
imagePullPolicy: Always
ports:
- name: metrics
containerPort: 9400
securityContext:
privileged: true
resources:
requests:
cpu: 50m
memory: 512Mi
limits:
cpu: 500m
memory: 1Gi
---
apiVersion: v1
kind: Service
metadata:
name: dcgm-exporter
namespace: monitoring
labels:
app: dcgm-exporter
spec:
selector:
app: dcgm-exporter
ports:
- name: metrics
port: 9400
targetPort: metrics