Prometheus Exporter for Vultr

Metrics

Metrics are all prefixed vultr_

Name	Type	Description
block_storage_up	Counter	Number of Block Storage volumes
block_storage_size	Gauge	Size (GB) of Block Storage volumes
exporter_build_info	Counter	Build status (1=running)
exporter_start_time	Gauge	Start time (Unix epoch) of Exporter
kubernetes_cluster_up	Counter	Number of Kubernetes clusters
kubernetes_node_pool	Gauge	Number of Kubernetes cluster Node Pools
kubernetes_node	Gauge	Number of Kubernetes Cluster Nodes
load_balancer_up	Number of Load Balancers
load_balancer_instances	Number of Load Balancer instances
reserved_ips_up	Counter	Number of Reserved IPs

Image

• ghcr.io/dazwilkin/vultr-exporter:334032c9cba9edbde700edf1a5c4c5786104ca68

API Key

The Exporter needs access to your Vultr API Key

export API_KEY="[YOUR-API-KEY]"

Image

• ghcr.io/dazwilkin/vultr-exporter:334032c9cba9edbde700edf1a5c4c5786104ca68

API Key

The Exporter needs access to your Vultr API Key

export API_KEY="[YOUR-API-KEY]"

Go

export API_KEY="[YOUR-API-KEY]"

go run ./cmd/server \
--endpoint=0.0.0.0:8080 \
--path=/metrics

Container

API_KEY="[YOUR-API-KEY]"

IMAGE="ghcr.io/dazwilkin/vultr-exporter:334032c9cba9edbde700edf1a5c4c5786104ca68"

podman run \
--interactive --tty --rm \
--publish=8080:8080 \
--env=API_KEY=${API_KEY} \
${IMAGE} \
  --endpoint=0.0.0.0:8080 \
  --path=/metrics

Kubernetes

NOTE If running vult-exporter on VKE, ensure that you're API_KEY includes the public IP addresses of the cluster's nodes as these will be originating Vultr API requests. I think these access control changes can't be done programmatically.

API_KEY="[YOUR-API-KEY]"

IMAGE="ghcr.io/dazwilkin/vultr-exporter:334032c9cba9edbde700edf1a5c4c5786104ca68"

NAMESPACE="exporter"

kubectl create namespace ${NAMESPACE}

kubectl create secret generic vultr \
--namespace=${NAMESPACE} \
--from-literal=apiKey=${API_KEY}

echo "
apiVersion: v1
kind: List
metadata: {}
items:
  - kind: Service
    apiVersion: v1
    metadata:
      labels:
        app: vultr-exporter
      name: vultr-exporter
    spec:
      selector:
        app: vultr-exporter
      ports:
        - name: http
          port: 8080
          targetPort: 8080
  - kind: Deployment
    apiVersion: apps/v1
    metadata:
      labels:
        app: vultr-exporter
      name: vultr-exporter
    spec:
      replicas: 1
      selector:
        matchLabels:
          app: vultr-exporter
      template:
        metadata:
          labels:
            app: vultr-exporter
        spec:
          containers:
            - name: vultr-exporter
              image: ${IMAGE}
              command:
              - /server
              args:
              - --endpoint=0.0.0.0:8080
              - --path=/metrics
              env:
                - name: API_KEY
                  valueFrom:
                    secretKeyRef:
                      name: vultr
                      key: apiKey
                      optional: false
              ports:
                - name: metrics
                  containerPort: 8080
          restartPolicy: Always
" | kubectl apply --filename=- --namespace=${NAMESPACE}

# Use your preferred HTTP Load-balancer
kubectl port-forward deployment/vultr-exporter 8080:8080 \
--namespace=${NAMESPACE}

# To use a Vultr Load balancer
# Replaces the service created above w/ a Vultr Load balancer
echo "
apiVersion: v1
kind: Service
metadata:
  annotations:
    service.beta.kubernetes.io/vultr-loadbalancer-protocol: "http"
  name: vultr-exporter
spec:
  type: LoadBalancer
  selector:
    app: vultr-exporter
  ports:
    - name: http
      port: 80
      targetPort: 8080
" | kubectl apply --filename=- --namespace=${NAMESPACE}

Raspberry Pi 4

if [ "$(getconf LONG_BIT)" -eq 64 ]
then
  # 64-bit Raspian
  ARCH="GOARCH=arm64"
  TAG="arm64"
else
  # 32-bit Raspian
  ARCH="GOARCH=arm GOARM=7"
  TAG="arm32v7"
fi

podman build \
--build-arg=GOLANG_OPTIONS="CGO_ENABLED=0 GOOS=linux ${ARCH}" \
--build-arg=COMMIT=$(git rev-parse HEAD) \
--build-arg=VERSION=$(uname --kernel-release) \
--tag=ghcr.io/dazwilkin/vultr-exporter:${TAG} \
--file=./Dockerfile \
.

Prometheus

global:
  scrape_interval: 1m
  evaluation_interval: 1m

rule_files:
- "/etc/alertmanager/rules.yml"

  # Vultr Exporter
- job_name: "vultr-exporter"
  static_configs:
  - targets:
    - "localhost:8080"

Alertmanager

groups:
- name: vultr_exporter
  rules:
  - alert: vultr_kubernetes_cluster_up
    expr: vultr_kubernetes_cluster_up{} > 0
    for: 6h
    labels:
      severity: page
    annotations:
      summary: Vultr Kubernetes Engine clusters

Sigstore

vultr-exporter container images are being signed by Sigstore and may be verified:

cosign verify \
--key=./cosign.pub \
ghcr.io/dazwilkin/vultr-exporter:334032c9cba9edbde700edf1a5c4c5786104ca68

NOTE cosign.pub may be downloaded here

To install cosign, e.g.:

go install github.com/sigstore/cosign/cmd/cosign@latest

References

• Vultr API: List all Kubernetes Clusters
• govultr SDK
• pkg.go.dev: Cluster
• pkg.go.dev: NodePool

Similar Exporters

• Prometheus Exporter for Azure
• Prometheus Exporter for Fly.io
• Prometheus Exporter for GCP
• Prometheus Exporter for Koyeb
• Prometheus Exporter for Linode

---