Windows Home Server — Part 12: Replacing Datadog & CloudWatch (Bare-Metal Observability)

Renting cloud observability infrastructure (like Datadog, New Relic, or AWS CloudWatch) introduces significant licensing and custom metric ingestion costs.

We can run an enterprise-grade observability stack natively on Windows for free. We will install Prometheus alongside Windows Exporter to collect hardware metrics (CPU, RAM, disk read/write times, and network bandwidth). We will then install Grafana to visualize these metrics, and set up Loki with Promtail to aggregate application logs from our C:\Server\logs\ folder—all running natively in the background.

---

1. Metric Collection: Prometheus & Windows Exporter

Windows Exporter (formerly WMI Exporter) runs as a background service on Windows, collecting operating system metrics and exposing them on port 9182 for Prometheus to read.

Step 1: Install Windows Exporter

Download and install it via winget:

winget install --id Prometheus.WindowsExporter -e --accept-source-agreements --accept-package-agreements

This automatically registers and starts a Windows Service named windows_exporter listening on http://localhost:9182/metrics.

Step 2: Install Prometheus

Download and extract Prometheus for Windows:

New-Item -ItemType Directory -Force -Path "C:\Server\prometheus"
Invoke-WebRequest -Uri "https://github.com/prometheus/prometheus/releases/download/v2.50.0/prometheus-2.50.0.windows-amd64.zip" -OutFile "C:\Server\prometheus\prometheus.zip"
Expand-Archive -Path "C:\Server\prometheus\prometheus.zip" -DestinationPath "C:\Server\prometheus" -Force

# Clean up paths
Move-Item "C:\Server\prometheus\prometheus-2.50.0.windows-amd64\prometheus.exe" -Destination "C:\Server\bin\prometheus.exe" -Force
Move-Item "C:\Server\prometheus\prometheus-2.50.0.windows-amd64\promtool.exe" -Destination "C:\Server\bin\promtool.exe" -Force
Remove-Item "C:\Server\prometheus\prometheus-2.50.0.windows-amd64" -Recurve -Force
Remove-Item "C:\Server\prometheus\prometheus.zip" -Force

Step 3: Configure Prometheus Scrape Targets

Create C:\Server\prometheus\prometheus.yml:

global:
  scrape_interval: 15s
  evaluation_interval: 15s

scrape_configs:
  # Host metrics (Windows Exporter)
  - job_name: "windows-host"
    static_configs:
      - targets: ["localhost:9182"]

  # Uptime Kuma metrics
  - job_name: "uptime-kuma"
    metrics_path: "/metrics"
    static_configs:
      - targets: ["localhost:3001"]

Step 4: Wrap Prometheus in NSSM

Create the service:

nssm install Prometheus C:\Server\bin\prometheus.exe "--config.file C:\Server\prometheus\prometheus.yml --storage.tsdb.path C:\Server\data\prometheus_data"
nssm set Prometheus AppDirectory "C:\Server\prometheus"
nssm start Prometheus

---

2. Visualization Tier: Grafana

Grafana connects to Prometheus and presents metrics in clean dashboards.

Step 1: Install Grafana

Download and run the installer:

# Install Grafana via winget
winget install --id GrafanaLabs.Grafana -e --accept-source-agreements --accept-package-agreements

This installs Grafana as a system service automatically. Open your browser and navigate to http://localhost:3000 (default credentials: admin / admin), and click Add your first data source → select Prometheus → set URL to http://localhost:9090 → click Save & test.

---

3. Log Aggregation: Loki & Promtail

Instead of parsing files locally, we will use Loki to index logs, and Promtail to parse them and forward them to Loki.

Step 1: Download Loki and Promtail

Run in PowerShell:

New-Item -ItemType Directory -Force -Path "C:\Server\loki"
New-Item -ItemType Directory -Force -Path "C:\Server\promtail"

# Download Loki
Invoke-WebRequest -Uri "https://github.com/grafana/loki/releases/download/v2.9.4/loki-windows-amd64.exe.zip" -OutFile "C:\Server\loki\loki.zip"
Expand-Archive -Path "C:\Server\loki\loki.zip" -DestinationPath "C:\Server\loki" -Force
Move-Item "C:\Server\loki\loki-windows-amd64.exe" -Destination "C:\Server\bin\loki.exe" -Force
Remove-Item "C:\Server\loki\loki.zip" -Force

# Download Promtail
Invoke-WebRequest -Uri "https://github.com/grafana/loki/releases/download/v2.9.4/promtail-windows-amd64.exe.zip" -OutFile "C:\Server\promtail\promtail.zip"
Expand-Archive -Path "C:\Server\promtail\promtail.zip" -DestinationPath "C:\Server\promtail" -Force
Move-Item "C:\Server\promtail\promtail-windows-amd64.exe" -Destination "C:\Server\bin\promtail.exe" -Force
Remove-Item "C:\Server\promtail\promtail.zip" -Force

Step 2: Configure Loki

Create C:\Server\loki\loki-config.yml:

auth_enabled: false

server:
  http_listen_port: 3100

common:
  path_prefix: C:\Server\data\loki_data
  storage:
    filesystem:
      chunks_directory: C:\Server\data\loki_data\chunks
      rules_directory: C:\Server\data\loki_data\rules
  replication_factor: 1
  ring:
    kvstore:
      store: inmemory

schema_config:
  configs:
    - from: 2020-10-24
      store: tsdb
      object_store: filesystem
      schema: v13
      index:
        prefix: index_
        period: 24h

Step 3: Configure Promtail to Tail Logs

Create C:\Server\promtail\promtail-config.yml. This script instructs Promtail to tail log files in C:\Server\logs\ and send them to Loki:

server:
  http_listen_port: 9080
  grpc_listen_port: 0

positions:
  filename: C:\Server\data\promtail_positions.yaml

clients:
  - url: http://localhost:3100/loki/api/v1/push

scrape_configs:
  - job_name: server-logs
    static_configs:
      - targets:
          - localhost
        labels:
          job: app-logs
          __path__: C:\Server\logs\*.log

Step 4: Wrap Loki and Promtail in NSSM

Create the services:

# Wrap Loki
nssm install Loki C:\Server\bin\loki.exe "-config.file C:\Server\loki\loki-config.yml"
nssm set Loki AppDirectory "C:\Server\loki"
nssm start Loki

# Wrap Promtail
nssm install Promtail C:\Server\bin\promtail.exe "-config.file C:\Server\promtail\promtail-config.yml"
nssm set Promtail AppDirectory "C:\Server\promtail"
nssm start Promtail

Step 5: Add Loki Data Source in Grafana

1. Go to Grafana → Data Sources → Add new data source → select Loki.
2. Set URL to http://localhost:3100 and click Save & test.
3. Go to the Explore tab in Grafana, select Loki, and run the query {job="app-logs"}. You will see a live log stream from all Caddy, Node, Python, and DB logs—eliminating Datadog and CloudWatch fees.

---

In the next part, we will secure client access and create private overlay networks for free using Tailscale.

Proceed to Part 13: Replacing Paid VPNs & Team Access (Private WireGuard & Tailscale) →