Preface#

In a previous post, I introduced how to deploy Scrutiny with Docker to monitor your hard drives. Now we’ll further refine and improve the deployment process.

Why Scrutiny#

Scrutiny is a Docker-based disk health monitoring tool. It collects disk information via SMART (Self-Monitoring, Analysis and Reporting Technology) data and presents trends, alerts, and other metrics through a web interface.

Key advantages include:

Supports multiple drive types (HDD, SSD, NVMe)
Visual dashboards with trend charts for temperature, health status, read/write errors, and more
Supports InfluxDB as time-series storage for long-term monitoring
Can integrate with other systems via API

Deployment#

Install Scrutiny#

1
mkdir -p /opt/scrutiny/{influxdb,config} && cd /opt/scrutiny
2
nano docker-compose.yml

1
version: '2.4'
2

3
services:
4
  scrutiny-influxdb:
5
    image: influxdb:2.2
6
    container_name: scrutiny-influxdb
7
    ports:
8
      - 8086:8086
9
    restart: unless-stopped
10
    volumes:
11
      - ./influxdb:/var/lib/influxdb2
12
    environment:
13
      TZ: Asia/Kolkata # CHANGE THIS
14
    healthcheck:
15
      test: ["CMD", "curl", "-f", "http://localhost:8086/health"]
16
      interval: 5s
17
      timeout: 10s
18
      retries: 20
19

20
  scrutiny-web:
21
    image: 'ghcr.io/analogj/scrutiny:master-web'
22
    container_name: scrutiny-web
23
    ports:
24
      - 8080:8080
25
    restart: unless-stopped
26
    volumes:
27
      - ./config:/opt/scrutiny/config
28
    environment:
29
      SCRUTINY_WEB_INFLUXDB_HOST: scrutiny-influxdb
30
      TZ: Asia/Kolkata # CHANGE THIS
31
    depends_on:
32
      scrutiny-influxdb:
33
        condition: service_healthy
34
    healthcheck:
35
      test: ["CMD", "curl", "-f", "http://localhost:8080/api/health"]
36
      interval: 5s
37
      timeout: 10s
38
      retries: 20
39
      start_period: 10s

Start Scrutiny#

1
docker-compose up -d

The Scrutiny web service runs on port 8080 by default
InfluxDB runs on port 8086 by default
You can open a browser and visit http://your-server-ip:8080 to access the Scrutiny web interface

PVE Collection#

Run the following on the PVE node:

1
cd /opt/scrutiny
2
nano scrutiny-collector-install.sh

Replace API_ENDPOINT with the IP address of your Scrutiny container.

1
#!/bin/bash
2

3
API_ENDPOINT="http://192.168.XX.XX:8080"
4

5
apt install smartmontools -y
6

7
INSTALL_DIR="/opt/scrutiny"
8
BIN_DIR="$INSTALL_DIR/bin"
9
LATEST_RELEASE_URL=$(curl -s https://api.github.com/repos/AnalogJ/scrutiny/releases/latest | grep "browser_download_url.*scrutiny-collector-metrics-linux-amd64" | cut -d '"' -f 4)
10

11
mkdir -p $BIN_DIR
12
curl -L $LATEST_RELEASE_URL -o $BIN_DIR/scrutiny-collector-metrics-linux-amd64
13
chmod +x $BIN_DIR/scrutiny-collector-metrics-linux-amd64
14

15
mkdir -p /root/scrutiny
16
cat << EOF > /root/scrutiny/scrutiny.sh
17
#!/bin/bash
18
/opt/scrutiny/bin/scrutiny-collector-metrics-linux-amd64 run --api-endpoint "$API_ENDPOINT" 2>&1 | tee /var/log/scrutiny.log
19
EOF
20
chmod +x /root/scrutiny/scrutiny.sh
21

22
cat << 'EOF' > /etc/systemd/system/scrutiny.service
23
[Unit]
24
Description=Scrutiny Collector
25
Requires=scrutiny.timer
26

27
[Service]
28
Type=simple
29
ExecStart=/root/scrutiny/scrutiny.sh
30
User=root
31
EOF
32

33
cat << 'EOF' > /etc/systemd/system/scrutiny.timer
34
[Unit]
35
Description=Timer for the scrutiny.service
36

37
[Timer]
38
Unit=scrutiny.service
39
OnBootSec=5min
40
OnUnitActiveSec=60min
41

42
[Install]
43
WantedBy=timers.target
44
EOF
45

46
systemctl enable scrutiny.timer
47
systemctl start scrutiny.timer
48
systemctl status scrutiny.timer

Run the Script#

1
chmod +x scrutiny-collector-install.sh
2
./scrutiny-collector-install.sh

Monitoring Disk Health with Scrutiny in a Proxmox VE Environment