iDRAC Inlet Temperature Monitor

A production‑ready, offline‑friendly iDRAC temperature monitoring system with a modern web UI, combined email alerts, spike‑suppression logic, historical graphing, and per‑server log export.

Designed for datacenter / lab environments where:

• Internet access may be restricted
• Accuracy and alert discipline matter
• Operators need clear visibility, not noise

---

Key Features

Multi‑Server Monitoring

• Monitor multiple Dell iDRAC endpoints at once
• Same credentials supported across all servers
• Uses Redfish API over HTTPS

Real‑Time Dashboard (Web UI)

• Instant temperature visibility on page load
• Server cards with:
  • Current inlet temperature
  • Health status (NORMAL / WARNING / CRITICAL)
  • Last update timestamp
• All primary information fits on ONE desktop screen
• No scrolling unless logs are opened

Historical Graph (Per Server)

• View last 1 hour temperature trend
• Server‑selectable via dropdown
• Fixed‑height chart (no stretching / layout issues)
• Works fully offline using a locally hosted Chart.js

Integrated Logs (User‑Friendly)

• View recent temperature logs directly in the browser
• Filter logs by selected server
• Logs are hidden by default (shown only on demand)
• Download CSV logs:
  • Per server
  • Or all servers combined

Smart Email Alerting (No Spam)

• One single email, not one per server
• Clear HTML table with all servers included
• Two types of emails:
  • Hourly summary
  • Sustained alert

Spike‑Suppression Logic (Critical Feature)

Prevents false alerts caused by short temperature spikes.

Behavior:

1. Sudden spike (e.g. 25 °C → 30 °C)
   → No alert
2. Temperature remains high for 5 continuous minutes
   → Alert is sent
3. Temperature remains high after alert
   → No repeated alerts
4. After 30 minutes still high
   → Another alert
5. Temperature returns to normal
   → Alert system resets

This mirrors real NOC / datacenter alert practice.

---

Project Structure

.
├── app.py                     # Flask backend + monitor logic
├── .env                       # Configuration and credentials
├── storage/
│   ├── temperature.log        # Human‑readable log
│   ├── idrac_log.csv           # CSV data for graphs & export
│   └── app.log                # Application log
├── templates/
│   └── dashboard.html         # Web UI
└── static/
    └── chart.umd.min.js       # Chart.js (local, offline‑safe)

---

Configuration (.env)

Example:

IDRAC_SERVERS=10.129.16.81,10.129.16.82,10.129.16.84
IDRAC_USER=root
IDRAC_PASS=yourpassword

WARNING_TEMP=25
CRITICAL_TEMP=30

SAMPLE_INTERVAL_SEC=10

MAIL_FROM_ADDRESS=noreply@example.com
MAIL_FROM_NAME="iDRAC Monitor"
EMAIL_TO=admin1@example.com,admin2@example.com
MAIL_HOST=mail.example.com
MAIL_PORT=25

---

Web UI Overview

Temperature Cards (Top Priority)

• Shows current inlet temperature
• Color‑coded:
  • ✅ Green – NORMAL
  • ⚠️ Orange – WARNING
  • 🔴 Red – CRITICAL
• This is what operators should look at first

Graph Section

• Select a server from dropdown
• Shows last 1 hour temperature history
• Fixed size, stable layout
• Updated manually (no auto‑reset)

Logs Section

• Hidden by default
• View logs for the selected server
• Scrollable only when opened
• Download logs as CSV

---

API Endpoints (For Developers)

Endpoint	Method	Description
/	GET	Dashboard UI
/api/state	GET	Live state of all servers
/api/history?host=IP	GET	1‑hour history for graph
/api/logs?host=IP	GET	Recent logs for selected server
/api/export?host=IP	GET	Download CSV logs

---

Email Behavior

Hourly Summary

• Sent exactly on HH:00
• One email only
• Includes all servers in a table

Alert Email

• Sent only after sustained high temp (≥ 5 minutes)
• Sent once per incident
• Re‑sent only if condition persists for 30 minutes
• Automatically resets when temperature normalizes

---

Offline / Restricted Network Ready

• No external CDN usage
• Chart.js is served locally from /static
• No internet required after setup
• Suitable for air‑gapped environments

---

Running the Application

python app.py

Access the dashboard:

http://localhost:5000

---

Why This Design Works

• User‑first UI: temperature first, logs second
• Alert discipline: no spam, no noise
• Operational realism: mirrors real datacenter monitoring logic
• Maintainable: clean Flask backend, simple frontend
• Safe: no reliance on external networks

---

Possible Future Enhancements

• Threshold lines on graph (warning / critical)
• Multi‑server combined graph
• 6h / 24h history selector
• Attach graph image to emails
• Authentication for dashboard access

---

License / Usage

Internal tooling / educational use.
Adapt and extend as needed for your environment.