CloudOps Sentinel: Docker Fleet & Infrastructure Cockpit
Full-stack operational monitoring dashboard providing real-time server cluster metrics, Docker container lifecycle controls, and automated database retention routines.
Executive Overview & Context
CloudOps Sentinel is a full-stack infrastructure cockpit engineered to monitor fleet metrics, supervise Docker containers, and automate backup cycles. Features real-time CPU/memory meters, Docker container start/stop/restart management, automated SQLite database retention with pre-cleanup SQL dumps, and Trivy CVE report ingestion.
Project Objectives
1. Build authenticated telemetry endpoints interfacing with the Linux kernel and Docker socket. 2. Render live CPU, RAM, disk, and container lifecycle telemetry in a React dashboard. 3. Automate SQLite database retention routines with pre-cleanup SQL dumps. 4. Ingest and display Trivy vulnerability trends.
Challenges & Pain Points
Replaced disjointed command-line utilities (docker ps, htop, df, systemctl) with a consolidated, secure web cockpit for swift diagnosis during incident responses.
Engineered Implementation
Built a Node.js/Express backend communicating directly with /proc system statistics and the Docker UNIX domain socket (/var/run/docker.sock). Connected to a React single-page frontend with responsive metric cards, chart visualizations, and process controls.
Key Engineering Features
System Topology & Traffic Flow
React dashboard client → Role-gated Node.js REST API → Linux /proc telemetry & Docker Engine socket (/var/run/docker.sock) → Local SQLite persistent store.
Sentinel Fleet Telemetry Architecture
Role-gated monitoring console with Docker socket IPC and local SQLite persistence
Sentinel React UI
Nginx Ingress Proxy
PM2 Process Supervisor
Express Monitoring API
Docker Engine Daemon
Persistent SQLite Store
GitHub Actions CI/CD
Email & Backup Reporter
| # | From → To | Flow / Action | Protocol / Type |
|---|---|---|---|
| 1 | Sentinel React UI→Nginx Ingress Proxy | HTTPS Admin Route | HTTPS / TLS 1.3 |
| 2 | Nginx Ingress Proxy→PM2 Process Supervisor | Supervised Proxy Pass | HTTP / localhost:5000 |
| 3 | PM2 Process Supervisor→Express Monitoring API | Process Management | Node.js IPC |
| 5 | Express Monitoring API→Docker Engine Daemon | Engine Socket IPC | UNIX Domain Socket |
| 4 | Express Monitoring API→Persistent SQLite Store | Telemetry & Logs | WAL Mode SQL |
| 6 | GitHub Actions CI/CD→PM2 Process Supervisor | CI/CD SSH Deploy | SSH (Port 22) |
| A | Email & Backup Reporter→Express Monitoring API | Alert Dispatch & Backupsecondary | SMTP 465 / Mail |
Inspect Detailed Cloud & Security Architecture
Explore full multi-layer topology, live active ports, security controls, and verified repository evidence in the Architecture Explorer.
Implementation Details & Workflow
Implemented Node.js child-process supervisors and stream readers to process /proc/stat, /proc/meminfo, and Docker socket events without CPU-intensive polling.
CI/CD & Automation Pipeline
Continuous integration runs automated Dockerfile linting (Hadolint), vulnerability scans with Trivy, and unit tests validating API endpoint authentication.
Security & Hardening Controls
Secured access to /var/run/docker.sock by running the service within a dedicated group with least-privilege POSIX permissions. Implemented JWT token expiration and rate limiting.
Technologies & Cloud Services
Measurable Results
Sub-second container state discovery and metric refresh without noticeable CPU overhead on host infrastructure.
Preventing SQLite database bloat over long monitoring runs was achieved by implementing an automated 30-day retention pruning job with automatic vacuuming.
