Home / Builds / IT Infrastructure
IT Infrastructure & Server Management

Anikira Network Deployment

A comprehensive showcase of bare-metal server provisioning, Docker containerization, reverse proxy routing, and server orchestration designed to securely support high-volume connections.

Ubuntu 20.04 LTS
Docker & Portainer
NGINX Proxy Manager
WireGuard VPN
Pterodactyl Panel
Grafana/Prometheus

I. Bare-Metal Provisioning & OS Configuration

Starting from a raw Contabo KVM deployment, I provisioned and secured an Ubuntu 20.04 LTS environment. This included setting up SSH key authentication, UFW firewall rules, and optimizing kernel parameters for a high-load AMD EPYC architecture.

root@vmi956512:~
Last login: Fri Dec 9 11:03:41 2022 from 158.62.43.184
root@vmi956512:~# neofetch
root@vmi956512.contaboserver.net ----------------------------------- OS: Ubuntu 20.04.4 LTS x86_64 Host: KVM/QEMU (Standard PC (i440FX + PIIX, 1996) pc-i440fx-5.2) Kernel: 5.4.0-122-generic Uptime: 95 days, 6 hours, 39 mins Packages: 652 (dpkg) Shell: bash 5.0.17 Terminal: /dev/pts/0 CPU: AMD EPYC 7282 (8) @ 2.794GHz GPU: 00:02.0 Vendor 1234 Device 1111 Memory: 14374MiB / 30096MiB
root@vmi956512:~#
18% CPU
14.45 GB / 29.39 GB RAM ↑ 0.19 Mb/s ↓ 0.47 Mb/s ⏱ 95 days

Fig 1. Live terminal verification of bare-metal resource provisioning and uptime stability.

II. Docker Containerization & Orchestration

Instead of running services on bare metal, I engineered a fully containerized environment managed via Portainer. This allowed for isolated deployments of VPNs, reverse proxies, and monitoring nodes, ensuring that a crash in one service would not bring down the entire network.

Containers
Name ↑↓ State ↑↓ Image ↑↓ IP Address ↑↓ Published Ports
portainer running portainer/portainer-ce:latest 172.17.0.2 8000:8000, 9443:9443
NGINXProxyManager running jc21/nginx-proxy-manager:latest 172.17.0.3 443:443, 80:80, 81:81
wg-easy running weejewel/wg-easy 172.21.0.2 51820:51820, 51821:51821
prometheus running prom/prometheus 172.26.0.2 9090:9090
grafana running grafana/grafana 172.26.0.3 3000:3000
dbb33845-bda7... exited ghcr.io/pterodactyl/yolks:java_17 - -
uptime-kuma healthy louislam/uptime-kuma:1 172.17.0.4 3001:3001

Fig 2. Live Docker container environment demonstrating an organized, microservices-based infrastructure.

III. Traffic Routing & Reverse Proxy Topology

To securely expose internal services (like Grafana, Pterodactyl, and web dashboards) to the public internet, I engineered a network topology routing Cloudflare DNS traffic through an NGINX Reverse Proxy, terminating SSL and protecting internal IP addresses.

Public Internet
Cloudflare Proxy & DNS
NGINX Reverse Proxy
Portainer Admin Port 9443
Grafana Analytics Port 3000
Pterodactyl Panel Port 80
Uptime Kuma Port 3001

Fig 3. Visualization of external traffic termination and internal port mapping.

IV. Game Server Orchestration

To securely allocate resources to specific gaming instances (like a Minecraft survival network), I deployed a Pterodactyl daemon node. This allowed for exact RAM/Disk allocation and isolated environments per server instance.

Nodes All nodes available on the system.
Name Location Memory Disk Servers SSL
PH1 PH-1 10240 MiB 460800 MiB 0

Fig 4. Node resource allocation dashboard within the Pterodactyl management ecosystem.

V. Systems Monitoring & Telemetry

Maintaining server health requires proactive monitoring. I connected Prometheus node-exporters to a centralized Grafana dashboard, providing real-time telemetry on CPU loads, RAM utilization, I/O wait times, and network bandwidth across the cluster.

General / Resource Overview

Last 12 hours
node-exporter: Overall total 5m load & average CPU used%
CPU Cores
Current: 8
node-exporter: Overall total memory & average memory used%
Total
Current: 29.4 GiB
node-exporter: Overall total disk & average disk used%
Total
Current: 193.8 GiB
Resource Details: [vmi956512.contaboserver.net]
Uptime
1 day
CPU Cores
8
Total RAM
29 GiB
DeviceFilesystemSizeAvailUsed
/dev/sda3 ext4 194 GiB 174.6 GiB 5.1%
CPU iowait
0.00%
Free inodes:/
12.7 Mil
CPU Basic
Network bandwidth usage per second All

Fig 5. Live Grafana dashboard reflecting AMD EPYC resource loads, network bandwidth, and active filesystem utilization.

VI. Scalability & Operational Impact

Technical systems are ultimately built to serve people. This architecture supported seamless, low-latency engagement, allowing the community to organically scale and monetize over a sustained period.

Active User Community
99.9% Server Uptime
Zero Data Breaches

Fig 6. Organic growth and stability metrics achieved over 2+ years of continuous operation.

The Execution Standard

Running a multi-server network demands a deep understanding of Linux environments, network security, and resource allocation. By building this infrastructure from the ground up, I ensure platforms are fast, secure, and ready to scale effortlessly.