Infrastructure Explorer
Interactive force-directed topology of a dual-galaxy network
🌌 2 Galaxies
🖥️ 7 Systems
⚡ 98 Cores
💾 110TB+ Storage
🔗 Tailscale Mesh VPN
Challenges
Learn to navigate the force-directed topology graph.
○ Click any node to view its details
- Find a system node in the graph
- Click on it to select it
- The details panel will appear with system information
Click any node in the graph 💡 Nodes are the colored shapes — each represents a system in the infrastructure.
○ Switch to the Radial layout
- Find the layout dropdown in the controls toolbar
- Select "Radial" from the options
- Watch the graph rearrange into a radial pattern
Use the layout dropdown to switch to Radial 💡 The controls toolbar is above the graph. The layout dropdown shows the current layout name.
○ Toggle fullscreen mode
- Find the fullscreen button in the controls toolbar
- Click it to expand the graph
- Press Escape or click again to exit
Click the fullscreen button (expand icon) 💡 Fullscreen gives you more space to explore the topology.
○ Click the background to deselect
- Select a node first
- Then click on an empty area of the graph
- The details panel will close
Click empty space in the graph 💡 Clicking the background resets the selection and clears highlighting.
○ Find the gateway node connecting both galaxies
- Look for a node with many connections
- It should have Tailscale (dashed) edges crossing to the other cluster
- Click it to confirm its role
Find and click the node that bridges Milky Way and Andromeda 💡 The gateway handles cross-site routing via Tailscale subnet routing.
Explore the two physical sites and their connections.
○ Identify the two galaxy clusters in the graph
- Look for two distinct groups of nodes
- Milky Way nodes have cyan-tinted borders
- Andromeda nodes have pink-tinted borders
Observe the two clusters and their color coding 💡 The clustered layout separates the two galaxies visually.
○ Count the systems in each galaxy
- Count nodes in the Milky Way cluster (cyan borders)
- Count nodes in the Andromeda cluster (pink borders)
- Compare the counts
Count nodes per cluster 💡 Milky Way has more systems but Andromeda has more compute power.
Expected:
Milky Way: 8 (including network devices), Andromeda: 5
○ Find the Tailscale mesh edges
- Look for dashed purple lines in the graph
- These represent encrypted VPN tunnels between sites
- Count how many cross-site connections exist
Identify the dashed purple edges connecting the two clusters 💡 Tailscale edges are visually distinct — dashed and purple.
○ Identify the subnet router node
- Click nodes in Andromeda until you find one with "Tailscale Router" in its services
- This system advertises its local subnet to the remote site
- Note how many Tailscale edges connect to it
Find the node that acts as Tailscale subnet router 💡 The subnet router allows Milky Way systems to reach Andromeda local IPs directly.
○ Calculate total CPU cores across all systems
- Click each node and note its core count from the details panel
- Sum them across both galaxies
- Compare core distribution between sites
Click each system node and sum the cores 💡 Network devices (firewall, switch) also have cores, but much fewer.
Expected:
98 cores total
Understand what each node type represents in the topology.
○ Find all hypervisor nodes
- Look for hexagon-shaped nodes (larger, orange-tinted)
- Click each to confirm it runs Proxmox VE or similar
- Note which galaxy each hypervisor belongs to
Find nodes with hexagon shape — these are hypervisors 💡 Hypervisors run virtual machines and containers. They are the compute backbone.
○ Click the gateway node and review its services
- Find the star-shaped cyan node
- Click it to open the details panel
- Read through its services list
Click the star-shaped node (gateway) and review its services 💡 The gateway runs Gitea, Prometheus, Grafana, and more — it is the services hub.
○ Locate all storage nodes
- Look for rectangle-shaped green nodes
- These are NAS devices (Synology, Unraid)
- Click each to see storage capacities
Find all rectangle-shaped nodes — these are storage systems 💡 Storage nodes have TB capacities in their services lists.
○ Find the media server
- Look for a rounded-rectangle pink node
- It should be in Andromeda
- Click it to confirm it runs Plex
Find the media node and confirm its services 💡 The media server runs as an LXC container on one of the hypervisors.
○ Identify the network devices
- Look for diamond-shaped amber nodes
- These are the firewall and managed switch
- Click each to see their network services
Find diamond-shaped nodes — these are network infrastructure 💡 The firewall manages VLANs, DHCP, DNS. The switch handles L2 VLAN filtering.
Trace connections between systems across sites.
○ Distinguish local LAN from Tailscale edges
- Solid thin lines = local LAN connections
- Dashed purple lines = Tailscale VPN tunnels
- Identify which connections stay local vs cross-site
Observe the edge styles and categorize them 💡 Local LAN edges connect systems within the same galaxy. Tailscale edges cross between galaxies.
○ Trace the path from workstation to remote hypervisor
- Start at the workstation node in Milky Way
- Follow its Tailscale edge to the remote hypervisor
- Note the network hop: workstation → Tailscale → remote site
Trace the connection from Capella-Outpost to Tarn-Host 💡 The workstation connects directly to the remote hypervisor via Tailscale mesh.
○ Find the backup connection between NAS units
- Look for a dotted green line in the Andromeda cluster
- This represents a backup replication path
- Click both endpoints to see what data flows between them
Find the dotted green edge labeled "Hyper Backup" 💡 Backup connections are dotted green lines. They flow from primary to backup storage.
○ Identify the star topology in Milky Way
- Find the switch (diamond node) in Milky Way
- Count how many local LAN edges connect to it
- This shows the physical star topology through the managed switch
Count edges connecting to the Nexus switch node 💡 In a star topology, all hosts connect through a central switch. The firewall connects to the switch as well.
○ Count how many cross-site connections exist
- Find all dashed purple (Tailscale) edges
- Count them
- Note which Milky Way systems have direct cross-site access
Count all Tailscale (dashed purple) edges in the graph 💡 Not every system needs a direct Tailscale connection — subnet routing handles the rest.
Expected:
3 Tailscale mesh connections
Read and interpret real-time system health data.
○ Check if the data source is Live or Demo mode
- Look at the controls toolbar for the status indicator
- Green dot = Live Data (connected to real infrastructure)
- Orange dot = Demo Mode (showing static data)
Find the status badge in the controls toolbar 💡 Live data is only available when connected to the local network.
○ Find the system with the highest CPU usage
- Click each online system node
- Check the Live Metrics section in the details panel
- Compare CPU percentages across systems
Click nodes and compare CPU usage in the details panel 💡 CPU bars appear in the details panel when live metrics are available.
○ Identify any offline systems
- Look for nodes that appear dimmed or faded
- Offline nodes have reduced opacity and red borders
- Click them to confirm their status
Look for dimmed nodes — these are offline systems 💡 Offline systems still appear in the topology but are visually muted.
○ Compare resource utilization between galaxies
- Check CPU/RAM for each Milky Way system
- Check CPU/RAM for each Andromeda system
- Determine which site is under more load
Click systems in each galaxy and compare their metrics 💡 Andromeda has more raw compute but may have different workload patterns.
○ Observe how node borders change with load
- Look for systems with active workloads
- Nodes with high CPU show thicker, warmer-colored borders
- Nodes with normal load have standard borders
Watch for visual changes in node borders indicating CPU load 💡 The border color and width subtly shift to indicate system health.
Evaluate the topology for single points of failure and redundancy.
○ Identify which node is most connected
- Look at each node and count its edges
- The most connected node has the most lines radiating from it
- Consider what happens if this node goes offline
Find the node with the highest edge count 💡 Central nodes in each galaxy are often the most connected — and the biggest risk.
○ Find single points of failure in the topology
- Identify nodes whose removal would disconnect other systems
- Consider the switch in Milky Way and the hypervisor in Andromeda
- Assess whether each galaxy can operate if these fail
Look for nodes that, if removed, would isolate other systems 💡 The switch (Nexus) and the primary hypervisor in each galaxy are critical.
○ Assess what happens if Tarn-Host goes offline
- Click Tarn-Host to see all its services
- Note it is both hypervisor AND Tailscale subnet router
- Determine what breaks: cross-site access, media server, local VMs
Click Tarn-Host and trace all dependencies 💡 Tarn-Host is the most critical node in Andromeda — it hosts VMs and routes all cross-site traffic.
○ Evaluate backup redundancy paths
- Trace the backup edge from primary to backup NAS
- Check if backups cross the Tailscale bridge or stay local
- Assess 3-2-1 backup rule compliance
Follow backup edges and assess geographic distribution 💡 Backups staying within one galaxy means a site loss could affect both copies.
○ Propose where to add a new system for better resilience
- Identify which galaxy has fewer redundant systems
- Consider adding a secondary gateway or backup target
- Think about geographic distribution for disaster recovery
Analyze gaps in redundancy and propose an addition 💡 Milky Way has no local NAS — all storage is in Andromeda. Is that a risk?
What You Can Do
- Force-directed topology graph with multiple layouts
- Live metrics from real infrastructure (when available)
- Click-to-explore system details and services
- Tailscale mesh and cross-site connection visualization
- Storage and backup architecture overview
Learning Objectives
Beginner
- Navigate the graph and click nodes
- Identify the two galaxy clusters
Intermediate
- Understand system roles and types
- Trace network paths between sites
Advanced
- Read live metrics and identify load patterns
- Compare resource usage across sites
Expert
- Assess single points of failure
- Design resilience improvements
Demo Mode
Initializing topology graph...
Select a system
Click on any node in the graph to view its details, services, and connections.
Network Legend
Node Types
Hypervisor
Gateway
Workstation
NAS / Storage
Media Server
Network Device
Connections
Local LAN
Tailscale VPN
Backup Replication
Service Dependency
Galaxy Colors
Milky Way (Primary Site)
Andromeda (Remote Site)
How It Works
🏠
Dual-Site Architecture
Two physical locations connected via encrypted VPN. Each site operates independently with full local connectivity.
🔐
Zero-Trust Networking
Tailscale provides encrypted, authenticated connections between all nodes. No open ports, no exposed services.
🔄
Subnet Routing
Tarn-Host acts as a subnet router, allowing Milky Way systems to reach Andromeda local IPs transparently.