How do hundreds of devices share ONE public IP? Built a NAT/PAT lab from scratch in Packet Tracer to see it in action.
Ever wonder how your entire house or office gets by on a single public IP? It’s NAT, and specifically PAT (Port Address Translation). I just built this from scratch in Cisco Packet Tracer to see exactly how it works under the hood.
The Topology:
PC0 & PC1 > Switch > R1 (NAT) > R2 (ISP) > Server
What I configured:
- Interface IPs and descriptions (LAN: 192.168.10.0/24, Outside link: 203.0.113.0/30)
- Tagged interfaces: `ip nat inside` on the LAN side, `ip nat outside` on the ISP side
- Default route (`0.0.0.0/0`) pointing to the ISP router
- Standard ACL to define allowed internal hosts, tied to: `ip nat inside source list 1 interface gig0/1 overload`
- Saved it all with `copy running-config startup-config`
Verification:
- Successfully pinged the ISP server from the internal PCs.
- Ran `show ip nat translations` to watch it work. Seeing the unique port numbers map to the single public IP is where the concept really clicks.
The Takeaway:
NAT gets private addresses to the internet. PAT takes it a step further by using unique source port numbers to track each individual session, allowing hundreds of devices to share one public IP without traffic collisions.
Lab is configured, tested, and working. On to the next one.
*Note:* If you’re currently studying for your CCNA or just building your network engineering fundamentals, I’ve been working through the Network Engineer Complete Bundle (L1 to L3) from CodeLivly. It covers this exact type of architecture from the ground up without the usual textbook fluff. It’s a solid resource if you want structured, practical labs: