The scenario
Host-A (10.0.1.10) reaches Host-B (10.0.4.10) through PE1, P1, P2 and PE2. Toward Host-B, PE1 pushes 16001, P1 swaps it for 16002, P2 swaps that for 16003, and PE2 pops 16003 and delivers plain IP. Replies take a separate LSP: PE2 pushes 16011, P2 swaps it for 16012, P1 swaps that for 16013, and PE1 pops it. P1 and P2 have routes only to their own links and no route to 10.0.1.0/24 or 10.0.4.0/24; no router runs OSPF.
What you will work out
See how a label is swapped at every transit router, why each label only means something on one link, and what a static LSP cannot do when a core link fails.
Why this matters
Provider backbones often put several core routers between two PEs. Every one of them does the same small job, one exact-match label lookup and a swap, whatever the path length, and none of them needs the customer's routes. The flip side is that the LSP is only as good as every entry along it, which is why real networks let a protocol build those entries and fix them when a link fails.
Running it
MPLS: Multi-Hop Label Path opens on a canvas, already wired and addressed. Trace a packet through it hop by hop, or run the same topology live on real Linux network namespaces. How the two modes differ.
Open MPLS: Multi-Hop Label Path
Runs in your browser. No install, no signup, and the simulation starts instantly.
Open the lab →