Path Tracing Logic

The first InfraLynx path-tracing engine is intentionally deterministic and breadth-first. Its purpose is to define the traversal contract, not to solve every networking scenario in the first iteration.

Trace Inputs

• start node ID
• target node ID
• allowed edge kinds
• maximum depth

Trace Behavior

1. Build an adjacency index from topology edges.
2. Sort outbound edges into a stable order.
3. Traverse with breadth-first search.
4. Stop when the target node is reached or the depth limit is exceeded.

Determinism Rules

• the same edge set must produce the same traversal order
• visited node tracking prevents loops
• depth limiting bounds recursion and runtime cost
• edge-kind filters let callers trace only the layers they care about

Near-Term Use Cases

• validate direct L2 and L3 service reachability
• prove that an interface-to-address binding can be traversed
• document expected behavior for later topology and path APIs

Deferred Work

• weighted path selection
• multi-path tracing
• policy-based routing interpretation
• large-topology performance indexing