IPv4 and IPv6

IPv4 and IPv6 are the base forwarding models used by the rest of the engine. They provide deterministic route lookup, connected and static routes, recursive next-hop resolution, TTL / hop-limit handling, and VRF-aware forwarding.

Support level

Area	Level	Notes
IPv4 forwarding	Supported	Longest-prefix match, connected routes, static routes, administrative distance, TTL decrement.
IPv6 forwarding	Supported	IPv6 RIB, static routes, hop-limit decrement, connected prefixes.
VRF-aware lookup	Supported	Per-VRF route tables and interface binding for routed scenarios.
Recursive next hop	Supported	Static and protocol routes can resolve through another route.
Fragmentation / PMTUD	Partial	Packet size checks exist; full vendor fragmentation behaviour is not the goal.
Transport stack	Behaviour model	TCP and UDP handlers support higher-level protocol tests, not a full socket stack.

Standards coverage

Standard	Coverage	Notes
RFC 791	Behaviour model	IPv4 addressing, TTL, forwarding, protocol demux, and route lookup semantics.
RFC 8200	Behaviour model	IPv6 forwarding, hop limit, next-header handling, and address semantics.
RFC 1812	Partial	Router behaviour relevant to forwarding, ICMP errors, and TTL expiry.
RFC 2784	Supported	GRE-over-IPv4 tunnel forwarding is covered on the IPsec / GRE page.

Feature matrix

Feature	Status	Notes
Connected routes	Supported	Installed from interface addresses.
Static routes	Supported	Prefix, next-hop, egress-interface, distance, and VRF forms.
Default route	Supported	IPv4 `0.0.0.0/0` and IPv6 `::/0`.
Longest-prefix match	Supported	Stable lookup order with distance and metric tie-breaks.
Recursive lookup	Supported	Next-hop reachability is resolved through the RIB.
Null / discard route	Partial	Accepted in config surfaces where implemented; runtime behaviour is simplified.
TTL / hop-limit expiry	Supported	Produces deterministic ICMP-style failure for diagnostics.
DSCP preservation	Supported	QoS may mark DSCP; forwarding carries the packet metadata.
VRF leaking	Config model	Route-target and route-policy-driven leaking is handled in BGP/MPLS scenarios.

Vendor command matrix

Command	IOS-style	Junos-style	VyOS-style	Notes
`ip route 192.0.2.0 255.255.255.0 10.0.0.1`	Supported	n/a	Supported	Canonical IPv4 static route.
`ipv6 route 2001:db8::/64 2001:db8:1::1`	Supported	n/a	Supported	Canonical IPv6 static route.
`ip route vrf BLUE ...`	Supported	Partial	Partial	Per-VRF static routes.
`show ip route`	Supported	Vendor-shaped view	Vendor-shaped view	IPv4 RIB.
`show ipv6 route`	Supported	Vendor-shaped view	Vendor-shaped view	IPv6 RIB.

Behaviour notes

Forwarding is replay-oriented. A route lookup is a pure function of packet metadata, interface state, and the current RIB snapshot. Equal candidates use stable ordering so two runs of the same topology do not drift.

The IP layer does not try to emulate every ASIC, CEF, kernel, or vendor FIB quirk. It gives routing protocols a deterministic substrate and gives labs a clear answer about reachability.

Examples

Canonical example

configure terminal
interface GigabitEthernet0/0
 ip address 10.0.12.1 255.255.255.252
 ipv6 address 2001:db8:12::1/64
ip route 192.0.2.0 255.255.255.0 10.0.12.2
ipv6 route 2001:db8:200::/64 2001:db8:12::2
end
show ip route
show ipv6 route

Vendor styles

IOS-style

ip route 192.0.2.0 255.255.255.0 10.0.12.2
ipv6 route 2001:db8:200::/64 2001:db8:12::2
show ip route

Junos-style

set routing-options static route 192.0.2.0/24 next-hop 10.0.12.2
set routing-options rib inet6.0 static route 2001:db8:200::/64 next-hop 2001:db8:12::2
show route protocol static

VyOS-style

set protocols static route 192.0.2.0/24 next-hop 10.0.12.2
set protocols static route6 2001:db8:200::/64 next-hop 2001:db8:12::2
show ip route

Known limits

Full TCP congestion behaviour, kernel socket APIs, hardware ECMP hashing, fragment reassembly corner cases, and vendor CEF adjacency internals are not modelled.