Users Guide

not apply to server-side L2 VLT ports because they do not connect to any PIM routers. These VLT ports can be members of multiple PIM-
enabled L3 VLANs for compatibility with IGMP.
To route traffic to and from the multicast source and receiver, enable PIM on the L3 side connected to the PIM router using the ip pim
sparse-mode command.
Each VLT peer runs its own PIM protocol independently of other VLT peers. To ensure the PIM protocol states or multicast routing
information base (MRIB) on the VLT peers are synced, if the incoming interface (IIF) and outgoing interface (OIF) are Spanned, the
multicast route table is synced between the VLT peers.
To verify the PIM neighbors on the VLT VLAN and on the multicast port, use the show ip pim neighbor, show ip igmp
snooping mrouter, and show running config commands.
You can configure virtual link trunking (VLT) peer nodes as rendezvous points (RPs) in a Protocol Independent Multicast (PIM) domain.
If the VLT node elected as the designated router fails and you enable VLT Multicast Routing, multicast routes are synced to the other peer
for traffic forwarding to ensure minimal traffic loss. If you did not enable VLT Multicast Routing, traffic loss occurs until the other VLT peer
is selected as the DR.
VLT Routing
Layer 2 protocols from the ToR to the server are intra-rack and inter-rack. No spanning tree is required, but interoperability with spanning
trees at the aggregation layer is supported. Communication is active-active, with no blocked links. MAC tables are synchronized between
VLT nodes for bridging and you can enable IGMP snooping.
Because VLT ports are Layer 2 ports and not IP interfaces, VLT Unicast and VLT Multicast routing protocols do not operate directly on
VLT ports. You must add the VLT ports as a member of one or more VLANs and assign IP addresses to these VLANs. VLT Unicast and
VLT Multicast routing protocols require VLAN IP interfaces for operation. Protocols such as BGP, ISIS, OSPF, and PIM are compatible
with VLT Unicast Routing and VLT Multicast Routing.
Spanned VLANs
Any VLAN configured on both VLT peer nodes is referred to as a Spanned VLAN. The VLT Interconnect (VLTi) port is automatically added
as a member of the Spanned VLAN. As a result, any adjacent router connected to at least one VLT node on a Spanned VLAN subnet is
directly reachable from both VLT peer nodes at the routing level.
VLT Unicast Routing
VLT unicast routing locally routes packets destined for the L3 endpoint of the VLT peer. This method avoids sub-optimal routing. Peer-
routing syncs the MAC addresses of both VLT peers and requires two local DA entries in TCAM. In case a VLT node is down, a timer that
allows you to configure the amount of time needed for peer recovery provides resiliency. You can enable VLT unicast across multiple
configurations using VLT links. You can enable ECMP on VLT nodes using VLT unicast.
VLT unicast routing is supported on both IPv4 and IPv6. To enable VLT unicast routing, both VLT peers must be in L3 mode. Static route
and routing protocols such as RIP, OSPF, ISIS, and BGP are supported. However, point-to-point configuration is not supported. To enable
VLT unicast, the VLAN configuration must be symmetrical on both peers. You cannot configure the same VLAN as Layer 2 on one node
and as Layer 3 on the other node. Configuration mismatches are logged in the syslog and display in the show vlt mismatch command
output.
If you enable VLT unicast routing, the following actions occur:
• L3 routing is enabled on any new IP address / IPv6 address configured for a VLAN interface that is up.
• L3 routing is enabled on any VLAN with an admin state of up.
NOTE: If the CAM is full, do not enable peer-routing.
NOTE: The peer routing and peer-routing-timeout is applicable for both IPv6/ IPv4.
Configuring VLT Unicast
To enable and configure VLT unicast, follow these steps.
1. Enable VLT on a switch, then configure a VLT domain and enter VLT-domain configuration mode.
CONFIGURATION mode
vlt domain domain-id
2. Enable peer-routing.
Virtual Link Trunking (VLT)
833