Dell EMC SmartFabric OS10 - Virtual Link

Trunking
Reference Architecture Guide
H18362.2

Abstract
This reference architecture guide provides an overview of Virtual Link Trunking (VLT)
and best practices for VLT deployment within the Dell EMC SmartFabric OS10 operating
system.

Dell Technologies Solutions

November 2021
Notes, cautions, and warnings

NOTE: A NOTE indicates important information that helps you make better use of your product.

CAUTION: A CAUTION indicates either potential damage to hardware or loss of data and tells you how to avoid
the problem.

WARNING: A WARNING indicates a potential for property damage, personal injury, or death.

© 2021 Dell Inc. or its subsidiaries. All rights reserved. Dell, EMC, and other trademarks are trademarks of Dell Inc. or its subsidiaries. Other
trademarks may be trademarks of their respective owners.
 Contents

Chapter 1: Introduction................................................................................................................. 4
Dell Technologies vision..................................................................................................................................................... 4
Typographical conventions................................................................................................................................................4

Chapter 2: Virtual Link Trunking Overview.................................................................................... 5

VLT operation....................................................................................................................................................................... 5
VLT architecture..................................................................................................................................................................5

Chapter 3: Dell EMC Virtual Link Trunking Layer 2 Reference Architecture................................... 7

Spanning tree protocol with VLT.....................................................................................................................................9

Chapter 4: Dell EMC Virtual Link Trunking Layer 3 Reference Architecture.................................. 10

EVPN support for VLT......................................................................................................................................................10
VRRP in VLT........................................................................................................................................................................10
VLT unicast routing............................................................................................................................................................ 11
Peer routing..........................................................................................................................................................................11
Peer routing timeout......................................................................................................................................................... 12
eVLT in Dell EMC SmartFabric OS10............................................................................................................................ 12
IP Anycast Gateway support...........................................................................................................................................13

Chapter 5: Dell EMC Virtual Link Trunking Features.....................................................................15

LACP fallback in VLT.........................................................................................................................................................15
Graceful LACP with VLT.................................................................................................................................................. 16
Peer liveliness check......................................................................................................................................................... 16
Delay-restore for orphan ports....................................................................................................................................... 17
Best practices for VLT deployment............................................................................................................................... 17

Appendix A: References............................................................................................................... 18
Dell Technologies Networking Guides...........................................................................................................................18
Support and feedback.......................................................................................................................................................18
Terminology......................................................................................................................................................................... 18

Contents 3
 1
Introduction
In the last two decades, applications have evolved from single monoliths to the highly distributed microservices model. The
scale of the traffic supported in a modern-day infrastructure is huge. All new applications demand a greater network bandwidth,
and the interconnect speeds are increasing at a rapid pace. To provide non-stop services, modern data centers require high
availability (HA) designs for providing non-stop services. The HA concept should be followed in both software and hardware
designs.
For software design, the software should communicate and synchronize data with the connected Top-of-Rack (ToR) or leaf
switch. This way, if one of the switches goes down, either due to planned maintenance or outage, the other switch will take
over. For hardware design, hosts with two NICs should be dual-attached with one link to two leaf switches in a port-channel.
Dell Technology calls this protocol Virtual Link Trunking or VLT. In this protocol, the end host functions as if it were connected
to a single switch with a bond.
VLT enables networks with high availability and resiliency. VLT has been deployed at various locations, and customers are
satisfied with its seamless functionality for the traffic flow and their business needs.
This document explains the dual-node VLT deployment strategies with its associated network reference architecture with
an emphasis on best practices and references for network scenarios. This document also covers the configuration and
troubleshooting of VLT using relevant show commands and different outputs.
NOTE: This guide might contain language that is inconsistent with the current guidelines for Dell Technologies content. Dell
Technologies plans to update the guide over subsequent future releases to revise the language accordingly.

Dell Technologies vision

The vision at Dell Technologies is to be the essential technology company for the data era. Dell Technologies ensures
modernization for today's applications and the emerging cloud-native world. Our Networking team is committed to disrupting
the fundamental economics of the market with an open strategy that gives you the freedom of choice for networking operating
systems and top-tier merchant silicon. The Dell Technologies strategy enables business transformations that maximize the
benefits of collaborative software and standards-based hardware, including lowered costs, flexibility, freedom, and security. Dell
Technologies provides further customer enablement through validated deployment guides that demonstrate these benefits while
maintaining a high standard of quality, consistency, and support.

Typographical conventions
Monospace text CLI examples
Underlined monospace text CLI examples that wrap the page, or to highlight information in CLI output
Italic monospace text Variables in CLI examples
Bold text UI fields and information that is entered in the UI

4 Introduction
 2
Virtual Link Trunking Overview
With the mandatory need for high availability in modern data centers and enterprise networks, VLT plays a vital role with rapid
convergence, seamless traffic flow, efficient load balancing, and loop-free mechanics.
VLT configured in the Dell EMC SmartFabric OS10 operating system functions by aggregating two identical physical switches to
form one single logical switch.
Two switches appear as a single switch on the network with all the links in VLT that can carry traffic across many physically
diverse topologies. Implementing VLT enables multiple parallel paths between nodes without loops and includes traffic load
balancing when alternate paths exist. The outcome is high availability and high resilience for connected access, core switches,
and clients during switch firmware upgrades. If a link fails, traffic continues to flow with minimal disruption.
VLT provides the following benefits:
● Provides node-level redundancy and load sharing by using the same port-channel terminating on multiple upstream nodes.
● Provides a loop-free topology.
● Eliminates STP-blocked ports.
● Optimizes bandwidth utilization by using all available uplink bandwidth.
● Guarantees fast convergence if either a link or device fails.
● Enhances optimized forwarding with Virtual Router Redundancy Protocol (VRRP).
● Optimizes routing with VLT peer routing for Layer 3 VLANs.
● Enables Layer 3 routing protocol support in VLT LAGs with scaled Layer 3 VLAN support.
● Provides link-level resiliency.
● Assures high availability without taking the network offline during firmware upgrades.

VLT operation
Both the VLT nodes of a domain always continue to forward data plane traffic in active/active mode. With the instantaneous
synchronization of MAC and ARP entries, both the nodes remain active/active and continue to forward the data traffic
seamlessly.
The VLT implementation in Dell EMC SmartFabric OS10 has been modified for optimal operation based on the internal
architecture and the base operating system.

VLT architecture
The VLT-Fabric Manager (VLT- FM) module implements the core VLT protocol functionalities and VLT fabric orchestration.
Other modules in the system use the services of VLT-FM to achieve the complete VLT solution.
The VLT fabric consists of two nodes providing a logical single switch view to the connected devices. However, each of the
VLT peers maintains separate control and data planes and can be configured individually for a port, protocol, and management
behaviors.
The VLT design architecture is based on the VLT fabric manager process, controlling the VLT neighbor, and overall fabric
orchestration. The protocol used to communicate between VLT peers is defined as VLT Control Protocol (VLT-CP). The
Node Discovery Service module, or NDS, handles the topology discovery and management. On receiving the topology update
message, VLT-FM builds raw topology and initiates a loop-prevention algorithm to construct the overlay topology.
The VLT application elects the primary node that is based on the lower MAC address; however, with the primary-priority
command, the node with the least primary priority becomes the primary node. This election is not preempted, which means that
the primary role does not change until the nodes are rebooted or the VLT process is restarted whenever there is a change in
priority.
A VLT interconnect (VLTi) synchronizes states between VLT peers. Dell EMC SmartFabric OS10 automatically adds VLTi ports
to VLANs spanned across VLT peers, but does not add VLTi ports to VLANs configured on only one peer.

Virtual Link Trunking Overview 5

● The VLTi synchronizes Layer 2 and Layer 3 control-plane information across the two nodes. The VLTi is used for data traffic
only when there is a link failure that requires VLTi to reach the final destination.
● LLDP, flow control, port monitoring, and jumbo frame features are supported on a VLTi. By default, VLTi ports are set to the
maximum supported MTU value.
● MAC, ARP, IPv6 neighbors learned over VLANs on VLT peer nodes synchronize using the VLTi.
The VLT-related information between the nodes is exchanged through the specific reserved VLAN (VLAN 4094). The VLT
database (VLT DB) stores the VLT control information exchanged between the VLT nodes. The local database (Local DB) stores
the MAC and ARP table entries.
The design and operation of internal communication details and failure handling are beyond the scope of this document.

6 Virtual Link Trunking Overview

 3
Dell EMC Virtual Link Trunking Layer 2
Reference Architecture
Dell EMC Virtual Link Trunking allows Layer 2 information to be shared across the two nodes. For this purpose, the same VLT
domain-id should be configured on both VLT nodes. For rapid convergence and optimal service, the same VLT MAC address
should be configured on both the nodes using the vlt-mac command (optional). In the absence of vlt-mac configuration, if
the primary VLT node goes down, the VLT port-channel on the secondary node flaps, causing slight traffic disruption.

Figure 1. Simple Layer 2 VLT

The primary node election's priority is based on the lower system mac-address of the switch; however, with the primary-
priority command, the VLT node with the least configured priority takes over as the primary node. This election will not be
preempted. If the primary node is reloaded, it is assigned the secondary role. The role change avoids disruptions in traffic flow
due to the election process.
The election happens only during the initial configuration or when VLT is initially launched. The VLT role election has no
significance for the data traffic flowing through the VLT domain. It is only for the control protocol exchange and handles
potential split-control failure scenarios.
VLAN ID 4094 is assigned automatically and internally reserved as a control VLAN to exchange VLT-related information between
the nodes. The IPv6 address that is automatically assigned within the reserved range is mapped for VLAN 4094 for reachability
between the VLT nodes.
For the VLT interconnect (VLTi) link, once the discovery interfaces are configured on both nodes, port-channel 1000 is
automatically configured, mapping the physical discovery interfaces. The ports should be configured as no switchport from the
default Layer 2 mode while configuring the discovery interfaces.
Every thirty seconds, heartbeat messages are sent between the VLT nodes to check the liveliness of the peers and to handle
the VLTi failure scenario. The heartbeat interval value is configurable and reaches the peer through the backup destination.
Similarly, keep-alive messages (non-configurable) are sent through the VLTi port-channel.
For VLT port-channels, the user should explicitly assign the vlt-port-channel id to the configured port-channel on both of the
nodes. This port-channel identifier should be the same for both of the nodes.
Important points to remember:
● VLT port-channel interfaces must be switch ports.
● Dell EMC strongly recommends that the VLTi (VLT interconnect) be a static LAG and that you disable LACP on the VLTi.
● If you include RSTP/PVST, and it must be configured before VLT is configured.
● VLT Heartbeat is supported only on default VRFs.
● If you enable IGMP snooping on the VLT peers, ensure the value of the delay-restore command is not less than the
query interval.

Dell EMC Virtual Link Trunking Layer 2 Reference Architecture 7

● Ensure that the spanning tree root bridge is at the Aggregation layer.
● VLT interconnect over 1G ports is not supported.
● Each VLT domain has a unique MAC address that you create.
● In a VLT domain, the peer switches must run the same Dell Networking OS software version.
● If you replace a VLT peer node, preconfigure the switch with the VLT system MAC address, unit-id, and other VLT
parameters before connecting it to the existing VLT peer switch using the VLTi connection.

In this sample configuration, the 1/1/1 and 1/1/2 interfaces form the discovery interface/VLTi (Po-1000) on the VLT Peer1.
Similarly, the 1/1/1 and 1/1/2 interfaces form the discovery interface on VLT Peer2. The VLT backup destination is mapped to
the remote peer's management IP address. Port-channel 1 is configured on both VLT nodes as interface 1/1/10 and 1/1/11. VLAN
101-110 is mapped to port-channel 1.

Figure 2. Sample configuration

The configurations for VLT-Peer1 and VLT-Peer2 are the same except for the back-up destination IP address. The sample
configuration for VLT-Peer1 is as follows:

VLT-Peer1# configure terminal

VLT-Peer1(config)# interface ethernet 1/1/1
VLT-Peer1(conf-if-eth1/1/1)# no shutdown
VLT-Peer1(conf-if-eth1/1/1)# no switchport
VLT-Peer1(conf-if-eth1/1/1)# exit
VLT-Peer1(config)# interface ethernet 1/1/2
VLT-Peer1(conf-if-eth1/1/2)# no shutdown
VLT-Peer1(conf-if-eth1/1/2)# no switchport
VLT-Peer1(conf-if-eth1/1/2)#exit
VLT-Peer1(config)# vlt-domain 1
VLT-Peer1(conf-vlt-1)# primary-priority 1
VLT-Peer1(conf-vlt-1)# vlt-mac de:11:de:11:de:11
VLT-Peer1(conf-vlt-1)# discovery-interface ethernet 1/1/1
VLT-Peer1(conf-vlt-1)# discovery-interface ethernet 1/1/2
VLT-Peer1(conf-vlt-1)# backup destination 10.16.208.184
VLT-Peer1(config)# interface range vlan 101-110
VLT-Peer1(conf-range-vl-101-110)#exit
VLT-Peer1(config)# interface port-channel 1
VLT-Peer1(conf-if-po-1)# switchport mode trunk
VLT-Peer1(conf-if-po-1)# switchport trunk allowed vlan 101-110
VLT-Peer1(conf-if-po-1)#end

8 Dell EMC Virtual Link Trunking Layer 2 Reference Architecture

Spanning tree protocol with VLT
Even though the VLT enabled topology provides a loop-free environment in the VLT domain, some non-VLT member ports
can end up in a loop scenario. Rapid spanning tree protocol (RSTP) and rapid per-VLAN spanning tree+ (RPVST+) modes of
spanning tree protocol in VLT are supported to provide a solution for this scenario. Depending on the network topology, the
primary node of the VLT domain can be designated as the root bridge.
NOTE: Dell Technologies recommends that you enable a spanning tree (RSTP or RPVST+) before enabling VLT to avoid
unintentional loops in the network.
The RPVST+ is enabled by default in Dell EMC SmartFabric OS10. There is no need to configure RPVST+ unless the default
spanning tree mode is changed.

VLT-Peer1(config)# spanning-tree mode rapid-pvst

The RSTP is enabled by using the following command.

VLT-Peer1(config)# spanning-tree mode rstp

For Multiple Spanning Tree Protocol (MSTP) on VLT nodes, configure both VLT peer nodes in the same MST region to avoid
network loops. Ensure that the VLAN-to-instance mappings, region name, and revision ID are the same on both VLT peer nodes.

VLT-Peer1(config)# spanning-tree mode mst

VLT-Peer1(config)# spanning-tree mst configuration
VLT-Peer1(conf-mst)# instance 1 vlan 2-10
VLT-Peer1(conf-mst)# revision 10
VLT-Peer1(conf-mst)# name ExampleMSTregion

Dell EMC Virtual Link Trunking Layer 2 Reference Architecture 9

 4
Dell EMC Virtual Link Trunking Layer 3
Reference Architecture
With Dell EMC Virtual Link Trunking, customers can enjoy the benefits of High availability in Layer 3 networks. For this purpose,
several configuration environments are supported in VLT, such as VRRP, IP Anycast Gateway, and eVLT. These protocols are
necessary to assist the VLT implementation in the Layer 3 environment. This way, an active/active mode can be supported in
Layer 3 as well. The following figure shows sample Layer 3 VLT topology.

Figure 3. Layer 3 VLT with multiple partner switches

EVPN support for VLT

VLT can also be used in combination with network virtualization technology such as EVPN. Primarily, EVPN uses RT-1 and RT-4
message types to handle multihomed nodes. RT-1 tells the network which switches are attached to which Ethernet segments.
It carries the LACP identifier of the host as Ethernet Segment ID (ESI). When other Ethernet Segments receive the BGP
updates of the RT-1 advertisement, they can determine which of their attached hosts belong to the same virtual network. RT-4
elects one of the VLT peers as the designated forwarder for multi-destination frames as it carries the mapping of the Ethernet
Segment to the switch serving the segment. More details about Dell EMC SmartFabric OS10 BGP EVPN can be found in the Dell
EMC Networking Virtualization Overlay with BGP EVPN document.

VRRP in VLT
With VRRP configured in both of the VLT nodes, the active/active mode is internally enabled by default, ensuring seamless
traffic flow. The active/active mode is activated when the VLT VLANs are configured with a VRRP group. For practical
purposes, VRRP offers a single virtual IP as a default gateway for its access clients.
VRRP should be enabled on host-facing VLANs. The gateway for the hosts should be the virtual IP of the respective VRRP
group. Since VRRP establishes an active/active mode in VLT, the VLT nodes route traffic from the hosts.

10 Dell EMC Virtual Link Trunking Layer 3 Reference Architecture

VLT unicast routing
VLT unicast routing requires that both VLT peers be in Layer 3 mode. Static route and routing protocols such as RIP, OSPF,
ISIS, and BGP are 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.

Peer routing
VLT peer routing enables optimized routing where packets destined for the Layer 3 endpoint of the VLT peer are locally routed.
VLT supports unicast routing of both IPv4 and IPv6 traffic.
To enable VLT unicast routing, both VLT peers must be in Layer 3 mode. The VLAN configuration must be symmetrical on both
peers. Users cannot configure the same VLAN as Layer 2 on the first node and as Layer 3 on the other node.
IPv4 routing provides forwarding of packets to a destination IP address, based on a routing table. This routing table defines how
packets are routed (dynamically, broadcasted directly, or using proxy ARP), and what type of information is included with the
packets.
Peer routing is not restricted to a VLAN with routing protocols. Instead, it is enabled domain wide. Peer routing ensures active/
active routing for the traffic path. To route traffic on behalf of other VLT peers, enable peer routing on both nodes. Peer routing
synchronizes the router MAC address (local destination MAC address) within the VLT domain and ensures seamless routing by
the peer node.

Figure 4. Peer routing

Dell EMC Virtual Link Trunking Layer 3 Reference Architecture 11

Peer routing timeout
If the peer node fails, the peer-routing-timeout command enables the availability of peer routing features. It retains the
corresponding local destination MAC address until the timer expires. For customers who would like to be aware of the node
failure and its associated traffic, the peer-routing-timeout command would cease routing on behalf of the peer after the
timer's expiration. This timer does not need to be configured for standard, high-availability requirement deployments.
The delay-restore timer command has a default value of 90 seconds. After a VLTi flap or node reboot, the VLT
port-channel is held in a down state in the secondary VLT node until the delay-restore time has passed. This function allows
the MAC and ARP address table synchronization to complete between the nodes and the control protocols to converge before
reinstating the VLT port-channel for forwarding the traffic. When the VLT port-channel is brought to a working state, the
forwarding tables are populated in the hardware, allowing traffic to flow without disruption. In a highly scaled setup, it is
recommended that the timer is increased to ensure that all of the tables are synchronized, and that the protocol converges
before opening the VLT port-channel to avoid traffic issues.

eVLT in Dell EMC SmartFabric OS10

Data Center Interconnect (DCI) between VLT domains is supported in Dell EMC SmartFabric OS10 with the VRRP feature
across the VLT domains (eVLT). With Dell EMC SmartFabric OS10, VRRP can be configured for dual VLT domains to form an
eVLT topology, as shown in the following figure. The same VRRP group-id enables a common virtual IP address across both the
VLT domains. Therefore, VM migration from DC-1 to DC-2 can be initiated without any dependency while retaining the VM IP
address and gateway IP address.
The VRRP group for the configured VLANs should be the same VRRP group id in both VLT domains. With this implementation,
only one of the nodes is selected as the VRRP Master, where the other three nodes each act as a standby for that VRRP
group. For example, the virtual IP address 192.168.1.254 is retained across both of the DCs. If a VM with a default gateway
of 192.168.1.254 is migrated from DC-1 to DC-2, the VM can retain its IP address and continue to forward and receive traffic
without changes.

12 Dell EMC Virtual Link Trunking Layer 3 Reference Architecture

Figure 5. DCI using eVLT with VRRP

Upstream devices from the VLT domains can have static routing, or any other dynamic routing protocols, to support the N-S
traffic. The host can still be reached if the traffic to the VM host lands at any upstream DC, ensuring high availability for the
application and services.

IP Anycast Gateway support

The Anycast IP-based Layer 3 gateway solution is a lightweight gateway router redundancy protocol that is enabled only on
VLANs. It does not exchange any control packets between routers and supports only the active/active data plane forwarding
mode. The primary node acts as the Anycast Gateway IP address owner in the VLAN and will respond to the ARP requests for
gateway anycast IP.
NOTE: The IP Anycast Gateway feature only works when VLT configurations are present in the switch. If you enable this
feature without VLT domain configuration, the anycast gateway configuration remains inactive. Be sure to configure the
same IP anycast IPv4 or IPv6 address and the same virtual global MAC address on both VLT nodes. Anycast IP gateway
routing and VRRP are mutually exclusive and cannot be configured on the same VLAN simultaneously. Customers can
enable IP Anycast Gateway for up to 512 Layer 3 VLANs.

Dell EMC Virtual Link Trunking Layer 3 Reference Architecture 13

Figure 6. IP Anycast Gateway support

14 Dell EMC Virtual Link Trunking Layer 3 Reference Architecture

 5
Dell EMC Virtual Link Trunking Features
LACP fallback in VLT
LACP fallback in a VLT configuration allows VLT peers to control member ports of a port-channel. It includes the following
characteristics:
● The LACP fallback feature is applied to support PXE boot from servers with NIC teaming or bonding.
● Use the lacp fallback enable command on the VLT port-channels on both of the VLT nodes that are connected to
servers.
● A member port in the VLT port-channel is made active if the LACP PDUs are not received from the server within the LACP
fallback timeout period, which is set to 15 seconds by default.
● LACP fallback in VLT allows the PXE boot server to establish a connection over a single port, download the boot image, and
continue the boot process.
● After the server boots, it establishes the LACP port-channel, which allows the VLT port-channels to function like a regular
VLT port-channel after the LACP PDUs are received.
● PXE boot uses untagged packets. Configure the VLT port-channel as an untagged member in the respective VLAN to reach
the PXE/DHCP server.
● During LACP fallback, one VLT port-channel member port is kept active, while other ports are kept in an inactive state.
The active port is selected based on the following order:
1. The port with the lowest LACP port priority is kept active.
2. The lowest port number is kept active if the port priority is the same or the default port.
3. The port on the VLT node with the lowest system MAC address is kept active if the port numbers are the same in both VLT
nodes.

Figure 7. LACP fallback

Dell EMC Virtual Link Trunking Features 15

Graceful LACP with VLT
When a VLT node is reloaded, all interfaces including VLT port-channel interfaces go down. Top-of-Rack (ToR) devices
connected at the other end of the VLT port-channel interfaces could take a considerable amount of time to detect the interface
status change and switch the traffic towards the other active VLT node. Using LACP PDUs, the graceful LACP feature enables
VLT nodes to inform ToR devices before taking down the member ports of its VLT port-channel interfaces. Thus, the graceful
LACP feature enables the ToR devices to switch the traffic to the other active VLT node.
Graceful LACP is supported in these scenarios:
● When a VLT node is reloaded
● When the secondary VLT node detects that the VLTi link is down but the heartbeat is functional
Graceful LACP is enabled by default and you cannot disable it.

Figure 8. Graceful LACP

Peer liveliness check

The VLT peer liveliness mechanism checks for the availability of the peer node. The system sends periodic keep-alive messages
to detect the liveliness of the peer node. Use a different link, other than the VLTi, for the peer liveliness check. This link is
referred to as the VLT backup link. Dell Technologies recommends using the OOB management network connection for the VLT
backup link.
If the VLTi goes down, the backup link helps differentiate the VLTi link failure from a peer node failure. If all links in the VLTi fail,
the VLT nodes exchange node liveliness information through the backup link.
Based on the node liveliness information:
● If only the VLTi link fails, but the peer is alive, the secondary VLT peer shuts down its VLT ports.
● If the primary VLT node fails, then both the VLTi and heartbeat fail, and the current secondary peer takes over the primary
role.
Configure the VLT backup link using the backup destination {ip-address | ipv6 ipv6–address} [vrf
management] [interval interval-time]. The interval range is from 1 to 30 seconds. The default interval is 30

16 Dell EMC Virtual Link Trunking Features

seconds. Irrespective of the configured interval, when the VLTi link fails, the system checks for the heartbeat connection
without waiting for the timed intervals, thus allowing faster convergence.

OS10(config)# vlt-domain 1
OS10(conf-vlt-1)# backup destination 10.16.151.110 vrf management interval 20

Delay-restore for orphan ports

The delay-restore feature for non-VLT ports is used to delay the bringup of non-VLT ports. There is enough time available for
the protocols and features to converge. If there is no such mechanism available, then traffic may be blocked as a result. This
feature is similar to VLT delay-restore mechanism for VLT LAGs, which is already supported.
The delay-restore mechanism applies to the following two scenarios:
● VLT peer reload – When the reloaded node joins as a secondary node in the VLT domain. VLT LAGs are brought up only
after the delay-restore timer expires.
● VLTi link failure – When the VLT heartbeat is configured and is up and running, if the VLTi link goes down, the secondary
VLT peer brings down its local VLT LAGs. All devices connected to the VLT LAG interfaces are expected to send or receive
traffic through the VLT primary device. When the VLTi link comes back up, the secondary VLT peer does not immediately
bring its VLT LAG up. They are brought up only after VLT delay-restore timer expires, assuming that heartbeat is up.
To configure a delay-restore for orphan ports, use the following commands:

OS10#configure terminal
OS10(config)#interface ethernet 1/1/1
OS10(conf-if-eth1/1/1)vlt delay-restore orphan-port enable
OS10(conf-if-eth1/1/1)exit
OS10(conf)vlt-domain 1
OS10(conf-vlt-1)delay-restore 150

Best practices for VLT deployment

When deploying VLT, consider the following best practices:
● Both VLT peers require the same Dell EMC SmartFabric OS10 version within the production environment; however, a
mismatch in versions is acceptable during the upgrade procedure.
● Backup destination, or heartbeat connectivity, is recommended through the management network.
● Enable spanning tree protocol to avoid unintentional loops in the network.
● Configure identical vlt-mac on both VLT peers to avoid VLT port-channel flap during failure scenarios, such as VLTi flap and
node reboot.
● Increase the delay restore timer in a scaled VLAN configuration.
● Always deploy more than one link for the VLTi port-channel.
● Implement dynamic LACP for VLT port-channels.
● Implement VLT nodes as the root bridge and the backup root bridge.
● Use the default active/active VRRP mode in VLT-VLANs.
● Configure symmetrical configurations on VLT port-channels and VLT-VLANs on both VLT nodes.
● Identify and correct any mismatches reported by the show vlt mismatch command.
● For high availability, connect VLT port-channels to the access nodes. Orphan ports that are connected directly to servers or
switches will fail if the connected node goes down.

Dell EMC Virtual Link Trunking Features 17

 A
References
Dell Technologies Networking Guides
For additional information, see the following documentation:
Dell EMC PowerSwitch Guides
Dell EMC Networking Layer 3 Leaf-Spine Deployment and Best Practices with OS10
Dell EMC Networking Virtualization Overlay with BGP EVPN
Dell Digital Locker (for the latest Dell EMC SmartFabric OS10 version)
OS10 Demo and Labs

Support and feedback

For technical support, visit http://www.dell.com/support or call (USA) 1-800-945-3355.
We encourage readers to provide feedback on the quality and usefulness of this publication by sending an email to
Dell_Networking_Solutions@Dell.com.

Terminology
VLT domain - The domain includes VLT peer devices, VLT interconnects, and all port-channels in the VLT connected to the
attached devices. It is also the configuration mode that you must use to assign VLT global parameters.
VLT interconnect (VLTi) - The link between VLT peer switches used to synchronize operating states.
VLT peer device - A pair of devices connected using a dedicated port-channel—the VLTi. You must configure VLT peers
separately.
Discovery interface - Interfaces on VLT peers in the VLT interconnect (VLTi) link.
VLT MAC address - Unique MAC address that you assign to the VLT domain. A VLT MAC address is a common address in both
VLT peers. If you do not configure a VLT MAC address, the primary peer's MAC address is used as the VLT MAC address across
both peers.
VLT node priority - The priority based on which the primary and secondary VLT nodes are determined. If priority is not
configured, the VLT node with the lowest MAC address is elected as the primary VLT node.
VLT port-channel - A combined port-channel between an attached device and the VLT peer switch.

18 References