Dell SmartFabric Services User Guide

Release 10.6.0

July 2025
Rev. A03
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.

Copyright © 2025 Dell Inc. All Rights Reserved. Dell Technologies, Dell, and other trademarks are trademarks of Dell Inc. or its subsidiaries.
Other trademarks may be trademarks of their respective owners.
 Contents

Chapter 1: About this guide........................................................................................................... 6

Text and Syntax Conventions.......................................................................................................................................... 6
Related Documents.............................................................................................................................................................6
Documentation Feedback.................................................................................................................................................. 7
Acronyms............................................................................................................................................................................... 7

Chapter 2: Change history.............................................................................................................8

Chapter 3: SFS fundamentals........................................................................................................9

SFS overview........................................................................................................................................................................9
SFS UI............................................................................................................................................................................. 10
Supported network topologies....................................................................................................................................... 10
SFS supported platforms................................................................................................................................................. 10
SFS and supported solutions.......................................................................................................................................... 10
Creating a network fabric................................................................................................................................................. 11
Server discovery and onboarding................................................................................................................................... 11
Dynamic server onboarding........................................................................................................................................ 11
Static server onboarding............................................................................................................................................ 13
Server profile and server interface profile............................................................................................................. 13
Fabric backup and restore............................................................................................................................................... 14

Chapter 4: Setting up SFS........................................................................................................... 15

Prerequisites....................................................................................................................................................................... 15
SFS configuration notes...................................................................................................................................................15
Configure SFS..................................................................................................................................................................... 17
Enable SFS on PowerSwitches................................................................................................................................. 17
Verify the switch operating mode........................................................................................................................... 20
Complete the fabric setup........................................................................................................................................ 20

Chapter 5: Managing SmartFabric with the SFS UI...................................................................... 21

Access fabric setup options............................................................................................................................................ 21
SmartFabric overview ............................................................................................................................................... 23
View fabric topology................................................................................................................................................... 24
Manage switches in a fabric..................................................................................................................................... 24
Configure server interface profile........................................................................................................................... 28
Configure and manage uplinks................................................................................................................................. 35
Configure networks and routing configuration.................................................................................................... 44
Configure global settings for SmartFabric............................................................................................................ 53
Life-cycle management..............................................................................................................................................56
View fabric compliance status..................................................................................................................................57

Chapter 6: SFS with VxRail......................................................................................................... 59

Supported network topologies.......................................................................................................................................59
Hardware and software requirements......................................................................................................................... 59

Contents 3
 Supported switches.................................................................................................................................................... 60
SFS personalities............................................................................................................................................................... 60
Support matrix................................................................................................................................................................... 60
Fabric operations and life cycle management............................................................................................................ 60

Chapter 7: SFS with PowerEdge MX............................................................................................ 61

Chapter 8: SFS for PowerScale (Isilon) back-end fabric.............................................................. 62

Supported network topologies.......................................................................................................................................62
Hardware and software requirements......................................................................................................................... 62
Supported switches.................................................................................................................................................... 62
PowerScale requirements............................................................................................................................................... 63
Support matrix................................................................................................................................................................... 63
Replace a switch in SmartFabric mode........................................................................................................................63

Chapter 9: SmartFabric commands..............................................................................................67

smartfabric l3fabric enable..............................................................................................................................................67
show logging smartfabric ...............................................................................................................................................68
show smartfabric cluster.................................................................................................................................................68
show smartfabric cluster member................................................................................................................................ 69
show smartfabric configured-server............................................................................................................................ 70
show smartfabric configured-server configured-server-interface....................................................................... 70
show smartfabric details.................................................................................................................................................. 71
show smartfabric discovered-server............................................................................................................................ 72
show smartfabric discovered-server discovered-server-interface.......................................................................72
show smartfabric networks............................................................................................................................................ 73
show smartfabric nodes...................................................................................................................................................73
show smartfabric personality......................................................................................................................................... 74
show smartfabric uplinks................................................................................................................................................. 75
show smartfabric upgrade-status................................................................................................................................. 77
show smartfabric validation-errors............................................................................................................................... 78
show switch-operating-mode.........................................................................................................................................79

Chapter 10: Best Practices.......................................................................................................... 80

Recommended upgrade sequence for SFS and solution deployments ............................................................... 80
Upgrade sequence in deployments with OMNI .................................................................................................. 80
Upgrade sequence in deployments without OMNI ............................................................................................ 80
Frequently asked questions............................................................................................................................................. 81

Chapter 11: Migration of SFS VxRail deployment from L2 to L3 personality..................................82

SFS VxRail integrated personalities.............................................................................................................................. 82
Migration guidelines.......................................................................................................................................................... 82
SFS VxRail deployment.................................................................................................................................................... 82
Migration procedure for deployments with untagged or tagged Client Management Network setup........ 83

Chapter 12: SFS concepts........................................................................................................... 88

REST_USER account....................................................................................................................................................... 88
Internal fabric components and networks...................................................................................................................88
MSTP Support on L3 personality................................................................................................................................... 91

4 Contents
 Networks..............................................................................................................................................................................91
Uplinks..................................................................................................................................................................................92
Routing profiles............................................................................................................................................................93
Uplink bonding options............................................................................................................................................... 93

Appendix A: Identify the MAC address of QLogic interface for server onboarding.......................94

Appendix B: Supported OS10 CLI commands in SmartFabric mode............................................. 96

Contents 5
 1
About this guide
This guide provides information regarding the integration of SmartFabric Services (SFS) with Dell VxRail, Dell PowerEdge MX,
Dell Isilon/Dell PowerScale devices, and other supported solutions. It covers the following details:
● SFS concepts and its components for leaf and spine deployment.
● Description, configuration information, limitations, and restrictions of SFS for each solution.
● Command reference information for all the SFS commands.
To use this guide, you must have a good knowledge of Layer 2 (L2) and Layer 3 (L3) networking technologies, and data center
deployments.
This document may contain language that is not consistent with current guidelines of Dell Technologies. There are plans to
update this document over subsequent releases to revise the language accordingly.

Text and Syntax Conventions

This guide uses the following conventions to describe text and command syntax.

Bold text UI elements that you click or select.

> (right angle Hierarchy of menu selections.
bracket)
Keyword Keywords are in Courier (a monospaced font) and must be entered in the CLI as listed.
parameter Parameters are in italics and require a number or word to be entered in the CLI.
{X} Keywords and parameters within braces must be entered in the CLI.
[X] Keywords and parameters within brackets are optional.
x|y Keywords and parameters separated by a bar require you to choose one option.

Related Documents
Use the following documentation in addition to this guide to get complete information about the SmartFabric Services
capabilities:

Table 1. Related Documents

Related Documentation Link
● Dell SmartFabric OS10 User Guide SmartFabric OS10 Documentation
● Dell SmartFabric OS10 Installation, Upgrade, and
Downgrade Guide
VxRail Documentation VxRail Networking Solutions
Networking Solutions Support Matrix Support Matrix
PowerEdge MX Documentation PowerEdge MX Manuals and Documents
PowerScale/Isilon Documentation PowerScale OneFS Info Hub
OpenManage Network Integration Documentation OMNI Documentation

6 About this guide

Dell Demo Center
Dell Demo Center is a scalable, cloud-based service that provides 24/7 self-service access to virtual labs, hardware labs, and
interactive product simulations. Several interactive demos are available on the Demo Center. Contact your Dell account team to
get access to the Demo Center.

Documentation Feedback
Dell Technologies strives to provide accurate and comprehensive documentation and welcomes your suggestions and
comments. You can provide feedback in the following ways:
● Online feedback form—Rate the documentation or provide your feedback on any of the product documentation pages at
Dell Support page.
● Email—Send your feedback to [email protected]. Include the document title, release number, chapter title,
and section title of the text corresponding to the feedback.
To get answers to your questions related to Dell Networking Solutions through email, chat, or call, go to Dell Technologies
Technical Support page.

Acronyms
The following acronyms are used throughout this guide:

Table 2. Acronyms
Acronym Expansion
API Application Programmable Interface
BGP Border Gateway Protocol
CLI Command-Line interface
UI User Interface
IOM Input/Output Module
LACP Link Aggregation Control Protocol
L2 Layer 2
L3 Layer 3
RPVST+ Rapid Per-VLAN Spanning-Tree Plus
SFS SmartFabric Services
STP Spanning-Tree Protocol
VLAN Virtual LAN
VLTi Virtual Link Trunking interconnect
VXLAN Virtual extensible LAN
VTEP VXLAN tunnel endpoint

About this guide 7

 2
Change history
The following table provides an overview of the changes to this guide from a previous release to the OS10.6.0 release. For more
information about the new features, see the respective sections.

Table 3. New in 10.6.0.5

Revision Date Feature Description
A03 2025-07-15 Configure networks in Delete multiple networks
SmartFabric simultaneously across all
network types.
Edit networks and ports in a Delete multiple server
server interface profile interface network profiles
simultaneously.
Enable switch ports Refresh the displayed data
for unused ports, ensuring the
port status is accurate and up
to date.
Replace a switch in Perform switch replacement
SmartFabric mode within the PowerScale
SmartFabric environment
using curl requests.

Table 4. New in 10.6.0.2

Revision Date Feature Description
A02 2025-04-11 Change the REST_USER Updated the details specific
password of the SFS fabric to the REST_USER password.
A01 2025-02-28 Edit networks for multiple This feature enables add,
server interfaces within a remove, and override VLAN
server profile configurations for tagged
networks across different
interfaces with the server
interface profile.
Change the REST_USER Change the REST_USER
password of the SFS fabric password of the SFS instance
directly from the SFS UI.

Table 5. New in 10.6.0.1

Revision Date Feature Description
A00 2024-12-18 Enhancement to create a This feature enables you
Layer 3 general-purpose to directly create a Layer
network 3 general-purpose network,
eliminating the requirement to
first create and convert a
Layer 2 network

8 Change history
 3
SFS fundamentals
This chapter provides information about the fundamentals of SFS including overview, supported topologies and platforms,
network fabric formation, and its supported solutions.

SFS overview
SFS is a SmartFabric OS10 feature that provides network fabric automation and API-based programming capabilities. SFS
has different personalities that can be integrated with systems including VxRail, PowerScale, generic PowerEdge servers,
PowerStore, storage, and MX servers. SFS integrated with these solution-specific deployments delivers autonomous fabric
deployment, expansion, and life cycle management.
SFS has two types: SFS for PowerEdge MX and SFS for leaf and spine. The following sections focus on concepts that are
related to SFS for leaf and spine:
SFS for leaf and spine is supported on S-series and Z-series Dell PowerSwitches. See Support matrix for a complete list of
supported platforms. SFS for leaf and spine has two personalities:
L2 single rack personality

NOTE: This personality is not available for deployments after OS10.5.0.5 release. All single rack and multirack deployments
from 10.5.0.5 release and later use the L3 personality.
SFS deployments on OS10 releases from 10.4.1.4 to OS10.5.0.5 support only L2 single rack.
● Provide fabric automation for a single pair of leaf switches.
● SFS deployment is limited to a single rack and cannot be expanded to a multirack deployment.

NOTE: When you upgrade switches with this personality enabled, they operate in the L2 single rack personality only.

● SFS L2 single rack personality is enabled by running a Python script in the OS10 Linux shell. See Enable L2 personality for
more information.
L3 multi rack personality

All SFS deployments using OS10.5.0.5 and later releases support SFS L3 personality and the capabilities are as follows:
● Provides fabric automation for leaf and spine.
● Allows SFS deployment in a single rack and expand to multirack as required.
● Allows you to enable SFS using CLI, API, or UI.

NOTE: The support to enable SFS using UI is not available from 10.5.3.0 release.

SFS fundamentals 9
 NOTE: In an MX-based deployment, the fabric is configured using OpenManage Enterprise-Modular (OME-M) UI, see Dell
OpenManage Enterprise-Modular Editions User Guide available in PowerEdge MX Documentation.
See the Solutions Support Matrix for a complete list of solutions that can be onboarded onto the fabric.
This guide covers the following SFS qualified solutions:
● SFS deployment with VxRail
● SFS deployment with PowerEdge MX
● SFS deployment with PowerScale (Isilon)

SFS UI
OS10 has support for SFS UI to set up the initial SFS configurations in a L3 leaf and spine topology. You access the SFS UI using
the latest version of the following browsers:
● Google Chrome
● Mozilla Firefox
● Microsoft Edge
For more information about SFS UI, see Access fabric setup configuration.

Supported network topologies

Following are the supported network topologies for SFS L3 multirack deployment:
● One leaf switch pair without spine switch
● Multiple leaf switch pairs with a two or more spine switches

SFS supported platforms

SFS is supported on the selected S-series and Z-series PowerSwitches for leaf and spine deployments. The platform support
varies depending on the solutions such as VxRail, PowerScale, and so on. The supported platforms are listed in Supported
Switches section under respective solution chapters.

SFS and supported solutions

SFS has different levels of integration with the following qualified solutions. See the Solutions Support Matrix for the latest
supported versions for all the qualified solutions.

Table 6. SFS and supported solutions

Qualified solutions Dynamic discovery Onboarding type
VxRail Yes * Dynamic or Import from Fabric *
NOTE: Use Import from Fabric
option to manually onboard the
VxRail nodes.

PowerStore X/T Yes Import from Fabric

PowerMax Yes Import from Fabric
Other third-party devices Yes ● Import from Fabric, if NIC supports
LLDP.
● Static, if the NIC does not support
LLDP.
PowerEdge MX N/A

10 SFS fundamentals
Table 6. SFS and supported solutions (continued)
Qualified solutions Dynamic discovery Onboarding type

NOTE: In PowerEdge MX, the servers are discovered and onboarded through
OME-Modular.

* From the VxRail versions 7.0.450 through 8.0.2xx, LLDP is disabled by default. For more information about the impact and
workaround for SFS and OMNI, see the section Dynamic Discovery of VxRail servers update: Impact and Workaround in the
OMNI Release Notes, Release 3.6.
NOTE: Any third-party device that supports standard Ethernet ports and protocols can be onboarded onto the solution
using the onboarding procedure, see Onboard a server.
Dynamic Discovery—Devices that support dynamic discovery send an LLDP TLV. Devices that do not send the LLDP TLV
must be manually added to the fabric.
Onboarding—Onboarding is the process of adding devices to the fabric through the server interface profiles creation.
Supported devices are automatically populated in the SFS UI and OMNI by MAC address, switch, and switch port number
for onboarding to the fabric. For VxRail, the SFS and VxRail Manager can automate the onboarding process. From the VxRail
versions 7.0.450 through 8.0.2xx, the SFS automates the onboarding process. You can also manually onboard the VxRail using
Import from Fabric option. PowerStore and PowerMax systems support dynamic discovery, and you can onboard the server
using the Import from Fabric feature in SFS and OMNI, see Import SmartFabric discovered server interfaces. Other devices
are manually onboarded by specifying the switch and switch port number for each interface, see Create server interface profile.

Creating a network fabric

When you enable SFS on the switches in leaf and spine architecture, a single network fabric is built with all the discovered
switches using industry standard L2 and L3 protocols. SFS supports the following automation capabilities in a leaf and spine
topology:
● Elects one switch from the fabric as the leader switch. From the 10.5.3.0 release, spine switch is also part of the SFS leader
election.
● Creates the infrastructure VLANs on the nodes.
● Allocates all the necessary internal IP addresses for leaf and spine configurations.
● Autoconfigures necessary BGP for all the relevant leaf and spine switches.
● Enables leaf and spine for underlay and overlay.
● Provides workload orchestration of server discovery and binding server profiles to networks.
See Internal components and entities for more information about infrastructure networks and fabrics that are created by SFS.
Also, you can use SFS show commands to view and verify fabric-related configuration.
After the fabric is created, you can configure uplinks, jump host, and onboard servers to the fabric using the SFS UI.

Server discovery and onboarding

SFS discovers and onboards a server based on the LLDP from the server when connected to the fabric.
Onboarding a server involves creating a server profile and assigning the networks (VLANs) to a specific switch port connected
to the server NIC port.
You can onboard a server in the following ways:
● Dynamic onboarding
● Static onboarding

Dynamic server onboarding

When the servers are connected to the fabric, SFS discovers the servers automatically.
From 10.5.3.0 release, SFS dynamically discovers PowerMax devices using the LLDP TLV received from these devices and
configures the discovered interfaces as the untagged member of the VLAN 4091—default client management network.

SFS fundamentals 11
 NOTE: This feature is applicable only for L3 personality.

SFS discovers a host as a known server TLV in LLDPDUs sent through the connected ports. Following are the list of known
servers discovered by SFS:
● VxRail

NOTE: By default, SFS discovers and onboards the VxRail server automatically. From 10.5.3.0 release, SFS provides
an option to onboard the VxRail servers manually by disabling the VxRail Manager Integration option. For more
information, see Edit default fabric settings. This option is applicable only for SFS L3 personality.

NOTE: Starting from the VxRail versions 7.0.450 through 8.0.2xx, LLDP is disabled by default. For more information
about the impact and workaround for SFS and OMNI, see the section Dynamic Discovery of VxRail servers update:
Impact and Workaround in the OMNI Release Notes, Release 3.6.

● PowerStore X
● PowerStore T
● PowerMax
When a known server is discovered on the server-facing port, SFS applies the server profiles or server interface profile
configurations.

Dynamic discovery of unknown servers

With OS10.5.2.2 and later releases, SFS dynamically discovers unknown servers using the standard LLDPDUs sent out through
the connected ports. Unknown server is a host that does not send a Dell custom TLV in LLDP frame. Upon discovery, the client
management network is configured by default.
NOTE: If the LLDP TLVs are advertised with both bridge and routing capabilities, SFS considers that the interface is
connected to a switch or router. Hence, the client management network, VLAN 4091, is not configured on that interface.
For dynamic discovery, the Port-id TLV in the LLDP packet is mandatory. SFS matches the port-ID TLV value in the LLDP
packet against the Interface field in the server interface profile to onboard the unknown server. After the server interface
profile is configured, the onboarding criteria for an unknown server are the same as onboarding a known server.

NOTE: Ensure that the LLDP port-ID TLV value of the unknown server is unique to avoid misconfiguration.

● If both known and unknown LLDP neighbors exist on the same interface, SFS discovers these servers as known and
unknown servers. During onboarding, if the server profile matches with both the known and unknown discovered servers on
the same physical interface or with the same server interface profile, the order of onboarding is as follows:
○ Known discovered server
○ Unknown discovered server
When you offboard the known server, SFS checks for the unknown server and onboards it if available.
● If both known and unknown LLDP neighbors exist with the same port ID, SFS discovers both the known and unknown
servers. During the onboard process, the order of onboarding is as follows:
○ Known discovered server
○ Unknown discovered server
When you offboard the known server, SFS checks for the unknown server and onboards the server if available.
● If there is a static onboarding server profile that is configured on an interface where the known and unknown LLDP
neighbors exist, SFS discovers both the LLDP neighbors. During onboarding if there are conflicts on the physical interface or
server interface profiles, the priority order of onboarding is as follows:
○ Static onboarding
○ Known discovered server
○ Unknown discovered server
When you offboard a statically onboarded server, SFS uses the priority order to search the known servers list first, then the
unknown servers list to onboard the device.

12 SFS fundamentals
Static server onboarding
SFS supports static onboarding of server on assigned ports instead of LLDP-based discovery. This option is used for onboarding
servers that are not discovered by SFS.
To statically onboard a server, you must assign an interface of the leaf switch to which the server is onboarded. See Onboard
nondiscovered server interfaces section in Onboard a Server for more information about static onboarding using SFS UI.
Configuration notes:
● STP is disabled on the server-connected ports.
● The bonding can be static or LACP.
● All existing bonding modes are supported on a statically onboarded server.
● VXLAN does not support STP on access ports and it is not applicable for L3 Fabric. For the VXLAN type of network, you
cannot configure STP; the network topology must remain loop free.
● All existing network types are allowed to be onboarded on statically onboarded servers.
● Since onboarding is static, when server is moved there is no support for moving the configurations along with the server.
● The port-role for the statically onboarded server is EndHost or GenericEndHost.
● When the server profile or server interface profile is deleted, all the impacted interfaces are brought to default configuration.

Server profile and server interface profile

Server profile and server interface profiles are logical entities of SFS. You configure these entities when onboarding a server to
the fabric.
Server interface profile represents the server NIC ports that are connected to the leaf switches. For a server interface profile,
a server interface ID must be configured using the MAC address of the server NIC port. SFS uses the server interface ID to
identify the server NIC port and configure the required networks on the server NIC ports.
Server profile is a list of server interface profiles with a common bonding type. Supported bonding types are AutoDetect and
LACP. You can configure a server profile using SFS UI, see Server onboarding.

Bonding technology and NIC bonding

Bonding technology or type is a property of server profile and the supported bonding types are Autodetect and LACP. NIC
bonding is a server interface profile property that determines whether the server interface profile to be in port channel or
normal interface.
The table describes the behavior of SFS depending on the option of Bonding type and NIC bonding selected when configuring
the server interface profile:

Table 7. SFS behavior for bonding type and NIC bonding

Bonding technology or type NIC bonding SFS behavior
AutoDetect Disable The VLAN is added to the physical
interface when LACP PDUs are not
received. After the LACP PDUs are
received, the VLAN is applied to port
channel interface.
AutoDetect Enable The system waits for LACP PDUs and
after the LACP PDUs are received,
the VLAN is applied to port channel
interface.
LACP Disable The VLAN is added to the physical
interface.
LACP Enable The system waits for LACP PDUs and
adds the VLAN only on the port channel
interface after it receives the LACP
PDUs.

SFS fundamentals 13
By default, the bonding type setting is AutoDetect and NIC bonding is disabled. Dell Technologies recommends that when you
configure the bonding type for a server profile as LACP, you must enable NIC bonding.

Fabric backup and restore

You can back up the SFS configuration to an external device and restore the fabric to a good state using the backed-up
configuration during a failure or error.
The backup functionality allows you to back up all the user configuration to a text file in JSON format. The REST endpoints are
available that returns a text file containing all the user configuration that was done through the REST interfaces.
The restore functionality allows you to restore the fabric to a last known good configuration. The REST endpoint is available that
enables you to stream the backed-up configuration. The restore action wipes the data completely and restores the configuration
on the switches from the backup file. The restore functionality is followed by leadership switchover. For the successful restore
operation to occur, the restore activity should be applied to the same set of switches in the fabric. Also, the SFS personality
should be the same when the backup was performed.
The backup and restore endpoints are accessible only for the users with sysadmin role. The REST payload is encrypted using
the SSL protocol.
You can also restore the configuration for a fabric from SFS UI, see Restore.

Table 8. HTTP Methods

Functionality HTTP Method URL Payload
Backup GET https://<ip-address>/ Output is a text as stream and
redfish/v1/Dnv/Backup Content-Type is application or
octet-stream.
Restore POST https://<ip-address>/ Content-Type: application or
redfish/v1/Dnv/Restore octet-stream and payload
contains the backup file.

14 SFS fundamentals
 4
Setting up SFS
This chapter explains the workflow to setup SFS including initializing SFS and enabling it on leaf and spine switches.

Prerequisites
Ensure that the following are met before enabling SFS on the switches:
● Configure the Out-of-band (OOB) management network on all leaf and spine switches. OOB management network enables
connections to the SFS UI. PowerSwitch S3048-ON or N3248TE-ON can function as an OOB management switch with the
OS10 factory default configuration. For more information about OOB management topology and connection details, see the
respective Solution Deployment Guides.
● Verify the SmartFabric OS10 version on the leaf and spine switches and update them with recommended version that is
listed in the Support Matrix.
● Change the default admin account password on each switch through OS10 CLI after the first login. You can change the
password in Full Switch or SmartFabric mode.
For more information about the OS10 CLI commands, see the Dell SmartFabric OS10 User Guide.

SFS configuration notes

This section lists important behaviors, considerations, and recommendations you must know before deploying SFS:

Recommendations
See Internal components and networks for more information about the SFS components and networks that are created by SFS.
Recommendations and notes regarding the SFS components are:
● Dell Technologies recommends that you do not change or disable the STP settings on switches in SmartFabric mode. Any
change to the STP settings results in a reboot of all the switches in the fabric and disabling STP settings cause loops.
● Dell Technologies recommends using uplinks from a leaf pair as a best practice. Uplinks from leaf switches to external
switches can be L2 or L3. All uplinks from spine switches to external switches must be L3.
● Except for PowerScale (Isilon) back-end deployment, the breakout feature is not supported on the leaf switches for the VLTi
or Leaf-Spine link connections. Use the default speed cables of the leaf switch when creating the Leaf-Spine link or VLTi
connections in a L3 fabric.
○ Leaf-Spine link example—If the default speed on the leaf switch port is 100G and the spine switch port is 400G, use
100G cable for Leaf-Spine link connection. Since the autobreakout feature is enabled on the spine switch, the port speed
of the spine switch is set to 100G automatically.
○ VLTi example—If the default speed of the leaf switches is 100G, use 100G cable for the VLTi connection.
● Ensure that the Designated Leader configuration is set for the leaf or spine switches in the existing fabric before expanding
the fabric to prevent the configuration loss. For more information about how to set the designated leader configuration, see
Designated Leader. Once a leader is elected, it initiates all applications to automatically build the network fabric, and the
leader virtual IP address is advertised for applications to automatically discover the fabric through inband networks.
● The autobreakout feature is not supported on leaf switches. If you want to breakout the ports to configure uplinks, server
connections, or jump hosts, use the SFS UI.
● Ensure that the default networks created by SFS do not conflict with any networks in the existing deployment. If so, change
the default networks using the instructions that are provided in Edit fabric settings.
● When you want to reboot any of the border leaf switches having an L3 uplink with eBGP peer configuration that is
connected to the external switch, Dell Technologies recommends shutting down the interface of the external switch that is
the connected to that leaf switch before rebooting. After reboot and the BGP session is established within the fabric, bring
up the interface of the external switch.

Setting up SFS 15
● Dell Technologies recommends not flapping the admin status of the server connected ports using the SFS UI, OMNI UI, and
REST API to avoid traffic loss on those ports after switch reboot.
● In SmartFabric mode, you can create VLAN using the interface vlan command through OS10 CLI, but you cannot
delete the VLAN from the CLI. Therefore, Dell Technologies recommends using the SFS UI to create, edit, or delete a VLAN.
● When changing the role (spine or leaf) of an existing switch in the fabric, Dell Technologies recommends disabling
SmartFabric mode on all switches in the fabric and enabling it again on all switches with the appropriate role.

Default settings
● SFS creates VLANs from 4000 to 4094 for internal use. Do not use these VLANs for general use.
● SFS creates default networks for the leaf and spine network configuration for the domain ID.
● By default, the autobreakout feature is enabled on the spine switches in SmartFabric mode. Autobreakout works only with
DAC and AOC cables.
● In SFS, all network QoS priorities are set to Iron by default, and it cannot be changed.

SFS behavior
SFS elects one switch from the fabric as a leader switch and designates the remaining switches as the backup switches. If a
leader reloads or fails, a new leader is elected from the backup switches using the keepalive information. The switches that are
configured as Designated leader have a higher priority to become the leader switch. If none of the switches are configured as
the designated leader, any switch can become the leader.
When you expand the fabric, the newly added switches may come up, form a fabric among themselves, and elect a leader before
they are connected to the existing fabric. When the new fabric merges with the existing fabric, SFS elects a new leader switch
for the combined fabric. If one of the new switches becomes the leader, it may overwrite the configuration in the existing
fabric. To prevent this overwrite, ensure that the designated leader flag is set for the current leader in the existing fabric before
expanding the fabric to prevent the configuration loss.

Preferred Leader behavior

Starting from OS10.5.5.0P1 release, the Preferred Leader flag is set automatically on all the leaf switches when you create an
uplink using SFS UI. The preferred leader flag setting can be viewed in the CLI of each switch using the show smartfabric
cluster command.
The provision to configure designated leaders on a fabric is available in the UI. For more information, see Designated leader.

Spanning tree considerations

● To avoid loops, SFS does not allow you to configure the same network on multiple uplinks.
● SFS supports RPVST+ and MSTP. The default spanning tree mode in SFS is RPVST+. Once the fabric is created, you can
change the STP mode using SFS UI, see Edit default settings.

NOTE: When you change the mode using SFS UI, the whole fabric goes through a reboot, and the new mode is set to
MSTP. The reboot action impacts the traffic flow in the cluster.

● By default, RPVST+ is enabled on the uplink interfaces for L3 VLAN networks. When RPVST+ is enabled on the uplink, the
total number of Port VLANs (PV) supported is 400.

BFD support
From OS10.5.3.0 release, SFS provides an option to enable or disable Bidirectional Forwarding Detection (BFD) for external BGP
peer routes. You can enable or disable BFD for an eBGP peer route policy when configuring uplinks, server interface profiles, and
routing profiles. By default, BFD is disabled.
NOTE: You cannot modify the BFD timers in SFS:
● For the S4100-ON series platform, the BFD interval is set to 500 milliseconds with a multiplier of 3 by default.
● For other platform switches, the BFD interval is set to 300 milliseconds with a multiplier of 3 by default.

16 Setting up SFS
If you want to edit the BFD configuration for a routing policy, you must delete the existing routing policy and create a policy with
the respective BFD configuration.

Configure SFS
By default, PowerSwitches boot in Full Switch mode when you power them up with SmartFabric OS10. This information explains
how to setup a fabric.
1. Enable SFS and set roles on the switches.
2. Connect to SFS UI and complete the relevant fabric configurations to setup the fabric. See SFS UI for more information
regarding login procedure and fabric configurations.
For SFS with PowerScale (Isilon) deployment, you must enable SFS with roles on all the switches using CLI. All other SFS
initial deployment operations using UI are not required. For more information regarding PowerScale deployments, see the related
documents available in PowerScale Info Hubs.

Enable SFS on PowerSwitches

You can enable SFS on PowerSwitches through CLI, UI, or REST API from OS10.5.0.5 and later versions. This option is
applicable only for SFS L3 leaf and spine personality.
NOTE: The support to enable SFS using UI is not available from 10.5.3.0 release.

After you enable SFS on all the switches in a leaf and spine deployment, a network fabric is created automatically with the
default fabric settings. See Internal fabric component and networks for more information about the default settings.

Enable SFS using CLI

To enable SFS on a switch using CLI, run the smartfabric l3fabric enable command and set a role. For more
information, see smartfabric l3fabric enable.
After you enable SFS on the switches and set a role, the system prompts for a confirmation to reload and boots in the
SmartFabric mode. Enter Yes to continue. In SmartFabric mode, the CLI is restricted to global switch management features and
monitoring. Using this command, enable SFS on all the switches with the corresponding role to create a fabric.
The following example shows how you can enable SFS on switches and set roles:
● Enable SFS on a switch with a spine role in CONFIGURATION mode.

OS10(config)# smartfabric l3fabric enable role SPINE

Reboot to change the personality? [yes/no]: yes

● Enable SFS on a switch with a leaf role in CONFIGURATION mode. In SFS, the two leaf switches are automatically
configured as a VLT pair. The Ethernet interfaces 1/1/4 and 1/1/5 are the VLTi interfaces.

OS10(config)# smartfabric l3fabric enable role LEAF vlti ethernet 1/1/4-1/1/5

Reboot to change the personality? [yes/no]: yes

If the VLTi interfaces are not contiguous, use the following syntax to enable VLT:

smartfabric l3fabric enable role LEAF vlti ethernet 1/1/25,1/1/29

● (Optional) Enter the domain ID for the SFS fabric in CONFIGURATION mode.

OS10(config)# smartfabric l3fabric enable role LEAF vlti ethernet 1/1/4-1/1/5 domain
101

OS10(config)# smartfabric l3fabric enable role SPINE domain 101

By default, domain ID is set to 100. You can edit the domain ID from the SFS UI. See Edit default fabric settings.

Setting up SFS 17
In Isilon or PowerScale back-end deployments, when enabling SFS in a leaf and spine topology, no VLTi configuration is required
for the leaf switch. To set the leaf role for a switch, use the smartfabric l3fabric enable role LEAF command
without the VLTi parameters.

Related Videos
Enable SFS on PowerSwitches using a CLI command:
Enable SFS using a CLI command

Enable SFS using RESTCONF API

You can enable SFS on OS10 switches using the RESTCONF API. For more information regarding general RESTCONF API
operations, see SmartFabric OS10 User Guide.

Description Enables SFS on the switches with leaf and spine roles.
RESTCONF endpoint /restconf/data/dell-smart-fabric:config-personality
JSON content
(spine) {
"dell-smart-fabric:config-personality": {
"service-enable":true,
"role": "SPINE",
}
}

Example
curl -k -u admin:admin -H 'Content-Type: application/json' -i -X POST
-d \
'{"dell-smart-fabric:config-personality":
{"service-enable":true,
"role":"SPINE"}
}'
https://100.104.26.104/restconf/data/dell-smart-fabric:config-
personality

The following example shows how to enable SFS on a leaf switch with an IP address (100.104.26.104) and VLTi interfaces as
1/1/5 and 1/1/6.

JSON content (leaf)

{
"dell-smart-fabric:config-personality": {
"service-enable":true,
"mode": "L3 Fabric",
"role": "LEAF",
"vlti": [
"ethernet1/1/5",
"ethernet1/1/6"
]
}
}

Example
curl -k -u admin:admin -H 'Content-Type:
application/json' -i -X POST -d \
'{"dell-smart-fabric:config-personality":{"service-
enable":true,"role":"LEAF",
"vlti": [
"ethernet1/1/5",
"ethernet1/1/6"
]
}
}'
https://100.104.26.104/restconf/data/dell-smart-fabric:config-
personality

Example—Usage of curl command on a Linux server is as shown:

18 Setting up SFS
Enable SmartFabric Services L2 personality using script
In SFS L2 personality, you can enable SFS only using API. All deployments from SmartFabric OS10.4.1.4 to OS10.5.0.x version
support only the single rack network fabric.
NOTE: The L2 personality is not available for new deployments after 10.5.0.5 release.

A python script is used to enable SFS. For more information about L2 personality commands and information, see VMware
Integration for VxRail Fabric Automation SmartFabric User Guide, Release 1.1 available in OMNI Documentation page.

Enable SFS using UI

Starting from OS10.5.3.0 release, the option to enable SFS mode using UI is not available. You can use the UI to enable SFS in
versions prior to 10.5.3.0.
To enable SFS using the UI, following the instructions:
1. Enable the RESTCONF API on the switch using the OS10 CLI. To enable RESTCONF API, use rest api restconf
command in CONFIGURATION mode. See SmartFabric OS10 User Guide for more information about RESTCONF API.
2. Open a browser session, go to https:// switch-mgmt-ip-address .
3. Log in to a switch using the credentials that are created to access an OS10 switch. The default username is admin and
password is admin.
4. Enable SFS on the switch using the Edit option that appears in the upper-right side of the page.
5. Enter the role of the switch and click OK to enable SFS.
● Spine:

Setting up SFS 19
 ● Leaf:

After you enable SFS on a switch, the system reloads to apply the configuration.
6. Repeat the steps 1 to 4 on all the remaining switches to enable SFS and set the role. All switches reload to form a fabric.

Verify the switch operating mode

To verify that the switches are in SmartFabric mode, run the show switch-operating-mode command on each switch.

OS10# show switch-operating-mode

Switch-Operating-Mode : SmartFabric Mode

For more information regarding the CLI, see the show switch-operating-mode command.

Complete the fabric setup

After enabling SFS, you must configure logical entities to complete the SFS setup.
1. Log in to SFS UI. For more information, see Deploy and manage a fabric.
2. Configure unique names for SFS components. For more information, see Update fabric names and descriptions.
3. (Optional) Configure breakout switch ports based on the deployment requirements. For more information, see Breakout
ports.
4. Configure uplinks for connectivity. For more information, see Create uplinks.
5. Configure Jump host for client management based on the requirement. For more information, see Configure Jump host.
6. Onboarding a server by applying server interface profile. For more information, see Server onboarding.
SFS UI also provides additional features that can be configured if needed. For more information, see Access fabric setup
options.

20 Setting up SFS
 5
Managing SmartFabric with the SFS UI
You can use SFS UI to set up a fabric in a leaf and spine topology. The SFS UI helps you with the SFS deployment operations
and management of the switches in a fabric.

Online help
SFS UI provides online help support for documentation which can be accessed from the SFS UI.
To see the appropriate help topic of a menu in the UI, choose a menu tab and click the help icon in the UI.

Access fabric setup options

Follow the instructions to access fabric setup options:

Log in to SFS UI
You can launch the SFS UI directly using the leader IP address. To get the IP address of the leader switch in the L3 fabric, use
the show smartfabric cluster command. For more information, see SFS commands.
From 10.5.6.1 release onwards, SmartFabric OS10 allows Multi-Factor Authentication (MFA) using RSA SecurID. MFA provides
an extra layer of security to the switches. For more information regarding MFA, see Dell SmartFabric OS10 User Guide.
Use the SFS UI to manage your SFS fabric:
1. Enter the leader switch IP address of the SFS fabric in your browser to launch the SFS UI. For example, https://
10.104.26.27.
2. Enter the credentials that are created to access an OS10 switch from the console or through a network connection. The
default username is admin, and the password is admin. You can also use any user-configured accounts with the sysadmin
role as credentials on the switch you are connected to.

Managing SmartFabric with the SFS UI 21

3. When MFA is enabled on the switches, the SFS UI prompts for the SecurID passcode authentication
once the first authentication factor (username and password) has been successfully completed.

You can also log in to SFS UI using the management IP address of any switch.
If you logged in to a nonleader switch in a fabric, the page displays the fabric mode, the role of the switch, and the IP address of
the leader. Click the link next to the leader IP address to log in to the leader switch. All fabric configurations can be performed
only from the SFS leader switch.

Upon successful login, the UI displays the summary information.

22 Managing SmartFabric with the SFS UI

The system validates the username and password using the REST token-based authentication. For more information, see the
REST Token-Based Authentication topic in the Dell SmartFabric OS10 User Guide.
The session is automatically logged out after the token time expires. The default token timeout value is 120 minutes. To
terminate the session manually, click Logout button at the upper right of the UI.

SmartFabric overview
Starting from 10.5.3.0 release, SFS UI displays a consolidated view of key metrics such as device status, device health, and the
latest fabric compliance errors for the fabric. The Overview dashboard displays the following metrics:
Device Status—Displays the status of the all the switches that are deployed in the SmartFabric instances along with the
number of switches in each status.
● Green—Indicates that the device is online.
● Red—Indicates that the device is offline.
Device Health—Displays the overall health of all the devices in the fabric.
Fabric Compliance—Displays the misconfiguration and compliance violations identified in the SmartFabric instance.
● Module—Name of the module in which the misconfiguration or compliance errors occurred.
● Status—Compliance status of each module in the fabric.
● Error Count—Number of errors in each module.
You can view the detailed list of all compliance errors in the SmartFabric instance from Serviceability page.

View node details

From Summary page, you can view the switch details in the fabric.

1. From Summary > Fabric Nodes, you can view the list of spine and leaf nodes that are deployed in the fabric.
2. Click the Fabric IDs to view the detailed information of the spine and leaf switches connected in the fabric. The details
include:
● Switch name
● Switch model
● OS10 version
● Switch role
● IP address of the switch
● Status of the switch (online or offline)
3. Click Domain at any time to update the fabric details.

Managing SmartFabric with the SFS UI 23

View fabric topology
The Topology tab displays the graphical topology of the network fabric for the selected SmartFabric instance. You can also
view the details of the switch in the fabric.
The L3 leaf and spine topology view that is created after you enable SFS. The topology view displays the switch icons with the
hostname and the service tag information under each node and the link connectivity between the switches. Mouse over a fabric
to see the detailed information about the leaf and spine switches, and the link connectivity.

Manage switches in a fabric

You can manage the spine and leaf switches available in a fabric.

Change port configuration on a switch

Change the port configuration such as auto negotiation or MTU of a port or multiple ports on a leaf or spine switch:
CAUTION: Changing the interface configurations can potentially cause a disruption in service. Ensure that you
are aware of the network settings and the remote peers that are connected to the interfaces before changing
the MTU, auto negotiation, admin status. If the configuration does not match the connected peer switch, it can
lead to connectivity issues.
1. Select Switches.
2. Select the spine or leaf switch by clicking the arrow to view more information.
3. Select a category (All Ports or Unused Ports).
4. Select a port from the category.
5. Click Edit.
6. Edit the following details:
● (Optional) Name
● (Optional) Description
● Admin status
● MTU
7. Click Edit.

24 Managing SmartFabric with the SFS UI

Configure auto negotiation status
You can enable or disable the auto negotiation on a single port or multiple ports. The auto negotiation option is not applicable for
port channel interfaces. If the selected list has a port channel interface when configuring auto negotiation, the system displays a
warning to clear the port channel interfaces from the selected list. Clear the port channel interface selection and try again.
To enable auto negotiation:
1. From All Ports, select a port or multiple ports.
2. Click Enable Auto Neg. The system displays a warning message.
3. Click Yes to confirm.
The system displays the stage-wise progress of the interface status.
To disable auto negotiation:
1. From All Ports, select a port or multiple ports.
2. Click Disable Auto Neg. The system displays a warning message.
3. Click Yes to confirm.
The system displays the stage-wise progress of the interface status.

Set MTU value

Set the maximum transmitting unit (MTU) for the port:
1. Select a port or multiple ports and click Set MTU.
2. Enter the MTU value and click Set. By default, the MTU value is set to 9216.
3. Click Yes to confirm.
The system displays the action success or failure message.

Enable switch ports

Dell Technologies recommends to:
● Enable the port status to operationally up before adding any devices to the port, if the port is disabled using SFS UI.

NOTE: Devices that are connected to the disabled port are not discovered.

● Ensure that the ports are UP before adding any switches, when you expand the leaf and spine fabric deployments.
● Ensure that the switch port is in UP, when onboarding a server to a leaf switch.
You can view and manage the unused ports in the switches. To enable or disable unused ports:
1. From Switches tab, select Unused Ports category to view the list of unused ports available in the leaf or spine switch.
2. Select a port or multiple ports from the list.
3. Click Enable Admin Status.
To disable the ports, select a port or multiple ports, and click Disable Admin Status. The system displays the change status
and update success message on completion.

On the Unused Ports tab, click the Refresh icon to update the displayed data and ensure that the status of unused ports is
accurate.

Configure breakout ports

You can configure breakouts for the Ethernet ports or port-group only on the leaf switches to connect to the external device or
jump host.
Use the following procedure to breakout switch ports:
NOTE: By default, the autobreakout feature is enabled in SmartFabric spine switches. SFS UI does not provide an option to
breakout ports in spine switches. In leaf switches, the autobreakout feature is disabled, so breakout configurations must be
done manually.
1. From Switches tab, click Leaf Switches.

Managing SmartFabric with the SFS UI 25

2. Select a leaf switch from the list for which you want to breakout.
3. Click Breakout Port category.
4. Select a port that you want to breakout and click Breakout Port.
NOTE: The existing configuration of the port is reset to default when you configure a breakout port.

5. Select the appropriate breakout option for the port from the list.
6. Click Submit. To view the details of the breakout ports, select a port to view the properties of the port.
The system displays breakout port configured successful or failure message.
To view the details of the breakout ports, select a port to view the properties of the port.
NOTE: If a server interface profile or network is associated with a port, breakout configuration fails. You must remove
the association before performing the breakout configuration.
When configuring breakout on any switch port that is already onboarded with server interface profile and associated with
networks, follow the procedure:

Table 9. Breakout procedure

Steps Action How to
1 Delete the server interface profile from a. Click the Server Interface tab and
the fabric to delete the networks select the server interface profile
associated with the service interface from the list.
profile. b. Click Delete.
2 Disable admin status of all the ports a. From Switches tab, select the
associated with the switch port-group. respective leaf switch > Unused
Ports.
b. Select all the ports associated with
the port-group and click Disable
Admin Status.
3 Configure breakout mode for the port. a. From Switches tab, select
respective leaf switch > Breakout
Port.
b. Select the interface and click
Breakout Port.
c. Select the breakout mode and click
Submit.

Configure FEC
Forward error correction (FEC) enhances data reliability. By default, FEC is enabled in SFS mode. For more information about
FEC, see SmartFabric OS10 User Guide. You can configure FEC on 25G, 50G, 100G, and 400G interfaces. The supported FEC
types are:
● CL74-FC—Supports 25G and 50G
● CL91-RS—Supports 100G
● CL108-RS—Supports 25G and 50G
● CL91-RS544—Supports 100G and 200G Pulse Amplitude Modulation 4-level (PAM4)
● CL134-RS—Supports 50G PAM4
● CL119-RS—Supports 200G and 400G PAM4
● Auto—Set the port to autonegotiate speed with a connected device
● Off—Disables FEC
NOTE: FEC is not supported on 10G and 40G speed types.

To configure FEC:
1. Select Switches.
2. Select the spine or leaf switch by clicking the arrow to view more information.
3. Select the All Ports category from spine or leaf switch.

26 Managing SmartFabric with the SFS UI

4. Select a port and click Configure FEC.
5. Select the FEC type from the list. The list displays the default FEC types applicable for the specific media, Auto, and Off
options.
Default FEC settings are determined by the inserted media type.
NOTE: The default FEC mode for 25G AOC and SR transceivers is CL108-RS. If a peer device does not support
CL108-RS but only CL74-FC, the ports do not come up. To enable the ports, configure the FEC type on the OS10 switch
to CL74-FC using the fec command.

6. Click Ok.

Configure jump port

A jump port is a designated port to which an external device such as a laptop can be connected. You can configure only one port
in a leaf switch port as a jump port for the external device to connect to L3 fabric. Select any available port that is not part of
an uplink, VLTi, and port connected to a server in fabric.
In VxRail deployment, a jump port is primarily used to bring up a VxRail cluster. By default, all VxRail nodes are placed in the
client control and client management networks. In VxRail deployment, the jump port is placed in the client management network
to reach the default VxRail Manager VM.
Use the following procedure to configure the jump port:
1. From Leaf Switches, select the leaf switch from the list.
2. Select the Jump Port category.
3. Click Add.
4. Enter the following details:
● Name—Name of the jump port.
● Interface Name—Select an interface of the leaf switch as the jump port.
● Untagged Network—Associate an untagged network with the jump port.
5. Click Ok.
The system displays a jump port creation success message.

Edit Jump port

You can edit the jump host configuration starting with the OS10.5.6.0 version.
1. From Leaf Switches, select the leaf switch from the list.
2. Click the Jump Port category.
3. Click Edit.
4. Edit the following options:
● Name—Name of the jump port.
● Interface Name—Change the interface that you want as a jump port from the available list.
● Untagged Network—Change the network that must be untagged.
5. Click Edit.
The system displays a jump port edit success message.
You can delete the jump port. Select the leaf switch for which you want to delete the configured jump port and click Delete.

Designated leader
From the SFS 10.5.4.4 release, you can configure the switches in the fabric as designated leaders. When the designated leader
is configured in a fabric, only the switches that are configured as designated leaders can be elected as the leader, and SFS
elects one of the designated switches to be the leader. When switches are configured as designated leaders, the preferred
leader flag is automatically set for those switches.

NOTE: The provision to configure preferred leader is no longer available in the SFS UI as of the 10.5.5.0P1 release.

Designated leader behavior

Managing SmartFabric with the SFS UI 27

● If a switch is configured to be a designated leader and it is not the current leader, the switch immediately becomes the
leader.

NOTE: During the leader switchover, the SFS instance may be unavailable for about four minutes. Data forwarding is
not affected.

● When the current leader is designated as the leader, the leader switchover does not occur.
You can remove the designated leader configuration from the switches at any time. If the designated leader feature is not
configured in the fabric or all the switches configured as designated leaders are offline, a new leader is elected from the list of
fabric switches.

Backoff timer
The backoff timer is used when the designated leader switches in the fabric are offline. If the designated leader switches are
offline, the fabric waits until the backoff timeout expires before electing a new leader from the nondesignated leader switches.
When a designated switch comes online, the nondesignated leader relinquishes control to the designated node, allowing the
designated node to become the leader.

Configuration notes
● Dell Technologies recommends configuring the spine switches as designated leader for better performance.
● Dell Technologies recommends configuring at least two or more designated leaders.
● Designated leader cannot be removed on a switch that is set as preferred leader.
NOTE: If the current leader switch (also configured as the designated leader) is removed from the role and a new switch is
assigned as the leader, SFS UI loses connectivity during the leader switchover. You must use the new leader's IP address to
log in to the SFS UI.

Configure designated leader

Use the following instructions to set the switches in the fabric as designated leaders and backoff timer from the UI:
1. Click Switches.
2. Click Configure Designated Leader.
3. Select a switch or multiple switches from the list that you want to assign as the designated leader.
4. Enter the backoff timer in minutes. By default, the backoff timer is set to 3 minutes.
The range of the timer can be set from 3 to 30 minutes.
5. Click Submit.

Configure server interface profile

See Server discovery and onboarding for more information about server onboarding and its types. You can onboard a server
statically or dynamically.

Onboard a server onto the fabric

You can onboard a server statically or dynamically. For more information about server onboarding types, see Server discovery
and onboarding. When entering the server interface ID, you must use only the MAC address of the host VM NIC as the server
interface ID when onboarding the server. OMNI automation works only with the MAC address as the server interface ID and
not with any name. By default, the auto breakout feature is enabled in SmartFabric spine switches. SFS UI does not provide an
option to breakout ports in spine switches.

Onboard discovered interfaces—Dynamic onboarding

Use the following procedure to dynamically onboard a discovered server:
1. From the Server Interface tab, click Create.

28 Managing SmartFabric with the SFS UI

2. Enter the server interface ID.
NOTE: You cannot configure a duplicate server interface ID. Use the MAC address of the host VM NIC as the server
interface ID without the colon (:).

3. Select the server profile. The template changes based on the server profile selection.
● Existing server profile—Select the server profile ID from the list.
● New server profile—Create a profile ID. Enter the unique string for the server profile ID and the bonding type.
4. Select one or multiple networks for tagged and untagged networks.
5. Select No for static onboarding.
6. Select Enable or Disable for NIC bonding.
7. Click Create.

Onboard nondiscovered server interfaces—Static onboarding

To statically onboard the Marvell QLogic interfaces, you must identify the correct LLDP MAC address. For more information
about the instructions, see Identify the MAC address of QLogic interface for server onboarding.
Use the following procedure to statically onboard the servers that are not discovered by SFS:
1. From the Server Interface tab, click Create.
2. Enter the server interface ID.
NOTE: Use the MAC address of the host VM NIC as the server interface ID without the colon (:). For example,
f8f21e2d78e0.

3. Select the server profile. The template changes based on the server profile selection.
● Existing server profile—Select the server profile ID from the list.
● New server profile—Create a profile ID. Enter the unique string for the server profile ID and the bonding type.
4. Select one or multiple networks that must be tagged and untagged.
5. Select Enable or Disable for NIC bonding.
6. Select Yes for static onboarding.
7. Select the leaf switch and the interface on which the server is connected.
8. Select the routing protocol. The template changes based on the routing protocol.
● None
● eBGP—Enter the routing policy name, peer interface IP address, peer ASN, description (optional), and BFD
configuration.
● Static Route—Enter the routing policy name, network address, prefix length, next hop IP address, and description
(optional).
9. Click Create.

Import SmartFabric discovered server interfaces

Automate onboarding of server interface profiles by importing profiles that are discovered by SFS. From the OS10.5.3.0 release,
Import from Fabric is available from SFS UI to import and onboard the discovered server interfaces to the fabric.
When known servers are connected to the fabric, SFS discovers the servers automatically and onboards the discovered servers
as part of this workflow. SFS discovers the hosts or servers as known using the originator field in the Dell custom LLDP TLV
sent by the servers. From OS10.5.2.2 release, SFS discovers unknown servers, and you can onboard the unknown servers using
the Import from Fabric option. Onboarding unknown servers is applicable for the SFS L3 leaf and spine personality. Use this
feature to onboard a known or unknown server.
● Known server—A known server is a host that sends a valid originator in Dell-specific (custom) TLV in LLDP frame that is
recognized by SFS.
● Unknown server—An unknown server is a host that sends LLDP frames that do not include the Dell-specific TLV.
1. Click Server Interface.
2. Click Import from Fabric. Discovered Server Interface window appears with the list of discovered interfaces.
NOTE: The interface that is already associated with a server interface profile is not listed in the discovery table.

Managing SmartFabric with the SFS UI 29

3. (Optional) Select the Edit icon at the end of each row to edit the server profile information of each interface. You can edit
the following:
● NIC bonding configuration
● Static onboarding configuration
4. (Optional) If needed, add server interface networks for the discovered server interface profiles. For more information about
adding networks, see Add networks section.
NOTE: This action overwrites the existing networks of all the server interface profiles.

5. (Optional) If needed, add the server profile for the interface. For more information about adding server profile, see Add to
Server Profile section.
6. Select one or multiple discovered interfaces and click Update.

Add to Server Profile

To add the discovered interfaces to a new or existing server profile:
1. Select one or more discovered interfaces and click Add to Server Profile.
2. Select the server profile to which you want to add the discovered server interfaces.
● Existing Server Profile—Select the Server Profile Id to associate the interface with the existing server profile and click
Associate.
● New Server Profile—Enter the Server Profile Id and Bonding Type to associate the interface with the new server
profile and click Associate.
The system displays the server interface profile association success message.

Add Networks
To add the networks to the discovered interfaces:
1. Select one or more interfaces from the list and click Add Networks.
2. Associate the networks with the discovered interfaces.
a. Select one or multiple networks for tagged networks.
b. Select a single network for untagged network.
c. Click Add.
NOTE: Adding the networks overwrites the existing networks of all the server interface profiles.

The system displays the server interface networks addition success message.
Remove from server profile
To remove the interface from the server profile, select one or more interfaces from the list, and click Remove from Server
Profile.

Edit networks and ports in a server interface profile

You can edit the network and port configuration in a server interface profile. You can also view the detailed information of a
server interface profile.

Edit networks on a server interface profile

Use the following procedure to edit the networks on a server interface profile:
1. Select the server interface ID from the list to view the detailed information.
You can edit the networks for up to 10 server interface IDs/interfaces at a time.
2. Click Edit Networks.
3. Select the changes that you want to apply to the untagged network.
When you edit the untagged networks, the selected network replaces the current untagged network that is associated with
the server interfaces.

30 Managing SmartFabric with the SFS UI

4. Select the changes that you want to apply to the tagged networks.
● Append/Remove—Select this option to add new tagged networks to or remove existing tagged networks from the list
for the selected interface.
○ Tagged Networks to be Appended—Select the tagged networks that you want to include to the existing list of
tagged networks.
○ Tagged Networks to be Removed—Select the tagged networks that you want to remove from the existing list of
tagged networks.
● Override—Select this option to replace the existing tagged networks for the selected interfaces.
○ Tagged Networks to be overridden—Select the tagged networks that you want to replace from the current list.
5. Click Edit.
CAUTION: Editing networks impacts network traffic and replaces existing network configuration. Dell
Technologies recommends performing this activity during a scheduled maintenance window.

The system highlights the differences in the configuration changes made to the tagged networks.
6. Click Ok after validating the edits.

Edit ports on a server interface profile

Use the following procedure to edit ports on a server interface profile:
1. Select the server interface ID from the list and click Edit Ports.
2. Edit the Static Onboarding Option and the NIC Bonding configuration for the profile. The template changes according to
the selection.
3. Click Edit.
The system displays the server interface profile update success message.

Delete server interface profiles

Starting with the 10.6.0.5 release, you can now delete multiple server interface profiles directly from the OMNI UI.
1. Choose the server interface profiles from the displayed list by checking the corresponding boxes, then click Delete.
The system displays a delete confirmation prompt.
2. Click Delete to confirm the removal of the server interface profiles.
The system displays the delete success message.

Create server template

The server template has network configuration settings that allow you to dynamically onboard a server or multiple servers
automatically. From the 10.5.4.4 release, you can create a server template. You can associate a server type with a server
template that contains tagged networks and an untagged network. When this particular type of server interface is discovered,
SFS configures the associated networks in the template to the server interface by default.
Before configuring the server template, ensure that the required networks have already been created.
You can create only one server template for a server type. The supported server types are VxRail, PowerMax, and PowerStore.
The server template feature should only be used when all nodes of the same type (VxRail, PowerMax, or PowerStore) shares all
the same networks.
NOTE: VxRail/ESXi host interfaces that have been onboarded using the server template feature do not work with OMNI-
vCenter automation. For OMNI-vCenter automation, a server interface profile for those interfaces must be created. When
you create a server interface profile for interfaces that have been onboarded using the server template, the networks
specified in the server interface profile take priority over the networks specified in the server template.

NOTE: Configuring the server interface profile takes precedence over the server template.

Managing SmartFabric with the SFS UI 31

Table 10. Server Template Behavior earlier to 10.5.6.1 release
Scenario Behavior
Create server template before server interface profile The server template is applied.
creation.
Create server template after server interface profile creation. The configured server template has no effect.
Edit server template before server interface profile creation. The modified server template is applied.
Edit server template after server interface profile creation. The modified server template has no effect.
Delete server template. You cannot delete the server template once it has been
created, and the UI displays an error message. Unable
to delete the server template because it
is currently associated with one or more
interfaces. Remove references to the server
template and retry the operation. To delete the
server template, you can follow the workaround:
1. If the server template type is not applicable for the server,
you can delete the server template directly.
2. If the server template is applicable for the server:
a. Shut down the interfaces that are connected to the
server
b. Reboot the leaf switch to which the server is
connected
c. Delete the server template
3. If the server interface profile is created, the server is
onboarded, and the server template is configured for the
server type:
a. Delete the server interface profile
b. Shut down the interfaces that are connected to the
server
c. Reboot the leaf switch to which the server is
connected
d. Delete the server template

To create a server template:

1. Click Server Interface.
2. Click Server Template.
3. Click Create.
4. Enter the following details:
● Name of the server template.
● Select the type of the server from the list.
● Select the tagged networks from the list.
● Select an untagged network from the list.

5. Click Create. The system displays a creation success message.

Create a server template with enhancements (10.5.6.1 and 10.5.6.4 releases)

Earlier to 10.5.6.1 release, server interface profiles for a server type have to be manually created based on the discovered server
list while onboarding the server. For more information, see Server discovery and onboarding.
Starting from 10.5.6.1 release, SFS has the capability to unify server templates and server interface profiles and reduce the user
input towards onboarding the servers. With the new behavior, when a server template is created, SFS automatically configures
the server interface profiles and server profiles based on an autoprovision flag setting in the server template.
Automation requires server interface profiles in order for the configuration to be propagated from the server. When a server
template is created with the autoprovision flag set to Yes, SFS generates profiles for servers discovered with LLDP by matching
the server type to that of the template. The profiles are generated using the default parameters and tagged and untagged
networks configured in the server template.

32 Managing SmartFabric with the SFS UI

 NOTE: Servers must send the required LLDP information to the port connected to the OS10 switches without which they
cannot be discovered and onboarded.
SFS also retains the conventional method of creating a server interface profile manually. If the servers already have the profiles
configured, SFS does not overwrite existing server interface profiles. When a server is removed from the fabric, SFS retains the
existing server interface profile and allows you to delete it if necessary.
When the autoprovision flag is set to Yes, a server profile is autogenerated for a server template, when the first
server gets onboarded using the server template. The server ID of the server profile is created using the template
<server_template_name>_<server_type>_Auto. If a server profile with the same name exists, SFS uses it instead
of autogenerating a profile. The bonding technology is set to AutoDetect.
The following values are used to autogenerate the server interface profile:
● Server interface ID—MAC address (without ":")
● Interface discovered—Yes
● NIC Bonding—Disable
● Static Onboarding—No
● Untagged network and tagged networks—Inherited from the server template configuration.

Earlier to 10.5.6.4 When the relevant server template is deleted or modified, the autogenerated server interface and
release server profiles are not deleted or updated.
From 10.5.6.4 When the server template is configured with the propagate network change setting enabled and when
release the server template is updated or deleted, the autogenerated server interface profiles can be updated
or deleted.

Starting from 10.5.6.4 release, an enhancement is available to propagate the networks to the server interface profile when the
server template is modified with network configurations. Depending on the autoprovision and network propagation parameters,
the following behaviors can be observed:

Table 11. Server template behavior—10.5.6.4 release

Autoprovision server Propagate network Autogenerate Update behavior Delete behavior
interface setting changes setting behavior
Yes Yes Server templates Any changes to the If a server template is
automatically generate network configurations deleted, the
the server interface in the server corresponding
profiles and server templates cascade to autogenerated server
profiles based on the the existing server interface profiles are
server types. interface profiles. also deleted based on
server ID
(servertemplatename_s
ervertype_Auto). The
server profile still exists
and needs to be deleted
manually, and the
interfaces move to the
default configuration.
Yes No Server templates Any changes to the If the server template
automatically generate networks in the is deleted, it has no
server interface profiles server templates are effect on the relevant
and server profiles. not cascaded to server interface profiles
the existing server and server profiles.
interface profiles.
No No Server templates do Update has no Delete has no
NOTE: Propagate not generate any server effect on the server effect on the server
network change interface profiles or interface profiles interface profiles
setting is disabled if server profiles. The because there are no because there are no
the auto provision networks in the server auto generated server auto generated server
templates are applied interface profiles. interface profiles.
server interface
as default networks
setting is not set. on the discovered
server interfaces when
there is no manually

Managing SmartFabric with the SFS UI 33

Table 11. Server template behavior—10.5.6.4 release (continued)
Autoprovision server Propagate network Autogenerate Update behavior Delete behavior
interface setting changes setting behavior
provisioned server
interface profiles.

To create a server template:

1. Click Server Interface.
2. Click Server Template.
3. Click Create.
4. Enter the following details:
● Name of the server template.
● Select the type of the server from the list.
● Select the autoprovision of server interface setting:

NOTE: This flag is available from 10.5.6.1 onwards.

○ Yes—The autoprovision flag is set, and the server interface profiles for the discovered interfaces are created
automatically based on the server type.
○ No—The autoprovision flag is not set, hence no server interface profiles are created automatically. The server
interface profile must be manually created with the relevant networks.
● Select the propagate network changes setting:

NOTE: This option is available from 10.5.6.4 onwards.

○ Yes—The propagate network flag is set. Any changes to the networks in the server templates propagate to the
existing server interface profiles.
○ No—The propagate network flag is not set. Any changes to the networks in the server templates do not propagate
to the existing server interface profiles.
● Select the tagged networks from the list.
● Select an untagged network from the list.

5. Click Create. The system displays a creation success message.

34 Managing SmartFabric with the SFS UI

Edit a server template
You can also edit the networks in the server template.
If you change the networks in the server template before creating the server interface profile, the changes are reflected in the
server interfaces. If the server interface profile for those interfaces is already configured, the server template changes are not
applied.

NOTE: You cannot change the autoprovision flag setting while editing the server template.

1. Select the template from the list and click Edit.

2. Edit the tagged and untagged network list.
3. Click Edit. The system displays edit success message.
You can also delete a server template.

Configure and manage uplinks

Uplinks enable the network fabric to communicate with the external network. Before creating an uplink, ensure that the external
network is configured with the L2 or L3 setup. Any ports available on the leaf switches may be used as uplinks, provided they
are compatible with the corresponding ports on the external switches.
SFS supports eBGP and static routes profiles and LACP and static for uplink bonding. For more information, see Uplinks section.

Configure L2 Uplink
You can create an uplink by selecting the fabric with a unique name, and select the interfaces, and networks to create a user
uplink. From the 10.5.6.0 release, you can specify the media type when you are configuring an uplink in the fabric.
Use the following procedure to create a L2 uplink:
1. Click Uplink > Create.
2. Enter the uplink details:
● Uplink Port Type—Select the uplink port type as L2.
● Name—Enter the name of the uplink.
● Media Type—Select the media type from the list:
○ Ethernet—Select this option when you want to configure an uplink with spanning-tree protocol enabled.
○ Ethernet - No Spanning Tree—Select this option when you want to configure an uplink without spanning-tree
protocol enabled.

NOTE: The Ethernet option only applies to a specific PowerEdge MX use case. To determine when to use the
Ethernet option, see the PowerEdge MX documentation. For all other deployments, including SFS for Leaf and Spine,
select Ethernet - No Spanning Tree option.

● Description (Optional)—Enter the description for that uplink.

3. Click Next.
4. Configure the port:
a. Select a rack in which you want to create the uplink.
b. Select one or more interfaces from each of the leaf switches to associate to the uplinks. If you want to split a port speed,
breakout the interface first before associating the interface to the uplinks. See Breakout Ports to configure breakout the
ports from SFS UI.
c. Select the port channel mode based on the configuration setup in the external network and click Next. To form a port
channel on the leaf switches, select an interface or interfaces that are of the same speed.
NOTE: Ensure that the corresponding ports on the external switches are configured with the same port channel mode.

5. Associate the networks with the selected interfaces:

a. If the network is already configured, select the untagged networks from the list.
b. If the network is not available, create a network using the Create Network—Enter the name, description, and VLAN ID.
Networks that are created using this wizard are created on the fabric as general purpose networks. For more information
about different types of networks in SFS, see Networks.

Managing SmartFabric with the SFS UI 35

 c. Select Yes or No appropriately to integrate the networks that are created automatically in the fabric through vCenter on
this uplink. The default selection is No.
When you select Yes, the uplink is created with the type marked as Default and the networks from the vCenter are
automatically appended to the L2 uplink during vCenter integration. For more information, see OpenManage Network
Integration User Guide.
NOTE: Dell Technologies recommends that only one uplink can be created with the Yes option for integrating the
networks on the uplink. Creating more than one uplink with network integration configuration enabled results in
unpredictable behavior. You can configure an uplink with network integration configuration enabled only when you
create it.

6. Click Finish.
The system displays a user uplink creation success message.

To avoid loops, SFS does not allow you to configure the same network on multiple uplinks.

Configure L3 VLAN uplink

Use the following procedure to create L3 VLAN uplink:
1. Click Uplink > Create.
2. Enter the uplink details:
● Select the uplink type as Layer 3.
● Select the L3 type as L3 VLAN.
● Enter the name and description (optional) for the L3 VLAN uplink.
3. Click Next.
NOTE: You can create L3 uplinks on both leaf and spine switches.

4. Associate the interfaces of the spine or leaf switches with the L3 uplink.
● Select the Switch group.
○ Spine—Select a spine switch and an interface or multiple interfaces of the spine switch to be associated with the
uplink.
○ Leaf—Select a leaf switch from the rack, and an interface or multiple interfaces of the leaf switch to be associated
with the uplink.
● Select the static or dynamic based on the configuration setup in the external network.
5. Click Next.
6. Associate the networks with the selected interfaces.
● Select if the network is a tagged or an untagged network.
● Enter the name, description (optional), VLAN ID, IP addresses, and prefix length.
7. Define a routing policy to associate with the uplink based on the external network connectivity setup.
● Static Route—A route policy template that contains a policy name, description (optional), the network address, prefix
length, and next hop IP address
● eBGP—A routing policy template that contains a policy name, description (optional), the BGP peer IP address, the peer
ASN, and BFD neighbor configuration.
NOTE: The network configurations reflect in the switch only after associating the network with an uplink or server
profile.

8. Click Finish.
The system displays uplink creation success message.

Configure L3 Routed uplink

Use the following procedure to create a L3 routed uplink:
1. Click Uplink > Create.
2. Enter the uplink details:

36 Managing SmartFabric with the SFS UI

 ● Select the uplink port type as Layer 3.
● Select the L3 type as L3 Routed Interface.
● Enter the name and description (optional) for the L3 routed uplink.
3. Click Next.
4. Associate an interface from the spine or leaf switch with the L3 Routed uplink.
a. Select the switch group.
● Spine—Select a spine switch and an interface from the spine to associate with the uplink.
● Leaf—Select the rack, switch, and an interface from the leaf switch to associate with the uplink.

NOTE: You can select only one interface from the spine or leaf switch for L3 Routed uplink.

5. Click Next.
6. Associate the networks with the selected interfaces by providing a name, description (optional), interface IP address, and
prefix length.
NOTE: You cannot associate a L3 Routed network with more than one uplink or server profile.

7. Define a routing policy to associate with the uplink based on the external network connectivity setup.
● Static Route—A route policy template that contains a network prefix and the next hop IP address.
● eBGP—A routing policy template that contains BGP peer IP address, peer AS number, and BFD configuration.
● Select the Fabric Interconnect checkbox if you are creating the uplink for multisite fabric interconnect purpose. See
Multisite fabric interconnect.
8. Click Finish.
The system displays uplink creation success message.

Edit an uplink
Prior to OS10.5.6.0 release, only the network and port configurations can be edited on an existing uplink. From 10.5.6.0 version,
you can edit all uplink-related configurations for an existing L2 or L3 uplink through a single edit.
1. Select the SFS instance > Uplink.
2. Select the fabric uplink from the list and click Edit.
3. Edit the following:
● Name
● Uplink networks (untagged and tagged)
● Uplink interfaces
● Port channel

NOTE: This option is not applicable for L3 routed interface.

4. Click Edit.
Modifying the link aggregation type removes the existing uplink session (if already formed) and a new session is added for
the new link aggregation type. This change affects the underlying traffic or route operation.

Delete an uplink
You can delete a user-created uplink:
1. Select the uplink from the displayed list and click Delete.
2. Click Delete to confirm.
NOTE: When you delete an uplink, the network and route profile that are associated with the uplink are not deleted.

The system displays the uplink interface delete success message.

Managing SmartFabric with the SFS UI 37

Multisite fabric interconnect
From 10.5.3.0 release, SFS allows you to configure fabric interconnect to connect fabrics across different sites. Using this
feature, you can configure the uplink, L3 routed network, IPv4 BGP peer policy, and BGP EVPN policy to interconnect the SFS
fabric connected through a direct link or an external network. You can configure multisite fabric interconnect using Configure
Multi-Site Fabric Interconnect wizard or you can individually create uplink, L3 routed network, and IPv4 BGP peer and BGP
EVPN policies for the fabrics in different sites depending on the requirement.
You can interconnect up to three SFS fabric sites.
SFS provides two options (direct and external) to configure multisite fabric interconnect depending on the interconnect
link between the SFS fabrics. You can configure fabric interconnect using Multi-Site Fabric Interconnect option for the
supported fabric interconnect links:
1. Direct—Following image is an example of direct interconnect between the multisite SFS fabrics.
● Direct link between the leaf switches across two sites:

● Direct link between the spine switches across two sites:

38 Managing SmartFabric with the SFS UI

2. External link—Following image is an example of external network interconnect between the multisite SFS fabrics.
● External link between the leaf switches across two sites:

● External link between the spine switches across two sites:

You must configure fabric interconnect configurations individually using Workflow 2:

● When you want to configure the multisite fabric interconnect for leaf and spine switches different from the supported fabric
interconnect links.
● When you want to customize the hop-count when creating fabric interconnect.

Prerequisites to setup multisite fabric interconnect

Ensure that the following requirements are met before configuring multisite fabric interconnect:
● Verify the physical connections on the fabric interconnect. SFS supports the fabric interconnects links (direct or external)
between leaf switches and spine switches.
● Ensure that the domain IDs of the fabrics are different.

Managing SmartFabric with the SFS UI 39

● Ensure that the ASNs of the switches are different.
● Launch the SFS UI of both the fabrics (Site 1 and Site 2) to collect the following information that you must enter during the
fabric interconnect setup:
○ Domain IDs of the Site 1 and Site 2 fabrics—From SFS UI, click Global Settings > Fabric Settings to view the domain
ID.
○ BGP ASN numbers of the switches deployed in the Site 1 and Site 2 fabrics—From SFS UI, click Global Settings >
Fabric Settings to view the ASN details.
○ Router ID of the switches deployed in the Site 1 and Site 2 fabrics—From SFS UI, click Global Settings > Fabric
Interconnect Details to view the Router ID address of each switch in the fabric.
● If the fabric interconnect link is external, ensure that you configure the external router connected to the edge rack switch
with the value of jump frame MTU as 9216.

Procedure to configure multisite fabric interconnect

The following are the high-level instructions to configure fabric interconnect:
1. Provide the domain ID for the SFS fabrics in Site 1 and Site 2. You can do any of the following:
● Enable SFS L3 personality on the leaf and spine switches with the domain ID. Domain ID is an optional parameter. If
you do not provide the domain ID when enabling SFS, it is set to the default domain ID 100. For more information, see
smartfabric l3fabric enable. Ensure that the mDNS option is set to Disable.
● Edit the domain ID for the SFS fabric if necessary. Domain ID ranges from 100 to 107.

NOTE: When you change the default fabric settings, all the switches in the fabric are reloaded. If you edit the
domain ID, the following values are autogenerated after a reload based on SFS recommendations:
○ Leaf and spine ASN
○ Fabric BGP subnet and mask
○ Fabric VTEP subnet and mask

NOTE: When you change the fabric VTEP subnet or any global settings (except the mDNS option) on one SFS site,
ensure that you update the BGP EVPN policy configured on the other sites with the correct peer router ID and peer
ASN details.

2. Configure uplink, L3 routed network, IPv4 BGP peer policy, and BGP EVPN policy in the Site 1 fabric using Multi-Site
Fabric Interconnect option or individually.
3. Configure uplink, L3 routed network, IPv4 BGP peer policy, and BGP EVPN policy in the Site 2 fabric using Multi-Site
Fabric Interconnect option or individually.

Workflow 1—Configure multisite fabric interconnect

You can configure the uplink, L3 routed network, IPv4 BGP peer route policy, and BGP EVPN policy to interconnect the SFS
fabric connected through a direct link or an external network. To configure multisite fabric interconnect on switches, use the
following instructions to configure the multisite fabric interconnect:
Configure multisite fabric interconnect—Site 1 fabric
When configuring the peer details for Site 1, provide the switch details of Site 2 fabric as the peer configuration.
1. Click Uplink > Multi-Site Fabric Interconnect.
2. In OS version less than 10.5.6.1, click the Configure Multi-Site Fabric Interconnect link to configure the multisite fabric
interconnect on switches.

40 Managing SmartFabric with the SFS UI

From OS version 10.5.6.1 or later, click Add to add the multisite fabric interconnect configuration.

Managing SmartFabric with the SFS UI 41

3. Select the switch group (Leaf or Spine). The template changes based on the selected switch group.
a. If the switch group is Leaf, select the rack from the list. The local fabric domain ID, local hostname, local BGP ASN, and
local loopback address are autopopulated.
4. Select the Fabric Interconnect Link (Direct or External). The template changes based on the selected link type.
5. Enter the domain ID of the peer fabric (Site 2). The system autopopulates the peer BGP ASN number based on the peer
domain ID.
6. Enter the following information:
a. Select the uplink interface of the switches that are connected to the respective peer-site switch interfaces.
b. Configure a Layer 3 routed network for the uplink.
● Local IP Address—Enter the IP address for the selected interface.
● Local Prefix Length—Enter the prefix details.
c. Configure the BGP peer policy for Site 1 fabric.
If the link is Direct:
i. Peer BGP ASN—Verify the peer BGP ASN number. The system autopopulates the peer BGP ASN number based on
the peer domain ID. If the BGP ASN number is different from the autopopulated number, change the ASN number
accordingly.
ii. Peer Interface IP—Enter the peer interface IP address.
iii. BFD—Select the BFD configuration. By default, BFD is disabled.
If the link is External:
i. Peer BGP ASN—Verify the peer BGP ASN number. The system autopopulates the peer BGP ASN number based on
the peer domain ID. If the BGP ASN number is different from the autopopulated number, change the ASN number
accordingly.
ii. Peer Interface IP—Enter the peer interface IP address.
iii. BFD—Select the BFD configuration. By default, BFD is disabled.
iv. External ASN—Enter the ASN number of the external device.
v. External Interface IP—Enter the interface IP address of the external device.
NOTE: SFS does not support the BGP password or BFD/BGP timer configuration.

d. Configure the BGP EVPN policy for Site 1 fabric.

● Peer Router ID—Enter the peer router ID.
● You can add up to two BGP EVPN policies for each switch. Click + to add a second loopback address.
NOTE: If the switch group is Leaf, you must enter all the relevant details for the listed leaf switches.

7. Click Submit.
The system displays a multisite fabric interconnect creation success message.
After submitting the configuration, SFS configures the following:
● Enables fabric interconnect.
● Disables mDNS.
● Creates a L3 routed network uplink.
● Creates a BGP peer route policy.
● Creates a BGP EVPN policy with the hop count set to 10 by default.
You can view the autoconfigured routed networks, uplinks, BGP peer route policies, and BGP EVPN policies in the respective
configuration tabs.
NOTE:
● Dell Technologies does not recommend editing or deleting any of the autoconfigured uplinks, L3 routed networks,
BGP peer route policies, and BGP EVPN policies that are generated during fabric interconnect configuration from the
respective configuration tabs.
● If you upgrade OS10 image from an earlier version to 10.5.6.1 or higher while the multisite fabric interconnect
configuration is still configured, the existing configuration details are not displayed under Multisite Fabric
Interconnect tab. However, you can view the configurations such as routed networks, uplinks, BGP peer route
policies, and BGP EVPN policies in the respective configuration tabs.
You cannot change the hop count in this workflow (Workflow 1). If you want to customize the hop count to a different value for
BGP EVPN:

42 Managing SmartFabric with the SFS UI

● Click the SFS instance > Network > Routing Configuration > BGP EVPN Configuration.
● Click Create to create a BGP EVPN policy.
● Apply the policy to the appropriate switch and delete any unused existing policy.
The uplink network, the eBGP peer route, and BGP EVPN policies are created, and their names are appended with the prefix
Fabric_Interconnect. To verify the configurations, you can go to the respective configuration menus to view the list of uplinks,
L3 routed networks, BGP peer route policy, and BGP EVPN policy that are created using this wizard.
From 10.5.6.1 onwards, you can view the consolidated view of the configured multisite fabric interconnect details. Select the
multisite fabric interconnection configuration to view the consolidated details of the configuration.

Edit multisite fabric interconnect configuration

From 10.5.6.1 onwards, you can edit the multisite fabric interconnect configuration.
After configuring the multisite fabric interconnect, you cannot change the Switch Group (leaf and spine), the Fabric
Interconnect Link, or the Peer Fabric Domain ID. To modify the Switch Group (leaf and spine), the Fabric Interconnect
Link, or the Peer Fabric Domain ID, delete the existing multisite fabric interconnect configuration, and configure it again.
1. Select the multisite fabric interconnect configuration from the list and click Edit.
2. Edit the following:
● Uplink interface of the switches that are connected to the respective peer-site switch interfaces.
● Layer 3 routed network (Local IP address and prefix length) for the uplink.
● BGP peer policy configuration (Peer BGP ASN, peer interface IP address, and BFD).
● BGP EVPN policy (Peer router ID).
3. Click Edit.
The system displays an edit success message.
Delete multisite fabric interconnect configuration
From 10.5.6.1 onwards, you can also delete the multisite fabric interconnect configuration:
1. Select the multisite fabric interconnection from the displayed list and click Delete.
2. Click Delete to confirm.
The system displays a delete success message.
Configure multisite fabric interconnect—Site 2 fabric
To configure multisite fabric interconnect on Site 2 fabric, follow the same steps to configure the L3 routed network, IPv4 BGP
peer route policy, and BGP EVPN policy. While configuring, provide the switch details of Site 1 fabric as the peer configuration.

Workflow 2—Configure uplinks, L3 routed networks, and BGP peer and EVPN policies
Alternative to the Workflow 1—Configure multisite fabric interconnect, you can also configure multisite fabric interconnect
by creating the configurations individually. Use this workflow to configure the multisite fabric interconnect for leaf and spine
switches that are different from the supported fabric interconnect links:
1. Disable VxRail Manager Integration option, see Edit fabric settings.
2. Create L3 routed interface uplink with eBGP peer route policy configuration, see Uplink. Ensure that you select the Fabric
Interconnect checkbox when creating the L3 routed uplink.
NOTE: You can also create eBGP peer route policy individually and associate with the uplink. Ensure that you select the
Fabric interconnect checkbox when creating the eBGP peer configuration.

3. Create BGP EVPN configuration, see Configure BGP EVPN. Ensure that you provide the appropriate Hop Count.

Managing SmartFabric with the SFS UI 43

Configure networks and routing configuration
You can configure networks and routing configuration.

Configure networks
You can configure the following types of networks:
● General purpose network (Layer 2 or Layer 3)
● L3 routed interfaces (Layer 3 only)
● Multi rack L3 VLAN (Layer 3 only)
● VLAN networks (Layer 2 or Layer 3)
● VXLAN networks (Layer 2 or Layer 3)
For detailed information about these network types, see Networks.

Configure general-purpose networks

Most VXLAN virtual networks are created as general-purpose networks using SFS. General-purpose networks can be Layer 2 or
Layer 3. When you create a general-purpose network, SFS creates a VLAN network along with the VXLAN virtual network. In a
general-purpose network, VXLAN network identifier (VNI) and VLAN ID are the same, and you can associate one VLAN with the
VNI across the fabric. If you delete a VLAN network, it automatically deletes the associated VXLAN network. For example, if you
create a general-purpose network with VLAN ID 50, SFS creates a VLAN 50 and associated VXLAN network with VNI 50 in the
SmartFabric. When you delete the VLAN network, both VLAN 50 and VXLAN VNI 50 are deleted.

Create a Layer 2 general-purpose network

To create a Layer 2 general-purpose network:
1. Click Networks > General Purpose Networks.
The page displays the list of the general-purpose networks that are already configured in the SmartFabric.
2. Click Create.
3. Select Layer 2 General Purpose Network for the network type from the list.
4. Enter the following details:
● Network ID.
● Network name. For example, network-201.
● VLAN ID. A number that ranges from 1 to 3999 (except 3939). For example, 201.
● Description.

5. Click Create.
The system displays a virtual network creation successful message.

44 Managing SmartFabric with the SFS UI

Create a Layer 3 general-purpose network
From 10.6.0.1 version, you can create a Layer 3 general-purpose network directly by eliminating the need to first create a Layer
2 network and then converting it to Layer 3 network.
To create a Layer 3 general-purpose network:
1. Click Networks > General Purpose Networks.
The page displays the list of the general-purpose networks that are already configured in the SmartFabric.
2. Click Create.
3. Select Layer 3 General Purpose Network for the network type from the list.
4. Enter the following details:
● Network name. For example, network-204.
● VLAN ID. A number that ranges from 1 to 3999 (except 3939). For example, 204.
● Description
● IP address
● Prefix length
● Gateway IP address
● Helper address

5. Click Create.
The system displays a virtual network creation successful message.

Edit a general-purpose network

You can edit the general-purpose network configuration and change the general-purpose network type between Layer 2 and
Layer 3.
1. Select a network from the list and click Edit.
2. Select the network type appropriately. Depending on the network type selection, the template changes.
3. Edit the details accordingly.
4. Click Edit.
The system displays a virtual network edit success message.
View the general-purpose network configuration
To view the details of the general-purpose networks, select a network from the list. The VLAN details of the specific network
including network ID, originator, network name, VLAN ID, QoS priority, network type, VLAN interface IP address details, prefix

Managing SmartFabric with the SFS UI 45

length, gateway IP address, and DHCP helper address. Portgroups that are created on the vCenter are displayed under
General Purpose Networks.

Delete a general-purpose network

When you delete a general-purpose network, both the VLAN and the VXLAN networks are deleted from SFS. To remove a
general-purpose network configuration:
Starting with the 10.6.0.5 release, you can now delete multiple general-purpose networks directly from the OMNI UI.
1. Select the general-purpose network or multiple networks and click Delete. The system displays the list of the server
interface profiles that are associated with the network.
2. Click Delete to confirm.
The system displays a network deletion success message.
To view the details of the general-purpose networks, select a network from the list. The VLAN details of the specific network
including network ID, originator, network name, VLAN ID, QoS priority, network type, VLAN interface IP address details, prefix
length, gateway IP address, and DHCP helper address. Portgroups that are created on the vCenter are displayed under
General Purpose Networks.

Configure L3 routed interfaces

You can create, edit, and delete Layer 3 routed interfaces.

Create L3 routed interface

Use the following procedure to create an L3 routed interface:
1. Click Network.
2. Select Networks > Layer 3 Routed Interfaces.
3. Click Create.
4. Enter the following details:
● Network ID
● Network name
● Description
● IP address
● Prefix length

46 Managing SmartFabric with the SFS UI

5. Click Create.
The system displays a network creation success message.

Edit L3 Routed Interface

Edit the network details:
1. Select the Network ID from the list and click Edit.
2. Edit any of the following details:
● Network name
● Description
● IP address
● Prefix length
3. Click Edit.
The system displays edit network success message.

Delete L3 Routed Interface

Starting with the 10.6.0.5 release, you can now delete multiple L3 routed interfaces directly from the OMNI UI.
1. Select the network ID or multiple IDs to remove and click Delete. The system displays the list of uplinks that are associated
with the network.
2. Click Delete to confirm.
The system displays a network deletion success message.

Configure multirack L3 VLAN

Starting from the 10.5.3.0 release, from SFS UI, you can configure L3 VLAN networks for the racks to which the servers are
connected. Using this feature, you can create a L3 VLAN network for each VLT pair (rack) with a different subnet. This network
is used for NSX-T overlay to create VTEP networks. Create, edit, and delete multirack L3 VLAN networks from SFS UI. You can
specify the IP address for each switch in a rack when creating a multirack L3 VLAN.

Create multirack L3 VLAN

Use the following procedure to create a multirack L3 VLAN:
1. Click Network.
2. Click Networks > Multi-Rack L3 VLAN.
3. Click Create.
4. Enter the following details:
● Network ID. For example, overlay.
● Network name.
● VLAN ID—A number from 1 through 3999 (except 3939). For example, 2713.
● Description.
5. Select the Specific IP Addresses checkbox to specify the IP address for each switch.
6. Enter the following details specific to the rack:
● IP addresses of the individual switches
● Prefix length
● Gateway IP address
● Helper IP addresses
7. Click Create.
The system displays a VLAN network creation success message.

Edit multirack L3 VLAN

1. Select a network ID from the list and click Edit.
2. Edit the configuration as necessary.
3. Click Edit.
The system displays edit network success message.

Managing SmartFabric with the SFS UI 47

Delete multirack L3 VLAN
Starting with the 10.6.0.5 release, you can now delete multiple multirack L3 VLANs directly from the OMNI UI.
1. Select a VLAN network or multiple networks from the displayed list and click Delete.
2. Click Delete to confirm the removal of the networks.
The system displays a network deletion success message.

Configure VLAN networks

VLAN networks with SFS are traditional Layer 2 or Layer 3 VLANs. VLAN networks are used to create networks that do not use
VXLAN. If the same Layer 3 VLAN ID is needed in multiple racks, use the Multi-Rack L3 VLAN instead. Create, edit, and delete
L2 or L3 VLAN networks for SmartFabric.

Create L2 VLAN network

Use the following procedure to create a Layer 2 VLAN network:
1. Click Network.
2. Click Networks > VLAN Networks.
3. Click Create.
4. Select the network type as Layer 2 VLAN Network.
5. Enter the following details:
● Network ID (For example, VLAN-600).
● Network name (For example, VLAN-600).
● VLAN ID. A number from 1 to 3999 (except 3939). For example, 600.
● Description
6. Click Create.
The system displays a VLAN network creation success message.

Create L3 VLAN network

1. Select Networks > VLAN Networks.
2. Click Create.
3. Select the network type as Layer 3 VLAN Network.
4. Enter the following details:
● Network ID
● Network name
● VLAN ID. A number that ranges from 1 to 3999 (except 3939).
● Description
● IP address
● Prefix length
5. Click Create to confirm.
The system displays a VLAN network creation success message.

Edit VLAN network

You can edit the configuration of the L2 VLAN network and change it to an L3 VLAN network:
1. Select a network ID from the list and click Edit.
2. Edit the network details or modify the L2 VLAN to L3 VLAN configuration as necessary.
3. Click Edit.

Delete VLAN network

Starting with the 10.6.0.5 release, you can now delete multiple VLANs directly from the OMNI UI.
1. Select a VLAN network or multiple VLAN networks from the displayed list and click Delete.
2. Click Delete to confirm the removal of the VLAN networks.
The system displays a network deletion success message.

48 Managing SmartFabric with the SFS UI

Configure VXLAN network
The VXLAN Network type in SFS includes advanced configuration options like the ability to customize the virtual network
identifier (VNI) and associate multiple VLANs to a single VNI.
Usually, you should use the General Purpose Network type for VXLAN networks instead. Create, edit, and delete L2 and L3
profile VXLAN network configurations through SFS UI. The purpose of VXLAN networks is to associate multiple L2 or L3 VLAN
networks to a single VXLAN network. Whereas a general purpose network does not have the flexibility to extend the VXLAN
network.
The VXLAN networks display a list of mapped VLANs. Select a VXLAN network to view details pertaining to that specific
network including network ID, VLAN ID, and network name.

Create L2 VXLAN network

Use the following procedure to create a Layer 2 VXLAN network:
1. Click Network.
2. Click Networks > VxLAN Networks. The page displays the list of the VXLAN networks that are configured in the
SmartFabric instance.
3. Click Create.
4. Select the virtual network type as Layer 2.
5. Enter the following details:
● Virtual network name
● VxLAN VNI. A value through 1 and 1677215.
● VLT VLAN ID. A value through 1 and 3999 (except 3939).
● (Optional) Description
6. Click Create.
The system displays a virtual network creation successful message.

Create L3 VXLAN network

From the OS10.5.3.0 release, you can configure the IPv6 virtual networks.
1. Click the SFS instance > Network.
2. Click Networks > VxLAN Networks. The page displays the list of the VXLAN networks that are configured in the
SmartFabric instance.
3. Click Create.
4. Select the Virtual Network Type as Layer 3.
5. Enter the following details:
● Virtual network name
● VxLAN VNI
● VLT VLAN ID
● Select the IPv4 checkbox to provide the IPv4 address, prefix, gateway IP address, and helper addresses.
● Select the IPv6 checkbox to provide the IPv6 address, prefix, gateway IP address, and helper addresses.
6. Click Create.
The system displays a virtual network creation successful message.

Associate multiple VLANs to a VXLAN network

Using the following steps, you can map multiple VLANs to a single VXLAN network:
1. Select a VXLAN network. The page displays the VXLAN details on the right pane.
2. Click the Create option available after the VXLAN details.
3. Enter the following details for the VLAN configuration:
● Network ID
● Network name
● VLAN ID
● Description

Managing SmartFabric with the SFS UI 49

4. Click Create.
The system displays edit network success message.

Edit VXLAN network

You can edit the configuration of the Layer 2 general purpose network and change it to a Layer 3 general purpose network. You
can edit the configuration of the VXLAN network:
1. Select a virtual network from the list and click Edit.
2. Edit the network configuration or modify the virtual network type.
3. Click Edit.
The system displays a virtual network edit success message.

Delete VXLAN network

To delete a VXLAN network, first delete the mapped VLAN or VLANs if associated, and delete the virtual network.
Starting from the 10.6.0.5 release, you can now delete multiple VXLAN networks directly from the OMNI UI.
1. Select the virtual network from the list.
2. Select the network that you want to remove and click Delete.
3. Click Delete to confirm.
Once you delete all the mapped VLANs, you can delete the VXLAN network.
Ensure that you delete all the mapped VLANs first before proceeding with the bulk delete action.

The system displays a network deletion success message.

Configure Routes
You can configure static routes, eBGP peer routes, and BGP EVPN configuration for a network.
When you create a route policy, the system configures the route policy name as the policy ID. In SFS, the route policy ID must
be unique and hence it does not allow you to create more than one policy with the same name. You can create multiple route
policies with the same configuration by providing different route policy name.
You can create and delete any routing policy configuration, but you cannot edit any routing policy. To modify an existing routing
policy, you must delete the policy and create a policy with the relevant configuration.

Configure static routes

Use the following procedure to configure static routes and associate the route to the switch.

Create a static route

1. Click Network > Routing Configuration.
2. Select Static Routes.
3. Click Create to add a new static route.
4. Enter the following details:
● Policy name
● Network address
● Prefix length
● Next hop IP address
● (Optional) Description
5. Click Create.
The system displays static route creation is successful.

50 Managing SmartFabric with the SFS UI

Add route to switch
The static route details display a list of mapped routes. Select a static route to view details pertaining to that specific route
including the switch ID.
1. Click Routing Configuration > Static Routes.
2. Select a static route that must be added to the switch.
3. Click Add Route to Switch that appears in the right pane.
4. Select the switch to map to this route.
5. Click Add.
The system displays the route added success message.

Edit a static route

From the 10.5.6.0 release, you can edit a static route.
1. Click the SmartFabric instance > Network > Routing Configuration.
2. Select Static Routes.
3. Click Edit.
4. Edit the following details:
● Network address
● Prefix length
● Next hop IP address
● (Optional) Description
5. Click Edit. The system displays static route creation is successful.
NOTE: Changing the Network IP address, prefix length, and next hop IP address removes the current policy neighbor
route, and a new route is added. This change affects the underlying traffic.
6. Click Ok to confirm.

Delete route from switch

1. Select the route to delete and click Delete Route.
2. Click Delete to confirm the removal of the route from the switch.
The system displays a route policy deletion success message.

Configure eBGP peer routes

You can configure eBGP peer routes for a network.
1. Click Network > Routing Configuration.
2. Click eBGP Peer Configuration.
3. Click Create to add an eBGP peer route.
4. Enter the following details:
● Policy name
● Peer interface IP address
● Peer ASN
● (Optional) Description
● BFD configuration
● Select the Fabric Interconnect checkbox if you are creating the uplink for multisite fabric interconnect purpose. See
Multisite fabric interconnect.
5. Click Create.
The system displays eBGP peer route creation is successful.
The eBGP peer details display a list of mapped routes. Select an eBGP route to view details pertaining to that specific route
including the switch ID.

Managing SmartFabric with the SFS UI 51

Add route to switch
The static route details display a list of mapped routes. Select a static route to view details pertaining to that specific route
including the switch ID.
1. Click Routing Configuration > Static Routes.
2. Select a static route that must be added to the switch.
3. Click Add Route to Switch that appears on the right pane.
The system displays virtual network edits success message.
4. Select the switch to map to this route.
5. Click Add.
The system displays the route added success message.

Delete route from switch

1. Select the route to delete and click Delete Route.
2. Click Delete to confirm the removal of the route from the switch.
The system displays route policy deletion success message.

Edit the eBGP peer configuration

You can edit the existing eBGP peer configuration. When modifying the peer IP address or ASN, the existing BGP neighbor
session gets removed and the new session gets created with the newly added peer IP address or ASN.

NOTE: Any changes to the existing route policy configuration impact the underlying traffic.

To edit the eBGP peer configuration:

1. Select the eBGP configuration from the list and click Edit.
2. Edit any of the configurations:
● Peer interface IP address
● Peer ASN
● Description
● BFD configuration
● Fabric Interconnect option
NOTE: Changing the IP address or ASN of a peer interface disrupts the BGP session and has an impact on the
underlying traffic.

3. Click Ok to confirm the edits.

The system displays the edit success message.

Configure BGP EVPN routes

Use the following procedure to configure a BGP EVPN routing policy:

Create BGP EVPN configuration

1. Click Network > Routing Configuration.
2. Click BGP EVPN Configuration.
3. Click Create.
4. Enter the following details:
● Policy Name—Enter the BGP EVPN policy name.
● Peer loopback address—Enter the peer loopback IP address.
● Peer ASN—Enter the peer ASN number.
● Hop count—Change the hop count as required. The default value of the hop count is 10.
● (Optional) Description—Enter the description for the policy.
5. Click Create.
The system displays eBGP peer route creation is successful.

52 Managing SmartFabric with the SFS UI

Add BGP EVPN route to the switch
1. Select a BGP EVPN route policy from the list.
2. Click Add Route to Switch.
3. Select the switch to which you want to add the route.
4. Click Add.
The system displays the route to switch addition success message.

Edit BGP EVPN configuration

From the 10.5.6.0 version, you can edit the BGP EVPN configurations.
1. Click Network > Routing Configuration.
2. Click BGP EVPN Configuration.
3. Click Edit.
4. Edit the following details:
● Peer loopback address
● Peer ASN
● Hop count
● Description
5. Click Edit. The system displays a message for confirmation.
NOTE: Changing the peer loopback IP address, peer ASN, or hop count disrupts the current BGP session and affects
the underlying traffic. The previous BGP session is replaced with a new session for the new peer's loopback IP address
or ASN, and traffic begins to converge over the new BGP overlay session.
6. Click Ok to confirm.

Delete eBGP route from switch

1. Select an eBGP route and click Delete Route.
2. Click Delete to remove the route from the switch.
The system displays a route policy deletion success message.

Configure global settings for SmartFabric

Starting from 10.5.3.0 release, you can configure SmartFabric switch services settings using SFS UI.
You can configure the following services on the SmartFabric switches:
● NTP
● DNS
● Syslog
● SNMP
NOTE: This feature is supported from SmartFabric OS10.5.2.2 and later versions, and applicable for SFS L3 leaf and spine
personality.

Configure NTP server

To configure an NTP server:
1. Click Global Settings > NTP.
The page displays the list of the NTP servers that are already configured in the fabric.
2. Click Add to configure an NTP server.
3. Enter the IP address or hostname of the NTP server.
4. Click Add.
To delete an NTP server, select an entry from the list and click Delete.

Managing SmartFabric with the SFS UI 53

Configure DNS server
To configure one or more DNS servers:
1. Click Global Settings > DNS.
The page displays the list of the DNS servers that are already configured in the fabric.
2. Click Add to configure one or more DNS servers.
3. Enter the IP address of the DNS server to configure a single DNS server setting. You can use the + button to add more DNS
servers.
4. Click Add.
The system displays the configuration success message.
NOTE: You cannot add a DNS server entry to the existing list. To add a DNS entry, you must delete all the existing entries
and add the new entry along with the previous entries.
Click Delete All to remove the configured DNS servers. This action deletes all DNS servers from the system.

Configure SNMP server

To configure or edit an SNMP server:
1. Click Global Settings > SNMP.
The page displays the list of the SNMP servers that are already configured in the fabric.
2. Click Add to configure an SNMP server.
3. Enter the IP address of the SNMP server.
4. Select the SNMP version.
a. For SNMP v2c, enter the community name.
b. For SNMP v3, enter the SNMP username and the security details. The security level configures SNMPv3 user
authentication and privacy settings:
● noauth—Do not authenticate users or encrypt SNMP messages; send messages in plain text.
● auth—Authenticate users in SNMP messages.
● priv—Authenticate users and encrypt or decrypt SNMP messages.

NOTE: Other SNMP settings such as view, group, engine ID, enable or disable traps, and contact information cannot
be configured from the SFS UI. You must configure these settings on each switch from the OS10 CLI. For more
information about how to configure SNMPv3 from the CLI, see Dell SmartFabric OS10 User Guide.
5. Click Add.
The system displays the configuration success message.
To delete the configured servers, select a server from the list and click Delete.

Configure syslog server

To configure and edit a syslog server:
1. Click Global Settings > Syslog.
The page displays the list of the syslog servers that are already configured in the fabric.
2. Click Add to configure syslog server.
3. Enter the IP address of the syslog server and log level.
4. Click Add.

Edit a syslog server

You can edit the log level for the syslog server.
1. Select the server from the list and click Edit.
2. Edit the log level of the server and click Update.
To delete the configured servers, select the server from the list and click Delete.

54 Managing SmartFabric with the SFS UI

Edit fabric settings
Use the following procedure to edit the default fabric settings:

NOTE: Any changes to the default fabric settings reboot all the switches in the network fabric.

1. Click Global Settings > Fabric Settings.

2. Click Edit.
3. Edit the values of the default settings:
● Domain ID
● Leaf and spine ASN
● Fabric BGP subnet and mask
● Fabric VTEP subnet and mask
● Client control VLAN
● Client management VLAN
● STP mode between MST and Rapid PVST
● mDNS
4. Click OK.
The system prompts for confirmation to continue. After you click OK, all the switches in the network fabric reload to apply
the fabric setting changes. The changed settings are applied only after a reboot.
When you edit the domain ID:
1. The fabric settings, fabric interconnect configuration, and fabric name are changed after a reload.
2. The interface breakout, networks, server profiles, and uplink configurations and its association are retained after a reload.
3. You must reconfigure BGP EVPN policy details for the external peer connectivity configurations such as multisite fabric
interconnect links, external router, NSX-T based on the changed ASN.

Manually onboard VxRail

To manually onboard the VxRail node onto the fabric:
1. Click Global Settings > Fabric Settings.
2. Click Edit.
3. Select Disable for mDNS.
4. Click Ok.
NOTE: The mDNS setting change does not reload the switches, and the change is applied to the latest selection
immediately.

Update default fabric, switch names, and descriptions

SFS assigns unique names for the network fabric, racks, and switches automatically. With 2.1 and later releases, you can edit the
default fabric and switch names, and descriptions of a SmartFabric instance. Use the following instructions to change the names
and descriptions:
1. Click Global Settings > Set Fabric & Switch name.
2. Click the Set Fabric & Switch name link.
3. Edit the name and description of the network fabric and click Next.
4. Edit the name and description of the rack and click Next.
5. Edit the name and description of the switches.
6. Click Finish.
NOTE: If you change the switch name in the UI, the hostname on the switch CLI is also updated.

Managing SmartFabric with the SFS UI 55

View fabric interconnect details
View the fabric interconnect details:
Click Global Settings > Fabric Interconnect details. This page displays the hostname of the switches and the respective
loopback address of the switches in the fabric. You can use this page to view the loopback address when configuring a multisite
fabric interconnection.

SupportAssist deprecation
CloudIQ Collector replaces the SupportAssist method of onboarding and collecting data for PowerSwitch systems starting
April 17, 2023. Dell Technologies recommends the CloudIQ Collector as the preferred method of onboarding PowerSwitch
systems to CloudIQ. As part of the ongoing SupportAssist depreciation process, the ability to Generate Access Key and Pin for
PowerSwitch has been removed from the Dell Technologies Support Page. Dell Technologies will no longer provide assistance
for defect and security remedies in SupportAssist after August 2023. Customers who are using SupportAssist are recommended
to move to CloudIQ Collector before the end of support for SupportAssist.

Life-cycle management
You can perform common life-cycle operations of upgrading the SmartFabric OS10 and restoring the SmartFabric.
From Life-cycle Management menu, you can:
● Restore SFS configuration data
● Perform an OS10 image upgrade

Change SmartFabric password

From 10.6.0.2 onwards, you can change the REST_USER password of the SFS instance:
1. Click Life Cycle Management > SmartFabric Password Change.
2. Enter the following details:
● Current Password—Enter the current password for the REST_USER.
● New Password—Enter the new password.
● Confirm New Password—Enter the new password again to confirm.
To enhance security, the new password must meet the following criteria:
● Must be at least nine characters long.
● Must contain at least one lowercase letter (a-z), one uppercase letter (A-Z), one numeric digit (0-9), and one special
character (!, @, #, $, %, ^, &, or *).
As you enter the new password, each requirement is validated in real time. Criteria that are met are highlighted in green.
NOTE: The password update cannot proceed unless all criteria are met. If any requirement is not met, an error message
appears. Continue changing the password until all criteria are met.
The system displays a password update success message.

3. Click Update Password.

Restore fabric configuration

Restore the SFS configurations on a fabric to a known good configuration with the backup configuration file stored on your
external device. Use the following procedure to restore:

1. Click Life Cycle Management > Restore.

2. Click Choose File.
3. Select the backup configuration JSON file that is stored externally.
4. Click Ok.
5. Select the check box to agree.

56 Managing SmartFabric with the SFS UI

 CAUTION: This action reboots all the switches in the fabric to apply the new configuration that cause traffic
disruption and connectivity. Dell Technologies recommends using the restore operation during a downtime
window.

6. Click OK to confirm. The switches are restored with the configuration in the JSON file.

Upgrade SmartFabric OS10 in switch

From the OS10.5.4.4 release onwards, you can upgrade OS10 software on the switches deployed in the fabric from SFS UI.
Before proceeding with the OS10 software upgrade, see the Support Matrix for the latest supported OS10 versions.
Before OS10.5.4.4 version, you can only upgrade the switches in fabric with an OS10 image using OMNI UI. For more
information, see the OpenManage Network Integration User Guide. From the OS10.5.4.4 release onwards, you can upgrade
OS10 software installed on the switches deployed in a fabric with the specific image from the remote location.
Upgrade the switches in a fabric with an OS10 image:
1. Select the switches that you want to upgrade and click Upgrade Node(s).
To upgrade all the switches in the SFS fabric, select all the switches and upgrade.

NOTE: You cannot stop the upgrade after it is triggered.

2. Select Remote to upgrade the OS10 software available in a remote server.

3. Enter the complete remote server URL with the image name.
The supported protocols are FTP, SFTP, HTTP, and SCP. For example, http://x.x.x.x/tftpboot/latest-build/
PKGS_OS10-Enterprise-10.5.3.2-installer-x86_64.bin.
For FTP, SFTP, and SCP remote servers, enter the username and password to connect to the remote server if required.
4. Click Next.
5. Select the appropriate options:
Download strategy—The strategy in which the latest OS10 image should be downloaded on the switches in the fabric.
● Sequence—The OS10 image is downloaded to the leader switch and sequentially downloaded to all other switches one
at a time.
● Parallel—The OS10 image is downloaded on all the switches in parallel.
Reboot strategy—The strategy in which the reboot should happen on the switches in the fabric.
● Sequence—The switches in the fabric reboot in sequence. In the sequence method, the traffic path through the
fabric is maintained throughout the upgrade process. However, it takes more time as the switches in the fabric reboot
sequentially.
● Parallel—The switches in the fabric reboot in parallel. In the parallel method, the upgrade process takes less time.
However, a traffic path through the fabric is not maintained throughout the upgrade process.
Schedule—Now is selected by default and it initiates the upgrade instantly.
6. Click Next.
7. Review the configurations and read the warnings before proceeding. If required, make the relevant changes.
8. Click Finish.
After you initiate the fabric upgrade, the system displays the detailed progress of the upgrade and the status for each
switch.
Once the upgrade is successful, the upgrade status column shows the status as success.

View fabric compliance status

From 10.5.3.0, you can view the fabric compliance status from SFS UI. SFS validates the health of the cluster, topology role,
underlay, overlay, network, server appliance discovery, uplink, policy, and VLT. SFS monitors the health in both the switch and
the whole fabric levels. SFS retrieves the fabric compliance status for the SFS instance and displays the noncompliance events
with details. SFS also provides recommended actions to eliminate the compliance violations or misconfigurations.
This feature is applicable for SFS L3 leaf and spine personality.
To view the fabric compliance errors:

Managing SmartFabric with the SFS UI 57

1. Click Serviceability > Fabric Compliance to view the latest compliance errors. The table lists the latest compliance events
with detailed information including switch name, service tag of the switch, status, error code, and the recommended action.
2. Click the information icon to view the recommended action for each compliance error.
3. Click Refresh to update the data and display the new compliance errors.
4. Click Download to download all the compliance errors. The downloaded zip file contains the fabric compliance errors in
CSV format.
You can also view the fabric compliance errors in the SmartFabric instance overview dashboard. Select Summary > Overview
to view the overview of fabric compliance status. The fabric compliance errors are grouped under different categories.

58 Managing SmartFabric with the SFS UI

 6
SFS with VxRail
SFS, used in leaf and spine network, creates a fully integrated solution between the fabric and a hyperconverged domain
infrastructure such as VxRail. When integrated with VxRail, SFS automates network setup, simplifying and accelerating the
deployment. Switches are automatically configured. When additional VxRail nodes are connected, the fabric identifies them as
VxRail nodes and automatically onboards the nodes to the required networks.

For more information regarding VxRail deployment-related documents, see VxRail Networking Solutions.

Supported network topologies

See Supported topologies section for information regarding the topologies of SFS with VxRail deployments.

Hardware and software requirements

The requirements to deploy VxRail with SFS are as follows:
Hardware components
● VxRail nodes
● PowerSwitches
Software components

SFS with VxRail 59

● SmartFabric OS10
● OMNI
● VxRail Manager
● VMware vCenter
For more information regarding detailed deployment requirements, see the Deployment Guides for respective releases.

Supported switches
SFS for leaf and spine is supported on S-series and Z-series PowerSwitches. See Support matrix for a complete list of
supported platforms.

SFS personalities
In the SFS-enabled network, VxRail deployment option includes L2 single rack or L3 multirack personalities. The table lists the
comparison between L2 and L3 fabric personalities:

Table 12. SFS personalities

L2 Single Rack personality L3 multirack personality
Single rack network fabric is supported for VxRail clusters. Multirack data center network fabric is supported that starts
For new SFS deployments, use the L3 leaf and spine fabric with a L3 single rack (L3 fabric profile) and which you can
personality as the SFS L2 personality is deprecated. expand to a multirack solution based on the demand.
Network fabric has two leaf switches in a single rack which Network fabric has up to 20 switches in a leaf and spine
you cannot expand. design that starts with a single rack which you can expand up
to eight racks.
All VxRail with SFS deployments from SmartFabric release All SmartFabric deployments with SmartFabric OS10.5.0.5 or
OS10.4.1.4 to OS10.5.0.5 support configuration with a single later.
pair of leaf switches for VxRail clusters.
Default uplink and jump host port are created as part of a You can create uplinks and jump host port through SFS UI or
fabric initialization, which you cannot modify after enabling OMNI after initial deployment.
SFS.
Enabled SFS by running a Python script in the OS10 Linux Enable SFS using CLI, API, or UI.
shell.
Existing deployments when upgraded to SmartFabric Dell Technologies recommends that you enable all new
OS10.5.0.5 continue to run in the L2 fabric profile and L3 deployments with L3 leaf and spine fabric personality. You
fabric capabilities are not available. If you upgrade switches cannot upgrade VxRail with SFS deployments to the new L3
with L2 personality to OS10.5.0.5, SFS operates with the leaf and spine fabric personality automatically.
VxRail L2 single rack personality.

Support matrix
See Networking Solutions Support Matrix regarding support matrices for SFS with VxRail across various releases.

Fabric operations and life cycle management

Dell OpenManage Network Integration (OMNI) enables you to configure and manage SFS-enabled PowerSwitches in different
deployments. You can use OMNI application to manage and operate one or more SFS instances either directly using the OMNI
UI through a web browser or through a vCenter plug-in. After initial deployments, Dell Technologies recommends that you use
OMNI UI to perform all fabric management and life cycle management activities.
For more information about OMNI, see OMNI Documentation.

60 SFS with VxRail

 7
SFS with PowerEdge MX
Dell PowerEdge MX is a unified, high-performance data center infrastructure providing the agility, resiliency, and efficiency to
optimize a wide variety of traditional and new emerging data center workloads and applications. In a Dell PowerEdge MX7000
infrastructure, the MX9116n fabric engine and MX5108n Ethernet switch support SFS.
For more information about the SFS and PowerEdge MX including architecture, deployment, configuration, operations, and
troubleshooting, see the Dell PowerEdge MX Networking Deployment Guide.

SFS with PowerEdge MX 61

 8
SFS for PowerScale (Isilon) back-end fabric
Dell PowerScale is a scale-out network-attached storage (NAS) platform that supports unstructured data workloads. All
PowerScale models are powered by the OneFS operating system. PowerScale uses PowerSwitches to provide the network.
SmartFabric OS10 with SFS, for PowerScale back-end fabric automates onboarding and network configuration of PowerScale
devices on a L3 leaf and spine fabric. Isilon OneFS interacts with back-end fabric formed by SFS. For more information, see
PowerScale Info Hub.

Supported network topologies

See Supported topologies section for information regarding the topologies of SFS. Only L3 personality is supported when
deploying SFS with PowerScale nodes.

Hardware and software requirements

The requirements to deploy PowerScale with SFS are as follows:
Hardware components
● PowerScale devices
● PowerSwitches
Software components
● SmartFabric OS10
● PowerScale OneFS

Supported switches
The following PowerSwitches are supported in PowerScale deployment:
● S4112F-ON
● S4148F-ON
● S5232F-ON
● Z9264F-ON
● Z9432F-ON
● Z9100-ON
● S5448F-ON (can be used as a spine switch only)

62 SFS for PowerScale (Isilon) back-end fabric

PowerScale requirements
Requirements specific to SFS with PowerScale deployment are as follows:
● By default, all PowerSwitches for PowerScale deployment are shipped with factory-installed OS10.

NOTE: PowerSwitches must be running SmartFabric OS10.5.0.5 or later software releases that support the PowerScale
with SFS deployment.

● In SFS with PowerScale deployment, the leaf nodes are not connected as a VLT pair. On the leaf switch, no VLTi
configuration is required while enabling SFS. For more information, see Enable SFS using CLI.
● You must enable SFS on the switches to setup a fabric. All other fabric operations are managed by Isilon OneFS.
● The switches are connected in a leaf and spine topology with BGP EVPN setup between the leaf switches.
● Leaf and spine switches belong to different autonomous systems (AS).
● PowerScale data traffic is forwarded on an untagged VXLAN network 4091.
● PowerScale appliances are detected through LLDP.
● All ports on which the PowerScale appliance is detected are added as untagged VXLAN access interfaces.
● All load balancing is achieved through BGP ECMP.
● By default, the autobreakout feature is enabled on the leaf and spine switches.

Support matrix
See Networking Solutions Support Matrix regarding support matrices for SFS with PowerScale across various releases.

Replace a switch in SmartFabric mode

You can replace an OS10 switch that is part of a SmartFabric in a PowerScale environment when an error occurs.
The following are the prerequisites:
1. The new switch must be the same model as the old switch.
2. Ensure that the OS10 version of the new switch matches the old switch and the rest of the cluster switches.
3. If applicable, back up the following configurations from the old switch to configure the new switch with the same details.
● Hostname
● Management IP address
● DNS and NTP IP addresses, if configured
● Spanning-tree mode, if changed from the default setting

NOTE: In SFS mode, RPVST+ is enabled by default.

● Other nonfabric commands

Perform the following common checks before and after replacing the switch:
1. Verify fabric health, fabric status, and fabric compliance before and after the node replacement.
2. Check the running-configuration using the show running-configuration command before and after the node
replacement to verify if the configuration from the old switch is correctly ported to the new switch.
3. The switch replacement procedure can also be performed when the leader node is faulty. SmartFabric elects a new leader
when the current leader goes down, and the SFS configuration is restored once the node replacement workflow completes.

SFS for PowerScale (Isilon) back-end fabric 63

The following procedure outlines how to replace a faulty (old) switch in SmartFabric mode:
1. Identify the switch that must be replaced.
2. Label each of the cables with the port numbers before disconnecting the cables. Power off the old switch and disconnect
the cables from the switch. Remove the old switch.
The old switch, once removed, is shown as offline in the fabric. If the removed node was the leader, a new node is elected as
the leader.

leaf1# show smartfabric nodes

Service-Tag Type Status Mode Chassis-Service-Tag Chassis-slot FabricId
------------------------------------------------------------------------------------------------
2LB10Q2 Z9100-ON ONLINE FABRIC
2LB00Q2 Z9100-ON OFFLINE FABRIC
2LHZZP2 Z9100-ON ONLINE FABRIC
2L5ZZP2 Z9100-ON ONLINE FABRIC
2LPZZP2 Z9100-ON ONLINE FABRIC
2L300Q2 Z9100-ON ONLINE FABRIC

3. Connect the cables to the new switch according to the labeling, excluding the connections to the PowerScale servers.
4. Configure the backed-up configurations on the new switch.
5. Enable SmartFabric Services on the new switch to add the new switch to the cluster.
Dell Technologies recommends configuring the fabric with the same domain ID as the cluster when it is not the default
domain ID.
6. Verify the status of the new switch.

64 SFS for PowerScale (Isilon) back-end fabric

 Ensure that the new switch is displayed as online and the old switch appears as an offline entry in the show smartfabric
nodes output. In the example output below, the old switch is 2LB00Q2 and the new switch is 4LB00Q4.

leaf1# show smartfabric nodes

Service-Tag Type Status Mode Chassis-Service-Tag Chassis-slot FabricId
------------------------------------------------------------------------------------------------
2LB10Q2 Z9100-ON ONLINE FABRIC
2LB00Q2 Z9100-ON OFFLINE FABRIC
2LHZZP2 Z9100-ON ONLINE FABRIC
2L5ZZP2 Z9100-ON ONLINE FABRIC
4LB00Q4 Z9100-ON ONLINE FABRIC
2LPZZP2 Z9100-ON ONLINE FABRIC
2L300Q2 Z9100-ON ONLINE FABRIC

7. Identify the IP address of the leader switch node using the show smartfabric cluster command.

leaf1# show smartfabric cluster

--------------------------------------------------------------------------------
CLUSTER DOMAIN ID : 100
VIP : fde2:53ba:e9a0:cccc:0:5eff:FE00:1100
ROLE : BACKUP
SERVICE-TAG : 4LB00Q4
LEADER-IPV4 : 10.10.100.12

8. Restore the server cable connections to the new switch.

Verify the connectivity between the PowerScale server and the leader node (VN4091 IPv6 address) by running the ping
command.

DNSCALE-7# ping6 vrf H3-1-1 fdfe:9042:c53d::5eff:fe00:101

PING fdfe:9042:c53d::5eff:fe00:101 (fdfe:9042:c53d::5eff:fe00:101) 56 data bytes
64 bytes from fdfe:9042:c53d::5eff:fe00:101: icmp_seq=1 ttl=64 time=2035 ms
64 bytes from fdfe:9042:c53d::5eff:fe00:101: icmp_seq=2 ttl=64 time=1025 ms
64 bytes from fdfe:9042:c53d::5eff:fe00:101: icmp_seq=3 ttl=64 time=0.663 ms
64 bytes from fdfe:9042:c53d::5eff:fe00:101: icmp_seq=4 ttl=64 time=0.317 ms
64 bytes from fdfe:9042:c53d::5eff:fe00:101: icmp_seq=5 ttl=64 time=0.338 ms
^C
----- fdfe:9042:c53d::5eff:fe00:101 ping statistics -----
5 packets transmitted, 5 received, 0% packet loss, time 4075ms
rtt min/avg/max/mdev = 0.317/610.625/2027.111/811.772ms, pipe 3

leaf1# show running-configuration interface virtual-network 4091

!
interface virtual-network4091
description virtual-network4091
no shutdown
ipv6 address fdfe:9042:c53d::5eff:fe00:101/64
leaf1#

9. Perform the switch replacement workflow using the curl request from the PowerScale node.
Curl requests can be run from the Linux prompt of either the PowerScale node or the leader node. In this example, the
execution is shown from the PowerScale node.
a. Run the command on the switch as the root user:

system "sudo -i"

b. Run the following curl command.

curl -X POST https://<LeaderIP>/redfish/v2/Dnv/Fabrics('<FabricID>')/

Fabrics.ReplaceNode
-H "Content-Type: application/json"
-d '{
"NewServiceTag": "<new node service tag number>",
"OldServiceTag": "<old node service tag number>"
}'

SFS for PowerScale (Isilon) back-end fabric 65

 To run the node replacement API, you can use either the IPv4 or IPv6 leader IP. Dell Technologies recommends using the
VIP obtained from the show smartfabric cluster output for the IPv6.

curl -k -X POST https://admin:[email protected]"/redfish/v2/Dnv/Fabrics('100’)/

Fabrics.ReplaceNode "
-d '{"NewServiceTag":"4LB00Q4”,"OldServiceTag":"2LB00Q2"}'
{
"OpaqueNum": "7700c972-c688-4f42-b4d8-aa000b8eee8f",
"Timestamp": 1739120127
}

curl -k -X POST https://admin:admin@[fde2:53ba:e9a0:cccc:0:5eff:fe00:1100]"/

redfish/v1/Dnv/Fabrics('100')
/Fabrics.ReplaceNode" -d '{"NewServiceTag":"4LB00Q4","OldServiceTag":"2LB00Q2"}'
{
"OpaqueNum": "a82f3f35-ebc6-413a-8dc0-da4adbd4e018",
"Timestamp": 1739478462
}

The node replacement process completes. Check the readiness of all nodes by verifying the switch status using show
smartfabric nodes output, and the configuration of the new switch is restored to match that of the old one.
NOTE: The offline entry of the old switch is removed automatically.

leaf1# show smartfabric nodes

Service-Tag Type Status Mode Chassis-Service-Tag Chassis-slot FabricId
---------------------------------------------------------------------------------------------
---
2LB10Q2 Z9100-ON ONLINE FABRIC
2LHZZP2 Z9100-ON ONLINE FABRIC
2L5ZZP2 Z9100-ON ONLINE FABRIC
4LB00Q4 Z9100-ON ONLINE FABRIC
2LPZZP2 Z9100-ON ONLINE FABRIC
2L300Q2 Z9100-ON ONLINE FABRIC

10. Ensure that the old switch configuration is correctly ported to the new switch by comparing the running configuration before
and after the switch replacement. Also, check for validation errors using show smartfabric validation-errors.

66 SFS for PowerScale (Isilon) back-end fabric

 9
SmartFabric commands
You can run show commands specific to SFS from the CLI to view fabric configuration information. The command output varies
depending on the SFS deployment.

smartfabric l3fabric enable

Enables SFS on the switches and creates a L3 network fabric.

Syntax smartfabric l3fabric enable role {LEAF [vlti ethernet node/slot/port] |

SPINE} [domain domain-ID]
Parameters role—Enter the role of the switch in the L3 fabric:
● LEAF [vlti ethernet node/slot/port]—Specify the role as LEAF for top of rack
switches and specify the VLTi ports that interconnect the leaf switches.

NOTE: Option to specify VLTi ports are not applicable for Isilon/PowerScale deployment.

● SPINE—Specify the role as SPINE for the switch that connects the leaf switches.
● [domain domain-ID]—Specify the domain ID of the switch. The domain ID can range from
100 to 107. By default, the domain ID is set to 100.
Default None
Command Mode CONFIGURATION
Usage Information Once you enable the L3 fabric and assign a role, the system prompts for confirmation to reboot
to apply the mode change. If you confirm with Yes, the switch reboots in SmartFabric mode and
a network fabric is created automatically with default fabric settings. From the 10.5.3.0 release,
you can enable the L3 fabric along with domain. The leaf and spine ASN and subnet prefixes are
autogenerated based on the domain ID. Use the no smartfabric l3fabric command to disable
the L3 fabric personality. When you disable the L3 fabric in the switch, the system reboots to
change the personality after confirmation. This command is accessible to users with sysadmin and
netadmin roles.
Example (Spine)
OS10(config)# smartfabric l3fabric enable role SPINE domain 101

Reboot to change the personality? [yes/no]: yes

Example (Leaf)
OS10(config)# smartfabric l3fabric enable role LEAF vlti
ethernet 1/1/4-1/1/5 domain 101

Reboot to change the personality? [yes/no]: yes

Example (disable
SFS on both leaf and OS10(config)# no smartfabric l3fabric
spine)
Reboot to change the personality? [yes/no]: yes

Supported Releases 10.5.0.3 or later

SmartFabric commands 67
show logging smartfabric
Displays important logs that are related to SFS modules.

NOTE: You must have access to sysadmin, secadmin, or netadmin roles to run this command.

Syntax show logging smartfabric 1-65535

Parameters <1-65535>—Number of recent messages to be displayed.
Default Not applicable
Command Mode EXEC
Usage Information You can run this command on all switches in SmartFabric mode and are provided with a high-level
view of the events happening in the SFS module. When you run this command on an active switch,
the system displays complete cluster-related logs. The logs include events that are related to cluster
formation and update, configuration changes, and on boarding events along with switch-specific
information. When you run this command on a backup switch, the system displays only the switch-
specific logs.
Example
MX9116N-B2# show logging smartfabric 100 | grep CAGT
2020-03-29 10:09:35.334 MX9116N-B2 [DNV-CAGT]
[chassis.get_lead_chassis] lead chassis not in chassis_data
2020-03-29 10:09:35.335 MX9116N-B2 [DNV-CAGT]
[chassis.get_lead_chassis] not receive mdns from lead MSM yet
2020-03-29 10:09:56.881 MX9116N-B2 [DNV-CAGT]
[app.process_cps_cluster_sync_event] received sync state 2
2020-03-29 10:09:56.881 MX9116N-B2 [DNV-CAGT] [ka.clr_priority]
2020-03-29 10:09:56.885 MX9116N-B2 [DNV-CAGT] [ka.set_priority] new
100 old 0
2020-03-29 10:09:58.014 MX9116N-B2 [DNV-CAGT] [ka.soft_reload_ka]
reloading ka...

Supported Releases 10.5.2.3 or later

show smartfabric cluster

Displays the basic cluster information of the switch or IOM.

Syntax show smartfabric cluster

Parameters None
Default None
Command Mode EXEC
Usage Information This command is supported in both Full Switch and SmartFabric modes.
Example (IOM)
MX9116N-A1# show smartfabric cluster

----------------------------------------------------------
CLUSTER DOMAIN ID : 119
VIP : fde1:53ba:e9a0:de14:0:5eff:fe00:1119
ROLE : BACKUP
SERVICE-TAG : 3GB1XC2
LEADER-IPV4 : 10.11.105.15
----------------------------------------------------------

Example (VxRail - L2
fabric) OS10# show smartfabric cluster

----------------------------------------------------------
CLUSTER DOMAIN ID : 100
VIP : fde2:53ba:e9a0:cccc:0:5eff:fe00:1100

68 SmartFabric commands
 ROLE : LEADER
SERVICE-TAG : B37HXC2
LEADER-IPV4 : 10.11.106.27
PREFERRED-LEADER :
----------------------------------------------------------

Example (VxRail - L3
fabric) OS10# show smartfabric cluster

----------------------------------------------------------
CLUSTER DOMAIN ID : 100
VIP : fde2:53ba:e9a0:cccc:0:5eff:fe00:1100
ROLE : LEADER
SERVICE-TAG : B37HXC2
LEADER-IPV4 : 10.11.106.27
PREFERRED-LEADER : true
----------------------------------------------------------

Supported Releases ● MX9116n and MX5108n—10.5.0.1 or later

● SFS-supported OS10 switches—10.5.0.3 or later

show smartfabric cluster member

Displays information about the switch in a cluster. Information includes service tag, IP address, status, role, switch type and
chassis model, and chassis service tag where the switch is connected.

Syntax show smartfabric cluster member

Parameters None
Default None
Command Mode EXEC
Usage Information When you run this command on a switch, the output that is displayed varies depending on the switch
role. For example, if you run this command on the leader switch, the output shows both the leader and
backup switch information. If you run this command on a backup switch, the output shows only the
leader switch information. This command is supported in both Full-Switch and SmartFabric modes.
Example (IOM)
MX9116N-A1# show smartfabric cluster member
Service-tag IP Address Status
Role Type Chassis-Service-Tag Chassis-Slot
-------------------------------------------------------------
9GB1XC3 fde1:53ba:e9a0:de14:e6f0:4ff:fe3e:45dd ONLINE
LEADER MX9116n SKY002L B1

Example (VxRail)
OS10# show smartfabric cluster member
Service-tag IP Address Status
Role Type Chassis-Service-Tag Chassis-Slot
--------------------------------------------------------------
3Z4ZZP2 fde2:53ba:e9a0:cccc:54bf:64ff:fee6:e462 ONLINE
BACKUP
3Z4ZZP1 fde2:53ba:e9a0:cccc:54bf:64ff:fee6:e463 ONLINE
BACKUP
BR2ZZP2 fde2:53ba:e9a0:cccc:3c2c:30ff:fe49:2585 ONLINE
BACKUP
B37HXC2 fde2:53ba:e9a0:cccc:e4f0:4ff:feb6:fdc3 ONLINE
LEADER
G17HXC2 fde2:53ba:e9a0:cccc:e4f0:4ff:feb6:e1c3 ONLINE
BACKUP

Supported Releases ● MX9116n and MX5108n—10.5.0.1 or later

● SFS-supported OS10 switches—10.5.0.3 or later

SmartFabric commands 69
show smartfabric configured-server
Displays list of all configured servers information in a fabric. Information includes ID, model type, slot, chassis model and service
tag, bonding technology, list of existing bond members, and server status.

Syntax show smartfabric configured-server

Parameters None
Default None
Command Mode EXEC
Usage Information This command is supported in both Full Switch and SmartFabric modes.
Example (IOM)
MX9116N-B1# show smartfabric configured-server
----------------------------------------------------------
Service-Tag : 00FWX20
Server-Model : PowerEdge MX740c
Chassis-Slot : 1
Chassis-Model : POWEREDGE MX7000
Chassis-Service-Tag : SKY002L
Is-Discovered : TRUE
Is-Onboarded : TRUE
Is-Configured : TRUE
**********************************************************
Bonding Technology : LACP
BondMembers:
Nic-Id : Switch-Interface
----------------------------------------------------------
NIC.Mezzanine.1A-1-1 3GB1XC2:ethernet1/1/1
NIC.Mezzanine.1A-2-1 9A2HEM3:port-channel1
----------------------------------------------------------

Supported Releases 10.5.1.0 or later

show smartfabric configured-server configured-

server-interface
Displays interface-level information of the configured servers. Information includes server ID, port ID, onboarded interface,
server status, fabric ID, native VLAN, network profiles, and bandwidth partition details.

Syntax show smartfabric configured-server configured-server-interface server-id

Parameters server-id—Enter a configured server ID information.
Default None
Command Mode EXEC
Usage Information This command is supported in both Full Switch and SmartFabric modes.
Example
MX5108N-B1# show smartfabric configured-server configured-server-
interface 004YX20 | no-more
----------------------------------------------------------
Server-Id : 004YX20
----------------------------------------------------------
Port-Id : NIC.Mezzanine.1B-2-1
Onboard-Interface :
Fabric-id :
Is-Discovered : FALSE
Is-Onboarded : FALSE
Is-Configured : TRUE
NicBonded : FALSE
Native-vlan : 0
Networks : c56d6202-0ec1-4fcd-b119-6abc761a1268

70 SmartFabric commands
 ----------------------------------------------------------
Port-Id : NIC.Mezzanine.1A-2-1
Onboard-Interface : 1G86XC2:ethernet1/1/3
Fabric-id :
Is-Discovered : TRUE
Is-Onboarded : FALSE
Is-Configured : TRUE
NicBonded : FALSE
Native-vlan : 0
Networks : c56d6202-0ec1-4fcd-b119-6abc761a1268
----------------------------------------------------------
Port-Id : NIC.Mezzanine.1B-1-1
Onboard-Interface : 2J86XC2:ethernet1/1/3
Fabric-id :
Is-Discovered : TRUE
Is-Onboarded : FALSE
Is-Configured : TRUE
NicBonded : FALSE
Native-vlan : 0
Networks : c56d6202-0ec1-4fcd-b119-6abc761a1268
----------------------------------------------------------
Port-Id : NIC.Mezzanine.1A-1-1
Onboard-Interface :
Fabric-id :
Is-Discovered : FALSE
Is-Onboarded : FALSE
Is-Configured : TRUE
NicBonded : FALSE
Native-vlan : 0
Networks : c56d6202-0ec1-4fcd-b119-6abc761a1268

Supported Releases 10.5.1.0 or later

show smartfabric details

Displays all details specific to the fabric. Details include name, description, ID, nodes that are part of the fabric, design type
associated with the fabric, and status detail of a fabric.

Syntax show smartfabric details

Parameters None
Default None
Command Mode EXEC
Usage Information This command is supported in both Full Switch and SmartFabric modes.
Example (IOM)
MX9116N-A1# show smartfabric details
----------------------------------------------------------
Name : A1-A2
Description :
ID : fc6c9051-f499-4816-a54a-25ef6fef2e33
DesignType : 2xMX9116n_Fabric_Switching_Engines_in_same_chassis
Validation Status: VALID
VLTi Status : VALID
Placement Status : VALID
Nodes : 3GB1XC2, 9A2HEM3
----------------------------------------------------------

Example (VxRail)
OS10# show smartfabric details
----------------------------------------------------------
Name : AutoFab-08ee685b-d6d6-5d0c-99d2-ae78f800d4b7
Description : Auto-Fabric Generator
ID : 08ee685b-d6d6-5d0c-99d2-ae78f800d4b7
DesignType : AutoFabricDesign--1
Validation Status: VALID

SmartFabric commands 71
 VLTi Status : VALID
Placement Status : VALID
Nodes : CAC00N2, AZY1234
----------------------------------------------------------
----------------------------------------------------------
Name : AutoFab-100
Description : Auto-Fabric Generator
ID : 100
DesignType :
Validation Status: VALID
VLTi Status : VALID
Placement Status : VALID
Nodes : AZY1234, CAC00N2, 9GTWNK2, FHTWNK2
----------------------------------------------------------

Supported Releases ● MX9116n and MX5108n—10.5.0.1 or later

● SFS-supported OS10 switches—10.5.0.3 or later

show smartfabric discovered-server

Displays information about all the discovered servers. Information includes server tag, model, slot, chassis model, and chassis
service tag.

Syntax show smartfabric discovered-server

Parameters None
Default None
Command Mode EXEC
Usage Information This command is supported in both Full Switch and SmartFabric modes.
Example
MX5108N-B1# show smartfabric discovered-server
----------------------------------------------------------
Server-Tag : 004YX20
Server-Model : PowerEdge MX740c
Server-Slot : 1
Chassis-Model : POWEREDGE MX7000
Chassis-Service-Tag : SKY002R
----------------------------------------------------------
----------------------------------------------------------
Server-Tag : 0002X20
Server-Model : PowerEdge MX740c
Server-Slot : 1
Chassis-Model : POWEREDGE MX7000
Chassis-Service-Tag : SKY0044
----------------------------------------------------------

Supported Releases 10.5.1.0 or later

show smartfabric discovered-server discovered-

server-interface
Displays interface-level information of all the discovered servers. Information includes port ID and switch interfaces on which
the server is onboarded.

Syntax show smartfabric discovered-server discovered-server-interface server-id

Parameters server-id—Enter a discovered server ID information.
Default None
Command Mode EXEC

72 SmartFabric commands
Usage Information This command is supported in both Full Switch and SmartFabric modes.
Example
MX9116N-B1# show smartfabric discovered-server discovered-server-
interface 00FWX
20
Nic-Id : Switch-Interface
------------------------------------------------------
NIC.Mezzanine.1A-1-1 3GB1XC2:ethernet1/1/1
NIC.Mezzanine.1A-2-1 9A2HEM3:ethernet1/1/1

Supported Releases 10.5.1.0 or later

show smartfabric networks

Displays a detailed description of the configured network profiles. The description includes network name, type, network ID, QoS
priority type, and VLAN.

Syntax show smartfabric networks

Parameters None
Default None
Command Mode EXEC
Usage Information This command is supported in both Full Switch and SmartFabric modes.
Example (IOM)
MX9116N-A1# show smartfabric networks
Name Type QosPriority NetworkId
Vlan
-------------------------------------------------------------------------------
v5 GENERAL_PURPOSE BRONZE 8f018a8c-c355-4d81-9bee-85cfedcf8d
5
network100-105 GENERAL_PURPOSE BRONZE deb0886c-4a9b-47f2-8220-55afcb1f17
100 - 105
fcor STORAGE_FCOE PLATINUM d1de8f16-ebd0-4b1a-9689-a802d23b2b
777
VLAN 1 GENERAL_PURPOSE SILVER 4bb446a3-702c-4a0f-abdd-07dd0c1477
1
v1 GENERAL_PURPOSE BRONZE 9f2bed94-9148-46d8-9df6-3b606c83a4
500

Example (VxRail)
OS10# show smartfabric networks
Name Type QosPriority NetworkId Vlan
-----------------------------------------------------------------------------
Client_Control_Network VXLAN IRON Client_Control_Network 3939
Client_Management_Network VXLAN IRON Client_Management_Network 4091

Supported Releases ● MX9116n and MX5108n—10.5.0.1 or later

● SFS-supported OS10 switches—10.5.0.3 or later

show smartfabric nodes

Displays comprehensive inventory information for all the nodes in the fabric. The information displayed includes service tag,
type, status, mode, fabric ID associated with the node, chassis service-tag, and chassis-slot.

Syntax show smartfabric nodes node-id node-id

Parameters node-id node-id—Specify the service tag of a switch to view detailed information of that switch.
Default None
Command Mode EXEC

SmartFabric commands 73
Usage Information This command is supported in both Full Switch and SmartFabric modes.
Example (IOM)
MX9116N-A1# show smartfabric nodes
Service-Tag Type Status Mode
Chassis-Service-Tag Chassis-Slot FabricId
---------------------------------------------------
3GB1XC2 MX9116n ONLINE FABRIC
SKY002L A1

9GB1XC3 MX9116n ONLINE FULL-SWITCH

SKY002L B1

9A2HEM3 MX9116n ONLINE FABRIC

SKY002L A2

Example (VxRail)
OS10# show smartfabric nodes
Service-Tag Type Status Mode
Chassis-Service-Tag Chassis-Slot FabricId
-------------------------------------------------
GGVQG02 S5232F-ON ONLINE FABRIC
7222c224-223c-5fa4-a244-
cd3ca1685550 (Name-Rack)
AZY1234 S5232F-ON ONLINE FABRIC
7222c224-223c-5fa4-a244-
cd3ca1685550 (Name-Rack)

Example (VxRail)
OS10# show smartfabric nodes node-id GGVQG02
----------------------------------------------------------
Node Name : Name-Leaf-2
Node Id : GGVQG02
Node Type : S5232F-ON
Node Status : ONLINE
Node Mode : FABRIC
Node Ready : true
Node Model : S5232F-ON
Replacement Node Id :
Chassis service tag :
Chassis slot :
Fabric : 7222c224-223c-5fa4-a244-cd3ca1685550 (Name-Rack)
Fabric node status : OPERATIONAL
Software Version : 10.5.2.xMRDEV
Hardware Version : X01
Serial Number : CN01WJVTCES0085G0037
----------------------------------------------------------

Supported Releases ● MX9116n and MX5108n—10.5.0.1 or later

● SFS-supported OS10 switches—10.5.0.3 or later

show smartfabric personality

Displays the personality of the SFS cluster.

Syntax show smartfabric personality

Parameters None
Default None
Command Mode EXEC
Usage Information Use this command to identify the fabric personality of the SFS cluster. The output varies depending
on the mode and role of the switch, and personality. When this command is run on a switch, the
system displays the personality and the role of the switch. In VxRail deployment, if vxrail is

74 SmartFabric commands
 displayed as personality it means a L2 fabric. This command is supported in both Full Switch and
SmartFabric modes.
Example (IOM)
MX9116N-A1# show smartfabric personality

Personality :None
Role :
VLTi :

Example (VxRail) Full Switch mode:

OS10# show smartfabric personality

Personality :None
Role :
VLTi :

SmartFabric Services mode:

TOR1# show smartfabric personality

Personality :vxrail
Role :
VLTi :ethernet1/1/29, ethernet1/1/30

OS10# show smartfabric personality

Personality :L3 Fabric

Role :LEAF
VLTi :ethernet1/1/5, ethernet1/1/6
Leaf1#

OS10# show smartfabric personality

Personality :L3 Fabric

Role :SPINE
VLTi :

Supported Releases ● MX9116n and MX5108n—10.5.0.1 or later

● SFS-supported OS10 switches—10.5.0.3 or later

show smartfabric uplinks

Displays all uplink-related information in the SFS. Information includes uplink name, description, ID, media type, native VLAN,
configured interfaces, and the network profile associated with the uplink.

Syntax show smartfabric uplinks

Parameters None
Default None
Command Mode EXEC
Usage Information This command is supported in both Full Switch and SmartFabric modes.
Example (IOM)
MX9116N-A1# show smartfabric uplinks
----------------------------------------------------------
Name : uplink to b1
Description :
ID : 2725707d-886a-41c6-9d0d-38c4115788ff
Media Type : ETHERNET
Native Vlan : 1
Untagged-network :
Networks : deb0886c-4a9b-47f2-8220-55afcb1f1756,

SmartFabric commands 75
 9f2bed94-9148-46d8-9df6-3b606c83a472
Configured-Interfaces : 9A2HEM3:ethernet1/1/42, 3GB1XC2:ethernet1/1/42
----------------------------------------------------------
----------------------------------------------------------
Name : u1
Description :
ID : e1c8169e-00dd-4a72-9e42-54485c049591
Media Type : FC
Native Vlan : 0
Untagged-network :
Networks : d1de8f16-ebd0-4b1a-9689-a802d23b2b26
Configured-Interfaces : 3GB1XC2:fibrechannel1/1/44:1
----------------------------------------------------------

Example (VxRail)
OS10# show smartfabric uplinks
----------------------------------------------------------
Name : FABRICUPLINKNew
Description : L3VxLAN780 Uplink
ID : L3VxLANUplink-780
Media Type : ETHERNET
Native Vlan : 0
Untagged-network :
Networks : Network780
Configured-Interfaces : CAC00N2:ethernet1/1/22:2
----------------------------------------------------------
----------------------------------------------------------
Name : L3VLANUplink
Description : Uplink On L3VLAN Network800
ID : L3VLANUplink-800
Media Type : ETHERNET
Native Vlan : 0
Untagged-network :
Networks : Network800
Configured-Interfaces : CAC00N2:ethernet1/1/22:1
----------------------------------------------------------
----------------------------------------------------------
Name : L2VxLANUplink
Description : Uplink On L2VxLAN Network770
ID : L2VxLANUplink-770
Media Type : ETHERNET
Native Vlan : 0
Untagged-network :
Networks : Network770
Configured-Interfaces : CAC00N2:ethernet1/1/22:3
----------------------------------------------------------
----------------------------------------------------------
Name : L2VxLANUplink
Description : Uplink On L2VxLAN Network771
ID : L2VxLANUplink-771
Media Type : ETHERNET
Native Vlan : 0
Untagged-network : Network771
Networks :
Configured-Interfaces : CAC00N2:ethernet1/1/22:4
----------------------------------------------------------
----------------------------------------------------------
Name : L3Uplink
Description : Uplink On L3 Network600
ID : L3RUplink-600
Media Type : ETHERNET
Native Vlan : 0
Untagged-network : Network600
Networks :
Configured-Interfaces : AZY1234:ethernet1/1/21:2
----------------------------------------------------------

Supported Releases ● MX9116n and MX5108n—10.5.0.1 or later

● SFS-supported OS10 switches—10.5.0.3 or later

76 SmartFabric commands
show smartfabric upgrade-status
Displays all the information about the upgrade status.

Syntax show smartfabric upgrade-status

Parameters None
Default Not applicable
Command Mode EXEC
Usage Information You can run this command on all the switches in the fabric. The output remains the same on all the switches. Thi
command is accessible only to the users with sysadmin, secadmin, or netadmin roles.
Example
MX9116N-A1# show smartfabric upgrade-status

Opaque-id : ea89b7c4-de00-4fc9-ad54-9ed5bf61e300
Upgrade Protocol : PUSH
Upgrade start time : 2021-02-12 01:51:29.261000
Status : SUCCESS
Nodes to Upgrade : SVC009F, SVC009A, 7H92Q03, SVC009C, 8PQXV23,
34HQXC2, 8PTXV23, D6RRNK2, BWGQXC2, 9KK1W23,
9J45W23, 8Q52W23
Reboot Sequence : 7H92Q03,SVC009C,9KK1W23,SVC009F,8Q52W23,8PTXV23,
D6RRNK2,BWGQXC2,SVC009A,34HQXC2,5WQQXC2,9J45W23,
8PQXV23

Node Current Current Status-Message

-Action -Status
-------------------------------------------------------------------------------
SVC009C DOWNLOAD SUCCESS Skipping ONIE update for the node SVC009C
since ONIE is already installed with latest
firmware.
SVC009F DOWNLOAD SUCCESS Skipping ONIE update for the node SVC009F

since ONIE is already installed wit

latest
firmware.
SVC009A DOWNLOAD SUCCESS Skipping ONIE update for the node SVC009A
since ONIE is already installed with latest
firmware.
8Q52W23 DOWNLOAD SUCCESS Skipping ONIE update for the node 8Q52W23
since ONIE is already installed with latest
firmware.
34HQXC2 DOWNLOAD SUCCESS Skipping ONIE update for the node 34HQXC2
since ONIE is already installed with latest
firmware.
BWGQXC2 DOWNLOAD SUCCESS Skipping ONIE update for the node BWGQXC2
since ONIE is already installed with latest
firmware.
9J45W23 DOWNLOAD SUCCESS Skipping ONIE update for the node 9J45W23
since ONIE is already installed with latest
firmware.
7H92Q03 DOWNLOAD SUCCESS Skipping ONIE update for the node 7H92Q03
since ONIE is already installed with latest
firmware.
9KK1W23 DOWNLOAD SUCCESS Skipping ONIE update for the node 9KK1W23
since ONIE is already installed with latest
firmware.
8PQXV23 DOWNLOAD SUCCESS Skipping ONIE update for the node 8PQXV23
since ONIE is already installed with latest
firmware.
D6RRNK2 ONIE-INSTALL SUCCESS [Action: REBOOT] Successfully rebooted.
8PTXV23 DOWNLOAD SUCCESS Skipping ONIE update for the node 8PTXV23
since ONIE is already installed with latest
firmware.

Supported Releases 10.5.2.4 or Later

SmartFabric commands 77
show smartfabric validation-errors
Displays validation-error information of the topology. Information includes error category, subcategory, description,
recommended action, severity, timestamp, EEMI, problem, and recommendation for each error.

Syntax show smartfabric validation-errors

Parameters None
Default None
Command Mode EXEC
Usage Information Use this command to view a list of topology validation errors with a detailed description about each
error. This command can be run on any IOM deployed in the same cluster. This command is supported
in both Full Switch and SmartFabric modes.
Example
OS10# show smartfabric validation-errors
----------------------------------------------------------
ErrorKey : V098911:ethernet1/1/4: Leaf-Spine Link Error
MessageID :
Description : LEAF to LEAF connection detected on
V098911:ethernet1/1/4
EEMI :
Category : FABRIC_ERROR
Subcategory : LEAF_SPINE_LINK_ERROR
Severity : SEVERITY_1
Recommended Action:Please verify link V098911:ethernet1/1/4
Timestamp : 1587488570
Problem Link
SourceNode :V098911
SourceInterface :ethernet1/1/4
DestinatioNode :V660EE3
DestinationInterface:
Recommended Link
SourceNode :V098911
SourceInterface :ethernet1/1/4
DestinatioNode :V660EE3
DestinationInterface:ethernet1/1/4
----------------------------------------------------------

----------------------------------------------------------
ErrorKey : V098911:ethernet1/1/3:VLT VLTi Error
MessageID :
Description : VLT LEAF to LEAF VLTi connection not detected on
V098911:ethernet1/1/3
EEMI :
Category : FABRIC_ERROR
Subcategory : LEAF_SPINE_LINK_ERROR
Severity : SEVERITY_1
Recommended Action:Please verify link V098911:ethernet1/1/3
Timestamp : 1587490907
Problem Link
SourceNode :V098911
SourceInterface :ethernet1/1/3
DestinatioNode :V09EE89
DestinationInterface:
Recommended Link
SourceNode :V098911
SourceInterface :ethernet1/1/3
DestinatioNode :V09EE89
DestinationInterface:ethernet1/1/9
----------------------------------------------------------

Supported Releases ● MX9116n and MX5108n—10.5.0.1 or later

● SFS-supported OS10 switches—10.5.0.3 or later

78 SmartFabric commands
show switch-operating-mode
Displays the current operating mode of a switch.

Syntax show switch-operating-mode

Parameters None
Command Mode EXEC
Usage Information Some OS10 switches operate in both Full Switch and SFS modes. Use this command to view or verify
the operating mode of a switch that is deployed in the SFS environment.
Example
OS10# show switch-operating-mode
Switch-Operating-Mode : SmartFabric Mode

Supported Releases 10.4.0E(R3S) or later

SmartFabric commands 79
 10
Best Practices
This chapter provides you a list of the best practices for SFS deployments with various solutions.

Recommended upgrade sequence for SFS and

solution deployments
This section covers the recommended upgrade sequence for SFS and its supported solution deployments. This recommended
order of upgrade is applicable for all SFS deployments.
Prerequisites
Ensure that you follow the prerequisites before proceeding with the upgrade.
● Check the release notes of each product for the version that you are upgrading to for any caveats or upgrade sequence
requirements.
● Verify if all the product version that you are upgrading to are compatible with the solutions support matrix, see Networking
Solutions Support Matrix.

Upgrade sequence in deployments with OMNI

This recommended order of upgrade is applicable for all SFS deployments with OMNI.
1. Upgrade OMNI. Follow the instructions provided in the OMNI User Guide and the OMNI Release Notes.
2. Upgrade OS10 using OMNI.
a. Before upgrading OS10, check the OS10 release notes of the respective OS10 version and the SmartFabric OS10
Installation, Upgrade, and Downgrade Guide available in SmartFabric OS documentation for any additional requirements.
b. Follow the SmartFabric OS upgrade procedure in the OMNI User Guide.

NOTE: The switches in the SmartFabric must be at the same version of OS10 when upgrading using OMNI. If not,
upgrade process may fail. If you want to upgrade the switches that have different OS10 versions installed, upgrade each
switch manually.

3. If needed, upgrade the solutions deployed with SFS such as VxRail, ESXi, vCenter, and so on.
a. Enable Maintenance mode for vCenter instance in OMNI before upgrading the attached solutions such as VxRail,
ESXi, vCenter, and so on. For more information about how to move the vCenter to maintenance mode, see vCenter
Maintenance mode section in the OMNI User Guide.
b. To upgrade the solutions, see the respective solution documentation for instructions.

Upgrade sequence in deployments without OMNI

This recommended order of upgrade is applicable for all SFS deployments that do not include OMNI.
1. Upgrade OS10 on the switches, see SmartFabric OS10 Installation, Upgrade, and Downgrade Guide available on the
SmartFabric OS10 documentation support page. Check the OS10 release notes of the respective OS10 version for any
additional requirements.
2. If needed, upgrade the solutions deployed with SFS such as VxRail, ESXi, and so on. See the respective solution
documentation for upgrade instructions.

80 Best Practices
Frequently asked questions
This chapter provides answers to frequently asked questions about SFS.
● Can I use different platforms in the same VLT leaf pair in a fabric?
No, In SFS deployments, both leaf switches in a given VLT pair must be on the same platform. See Networking Solutions
Support Matrix for the list of supported switches and typical roles in SFS deployments.
● Can I use the different switch port profiles available on S4148F-ON in SmartFabric mode?
Yes, all six profiles available on S4148F-ON can be used in SmartFabric mode as follows:
1. Set the appropriate port profile for the switch in Full-Switch mode.

CAUTION: Changing the switch port profile resets the system defaults and overwrites the existing switch
configuration. You must reconfigure the setting on a switch manually after applying the new port profile.

2. Reload the switch to apply the new port profile.

3. Enable SFS on the switch to change the switch mode to SmartFabric. The switch reboots in SmartFabric mode and uses the
set port profile.
NOTE: For more information about port profiles, see SmartFabric OS10 User Guide. This procedure is applicable only on the
S4148F-ON platform as S4148FE-ON and S4148U-ON switches are not supported in SFS deployments.
● How do I change the OS10 admin password in SmartFabric mode?
By default, admin is the username and password to log in to the OS10 CLI. The default OS10 admin password should be
changed on each switch after the first login. The procedure is the same for a switch in Full Switch mode or SmartFabric mode.
For more information about changing the password, see SmartFabric OS10 User Guide. The admin password does not have to
be the same password on all switches in an SFS deployment. When logging in to the SFS UI, use the credentials of the leader
switch. The leader switch may change and use the command show smartfabric cluster at any switch in the SmartFabric
to see the details of the leader switch.
● How can I upgrade firmware for a SFS-enabled switch?
The firmware can be upgraded using the image install command. For more information about installing firmware upgrade,
see SmartFabric OS10 User Guide.
● Can I have a peer switch with multiple BGP devices?
Yes, you can peer SFS switches with multiple BGP devices.
● Is OSPF supported for SFS?
No. OSPF is not supported in SFS with leaf and spine deployments.
● How many jump ports can I configure?
You can configure one jump port per leaf switch in a fabric.
● Can I change the VLTi interface after SFS has been enabled?
There is no option to change the VLTi interface in SmartFabric mode. If you want to change the VLTi interfaces, the leaf
switches must be changed from SmartFabric mode to Full-Switch mode using the no smartfabric l3fabric command.
This command deletes all the existing configuration on the switches. Enable SmartFabric mode back on the leaf switches and
specify the new VLTi interfaces with the smartfabric l3fabric enable role LEAF vlti ethernet command, see
here.
● If the switch has some configuration and the SFS feature is enabled, what happens to the existing configuration of a switch?
When moving from Full Switch mode to SFS mode, the switch retains management settings including management IP address,
management route, hostname, NTP server, IP name server, and RADIUS and TACACS+ configurations. However, all other
configurations are replaced by the SFS automated configuration. Before enabling SFS on the switch, you must manually back up
the configuration. For more information about startup configuration backup and restore through CLI, see Dell SmartFabric OS10
User Guide,
When reverting from SFS to Full Switch mode, the system retains the switch management settings, and removes the SFS
automated configuration and replaces it with the default configuration of the switch. Before switching, you must manually back
up the configuration using OMNI or RESTAPI.

Best Practices 81
 11
Migration of SFS VxRail deployment from L2
to L3 personality
Use the following procedure to migrate a network fabric from L2 single rack personality to L3 leaf and spine fabric personality.

SFS VxRail integrated personalities

● SFS for VxRail L2 single rack personality—enables an automated single rack network fabric (L2 fabric profile) for VxRail
clusters.
● SFS L3 leaf and spine fabric personality—enables a multi rack data center network fabric offering flexibility to start with
a L3 single rack (L3 fabric profile) and expand to a multi rack solution on demand. The L3 personality is integrated with
VxRail to enable single-site, multi rack VxRail deployments allowing VxRail nodes to be deployed in any rack without complex
underlay network configuration.
For more information about configuring VxRail L2 single rack personality, see VMware Integration for VxRail Fabric Automation
SmartFabric User Guide, Release 1.1. To understand the comparison between these two personalities, see VMware Integration
for VxRail Fabric Automation SmartFabric User Guide, Release 1.3.

Migration guidelines
The following guidelines should be followed before beginning with the migration:
CAUTION: Migrating from L2 to L3 fabric personality necessitates repeated reboots of leaf switches, resulting
in at least 2-hour outage. Dell Technologies recommends that you do this task during a scheduled maintenance
window.
● Before beginning the migration, you must upgrade the switches in the fabric to OS10.5.6.0 version. Check the OS10 release
notes of the respective OS10 version and the SmartFabric OS10 Installation, Upgrade, and Downgrade Guide available
in SmartFabric OS documentation for any additional requirements. Also, see the Support Matrix for the latest supported
versions of products in solution deployments.
● During the migration activity, do not change any hardware, such as VLTi connections or platforms.
● Before disabling SFS L2 personality, ensure that all VxRail clusters and VM application services are gracefully shut down.
● During the procedure, do not make any software upgrades through ONIE. Before taking the configuration backup, perform
any necessary upgrades in the L2 fabric.
The migration strategy differs based on whether OMNI is deployed externally or internally, and whether the Client Management
Network is tagged or untagged.

SFS VxRail deployment

The following figure shows a typical example deployment of SFS VxRail deployment with external and internal OMNI:

82 Migration of SFS VxRail deployment from L2 to L3 personality

Migration procedure for deployments with untagged
or tagged Client Management Network setup
The procedures for migrating from L2 to L3 for deployments with untagged or tagged Client Management Network setup are as
follows:
1. Note down the VxRail Management VLAN ID. This information is accessible through VxRail Manager. For more information,
see VxRail Appliance Series Documentation.
2. Back up the configuration using the REST API, GET: https://<LeaderIP>/redfish/v1/Dnv/Backup.
The following is an example:

curl -k -X GET https://admin:[email protected]/redfish/v1/Dnv/Backup

{
"_comment": "The node keyword has list of switches in the cluster.
The order of the node is all leaves grouped by vlt-fabric followed by the spines.

The time of backup is 07/26/2023, 14:37:46",

"personality": "vxrail",
"node": [
"GDLJPK2",
"5DLJPK8"
],
"data": {
"dnv-subscription-event/subscriptions/subscription,target": [
{
"dnv-subscription-event/subscriptions/subscription/subscription-id":
"0ee12430-317a-4f17-af97-0666fe52247c",
"dnv-subscription-event/subscriptions/subscription/description":
"Event Subscribtion from OMNI: 100.104.26.87",
"dnv-subscription-event/subscriptions/subscription/destination-uri":
"https://100.104.26.87:5443/delawareos10/rest/events/listener",
"dnv-subscription-event/subscriptions/subscription/event-type": [
1
],
"dnv-subscription-event/subscriptions/subscription/context":

Migration of SFS VxRail deployment from L2 to L3 personality 83

 "sf_100.104.72.183",
"dnv-subscription-event/subscriptions/subscription/protocol": 1,
"dnv-subscription-event/subscriptions/subscription/http-header": [
"Authorization:Basic RVZFTlRTX1VTRVI6NVNFT1NKUkU="
]
}
],
"dell-dnv-fabric/fabrics/fabric/uplinks,target": [
{
"dell-dnv-fabric/fabrics/fabric/fabric-id": "254",
"dell-dnv-fabric/fabrics/fabric/uplinks/uplink-id":
"AutoFabricUplink-254",
"dell-dnv-fabric/fabrics/fabric/uplinks/configured-interfaces": [
"{1}:ethernet1/1/26",
"{0}:ethernet1/1/26"
],
"dell-dnv-fabric/fabrics/fabric/uplinks/media-type": 1,
"dell-dnv-fabric/fabrics/fabric/uplinks/description": "Auto-Fabric
Uplink Generator",
"dell-dnv-fabric/fabrics/fabric/uplinks/uplink-type": 2,
"dell-dnv-fabric/fabrics/fabric/uplinks/lag-type": 1
}
],
"dell-dnv-user-config/info,target": [
{
"dell-dnv-user-config/info/user-config": 1
}
],
"dell-dnv-inventory/nodes/node,target": [
{
"dell-dnv-inventory/nodes/node/fabric-id": "254",
"dell-dnv-inventory/nodes/node/fabric-node-status": 2,
"dell-dnv-inventory/nodes/node/node-id": "{1}",
"dell-dnv-inventory/nodes/node/fabric-lacp-mac": "f0d4e2224912"
},
{
"dell-dnv-inventory/nodes/node/fabric-id": "254",
"dell-dnv-inventory/nodes/node/fabric-lacp-mac": "f0d4e2224912",
"dell-dnv-inventory/nodes/node/node-id": "{0}",
"dell-dnv-inventory/nodes/node/fabric-node-status": 2
}
],
"dnv-autolag/peer-id-map,target": [
{
"dnv-autolag/peer-id-map/fabric-id": "254",
"dnv-autolag/peer-id-map/partner-id": "2-AutoFabricUplink-254",
"dnv-autolag/peer-id-map/id": 101,
"dnv-autolag/peer-id-map/creation-tag": "2023-07-25.18:36:50,3"
}
],
"dell-dnv-rest-user/info,target": [
{
"dell-dnv-rest-user/info/user-name": "REST_USER",
"dell-dnv-rest-user/info/current-password":
"hViYkm2AzVQpsnVVgUYN826gbKuz9qLfsk0zK2g3PdSianLGS6MJQ3KhPGB27MqR"
}
],
"dnv-vrf/vrf-configs/vrf-config,target": [
{
"dnv-vrf/vrf-configs/vrf-config/vrf-name": "tenant4",
"dnv-vrf/vrf-configs/vrf-config/ipv4-exp-rt-route-target": "4:4"
},
{
"dnv-vrf/vrf-configs/vrf-config/vrf-name": "tenant3",
"dnv-vrf/vrf-configs/vrf-config/ipv4-exp-rt-route-target": "3:3"
},
{
"dnv-vrf/vrf-configs/vrf-config/vrf-name": "tenant1",
"dnv-vrf/vrf-configs/vrf-config/ipv4-exp-rt-route-target": "1:1"
},
{
"dnv-vrf/vrf-configs/vrf-config/vrf-name": "tenant2",
"dnv-vrf/vrf-configs/vrf-config/ipv4-exp-rt-route-target": "2:2"

84 Migration of SFS VxRail deployment from L2 to L3 personality

 }
],
"dell-dnv-inventory/interfaces/interface,target": [
{
"dell-dnv-inventory/interfaces/interface/node-id": "{1}",
"dell-dnv-inventory/interfaces/interface/interface-name":
"{1}:ethernet1/1/30",
"dell-dnv-inventory/interfaces/interface/interface-role": 3
},
{
"dell-dnv-inventory/interfaces/interface/node-id": "{0}",
"dell-dnv-inventory/interfaces/interface/interface-name":
"{0}:ethernet1/1/29",
"dell-dnv-inventory/interfaces/interface/interface-role": 3
},
{
"dell-dnv-inventory/interfaces/interface/node-id": "{0}",
"dell-dnv-inventory/interfaces/interface/interface-name":
"{0}:ethernet1/1/26",
"dell-dnv-inventory/interfaces/interface/interface-role": 12
},
{
"dell-dnv-inventory/interfaces/interface/node-id": "{0}",
"dell-dnv-inventory/interfaces/interface/interface-name":
"{0}:ethernet1/1/30",
"dell-dnv-inventory/interfaces/interface/interface-role": 3
},
{
"dell-dnv-inventory/interfaces/interface/node-id": "{1}",
"dell-dnv-inventory/interfaces/interface/interface-name":
"{1}:ethernet1/1/29",
"dell-dnv-inventory/interfaces/interface/interface-role": 3
},
{
"dell-dnv-inventory/interfaces/interface/node-id": "{1}",
"dell-dnv-inventory/interfaces/interface/interface-name":
"{1}:ethernet1/1/26",
"dell-dnv-inventory/interfaces/interface/interface-role": 12
}
],
"dell-dnv-fabric-design/fabric-designs/fabric-design,observed": [
{
"dell-dnv-fabric-design/fabric-designs/fabric-design/name":
"AutoFabricDesign--1",
"dell-dnv-fabric-design/fabric-designs/fabric-design/nodes": {
"0": {
"node-name": "Switch-A",
"chassis-name": "Chassis-X",
"node-placement-slot": 1,
"node-type": 1
},
"1": {
"node-name": "Switch-B",
"chassis-name": "Chassis-X",
"node-placement-slot": 1,
"node-type": 1
}
},
"dell-dnv-fabric-design/fabric-designs/fabric-design/links": {
"0": {
"source-node": "Switch-A",
"destination-node": "Switch-B",
"destination-physical-port": "ethernet1/1/30",
"source-physical-port": "ethernet1/1/30",
"link-id": "VLTi-0"
},
"1": {
"source-node": "Switch-A",
"destination-node": "Switch-B",
"destination-physical-port": "ethernet1/1/29",
"source-physical-port": "ethernet1/1/29",
"link-id": "VLTi-1"
}

Migration of SFS VxRail deployment from L2 to L3 personality 85

 }
}
],
"dell-dnv-plugin/config-reader,target": [
{
"dell-dnv-plugin/config-reader/config-read": 1
}
],
"dell-dnv-image-upgrade/latest-info,observed": [
{
"dell-dnv-image-upgrade/latest-info/enhanced-upgrade-support": 1
}
],
"dell-dnv-fabric/fabrics/fabric,target": [
{
"dell-dnv-fabric/fabrics/fabric/fabric-id": "254",
"dell-dnv-fabric/fabrics/fabric/fabric-nodes": [
"{0}",
"{1}"
],
"dell-dnv-fabric/fabrics/fabric/fabric-design": "AutoFabricDesign--1",
"dell-dnv-fabric/fabrics/fabric/fabric-design-mappings": {
"0": {
"physical-node": "{0}",
"design-node": "Switch-A"
},
"1": {
"physical-node": "{1}",
"design-node": "Switch-B"
}
},
"dell-dnv-fabric/fabrics/fabric/name": "AutoFab-254",
"dell-dnv-fabric/fabrics/fabric/description": "Auto-Fabric Generator",
"dell-dnv-fabric/fabrics/fabric/default-uplink": ""
}
]
}
}root@2rack-L2:~#

3. Copy the GET response output and store it as a JSON file.

4. Use the Graceful shutdown option to shut down the VxRail cluster and VM applications through vCenter. For more
information, see VxRail Software Documentation.
5. Disable SFS Layer 2 personality on the fabric with sfs_disable.py command from Linux Shell prompt. Run this command
on all the switches in the fabric.

root@2rack-L2:~# sfs_disable.py

Turning off SmartFabric Services will erase all the configuration performed through
SmartFabric Services and the switch will removed from SmartFabric cluster and the
switch
will now have to be manually configured.For workloads that were communicating with
the
SmartFabric Services Master need to be reconfigured for manual deployment.

Are you sure you want to continue with the changes by rebooting the OS10 node?.(y/n)

Enter y to continue. The switch reloads and restarts in Full-Switch mode.

6. Use the smartfabric l3fabric enable command with domain ID (100-107) to enable SFS with L3 personality on
the switches. The switches reboot in SmartFabric mode. For more information about the command, see Dell SmartFabric
Services User Guide.
The following are some examples:

OS10(config)# smartfabric l3fabric enable role SPINE domain 101

Reboot to change the personality? [yes/no]: yes

OS10(config)# smartfabric l3fabric enable role LEAF vlti

ethernet 1/1/4-1/1/5 domain 101
Reboot to change the personality? [yes/no]: yes

86 Migration of SFS VxRail deployment from L2 to L3 personality

7. If the REST_USER default password is modified from the default password (admin), delete the SFS instance in OMNI and
readd it using the default REST_USER password.
8. Log in to the SFS UI with admin credentials to edit and modify the Client Management VLAN ID to the value (VxRail
Management VLAN ID) that you saved in Step 1. For more information about how to edit the default settings, see Edit fabric
settings. Switches reboot in order to apply the fabric settings.
9. From the SFS UI, use the Restore feature to restore all configurations from the backed-up JSON file, see Restore. All L2
parameters are converted to L3, including the uplink. The switches reboot after the restore. The REST_USER password is
changed to the previously configured nondefault password.
10. Power On the VxRail cluster and VM application services.
11. Verify the traffic flow and restored configurations. You can extend the fabric by adding extra spine and leaf switches as
needed.
Follow the recovery procedures if you have any difficulty during the migration:
1. If the configuration restore process in SFS L3 personality fails, attempt the restore procedure once more.
2. If you want to switch back to L2 personality, move the switches to the Full-Switch mode by disabling the SFS mode
and enable the L2 personality using the sfs_enable_vxrail_personality.py command in the Linux shell. For more
information about enabling L2 personality, see VMware Integration for VxRail Fabric Automation SmartFabric User Guide,
Release 1.1. Use the Restore functionality in the UI to apply all the configuration from the backed-up JSON file. See Restore
for more information.
3. Contact Dell Technical Support if you are still experiencing problems with the migration.

Migration of SFS VxRail deployment from L2 to L3 personality 87

 12
SFS concepts
This chapter covers additional information that can support you with the fabric configuration tasks.

REST_USER account
When the fabric is formed, the SFS REST service is started on the leader node. Applications that are integrated with SFS use
the REST service for fabric operations. Communication is performed with the fabric using the IPv6 VIP assigned to the SFS
leader or using the IPv4 out-of-band Management IP of the leader.
A default REST_USER account is created to authenticate all REST queries. OMNI communicates with SmartFabric REST
Services through REST_USER account only. The default password is admin, and Dell Technologies recommends changing the
default password through Change SmartFabric password in SFS UI.
The system displays the XXXXXX: Internal-Err(CLUSTER): REST_USER is using the default password
message under Fabric Compliance page, during the following scenarios:
● REST_USER account having the default REST_USER password.
● REST_USER account having non default password but does not meet the required password criteria
Dell Technologies recommends changing the REST_USER account password immediately, as it is either set to the default or
does not comply with the password criteria. The password criteria are as follows:
● Must be at least nine characters long.
● Must contain at least one lowercase letter (a-z), one uppercase letter (A-Z), one numeric digit (0-9), and one special
character (!, @, #, $, %, ^, &, or *).
The REST_USER password never expires.

Internal fabric components and networks

SFS automatically creates the following components and networks:

Internal SFS components

SFS creates a VLT fabric automatically in the leaf and spine environment. VLT fabric is autoassigned with a fabric-ID, a
universally unique identifier (UUID). When a VLT fabric is created, the management IP addresses of the VLT peers are used
automatically to set up the VLT backup link. If the management IP address of the peers is changed after the fabric is created,
the VLT backup link is updated automatically.
SFS creates a network fabric with the leaf and spine switches automatically. The network fabric is autoassigned with a
fabric-ID, a name, and description. The fabric name and description are automatically assigned but can be changed through the
SFS UI. For more information, see Update fabric name and description.
Fabric links create a connection between the switches in a network fabric. Leaf-Spine link is a link formed between a leaf and
spine switch. All parallel links with the same connectivity are grouped to form a port channel interface. VLT interconnect (VLTi)
is a link formed between the two leaf switches in a same rack.

Internal virtual networks

The internal virtual networks created by SFS are:

VLAN 4000 for SFS SFS automatically configures VLAN 4000 on all the switches that are discovered in a fabric and uses
cluster control the network for all internal fabric operations. When a leaf or spine switch is discovered, the VLTi or
Leaf-Spine link ports are automatically added as tagged members.

88 SFS concepts
VLAN 4001 to 4079 SFS automatically configures the leaf and spine network using eBGP as the underlay routing protocol.
for leaf and spine SFS uses the reserved VLAN range from 4001 to 4079 with automatic IP addressing to set up the
connections peer connections. When SFS detects a Leaf-Spine link connection on either a leaf or spine switch, it
assigns the Leaf-Spine link to the untagged member of this VLAN. IP address from the reserved range
is used for this VLAN, and an eBGP session is started on the VLAN interface.
VLAN 4080 for SFS automatically configures VXLAN overlay networks with EVPN to extend networks between racks
Global untagged in a multirack deployment. VLAN 4080 with automatic IP addresses from the reserved range is used
VXLAN for VLTi links. VXLAN requires one VLAN to be assigned globally for untagged port-scoped VLAN
(Port, VLAN) pairs.
VLAN 4089—OS10 In SmartFabric mode, VLAN 4089 is the default VLAN and is reserved for OS10 internal use.
internal use
VLAN 4090—iBGP SFS automatically configures iBGP peering between a pair of leaf switches directly connected over
peering between VLTi links. VLAN 4090 is created automatically with IP addresses from the reserved range for enabling
leaf switches iBGP sessions between the VLT peer switches.
VLAN 4094—VLT SFS automatically creates VLAN 4094 on all leaf switches. This VLAN is used for all VLT control traffic
control VLAN between two VLT peer switches and is added on the VLTi ports on leaf switches.
VLAN 4091—Default SFS automatically configures an overlay network that is called a client_Management_Network.
client management When a device is automatically onboarded on to the network fabric, the device uses the VLAN mapped
network to this overlay network. This network is a native VLAN unless there is a policy specifying a different
native VLAN. VLAN 4091 is used as the default client management VLAN for the VXLAN network.
NOTE: You can change this VLAN to a specified VLAN through SFS UI.

VLAN 3939—Default SFS configures a second overlay network that is called Client_Control_Network for SFS-
client control integrated solutions. When a device such as VxRail is discovered, it is automatically added as a
network tagged member of this network. SFS enables leader advertisement and fabric discovery by integrated
solutions. The SFS leader virtual IP address for the VXLAN network is advertised. The VIP address
fde1:53ba:e9a0:cccc:0:5eff:fe00:1100 is fixed and not user configurable. VLAN 3939 is
used as the default client control VLAN for this VXLAN network for SFS-integrated solutions including
VxRail and PowerScale solutions. Although you can change the VLAN associated with the default
client management and control networks, Dell Technologies recommends not to change the VLANs for
VxRail deployments.

To check the networks that are created in SFS-integrated deployment, use show virtual-network command. The
following is the example output:

OS10# show virtual-network

Codes: DP - MAC-learn Dataplane, CP - MAC-learn Controlplane, UUD - Unknown-Unicast-Drop
Un-tagged VLAN: 4080
Virtual Network: 3939
Description: In-band SmartFabric Services discovery network
VLTi-VLAN: 3939
Members:
VLAN 3939: port-channel1000, ethernet1/1/12, ethernet1/1/13
VxLAN Virtual Network Identifier: 3939
Source Interface: loopback2(172.30.0.0)
Remote-VTEPs (flood-list): 172.30.0.1(CP)

Virtual Network: 4091

Description: Default untagged network for client onboarding
VLTi-VLAN: 4091
Members:
Untagged: ethernet1/1/12, ethernet1/1/13
VLAN 4091: port-channel1000
VxLAN Virtual Network Identifier: 4091
Source Interface: loopback2(172.30.0.0)
Remote-VTEPs (flood-list): 172.30.0.1(CP)

Reserved networks
SFS uses the reserved networks internally by default. If these networks conflict with any networks in the existing deployment,
change the default networks using the instructions that are provided in Edit Default Fabric Settings.

SFS concepts 89
Domain ID and ASN mapping
After you enable the SFS on the switches, SFS:
● Associates domain ID for the cluster if it is not assigned when enabling SFS. By default, the domain ID is set to 100.
● Autogenerates ASN for the leaf and spine switches in the fabric.
● Autogenerates subnet with prefix from domain ID.
NOTE: You can edit the domain ID from Fabric Settings page. Any change to the domain ID changes the ASN for leaf and
spine switches, and the subnet details.
The following table lists the ASN and BGP subnet, and VTEP subnet mapped with the domain ID:

Table 13. Domain ID to ASN mapping—OS10 release prior to OS10.5.4.0

Domain ID Leaf ASN Spine ASN Fabric BGP subnet Fabric VTEP subnet
and mask and mask
100 65011 65012 172.16.0.0/16 172.30.0.0/16
101 65015 65016 172.18.0.0/16 172.32.0.0/16
102 65017 65018 172.19.0.0/16 172.33.0.0/16
103 65019 65020 172.20.0.0/16 172.34.0.0/16
104 65021 65022 172.21.0.0/16 172.35.0.0/16
105 65023 65024 172.22.0.0/16 172.36.0.0/16
106 65025 65026 172.23.0.0/16 172.37.0.0/16
107 65027 65028 172.24.0.0/16 172.38.0.0/16

In OS10 version earlier to 10.5.4.0, SFS uses the public IP addresses in domains 101 to 107 for the VTEP subnets by default. You
may edit the default fabric settings to change the IP addresses from public to private from Fabric Settings page.
CAUTION: When you edit the fabric settings, all switches in that domain reboot simultaneously to apply the
changes.
From 10.5.4.0 release onwards, SFS uses private subnets for VTEP IP addresses for new fabric deployments as follows:

Table 14. Domain ID to ASN mapping—OS10.5.4.0 and later

Domain ID Leaf ASN Spine ASN Fabric BGP subnet Fabric VTEP subnet
and mask and mask
100 65011 65012 172.16.0.0/16 172.30.0.0/16
101 65015 65016 172.18.0.0/16 172.31.101.0/24
102 65017 65018 172.19.0.0/16 172.31.102.0/24
103 65019 65020 172.20.0.0/16 172.31.103.0/24
104 65021 65022 172.21.0.0/16 172.31.104.0/24
105 65023 65024 172.22.0.0/16 172.31.105.0/24
106 65025 65026 172.23.0.0/16 172.31.106.0/24
107 65027 65028 172.24.0.0/16 172.31.107.0/24

For the existing fabric deployment, even when you upgrade from the older OS10 version to OS10.5.4.0 or later, SFS still retains
the public VTEP IP addresses. You must edit the VTEP IP addresses to the private range from Fabric Settings page.

90 SFS concepts
MSTP Support on L3 personality
The default spanning tree mode in SFS is RPVST+. On PowerSwitches in SFS mode, RPVST+ is enabled globally and
automatically configured.
In L3 personality, STP is enabled on:
● SFS cluster control VLAN (VLAN 4000). The spine switches are configured to take over the STP root role.
● All user created VLANs.
In L3 personality, STP is disabled on:
● All inter leaf-spine VLANs and leaf-leaf VLANs (4001-4091).
● All server facing ports.
NOTE: Do not modify STP settings on switches in SFS L3 personality.

If you must interoperate switches that are controlled by SFS to external switches which are running RSTP or MSTP, SFS has an
API to change the global fabric STP mode to MSTP. SFS creates 2 reserved MSTIs:
● MST with instance id 63: Cluster control VLAN 4000 is part of this instance. Spine switches are configured to take over as
STP root for this MSTI.
● MST with instance id 62: All the SFS reserved VLANs (4001-4091) are part of this configuration. On this MSTI, spanning tree
is disabled.
● All user created VLANs are part of CST (default MST instance), which interoperates with RSTP. STP is enabled for this
MSTI.
You can change the mode to MSTP once the fabric is created. When you change the mode, the whole fabric goes through a
reboot cycle and the new mode is set to MSTP.

NOTE: Changing the STP mode impact the traffic flow in the cluster.

When the mode is changed, MSTI is created and VLANS are assigned to the MSTI. The CST is configured with STP priority such
that SFS-controlled switches have lower priority to become a root bridge.
There is no change on existing STP behavior for SFS-controlled entities because of this change. All other STP behaviors such as
disabling STP on server facing ports still holds good.
You can enable MSTP or revert to RPVST using the Edit Default Fabric Settings option available in the SFS UI. Select STP
Mode as MST or Rapid PVST For more information, see Edit Default Fabric Settings.
NOTE: The default spanning-tree priority value that is configured on a SFS-enabled switch for VLANs or instance 0 is:
default priority for RPVST is 32769 and instance 0 priority for MSTP is 61440.

Networks
This section describes the following type of networks that you can create and associate these networks with the entities
present in a fabric:
● General purpose networks
● VXLAN networks
● L3 VLAN networks
● L3 Routed networks
● Multirack L3 VLAN networks

NOTE: Before 10.5.3.0 release, you cannot configure Multirack L3 VLAN networks using SFS UI. This network type is
applicable for NSX-T deployment, and you can configure this network using OMNI UI. For more information about this
network type, see Configure multirack L3 VLAN networks.

You can create these networks using SFS UI and then associated these networks with servers profiles, uplinks, or interfaces for
traffic flow. For more information, see Create a network.
General purpose networks
General purpose networks can be categorized as L2 VLAN networks in SFS L2 personality and L2 VXLAN networks in SFS L3
personality. In L3 personality, when you create a general purpose network, SFS automatically creates a virtual network (VXLAN)
corresponding to a VLAN network. This virtual network has one-to-one mapping with the network, for each VLAN, there is a

SFS concepts 91
virtual network with VNI same as the VLAN ID. If you delete a VLAN network, it automatically deletes the associated VXLAN
network.
For example, if you create a L2 general purpose network with VLAN ID 50, SFS creates a VXLAN network with VNI 50 and
associates to VLAN 50.
VXLAN networks
VXLAN network extends L2 connectivity over an underlay L3 connected network. Association of VXLAN network to interface
creates a binding and associates this interface to VXLAN bridge. L3 VXLAN network supports asymmetric-IRB. Create a virtual
network template and a network template and associate the virtual network template to the network template.
● The virtual network template defines the VNET-ID.
● Network template defines the VLAN ID.
L3 VXLAN network is a VXLAN type of network that contains a list of IP addresses and an anycast IP address. Optionally,
you can specify DHCP relay addresses. L3 VXLAN network can be configured over a leaf switch. L3 VXLAN network can be
attached to an uplink. Each VLTi uplink interface contains an IP address that is allocated from the list of IP addresses that are
configured on the L3 VXLAN network.
L3 VLAN network
L3 VLAN network is used for L3 VLAN underlay. Specify:
● VLAN ID
● A pair of IP addresses to be assigned to the VLT pair
● VRRP gateway IP address for VIP
L3 VLAN network contains a list of IP addresses and a gateway IP address. Optionally, you can specify DHCP relay addresses.
You can configure and attach a L3 VLAN network to an uplink. Each VLTi uplink interface contains an IP address that is
allocated from the list of IP addresses that are configured on the L3 VLAN network.

NOTE: You must assign a L3 VLAN only to the server interface profile configured in a single rack.

L3 Routed network
L3 routed network is used to assign IP address on a single interface. L3 routed network contains a list of IP addresses and a
gateway IP address. Optionally, you can specify the DHCP relay addresses. L3 routed network can be configured on a leaf or
a spine switch. L3 routed network can be attached to an uplink. Each uplink interface contains an IP address that is allocated
from the list of IP addresses that are configured on the L3 routed network. An LACP port channel cannot be the remote end for
these uplinks. If the gateway IP address is specified and VRRP is enabled, the switches configure this IP address as the gateway
IP address. Attach this network to any uplink that has a single interface. This network can only be attached to a single entity.
Multirack L3 VLAN network
Multirack L3 VLAN network is a template that contains rack-specific IP configurations for a L3 VLAN. You can configure IPv4
attributes for each rack in an NSX-T deployment consisting of multirack leaf and spine topology. A rack holds a pair of switches
that are configured with VLT. The rack identifier is the same as the fabric ID. A rack containing a single switch without VLT does
not support a multi rack L3 VLAN network.
In an NSX-T topology that is connected to an SFS cluster, you can create a multi rack L3 VLAN network. Specify the network
identifier, L3 VLAN ID, and list of rack-specific IPv4 configurations for this network. Attach the created network as untagged or
tagged on any uplink or server interface of the rack. After you attach the network, SFS automatically creates a L3 VLAN and
applies the rack-specific IP configurations to the respective rack within the topology. You can modify the IPv4 configuration for
a specific rack in the multi rack L3 VLAN network.
NOTE: For faster traffic convergence, Dell Technologies recommends setting the BGP keepalive timer to 1 second and
hold-time to 3 seconds on the NSX-T tier-0 node when the BFD configuration is disabled or not supported. If BFD is
enabled, the BFD timer setting is used. For more information, see BFD support information in SFS configuration notes.

Uplinks
This information explains the types of uplinks that you can create in a fabric.

L2 uplinks
You can configure L2 uplinks only on the leaf switches in a fabric. L2 uplinks are a set of user-selected ports that belong to
same VLT leaf switches in which the L2 network is applied. The L2 uplink from the leaf switches is either an LACP or a static

92 SFS concepts
port channel. SFS creates a VLT port channel for connected ports. If the ports are from a single device, the VLT port channel
is a single armed VLT port channel. In case these ports exist on multiple switches, a VLT or port channel is formed across these
ports. This port channel can be made as an access point for a VXLAN L2 network.
To avoid loops, SFS does not allow you to configure the same network on multiple uplinks.

L3 uplinks
You can configure L3 uplinks on a leaf or spine switch. SFS supports L3 routed or L3 VLAN uplinks.
With L3 routed uplinks, point-to-point links are required between the switches with L3 uplinks and the external switches. With
L3 VLAN, all uplinks are in a port channel, and an IP address is assigned to the VLAN containing the port channel. Point-to-point
IP networks and addresses must be planned for each physical link in the L3 uplink. Each leaf switch in the fabric needs an IP
address on the external Management VLAN, and an anycast gateway address on the same VLAN. The virtual router or anycast
gateway address is shared by all leaf switches in the fabric.

Routing profiles
SFS supports eBGP and static routing profiles. You can create and delete the routing profiles, but you cannot edit any routing
policy. To modify, you must delete the existing routing policy and create a policy with the relevant configuration.
Static route—A static route profile is a routing template that contains a network prefix and the next hop IP address. During
uplink configuration, you can associate this profile to the uplinks created on one or more switches in the fabric. When this profile
is created, a route with the specified prefix and next hop IP address is configured on the switch.
eBGP route—An eBGP peer routing profile is a routing template that contains BGP remote addresses and the remote AS
number. A remote address can be an interface address or a loopback address. During uplink creation, you can associate this
policy to uplink created on one or more switches in a fabric. When this policy is created, a BGP session is configured on the
switch.
NOTE: SFS does not support OSPF or routing protocols other than eBGP.

Uplink bonding options

Following are the uplink bonding types supported:
● LACP
● Static bonding
LACP
In LACP uplink bonding, SFS configures the LACP port channel for the uplink using the LACP PDUs received from the remote
device. Networks that are attached to the uplink are associated with the LACP port channel that is created.
Static bonding
In static bonding, SFS configures a static port channel for the uplink and the networks that are attached on the uplink are
associated with the port channel that is created.

SFS concepts 93
 A
Identify the MAC address of QLogic interface
for server onboarding
NOTE: This instruction is applicable only for Marvell QLogic NIC interfaces.

If the QLogic interfaces are connected to the SFS fabric, OMNI automation creates the server interface profile for QLogic
interfaces using vCenter data. Multiple server interface profiles are created due to a mismatch between the base MAC address
and the LLDP MAC address for the same QLogic interface.
To statically onboard QLogic interfaces to the fabric, you must first determine the LLDP MAC address and set the server
interface profile using the LLDP MAC address. A computing algorithm can be used to determine which MAC address is correct
from the list of MAC addresses displayed in the output of the show lldp neighbors command.
The LLDP MAC address for a dual port configuration can be computed as follows:
● First Port LLDP0 MAC address—Add 10 to the base MAC address in Hexadecimal.
● Second Port LLDP1 MAC address—Add 11 to the base MAC address in Hexadecimal.
To identify the correct MAC address, you must check the iDRAC to determine the base MAC address of the interface. For more
information, see iDRAC9 Support Information.
If the interface is added to the vCenter vDS, the output of the show lldp neighbors command includes both the base
MAC and LLDP MAC addresses. In this case, the LLDP MAC address must be determined for server onboarding.
The following is the example output of the show lldp neighbors command.

OS10# show lldp neighbors

ethernet1/1/10 Not Advertised 34:80:0d:8f:2b:b8 34:80:0d:8f:2b:b8
ethernet1/1/10 Host46 34:80:0d:8f:2b:a8 vmnic4

The base MAC address of the interface can be determined from iDRAC. The following is the image from the iDRAC. The image
displays the base MAC address of the interface.

Here, the base MAC address of the interface is 34:80:0d:8f:2b:a8 and the LLDP Mac address is 34:80:0d:8f:2b:b8.
You must use 34:80:0d:8f:2b:b8 to create the server interface profile for static onboarding. You can verify the MAC
address that must be used for server onboarding using the algorithm. Since, it is connected to the first port, the LLDP MAC can
be computed by adding 10 to the base MAC address (34:80:0d:8f:2b:a8 + 10).
If the interface is not added to the vCenter vDS, the show lldp neighbors command displays the LLDP MAC address and
not the base MAC address in the output.
The following is the example output of the show lldp neighbors command.

OS10# show lldp neighbors

ethernet1/1/1 Not Advertised 34:80:0d:8c:b1:03 34:80:0d:8c:b1:03

94 Identify the MAC address of QLogic interface for server onboarding

The following is the image from the iDRAC. The image displays the base MAC address of the interface.

Similarly, here the base MAC address of the interface is 34:80:0d:8c:b0:f2. The LLDP MAC address is
34:80:0d:8c:b1:03. You must use the 34:80:0d:8c:b1:03 to create the server interface profile for static onboarding.
You can verify the MAC address that must be used for server onboarding using the algorithm. Since it is connected to the
second port, the LLDP MAC address can be computed by adding 11 to the base MAC address (34:80:0d:8c:b0:f2 + 11).

Identify the MAC address of QLogic interface for server onboarding 95

 B
Supported OS10 CLI commands in
SmartFabric mode
In addition to Full Switch mode, some CLI commands are supported in SmartFabric mode. For more information about how to
use the individual CLI commands, see Dell SmartFabric OS10 User Guide.
The following is the list of the CLI commands that can be used in SmartFabric mode:
● aaa accounting
● aaa authentication login

NOTE: In SmartFabric mode, the local authentication method must be configured along with TACACS or RADIUS
authentication.

● aaa authentication login mfa rsa-securid [console-exempt]

● aaa re-authenticate enable
● banner login
● boot protect enable username password
● clear counters interface
● clear ip {access-list | arp | bgp | ospf | route}
● clear ipv6 {access-list | ospf | route}
● clear lacp counters
● clear lldp {counters | dcbx | table}
● clear logging
● clear mac address-table dynamic {all | address | vlan vland-id | interface}
● clear qos statistics [interface | type]
● clear route-map pbr-statistics
● clear spanning-tree {counters | detected-protocol}
● clock set
● clock timezone
● configure terminal
● crypto ca-cert delete
● crypto ca-cert install
● crypto cdp add
● crypto cdp delete
● crypto cert generate request cert-file
● crypto crl delete
● crypto crl install
● debug mfa
● end
● exec-timeout
● exit
● fc alias
● fc zone
● fc zoneset
● host-description
● hostname
● interface mgmt
● ip name-server
● ip ssh server challenge-response-authentication

96 Supported OS10 CLI commands in SmartFabric mode

● ip ssh server cipher
● ip ssh server enable
● ip ssh server hostbased-authentication
● ip ssh server kex
● ip ssh server mac
● ip ssh server password-authentication
● ip ssh server port
● ip ssh server pubkey-authentication
● ip telnet server enable
● login-statistics enable
● logging audit enable
● logging console
● logging enable
● logging log-file
● logging monitor
● logging server
● management route
● member (alias)
● member (zone)
● member (zoneset)
● mfa rsa-server host
● ntp authenticate
● ntp authentication-key
● ntp master
● ntp server
● ntp trusted-key
● password-attributes
● password-attributes max-retry lockout-period
● radius-server host
● radius-server host tls
● radius-server retransmit
● radius-server timeout
● server url
● show alarms [details | history [summary] | index # | severity [critical | major | minor |
warning]]
● show alias
● show boot [detail]
● show candidate-configuration
● show clock
● show clock timezone
● show configuration
● show copy-file status
● show diag
● show fc alias
● show image [status]
● show interface
● show inventory [media]
● show lacp {counter [interface port-channel ID] | interface ethernet chassis/slot/
port[:subport] | neighbor [interface port-channel ID] | port-channel [interface port-
channel ID] | system-identifier}
● show lldp {errors | interface ethernet chassis/slot/port[:subport] | med | neighbors
[detail | interface ethernet chassis/slot/port[:subport]] | timers}
● show logging
● show mfa

Supported OS10 CLI commands in SmartFabric mode 97

● show ntp associations
● show ntp status
● show port-channel summary
● show port-group Chassis/Slot/PortGroup
● show running-configuration aaa
● show running-configuration mfa-rsa-server
● show running-configuration support-assist
● show startup-configuration
● show support-assist eula
● show support-assist status
● show system [brief | node-id id fanout-configured]
● show terminal monitor
● show trace
● show users
● show version
● show vlan
● show vlt mac-inconsistency
● snmp-server community
● snmp-server contact
● snmp-server enable traps
● snmp-server engineID
● snmp-server group
● snmp-server host
● snmp-server location
● snmp-server user
● snmp-server view
● source-interface
● street-address
● support-assist
● support-assist-activity
● system
● system identifier
● system-cli disable
● system-user linuxadmin disable
● system-user linuxadmin password
● tacacs-server host
● tacacs-server timeout
● territory
● tlv-select interface {ethernet chassis/slot/port[:subport] | traffic [ interface ethernet
chassis/slot/port[:subport]]}
● username password role
● username sshkey
● username sshkey filename
● userrole inherit
● vfabric
● write memory
● zoneset activate

98 Supported OS10 CLI commands in SmartFabric mode