close

keyboard_arrow_left

Table of Contents Expand all

list Table of Contents

English

Multihoming an Ethernet-Connected End System Design and Implementation

date_range 20-Dec-23

For an overview of multihoming an Ethernet-connected end system in this reference design, see the Multihoming Support for Ethernet-Connected End Systems section in Data Center Fabric Blueprint Architecture Components.

Figure 1 illustrates the multihomed Ethernet-connected end system in this procedure:

Figure 1: Ethernet-Connected Multihoming Example Overview

Configuring a Multihomed Ethernet-Connected End System using EVPN Multihoming with VLAN Trunking

EVPN multihoming is used in this building block to connect an Ethernet-connected end system into the overlay network. EVPN multihoming works by treating two or more physical multihomed links as a single Ethernet segment identified by an EVPN Ethernet Segment ID (ESI). The set of physical links belonging to the same Ethernet segment are treated as one aggregated Ethernet interface. The member links—much like member links in a traditional aggregated Ethernet interface—provide redundant paths to and from the end system while also ensuring overlay network traffic is load-balanced across the multiple paths.

LACP with the fast timer mode is used to improve fault detection and disablement of impaired members of an Ethernet segment. MicroBFD may also be used to further improve fault isolation but may not scale to support all end-system facing ports. Furthermore, support for microBFD must exist at the end system.

The reference design tested an Ethernet-connected server was connected to a single leaf or multihomed to 2 or 3 leaf devices to verify that traffic can be properly handled in multihomed setups with more than 2 leaf devices; in practice, an Ethernet-connected server can be multihomed to a large number of leaf devices.

To configure a multihomed Ethernet-connected server:

(Aggregated Ethernet interfaces only) Create the aggregated Ethernet interfaces to connect each leaf device to the server. Enable LACP with a fast period interval for each aggregated Ethernet interface.
Leaf 10:
```
set interfaces et-0/0/13 ether-options 802.3ad ae11
set interfaces et-0/0/14 ether-options 802.3ad ae11
set interfaces ae11 aggregated-ether-options lacp active
set interfaces ae11 aggregated-ether-options lacp periodic fast
```
Leaf 11:
```
set interfaces et-0/0/13 ether-options 802.3ad ae11
set interfaces et-0/0/14 ether-options 802.3ad ae11
set interfaces ae11 aggregated-ether-options lacp active
set interfaces ae11 aggregated-ether-options lacp periodic fast
```
Leaf 12:
```
set interfaces et-0/0/13 ether-options 802.3ad ae11
set interfaces et-0/0/14 ether-options 802.3ad ae11
set interfaces ae11 aggregated-ether-options lacp active
set interfaces ae11 aggregated-ether-options lacp periodic fast
```
Note:
The three leaf devices in this step use the same aggregated Ethernet interface name—ae11—and member link interfaces—et-0/0/13 and et-0/0/14— to organize and simplify network administration.

Avoid using different AE names at each VTEP for the same ESI as this will require configuring the LACP admin-key so that the end system can identify the multihomed links as part of the same LAG.

Configure each interface into a trunk interface. Assign VLANs to each trunk interface.

Note:

If you are connecting your end system to the leaf device with a single link, replace the interface name—for example, ae11—with a physical interface name—for example, et-0/0/13—for the remainder of this procedure.

Leaf 10:

content_copy zoom_out_map
set interfaces ae11 unit 0 family ethernet-switching interface-mode trunk
set interfaces ae11 unit 0 family ethernet-switching vlan members 100
set interfaces ae11 unit 0 family ethernet-switching vlan members 200
set interfaces ae11 unit 0 family ethernet-switching vlan members 300
set interfaces ae11 unit 0 family ethernet-switching vlan members 400

Leaf 11:

content_copy zoom_out_map
set interfaces ae11 unit 0 family ethernet-switching interface-mode trunk
set interfaces ae11 unit 0 family ethernet-switching vlan members 100
set interfaces ae11 unit 0 family ethernet-switching vlan members 200
set interfaces ae11 unit 0 family ethernet-switching vlan members 300
set interfaces ae11 unit 0 family ethernet-switching vlan members 400

Leaf 12:

content_copy zoom_out_map
set interfaces ae11 unit 0 family ethernet-switching interface-mode trunk
set interfaces ae11 unit 0 family ethernet-switching vlan members 100
set interfaces ae11 unit 0 family ethernet-switching vlan members 200
set interfaces ae11 unit 0 family ethernet-switching vlan members 300
set interfaces ae11 unit 0 family ethernet-switching vlan members 400

Configure the multihomed links with an ESI.
Assign each multihomed interface into the ethernet segment—which is identified using the Ethernet Segment Identifier (ESI)—that is hosting the Ethernet-connected server. Ensure traffic is passed over all multihomed links by configuring each link as all-active.

The ESI values must match on all multihomed interfaces.

Leaf 10 :
```
set interfaces ae11 esi 00:00:00:00:00:01:00:00:00:01
set interfaces ae11 esi all-active
```
Leaf 11:
```
set interfaces ae11 esi 00:00:00:00:00:01:00:00:00:01
set interfaces ae11 esi all-active
```
Leaf 12:
```
set interfaces ae11 esi 00:00:00:00:00:01:00:00:00:01
set interfaces ae11 esi all-active
```

Enable LACP and configure a system identifier.

The LACP system identifier must match on all multihomed interfaces.

Leaf 10:

content_copy zoom_out_map
set interfaces ae11 aggregated-ether-options lacp system-id 00:01:00:00:00:01

Leaf 11:

content_copy zoom_out_map
set interfaces ae11 aggregated-ether-options lacp system-id 00:01:00:00:00:01

Leaf 12:

content_copy zoom_out_map
set interfaces ae11 aggregated-ether-options lacp system-id 00:01:00:00:00:01

After committing the configuration, verify that the links on each leaf switch are in the Up state

Example:

content_copy zoom_out_map
user@leaf10# run show interfaces terse ae11 
Interface               Admin Link Proto    Local                 Remote
ae11                    up    up
ae11.0                  up    up   eth-switch

Verify that LACP is operational on the multihomed links.

content_copy zoom_out_map
user@leaf10# run show lacp interfaces ae11 
Aggregated interface: ae11
    LACP state:       Role   Exp   Def  Dist  Col  Syn  Aggr  Timeout  Activity
      et-0/0/13      Actor    No    No   Yes  Yes  Yes   Yes     Fast    Active
      et-0/0/13    Partner    No    No   Yes  Yes  Yes   Yes     Fast    Active
      et-0/0/14      Actor    No    No   Yes  Yes  Yes   Yes     Fast    Active
      et-0/0/14    Partner    No    No   Yes  Yes  Yes   Yes     Fast    Active
    LACP protocol:        Receive State  Transmit State          Mux State 
      et-0/0/13                 Current   Fast periodic Collecting distributing
      et-0/0/14                 Current   Fast periodic Collecting distributing

Enabling Storm Control

Storm control can be enabled as part of this building block. Storm control is used to prevent BUM traffic storms by monitoring BUM traffic levels and taking a specified action to limit BUM traffic forwarding when a specified traffic level—called the storm control level—is exceeded. See Understanding Storm Control for additional information on the feature.

In this reference design, storm control is enabled on server-facing aggregated Ethernet interfaces to rate limit broadcast, unknown unicast, and multicast (BUM) traffic. If the amount of BUM traffic exceeds 1% of the available bandwidth on the aggregated Ethernet interface, storm control drops BUM traffic to prevent broadcast storms.

To enable storm control:

Create the storm control profile that will be used to enable the feature. The interfaces that are configured using the storm control profile are specified in this step.
Leaf Device:
```
set interfaces ae11 unit 0 family ethernet-switching storm-control STORM-CONTROL
```
Set the storm control configuration within the profile.
In this reference design, storm control is configured to strategically drop BUM traffic when the amount of BUM traffic exceeds 1% of all available interface bandwidth.
```
set forwarding-options storm-control-profiles STORM-CONTROL all bandwidth-percentage 1
```
Note:
Dropping BUM traffic is the only supported storm control action in the Cloud Data Center architecture.

Note:
The storm control settings in this version of the reference design drop multicast traffic that exceeds the configured storm control threshold. If your network supports multicast-based applications, consider using a storm control configuration—such as the no-multicast option in the storm-control-profiles statement—that is not represented in this reference design.

Storm control settings in support of multicast-based applications will be included in a future version of this reference design.
To verify storm control activity, filter system log messages related to storm control by entering the show log messages | match storm command.
```
user@leaf10> show log messages | match storm
Sep 27 11:35:34 leaf1-qfx5100 l2ald[1923]: L2ALD_ST_CTL_IN_EFFECT: ae11.0: storm control in effect on the port
```

Multihoming a Ethernet-Connected End System—Release History

Table 1 provides a history of all of the features in this section and their support within this reference design.

Table 1: Release History
Release	Description
19.1R2	QFX10002-60C and QFX5120-32C switches running Junos OS Release 19.1R2 and later releases in the same release train support all features documented in this section.
18.4R2	QFX5120-48Y switches running Junos OS Release 18.4R2 and later releases in the same release train support all features documented in this section.
18.1R3-S3	QFX5110 switches running Junos OS Release 18.1R3-S3 and later releases in the same release train support all features documented in this section.
17.3R3-S1	All devices in the reference design that support Junos OS Release 17.3R3-S1 and later releases in the same release train also support all features documented in this section.

arrow_backward PREVIOUS Centrally-Routed Bridging Overlay Design and Implementation

NEXT arrow_forward Edge-Routed Bridging Overlay Design and Implementation

Data Center EVPN-VXLAN Fabric Architecture Guide

ON THIS PAGE

Multihoming an Ethernet-Connected End System Design and Implementation

Configuring a Multihomed Ethernet-Connected End System using EVPN Multihoming with VLAN Trunking

Enabling Storm Control

Multihoming a Ethernet-Connected End System—Release History

Stay in touch

Helped me install or use the product
Accelerated my deployment

Discuss the product with a co-worker
Make a purchase inquiry

Data Center EVPN-VXLAN Fabric Architecture Guide

ON THIS PAGE

Multihoming an Ethernet-Connected End System Design and Implementation

Configuring a Multihomed Ethernet-Connected End System using EVPN Multihoming with VLAN Trunking

Enabling Storm Control

Multihoming a Ethernet-Connected End System—Release History

Related Documentation

Stay in touch