CloudEngine 8800&7800&6800&5800 Series

Switches
V200R001C00

Configuration Guide - Interface

Management

Issue 05
Date 2017-05-08

HUAWEI TECHNOLOGIES CO., LTD.

Copyright © Huawei Technologies Co., Ltd. 2017. All rights reserved.
No part of this document may be reproduced or transmitted in any form or by any means without prior written
consent of Huawei Technologies Co., Ltd.

Trademarks and Permissions

and other Huawei trademarks are trademarks of Huawei Technologies Co., Ltd.
All other trademarks and trade names mentioned in this document are the property of their respective
holders.

Notice
The purchased products, services and features are stipulated by the contract made between Huawei and the
customer. All or part of the products, services and features described in this document may not be within the
purchase scope or the usage scope. Unless otherwise specified in the contract, all statements, information,
and recommendations in this document are provided "AS IS" without warranties, guarantees or
representations of any kind, either express or implied.

The information in this document is subject to change without notice. Every effort has been made in the
preparation of this document to ensure accuracy of the contents, but all statements, information, and
recommendations in this document do not constitute a warranty of any kind, express or implied.

Huawei Technologies Co., Ltd.

Address: Huawei Industrial Base
Bantian, Longgang
Shenzhen 518129
People's Republic of China

Website: http://e.huawei.com

Issue 05 (2017-05-08) Huawei Proprietary and Confidential i

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management About This Document

About This Document

Intended Audience
This document provides the basic concepts and configuration procedures of the interfaces
supported by the device.
This document is intended for:
l Data configuration engineers
l Commissioning engineers
l Network monitoring engineers
l System maintenance engineers

Symbol Conventions
The symbols that may be found in this document are defined as follows.

Symbol Description

Indicates an imminently hazardous situation

which, if not avoided, will result in death or
serious injury.

Indicates a potentially hazardous situation

which, if not avoided, could result in death
or serious injury.

Indicates a potentially hazardous situation

which, if not avoided, may result in minor
or moderate injury.

Indicates a potentially hazardous situation

which, if not avoided, could result in
equipment damage, data loss, performance
deterioration, or unanticipated results.
NOTICE is used to address practices not
related to personal injury.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential ii

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management About This Document

Symbol Description

NOTE Calls attention to important information,

best practices and tips.
NOTE is used to address information not
related to personal injury, equipment
damage, and environment deterioration.

Command Conventions
The command conventions that may be found in this document are defined as follows.

Convention Description

Boldface The keywords of a command line are in boldface.

Italic Command arguments are in italics.

[] Items (keywords or arguments) in brackets [ ] are optional.

{ x | y | ... } Optional items are grouped in braces and separated by

vertical bars. One item is selected.

[ x | y | ... ] Optional items are grouped in brackets and separated by

vertical bars. One item is selected or no item is selected.

{ x | y | ... }* Optional items are grouped in braces and separated by

vertical bars. A minimum of one item or a maximum of all
items can be selected.

[ x | y | ... ]* Optional items are grouped in brackets and separated by

vertical bars. Several items or no item can be selected.

&<1-n> The parameter before the & sign can be repeated 1 to n

times.

# A line starting with the # sign is comments.

Interface Numbering Conventions

Interface numbers used in this manual are examples. In device configuration, use the existing
interface numbers on devices.

Security Conventions
l Password setting

Issue 05 (2017-05-08) Huawei Proprietary and Confidential iii

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management About This Document

– When configuring a password, the cipher text is recommended. To ensure device

security, change the password periodically.
– When you configure a password in plain text that starts and ends with %^%#......%^
%# (the password can be decrypted by the device), the password is displayed in the
same manner as the configured one in the configuration file. Do not use this setting.
– When you configure a password in cipher text, different features cannot use the
same cipher-text password. For example, the cipher-text password set for the AAA
feature cannot be used for other features.
– After the system software is downgraded and the switch restarts with the
configuration of the higher version, AAA, VTY, serial interface login, and SNMP
user passwords become invalid. As a result, users fail to log in to the switch using
the passwords and the switch is disconnected from the network management
system.
To address this problem, take the following measures:
i. If no password is configured for the console port, log in to the device through
the console port, and reconfigure AAA and password for users such as VTY
and SNMP users. To improve security, the console port password is
recommended.
ii. If a password is configured for login through the console port, the password
becomes invalid after the downgrade and you cannot log in to the switch
through the console port. Perform the following steps:
1) Connect to the console port.
2) Power recycle the device. During the startup, enter Ctrl+B according to
the prompt to enter the BIOS menu. The default password is
[email protected].
3) Select 7.Modify console password to delete and change the console port
password.
4) Restart the device, log in to the device through the console port, and
reconfigure the password for AAA, VTY, or SNMP user.
l Encryption algorithm
Currently, the device uses the following encryption algorithms: DES, 3DES, AES, DSA,
RSA, DH, ECDH, HMAC, SHA1, SHA2, PBKDF2, scrypt, and MD5. The encryption
algorithm depends on the applicable scenario. Use the recommended encryption
algorithm; otherwise, security defense requirements may be not met.
– For the symmetrical encryption algorithm, use AES with the key of 256 bits or
more.
– When you need to use an asymmetric cryptography, RSA (2048-bit or longer key)
is recommended. In addition, use different key pairs for encryption and signature.
– For the digital signature, RSA (2048-bit or longer key) or DSA (2048-bit or longer
key) is recommended.
– For key negotiation, DH (2048-bit or longer key) or ECDH (256-bit or longer key)
is recommended.
– For the hash algorithm, use SHA with the key of 256 bits or more.
– For the HMAC algorithm, use HMAC-SHA2.
– DES, 3DES, RSA and AES are reversible encryption algorithm. If protocols are
used for interconnection, the locally stored password must be reversible.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential iv

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management About This Document

– SHA1, SHA2, and MD5 are irreversible encryption algorithm. When configuring a
password for local administrator, it is recommended that you use the SHA2
irreversible encryption algorithm.
– To prevent brute force cracking of the user password, the iteration algorithm is
added to the password on the basis of salts. The iteration algorithm uses PBKDF2
or scrypt key export algorithm.
– The ECB mode has a poor capability of defending against plaintext playback
attacks, so ECB is not recommended for password encryption.
– In SSH2.0, the symmetric cryptography using the CBC mode may undergo the
plaintext-recovery attack to cause a data leak. Therefore, the CBC mode is not
recommended for SSH2.0.
l Personal data
Some personal data may be obtained or used during operation or fault location of your
purchased products, services, features, so you have an obligation to make privacy
policies and take measures according to the applicable law of the country to protect
personal data.
l The terms mirrored port, port mirroring, traffic mirroring, and mirroring in this manual
are mentioned only to describe the product's function of communication error or failure
detection, and do not involve collection or processing of any personal information or
communication data of users.

Declaration
This manual is only a reference for you to configure your devices. The contents in the manual,
such as command line syntax, and command outputs, are based on the device conditions in
the lab. The manual provides instructions for general scenarios, but do not cover all usage
scenarios of all product models. The contents in the manual may be different from your actual
device situations due to the differences in software versions, models, and configuration files.
The manual will not list every possible difference. You should configure your devices
according to actual situations.
The specifications provided in this manual are tested in lab environment (for example, the
tested device has been installed with a certain type of boards or only one protocol is run on
the device). Results may differ from the listed specifications when you attempt to obtain the
maximum values with multiple functions enabled on the device.

Mappings between Product Software Versions and NMS

Versions
The mappings between product software versions and NMS versions are as follows.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential v

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management About This Document

CE8800&7800&6800&5800 NMS
series switches Product
Software Version

V200R001C00 eSight V300R006C00/iManager U2000 V200R016C50

(iManager U2000 V200R017C50 for CE6870EI and
CE7855EI)

Mappings between Product Software Versions and

Controller Versions
The mappings between product software versions and Controller versions are as follows.

CE8800&7800&6800&5800 Controller
series switches Product
Software Version

V200R001C00 Agile Controller - DCN V200R001C00

Change History
Changes between document issues are cumulative. Therefore, the latest document version
contains all updates made to previous versions.

Changes in Issue 06 (2017-09-12) for Product Version V200R001C00

This version has the following updates:
The documentation is updated according to product feature updates.

Changes in Issue 05 (2017-05-08) for Product Version V200R001C00

This version has the following updates:
The documentation is updated according to product feature updates.

Changes in Issue 04 (2016-12-31) for Product Version V200R001C00

This version has the following updates:
The documentation is updated according to product feature updates.

Changes in Issue 03 (2016-09-10) for Product Version V200R001C00

This version has the following updates:
The following information is modified:

Issue 05 (2017-05-08) Huawei Proprietary and Confidential vi

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management About This Document

l 2.2 Configuration Notes

Changes in Issue 02 (2016-07-20) for Product Version V200R001C00

This version has the following updates:
The following information is added:
l 2.7.7 Configuring FEC
The following information is modified:
l 2.2 Configuration Notes
l 2.6.1 Switching Interfaces to Layer 3 Mode
l 2.7.3 Configuring Unidirectional Single-Fiber Communication

Changes in Issue 01 (2016-04-15) for Product Version V200R001C00

Initial commercial release.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential vii

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management Contents

Contents

About This Document.....................................................................................................................ii

1 Basic Configuration for Interfaces............................................................................................. 1
1.1 Interface Overview......................................................................................................................................................... 1
1.2 Configuring Basic Interface Parameters.........................................................................................................................8
1.2.1 Entering the Interface View......................................................................................................................................... 8
1.2.2 Configuring an Interface Description.......................................................................................................................... 8
1.2.3 Configuring the Traffic Statistics Collection Interval................................................................................................. 9
1.2.4 Configuring the Delay Before the Protocol Status of an Interface Can Go Up.........................................................10
1.2.5 Enabling or Disabling an Interface............................................................................................................................ 11
1.2.6 Checking the Configuration of Basic Interface Parameters...................................................................................... 12
1.3 Maintaining Interfaces.................................................................................................................................................. 13
1.3.1 Checking Interface Diagnostic Information.............................................................................................................. 13
1.3.2 Clearing Interface Traffic Statistics........................................................................................................................... 13

2 Ethernet Interface Configuration............................................................................................. 15

2.1 Ethernet Interface Overview.........................................................................................................................................15
2.2 Configuration Notes..................................................................................................................................................... 16
2.3 Default Configuration...................................................................................................................................................25
2.4 Configuring MEth Management Interface Attributes...................................................................................................30
2.5 Configuring a Port Group............................................................................................................................................. 32
2.6 Configuring Common Optical/Electrical Interface Attributes..................................................................................... 34
2.6.1 Switching Interfaces to Layer 3 Mode...................................................................................................................... 34
2.6.2 Configuring Auto-Negotiation.................................................................................................................................. 36
2.6.3 Configuring the Interface Rate.................................................................................................................................. 37
2.6.4 Configuring Flow Control......................................................................................................................................... 40
2.6.5 Configuring the Delay in Reporting Physical Status Changes.................................................................................. 42
2.6.6 Configuring the Delay in Reporting an Interface Up Event...................................................................................... 43
2.6.7 Configuring an Interface to Transit to the Error-Down State When a Link Flapping Occurs...................................44
2.6.8 Configuring an Interface to Transit to the CRC Error-Down State When the Number of Received Error Packets
Exceeds the Threshold........................................................................................................................................................46
2.6.9 Configuring the IFG.................................................................................................................................................. 48
2.6.10 Configuring Traffic Statistics to Include the IFG and Preamble Statistics..............................................................48
2.6.11 Collecting Protocol Packet Statistics Based on Protocol Types.............................................................................. 49
2.6.12 Setting the Jumbo Frame Length Allowed on an Interface..................................................................................... 51

Issue 05 (2017-05-08) Huawei Proprietary and Confidential viii

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management Contents

2.6.13 Configuring the Outbound/Inbound Bandwidth Usage Alarm Threshold.............................................................. 52

2.7 Configuring Attributes for an Optical Interface........................................................................................................... 52
2.7.1 Configuring Interface Split........................................................................................................................................ 53
2.7.2 Configuring Interface Aggregation........................................................................................................................... 57
2.7.3 Configuring Unidirectional Single-Fiber Communication........................................................................................58
2.7.4 Configuring the Training Function............................................................................................................................ 60
2.7.5 Configuring an Interface to Transit to Error-Down State When the Receive Optical Power Is Low....................... 62
2.7.6 Pre-configuring the Transmission Medium Type on an Optical Interface................................................................ 64
2.7.7 Configuring FEC....................................................................................................................................................... 66
2.7.8 Configuring the Medium Attribute of a 25GE Interface........................................................................................... 68
2.8 Configuring Attributes for an Electrical Interface........................................................................................................69
2.8.1 Setting the Duplex Mode........................................................................................................................................... 69
2.9 Configuring Port Isolation............................................................................................................................................ 70
2.10 Maintenance................................................................................................................................................................73
2.10.1 Configuring VCT.....................................................................................................................................................73
2.10.2 Configuring Internal Loopback Detection...............................................................................................................74
2.10.3 Clearing Interface Statistics.....................................................................................................................................75
2.11 Configuration Examples............................................................................................................................................. 76
2.11.1 Example for Configuring Interface Isolation...........................................................................................................76

3 Logical Interface Configuration................................................................................................79

3.1 Logical Interface Overview.......................................................................................................................................... 79
3.2 Configuration Notes..................................................................................................................................................... 82
3.3 Configuring a Sub-interface......................................................................................................................................... 86
3.3.1 Configuring an Ethernet Sub-interface......................................................................................................................87
3.3.2 Configuring an Eth-Trunk Sub-interface...................................................................................................................88
3.4 Configuring a Loopback Interface................................................................................................................................89
3.5 Configuring a NULL Interface..................................................................................................................................... 90
3.6 Configuration Examples............................................................................................................................................... 91
3.6.1 Example for Configuring Ethernet Sub-interfaces to Implement Inter-VLAN Communication.............................. 91

Issue 05 (2017-05-08) Huawei Proprietary and Confidential ix

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 1 Basic Configuration for Interfaces

1 Basic Configuration for Interfaces

About This Chapter

1.1 Interface Overview

1.2 Configuring Basic Interface Parameters
1.3 Maintaining Interfaces

1.1 Interface Overview

Interface Type
Interfaces of a device are used to exchange data and interact with other network devices.
Interfaces are classified into management interfaces, physical service interfaces, and logical
interfaces.

l Management interfaces
Management interfaces are used to log in to devices. Users can use management
interfaces to configure and manage devices. Management interfaces do not transmit
service data. For the configuration of management interfaces, see the CloudEngine
8800&7800&6800&5800 Series Switches Configuration Guide - Basic Configurations.
Table 1-1 describes the management interfaces.

Table 1-1 Description of management interfaces

Interfac Description Application

e

Console A data connection equipment The console interface is connected to

interface (DCE) interface that complies the COM serial interface of a
with the EIA/TIA-232 standard. configuration terminal to set up an on-
site configuration environment.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 1

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 1 Basic Configuration for Interfaces

Interfac Description Application

e

MEth Complies with the The MEth interface can be connected

interface 10/100/1000BASE-TX to a network interface of a
standard. configuration terminal or network
management workstation to set up an
on-site or remote configuration
environment.

Mini Complies with the USB 2.0 The Mini USB interface can be
USB standard. connected to an operation terminal for
interface onsite configuration.

All the numbers of different types of management interfaces are displayed as MEth0/0/0
on the command line interface (CLI) of a switch. Types of management interfaces
supported by different models of switches are as follows:
– Only CE8860EI, CE6850HI, CE6855-48T6Q-HI and CE6850U-HI switches
support the Mini USB interface.
– Only the CE8860EI supports two management interfaces which are two MEth
management interfaces, as shown in Figure 1-1. You can only activate one of the
two MEth management interfaces at a time. By default, interface 1 is used.
– Only CE6850HI, CE6855-48T6Q-HI, and CE6850U-HI switches provide two
combo interfaces, as shown in Figure 1-2. Each combo interface consists of one
optical interface and one electrical interface. Either the optical interface or the
electrical interface can work at a time.
– Only CE6850HI, CE6855-48T6Q-HI, and CE6850U-HI switches provide two
combo interfaces, as shown in Figure 1-2. Each combo interface consists of one
optical interface and one electrical interface. Either the optical interface or the
electrical interface can work at a time.

Figure 1-1 Two management interfaces

SLOT

1 2 3 4
1 2
CONSOLE

ETH

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 2

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 1 Basic Configuration for Interfaces

Figure 1-2 Two combo interfaces

1 2 3 4 56 7 8 9 10
1 2 1 2 ID STAT STAT STAT STAT

PWR1
FAN1
CONSOLE
SYS
MST

060A-F
FAN-
060A-F
FAN-
USB

ACT

FAN2

PWR2
ETH

11 12
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 1 2 3 4 5 6

SYS MST ID CE6850-48S6Q-HI 40GE 1 2 3 4

Breakout

13 13 14 14 15 15

15 15 14 14 13 13

l Physical service interfaces

Physical interfaces exist on interface cards and transmit service data.
Physical interfaces are sometimes called ports. This document uses the term interface.
Table 1-2 describes the physical interfaces that the device supports.

Table 1-2 Description of physical interfaces

Interface Description

GE electrical interface A GE electrical interface works at the data link layer,

provides a maximum of 1000 Mbit/s transmission rate,
processes Layer 2 protocol packets, and implements
Layer 2 forwarding.

10GE electrical interface A 10GE electrical interface works at the data link
layer, provides a maximum of 10000 Mbit/s
transmission rate, processes Layer 2 protocol packets,
and implements Layer 2 forwarding.

10GE optical interface A 10GE optical interface works at the data link layer,
provides a maximum of 10000 Mbit/s transmission
rate, processes Layer 2 protocol packets, and
implements Layer 2 forwarding.

25GE optical interface A 25GE optical interface works at the data link layer,
provides a maximum of 25000 Mbit/s transmission
rate, processes Layer 2 protocol packets, and
implements Layer 2 forwarding.
Only the CE8860EI supports 25GE interfaces.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 3

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 1 Basic Configuration for Interfaces

Interface Description

40GE optical interface A 40GE optical interface works at the data link layer,
provides a maximum of 40000 Mbit/s transmission
rate, processes Layer 2 protocol packets, and
implements Layer 2 forwarding.
A 40GE optical interface can be used as an
independent interface or be split into four or two 10GE
optical interfaces. For details about interface split, see
Configuring Interface Split.

100GE optical interface A 100GE optical interface works at the data link layer,
provides a maximum of 100 Gbit/s transmission rate,
processes Layer 2 protocol packets, and implements
Layer 2 forwarding.
A 100GE optical interface can be split into four 25GE
optical interfaces. Only the CE8860EI, CE6870EI
support 100GE interfaces.

NOTE

By default, an Ethernet interface on the device works in Layer 2 mode. To enable Layer 3
functions on the interface, run the undo portswitch command on the interface.
l Logical interfaces
Logical interfaces are manually configured and do not physically exist. They can be used
to exchange data and transmit service data.
Table 1-3 describes the logical interfaces that the device supports.

Table 1-3 Logical interfaces

Interface Description Configuration Reference

Eth-Trunk An Eth-Trunk is formed by binding For details about how to

multiple Ethernet interfaces to configure an Eth-Trunk, see Link
provide more bandwidth and higher Aggregation Configuration in the
transmission reliability. CloudEngine
8800&7800&6800&5800 Series
Switches Configuration Guide -
Ethernet Switching.

VLANIF A VLANIF interface has Layer 3 For details about how to

interface features and enables VLANs to configure a VLANIF interface,
communicate after being assigned an see VLAN Configuration in the
IP address. CloudEngine
8800&7800&6800&5800 Series
Switches Configuration Guide -
Ethernet Switching.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 4

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 1 Basic Configuration for Interfaces

Interface Description Configuration Reference

Layer 2 On VXLAN networks, service access For details about how to

sub- points are Layer 2 sub-interfaces. configure a Layer 2 sub-
interface You can configure different flow interface, see Configuring a
encapsulation modes on the Layer 2 Service Access Point in the
sub-interfaces, so that the sub- CloudEngine
interfaces can process different data 8800&7800&6800&5800 Series
packets. Switches Configuration Guide -
DCN and Server Management.

Layer 3 A Layer 3 sub-interface is a virtual -

sub- interface configured on a main
interface interface to implement inter-VLAN
communication.
NOTE
l Before configuring a Layer 3 sub-
interface on a Layer 2 physical
interface or Layer 2 Eth-Trunk, run
the undo portswitch command to
switch the interface to a Layer 3
interface.
l The MAC address of a sub-interface
is the same as that of the main
interface.
l A sub-interface can be assigned an IP
address.
l Unless otherwise stated, all sub-
interfaces mentioned in this chapter
are Layer 3 sub-interfaces.

Stack A stack interface is a logical interface For details about how to

interface used to implement the stacking configure a stack interface, see
function. The stacking function Stack Configuration in the
virtualizes multiple switches into a CloudEngine
logical switch to provide massive 8800&7800&6800&5800 Series
data forwarding and improve network Switches Configuration Guide -
reliability in data centers. Device Management.

Loopback A loopback interface is always Up -

interface and can be configured with a 32-bit
subnet mask.
There is one unique loopback
interface called the InLoopBack0
interface, which is automatically
created at startup. The InLoopback0
interface is a special and fixed
loopback interface. It uses the fixed
loopback address 127.0.0.1/8 to
receive data packets destined for the
host where the InLoopBack0
interface resides. The loopback
address of the InLoopBack0 interface
is not advertised.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 5

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 1 Basic Configuration for Interfaces

Interface Description Configuration Reference

NULL A NULL interface used in route -

interface filtering because any network packets
sent to this interface are discarded.

Tunnel A tunnel interface has Layer 3 For details about how to

interface features, transmits packets, and configure a tunnel interface, see
identifies and processes packets IPv6 Transition Technology
transmitted over a tunnel. Configuration in the
CloudEngine
8800&7800&6800&5800 Series
Switches Configuration Guide -
IP Service and GRE
Configuration Guide in the
CloudEngine
8800&7800&6800&5800 Series
Switches Configuration Guide -
VPN.

FCoE If Fibre Channel over Ethernet For details about how to

interface (FCoE) is configured, devices configure an FCoE interface, see
communicate and transmit traffic FC and FCoE Configuration in
through FCoE interfaces. Therefore, the CloudEngine
you need to configure FCoE logical 8800&7800&6800&5800 Series
interfaces on the devices. Switches Configuration Guide -
DCN and Server Management.

Fabric A Fabric interface is used as the For details about how to

interface logical interface of an SVF system to configure a Fabric interface, see
forward SVF data or protocol SVF Configuration in the
packets. CloudEngine
8800&7800&6800&5800 Series
Switches Configuration Guide -
Virtualization Configuration.

NVE An NVE interface is the logical For details about how to

interface interface to establish VXLAN tunnels configure an NVE interface, see
with other NVE devices. VXLAN Configuration in the
CloudEngine
8800&7800&6800&5800 Series
Switches Configuration Guide -
DCN and Server Management.

VBDIF A VBDIF interface is the virtual For details about how to

interface interface based on a BD to support configure a VBDIF interface, see
Layer 3 features and implement VXLAN Configuration in the
communication between different CloudEngine
BDs, between BD and non-BD 8800&7800&6800&5800 Series
networks, and between BDs and Switches Configuration Guide -
Layer 3 networks. DCN and Server Management.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 6

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 1 Basic Configuration for Interfaces

Interface Description Configuration Reference

FC An FC interface is used to connect to For details about how to

interface an FC switch, an FC SAN storage, or configure an FC interface, see
a server with FC interfaces. FC and FCoE Configuration in
NOTE the CloudEngine
Only 10GE Ethernet interfaces on the 8800&7800&6800&5800 Series
CE6850U-HI support FC interfaces, and Switches Configuration Guide -
supported in V100R005C10 and later DCN and Server Management.
versions.

Ethernet An Ethernet interface is used for For detailed configuration, see

communication between the VRP Open System Usage Guide in the
system and open system. CloudEngine
8800&7800&6800&5800 Series
Switches .

Management Interface Numbering Rules

The switch supports console, MEth, and Mini USB management interfaces.
The following table lists the numbers of the management interfaces.

Table 1-4 Management interface numbers

Interface Number

Console interface console 0

MEth interface MEth 0/0/0

Mini USB interface console 0

Physical Interface Numbering Rules

The numbering rules for physical interfaces are as follows:
Physical interfaces on the CloudEngine 8800&7800&6800&5800 Series switch are numbered
in the format of stack ID/subcard ID/interface sequence number.
l stack ID: specifies the stack ID.
l subcard ID: specifies the ID of a subcard on the switch. On the CE8860EI, the value of
subcard ID ranges from 1 to 4. Other switch models do not support subcards and the
value of subcard ID is 0.
l interface sequence number: specifies the sequence number of an interface on the switch.
The value starts from 1.

Figure 1-3 Physical interface numbering diagram

1 3 5 ...
...
2 4 6 ...

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 7

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 1 Basic Configuration for Interfaces

There are two rows of service interfaces on the switch. These interfaces are numbered from
top to bottom and left to right starting with 1. For example, the first interface in the upper left
part is numbered 1/0/1.

Converted interfaces are numbered using the dimension numbering rule. For example, a 40GE
interface is numbered 40GE x/y/n and is split into four 10GE interfaces, the converted 10GE
interfaces are numbered 40GE x/y/n:1, 40GE x/y/n:2, 40GE x/y/n:3, and 40GE x/y/n:4.

1.2 Configuring Basic Interface Parameters

1.2.1 Entering the Interface View

Context
To configure an interface, enter the interface view.

Procedure
Step 1 Run:
system-view

The system view is displayed.

Step 2 Run:
interface interface-type interface-number

The interface view is displayed.

interface-type interface-number specifies the type and number of an interface.

NOTE

If the specified interface does not exist, this command creates the interface and displays the interface
view.

Step 3 (Optional) Run:

commit

The configuration is committed.

You do not need to run this command if you only want to enter the interface view.

----End

1.2.2 Configuring an Interface Description

Context
To facilitate device management and maintenance, you can configure descriptions for
interfaces. An interface description can contain the device where the interface is located,
interface type, and remote device. For example: To-[DeviceB]10GE-1/0/1 indicates that an
interface of this device is connected to 10GE1/0/1 of device B.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 8

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 1 Basic Configuration for Interfaces

Procedure
Step 1 Run:
system-view

The system view is displayed.

Step 2 Run:
interface interface-type interface-number

The interface view is displayed.

Step 3 Run:
description description

The description is configured for the interface.

By default, an interface has no description.
The interface description is displayed from the first non-space character.
Step 4 Run:
commit

The configuration is committed.

----End

1.2.3 Configuring the Traffic Statistics Collection Interval

Context
By setting the traffic statistics collection interval, you can collect and analyze packet statistics.
According to traffic statistics, you can take measures to prevent network congestion and
service interruption.
l When congestion occurs, you can set the statistics collection interval on an interface to
300 seconds or less (30 seconds if congestion worsens). Then observe traffic distribution
on the interface within a short period of time. Take measures to data packets that cause
congestion to control the rate of the packets.
l When the network bandwidth is sufficient and services are running properly, set the
statistics collection interval on an interface to more than 300 seconds. If traffic
parameters on an interface are out of the specified range, change the statistics collection
interval to observe the traffic statistics in real time.
NOTE

l The interval set in the system view takes effect on all the interfaces that use the default interval.
l The interval set in the interface view takes effect only on this interface.
l The interval set in the interface view takes precedence over the interval set in the system view.

Procedure
l Configure the global traffic statistics collection intervals in the system view.
a. Run:
system-view

The system view is displayed.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 9

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 1 Basic Configuration for Interfaces

b. Run:
set flow-stat interval interval-time

The global traffic statistics collection interval is set.

By default, the global traffic statistics collection interval is 300s.
c. Run:
commit

The configuration is committed.

l Configure the traffic statistics collection interval on an interface.
a. Run:
system-view

The system view is displayed.

b. Run:
interface interface-type interface-number

The interface view is displayed.

c. Run:
set flow-stat interval interval-time

The traffic statistics collection interval is set on the interface.

By default, the traffic statistics collection interval on an interface is 300s.
d. Run:
commit

The configuration is committed.

----End

1.2.4 Configuring the Delay Before the Protocol Status of an

Interface Can Go Up
Context
On a network with primary and backup links, when the primary link becomes faulty, service
traffic is switched from the primary link to the backup link. After the primary link is restored,
service traffic is switched back to the primary link. When the interface protocol status of the
primary link is Up, Layer 3 protocols such as routing and VRRP protocols are used to send
protocol packets and re-negotiate with the peer end. If the link status of the local or peer
interface does not meet the forwarding requirement, the protocol packets are discarded. This
makes the Layer 3 protocol negotiation mechanism ineffective and results in extra device
costs.
For example, the local end uses a 10GE interface for the primary link and the interface
protocol status is Up; the peer end uses a VLANIF interface and the interface protocol status
is Up. However, the member physical interface of the VLANIF interface is still in STP
blocking status of the link layer. When the routing protocol packet sent by the local 10GE
interface reaches the peer end, the peer end cannot respond to the packet because the peer
physical interface is in STP blocking status. You can run the protocol up-delay-time
command to configure the delay before the protocol status of an interface can go Up. This
enables the protocol status of the interface to go Up after its link status is recovered, which
prevents the protocol packets from being discarded.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 10

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 1 Basic Configuration for Interfaces

Procedure
Step 1 Run:
system-view

The system view is displayed.

Step 2 Run:
interface interface-type interface-number

The interface view is displayed.

Step 3 (On an Ethernet interface), run:

undo portswitch

The interface is switched to Layer 3 mode.

By default, an Ethernet interface works in Layer 2 mode.

If an Ethernet interface already has Layer 2 configuration, this command will fail to be
executed on the interface. Before running this command on the interface, delete all the Layer
2 configuration of the interface.
NOTE

If many Ethernet interfaces need to be switched to Layer 3 mode, run the undo portswitch batch
interface-type { interface-number1 [ to interface-number2 ] } &<1-10> command in the system view to
switch these interfaces to Layer 3 mode in batches.

Step 4 Run:
protocol up-delay-time time-value

The delay before the protocol status of the interface can go Up is configured.

By default, the physical interface delay is 0s, the VLANIF interface delay is 1s.

Step 5 Run:
commit

The configuration is committed.

----End

1.2.5 Enabling or Disabling an Interface

Context
After modifying parameters of an interface, run the shutdown and undo shutdown
commands, or run the restart command to make the modification take effect.

When an interface is not connected to a cable or a fiber, disable the interface using the
shutdown command to prevent exceptions caused by interference.

NOTE

l Running the shutdown and undo shutdown commands is equivalent to running the restart
command. Running the shutdown command does not modify or delete interface configurations.
l A NULL interface is always Up and cannot be enabled or disabled by commands.
l A loopback interface is always Up and cannot be enabled or disabled by commands.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 11

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 1 Basic Configuration for Interfaces

Procedure
l Disable an interface.
a. Run:
system-view

The system view is displayed.

b. Run:
interface interface-type interface-number

The interface view is displayed.

c. Run:
shutdown

The interface is disabled.

By default, an interface is enabled.
d. Run:
commit

The configuration is committed.

l Enable an interface.
a. Run:
system-view

The system view is displayed.

b. Run:
interface interface-type interface-number

The interface view is displayed.

c. Run:
undo shutdown

The interface is enabled.

By default, an interface is enabled.
d. Run:
commit

The configuration is committed.

----End

1.2.6 Checking the Configuration of Basic Interface Parameters

Procedure
l Run the display interface [ interface-type [ interface-number ] ] command to check
running status information about an interface, including the running status, basic
configuration, and packet forwarding of the interface.
l Run the display interface brief [ interface-type ] [ main ] or display interface brief
panel-order command to check brief interface status and configuration information,
including the physical status, protocol status, bandwidth usage in the inbound and
outbound directions during a certain period of time, and the numbers of sent and
received error packets.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 12

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 1 Basic Configuration for Interfaces

l Run the display interface ethernet brief main non-unicast command to check brief
multicast and broadcast packet information about all physical Ethernet main interfaces,
including the physical status, protocol status, and statistics on multicast and broadcast
packets in the inbound and outbound directions.
l Run the display ip interface [ interface-type interface-number ] command to check the
IP configuration of an interface.
l Run the display interface description [ interface-type [ interface-number ] ] command
to check the description of an interface.
l Run the display interface counters [ inbound | outbound ] [ interface-type [ interface-
number ] ] command to check traffic statistics on an interface.
l Run the display interface counters rate [ inbound | outbound ] [ interface-type
[ interface-number ] ] command to check the incoming or outgoing traffic rate of an
interface.

----End

1.3 Maintaining Interfaces

1.3.1 Checking Interface Diagnostic Information

Context
If an interface fault occurs, you can check diagnostic information about all faulty interfaces or
a specified interface, including alarms and causes of interface Down events, Up/Down
transitions, and error packets, to rapidly locate the fault.

Procedure
l Run:
display interface troubleshooting [ interface-type interface-number ]

Interface diagnostic information is displayed.

----End

1.3.2 Clearing Interface Traffic Statistics

Context
To monitor the status of an interface or locate faults on the interface, collect traffic statistics
on the interface. Before collecting traffic statistics on an interface within a period, clear the
existing traffic statistics on this interface.

NOTICE
Interface statistics cannot be restored after they are cleared. Confirm your action before you
perform the operations.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 13

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 1 Basic Configuration for Interfaces

Procedure
l Run the reset interface counters [ interface-type [ interface-number ] ] command to
clear the interface statistics.
----End

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 14

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

2 Ethernet Interface Configuration

About This Chapter

2.1 Ethernet Interface Overview

2.2 Configuration Notes
This section provides the points of attention when configuring Ethernet interfaces on
CE8800&7800&6800&5800 series switches.
2.3 Default Configuration
2.4 Configuring MEth Management Interface Attributes
2.5 Configuring a Port Group
2.6 Configuring Common Optical/Electrical Interface Attributes
2.7 Configuring Attributes for an Optical Interface
2.8 Configuring Attributes for an Electrical Interface
2.9 Configuring Port Isolation
2.10 Maintenance
You can detect the cable status, perform loopback tests on an Ethernet interface to check
whether the interface forwarding is normal, and clear traffic statistics on the Ethernet
interface.
2.11 Configuration Examples

2.1 Ethernet Interface Overview

The Ethernet interfaces described in this chapter refer to the physical Ethernet interfaces.
Interfaces are used to exchange data and interact with other network devices, and are
classified into physical interfaces and logical interfaces. Physical interfaces include the
service interfaces (such as GE service interfaces) responsible for service transmission and
management interfaces (such as MEth management interfaces) responsible for device
management. Logical interfaces are manually configured interfaces and can be used to

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 15

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

exchange data but do not exist physically. Logical interfaces, such as VLANIF and loopback
interfaces, are used to transmit services.

The Ethernet becomes the most important local area network (LAN) networking technology
because it is flexible, simple, and easy to implement. Ethernet interfaces are real physical
interfaces provided for users to operate and configure. Ethernet interfaces are used to receive
and send data.

Based on different service functions, the Ethernet interfaces of switches can be classified into
the following types:
l MEth management interfaces: used to log in to the device to perform configuration and
management. They are not responsible for service transmission.
l Service interfaces: used to receive and send service data.

Based on different interface rates, the Ethernet interfaces of switches can be classified into the
following types:
l GE interfaces
l 10GE interfaces
l 25GE interfaces
l 40GE interfaces
l 100GE interfaces

Based on different electrical attributes, the Ethernet interfaces of switches can be classified
into the following types:
l Electrical interfaces
l Optical interfaces

Based on different packet forwarding modes, the Ethernet interfaces of switches can be
classified into the following types:
l Layer 2 Ethernet interfaces: physical interfaces working at the data link layer. They can
only forward the received packets in Layer 2 switching mode, or join VLANs to forward
the packets in Layer 3 routing mode through VLANIF interfaces.
l Layer 3 Ethernet interfaces: physical interfaces working at the network layer. IP
addresses can be configured for these interfaces. They can forward the received packets
in Layer 3 routing mode. That is, they can send and receive the packets whose source
and destination IP addresses are located in different segments.

2.2 Configuration Notes

This section provides the points of attention when configuring Ethernet interfaces on
CE8800&7800&6800&5800 series switches.

Involved Network Elements

No other NEs are required.

License Support
An interface is a basic feature of a switch and is not under license control.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 16

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Interface Query Tool

To use the interface query tool for CE series switches, visit http://support.huawei.com/
onlinetoolsweb/DC/en/interface_query.html.
You can use this tool to obtain interface attributes such as the default auto-negotiation status,
rate, and support for the auto-negotiation and flow control auto-negotiation configuration
based on the interface type and connection medium.

Feature Dependencies and Limitations

The interface-related feature dependencies and limitations are as follows:

Auto-Negotiation
Auto-negotiation configuration on an interface includes auto-negotiation of the interface rate,
flow control, training function, FEC mode, and duplex mode.
Auto-Negotiation
l If you configure the negotiation disable command on an interface in Up state, the
interface state will change from Up to Down and then to Up. Therefore, exercise caution
when deciding to run this command.
l Electrical interfaces do not support rate decrease auto-negotiation. For example, two GE
electrical interfaces are connected using a network cable. If the network cable works
only at the rate of 100 Mbit/s, the interfaces cannot negotiate the rate to 1000 Mbit/s and
are Down.
l If the module is replaced after the auto-negotiation mode takes effect on the interface,
and the new module does not support the negotiation disable command, the interface
status and configuration vary depending on the system software version as follows:
– In versions earlier than V100R005C00, the interface goes Down. You can run the
undo negotiation disable command in the interface view to delete the original non-
auto-negotiation mode configuration to make the interface go Up again.
– In V100R005C00 and later versions, if the device conflict-policy keep-config
command is configured, the configuration depends on the pre-configured medium
type and the negotiation disable configuration on the interface is saved. Otherwise,
the configuration is lost. By default, the conflict policy is based on the installed
medium type.
l In V100R006C00SPC600, if a QSFP28 interface on the CE8860EI does not go Up after
a high-speed copper cable is installed and auto-negotiation is disabled, you can upgrade
the system software version, load the V100R006SPH002 or later version patch, or enable
the FEC function to make the interface go Up.
Interface Rate
l If you configure the speed auto command on an interface in Up state, the interface state
will change from Up to Down and then to Up. Therefore, exercise caution when deciding
to run this command.
l After having GE copper modules installed, 10GE optical interfaces on the CE8860EI and
CE6860EI cannot connect to 10M, 100M, and 1000M interfaces, 10GE optical interfaces
on the CE5810EI cannot connect to 10M and 100M interfaces, 10GE1/0/13 to
10GE1/0/48 on the CE6880-48S4Q2CQ-EI as well as 10GE optical interfaces on the
CE6880-24S4Q2CQ-EI cannot connect to 10M interfaces, and 10GE optical interfaces
on the other switches can connect to 10M and 100M interfaces.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 17

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

l If the module is replaced after the interface rate takes effect, and the new module does
not support the speed configuration, the interface status and configuration vary
depending on the system software version as follows:
– In versions earlier than V100R005C00, the interface goes Down. You can run the
undo speed command in the interface view to delete the original rate configuration
in auto-negotiation mode to make the interface go Up again.
– In V100R005C00 and later versions, if the device conflict-policy keep-config
command is configured, the configuration depends on the pre-configured medium
type and the speed configuration on the interface is saved. Otherwise, the
configuration is lost. By default, the conflict policy is based on the installed
medium type.
l In V100R006C00, 10GE electrical interfaces on the CE8860EI do not support the speed
and speed auto commands and cannot work at the rate of 100 Mbit/s. They can negotiate
with remote interfaces to work at the rate of 10 Gbit/s or 1 Gbit/s. In V200R001C00 and
later versions, 10GE electrical interfaces on the CE8860EI support the speed { 10000 |
1000 } and speed auto { 10000 | 1000 } commands. They cannot work at the rate of 100
Mbit/s.

Flow Control
l CE6880-48S4Q2CQ-EI and CE6880-48T4Q2CQ-EI do not support flow control and
flow control auto-negotiation.
l The CE6850HI, CE6850U-HI, CE6851HI, CE6860EI, CE7800, or CE8800 series
switches do not support flow control when the packet forwarding mode is set to cut-
through. To enable flow control, run the assign forward mode store-and-forward
command to set the packet forwarding mode to store-and-forward.
l Flow control and flow control auto-negotiation cannot be concurrently configured on an
Ethernet interface.
l Flow control does not take effect for traffic on the following loopback interfaces:
– Loopback interface on the TRILL gateway
– Loopback interface of VXLAN services

Training Function on an Interface

l 40GE and QSFP28 interfaces on the CE6855HI, CE6860EI, CE6870EI, CE6880EI,
CE7855EI and CE8850EI do not support the training disable configuration.
l The training function takes effect on an interface only when the interface works in auto-
negotiation mode. If you run the negotiation disable command to configure the interface
to work in non-auto-negotiation mode after the training function takes effect, the training
function will become ineffective.
l If a 40GE interface is configured as a physical member interface in a stack system, the
training function cannot be disabled. In V100R003C10 and earlier versions, if the
training function is disabled on a 40GE interface, the 40GE interface cannot be
configured as a physical member interface in a stack system. In V100R005C00 and later
versions, if a 40GE interface is configured as a physical member interface in a stack
system, the training disable configuration will be automatically deleted from the
interface.
l When a training-disabled 40GE interface of a CE8860EI is connected to a 40GE
interface of a CE6851-48S6Q-HI, CE6850-48T6Q-HI or CE6855HI, the interfaces do
not go Up. The interfaces go Up only after the training function is enabled on the 40GE
interface of the CE8860EI.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 18

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

l When a QSFP+ high-speed cable is installed on a 40GE interface on the CE88-D16Q

card, you can run the training disable command to disable the training function on a
group of interfaces. Each interface group contains two interfaces, such as interfaces 1
and 2, interfaces 3 and 4, and interfaces 15 and 16. The training function cannot be
configured on one interface. This limitation does not apply to a 100GE interface that
connects to a QSFP+ high-speed cable on the CE8860EI. When a QSFP+ high-speed
cable is installed on a 40GE interface, pay attention to the following points:
– The training function can be enabled on a group of 40GE interfaces only when both
interfaces connect to QSFP+ high-speed cables and are not configured with
interface split.
– To disable auto-negotiation on one of the 40GE interfaces, run the undo training
disable command to enable the training function on both interfaces, and run the
negotiation disable command to disable auto-negotiation. After auto-negotiation is
disabled, the training disable command cannot be configured.
– If the training disable command is configured on the two 40GE interfaces, and the
QSFP+ high-speed cable on one interface is replaced with a 40GE optical module,
the interface state changes to DOWN(Transceiver type mismatch). The interface
goes Up after the undo training disable command is run.
– If the training disable command is configured on the 40GE interfaces, the pre-
configured medium type cannot be changed.
– The training disable command is mutually exclusive with the loopback internal,
service type tunnel, and service type trill unicast commands.

FEC
l The default FEC mode on an interface varies according to the device model and
connection medium. You can run the display interface command to check the FEC
mode of an interface. Interfaces at both ends of a link must work in the same FEC mode;
otherwise, the interfaces do not go Up. If their FEC modes are different, configure the
same FEC mode on the interfaces when they work in non-auto-negotiation mode.
l On 100GE interfaces, FEC is mutually exclusive with auto-negotiation (except on the
CE8860EI) and the speed command configuration. If FEC has been configured, the
system automatically deletes the FEC configuration when auto-negotiation is enabled or
the speed command is configured.
l On 25GE interfaces (including 25GE interfaces split from 100GE interfaces), FEC is
mutually exclusive with auto-negotiation and the port mode { 10g | ge } command
configuration. If FEC has been configured, the system automatically deletes the FEC
configuration when auto-negotiation is enabled or the port mode { 10g | ge } command
is configured.
l When the negotiation status of interfaces at both ends of a link is not affected, pay
attention to the following points:
– By default, FEC is disabled on interfaces with QSFP28-100G-LR4 optical modules.
According to IEEE 802.3, you are not advised to enable FEC.
– By default, FEC is enabled on interfaces with QSFP28 optical modules excluding
the QSFP28-100G-LR4 optical module. You are not advised to disable FEC
because error packets may exist.

Duplex Mode

l Only 10GE electrical interfaces on the CE6850-48T4Q-EI support the duplex mode
configuration.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 19

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

l If the default rate and auto-negotiation mode are restored on a 10GE electrical interface,
the default duplex mode is also restored.

Interface Split and Aggregation

Interface Split
l After configuring interface split or merge on an interface of a CE switch, you need to
save the configuration and restart the switch to make the configuration take effect. The
original configuration on the interface will be lost. Therefore, exercise caution when
deciding to restart the switch.
l After 100GEx/1/3, 100GEx/2/4, 100GEx/3/7, or 100GEx/4/8 (x indicates the switch's
stack ID) on a CE88-D8CQ card has a 40GE medium installed and is split, and the
system software of the switch needs to be upgraded from V100R006C00 or
V200R001C00 to V200R002C50 or a later version, run the display lldp neighbor brief
command before the upgrade to check whether the neighbor relationships are consistent
with the actual physical cable connections. If the neighbor relationships are inconsistent
with the actual physical cable connections, reconfigure services on the converted
interfaces after the upgrade to ensure that service traffic can be properly forwarded.
l If 40GE1/0/4 on a CE6850-48S4Q-EI is split into 10GE interfaces, super virtual fabric
(SVF) cannot be configured on the CE6850-48S4Q-EI. If SVF is configured on a
CE6850-48S4Q-EI, 40GE1/0/4 on the CE6850-48S4Q-EI cannot be split into 10GE
interfaces.
l In V100R005C00, when a CE6810-48S4Q-EI functions as a leaf switch, its last 40GE
interface cannot be split.
In V100R005C10, when a CE6810-48S4Q-EI, CE6810-48S4Q-LI, or
CE6810-32T16S4Q-LI functions as a leaf switch, its last 40GE interface cannot be split.
This limitation does not apply to a CE7850-32Q-EI that functions as a parent switch.
In V100R006C00 and later versions, by default, when a CE6810-48S4Q-EI,
CE6810-48S4Q-LI, or CE6810-32T16S4Q-LI functions as a leaf switch, its last 40GE
interface cannot be split. This limitation does not apply to a CE7850-32Q-EI/
CE7855-32Q-EI that functions as a parent switch. If you need to split the last 40GE
interface, run the device exclude leaf-type ce5810ei command to prevent a CE5810EI
from joining an SVF system as a leaf switch.
l A 40GE or 100GE interface cannot be split if it has been added to a stack interface.
Converted interfaces (excluding the CE7850EI and CE7855EI) can be used for stack
connections.
l Converted 10GE interfaces on the CE6850HI, CE6850U-HI, CE6851HI, CE6860EI,
CE6880EI, CE7800, and CE8800 do not support the cut-through packet forwarding
mode.
l If a 40GE interface is not split and is connected to four 10GE interfaces on the remote
device using a 1-to-4 cable, the 40GE interface cannot go Up and the interface indicator
is off. The four 10GE interfaces on the remote device can go Up and the interface
indicators are steady on. However, the interfaces cannot work properly. After interface
split is correctly configured, the interfaces can work properly.
l If a stack member switch with interface split configuration on a 40GE interface is
replaced by a switch of the same model and same stack ID, the original 40GE interface
split configuration takes effect on the new switch and the service configuration on the
converted 10GE interfaces is retained. The configuration is automatically delivered to
the new member switch without manual intervention. However, the new member switch
takes a longer time to start than other member switches.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 20

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

l In V100R003C00 and later versions, converted interfaces are numbered by adding a

dimension to the original interface number. In versions earlier than V100R003C00,
converted interfaces are numbered by interface sequence. If the system software version
is downgraded from V100R003C00 or a later version to a version earlier than
V100R003C00, the original interface split configuration is lost. You need to reconfigure
the interface split function.
l In V100R003 and earlier versions, an interface is still in Up state after the port split
command is executed, and the interface split configuration takes effect after the switch
restarts. In V100R005 and later versions, an interface is split after the port split
command is executed. The converted interfaces are in Offline state and support service
configurations. After the switch restarts, the converted interfaces go Up and the
configurations delivered when the converted interfaces are in Offline state take effect.
Interface Aggregation
l Restart the switch to make the aggregated interface take effect.
l Only 40GE interfaces aggregated from 10GE interfaces on the CE6810-48S-LI can be
used for stack connection.
l When 10GE interfaces are connected using a dedicated 1-to-4 cable or optical fiber, the
cable number sequence must be the same as the interface number sequence. For
example, interfaces 10GE1/0/1 to 10GE1/0/4 are aggregated and connected using a 1-
to-4 optical fiber, optical fiber 1 must be connected to 10GE1/0/1 and optical fiber 2
must be connected to 10GE1/0/2. The other two optical fibers must be connected
similarly. Otherwise, the interfaces do not go Up.
l 40GE aggregated interfaces that have 40GE high-speed cables installed do not support
the training disable command.

Layer 3 Interface
l In V100R005C00 and earlier versions, physical Ethernet interfaces on switches
excluding the CE6810LI can be switched to Layer 3 mode using the undo portswitch
command in the interface view. You can configure IP addresses for Layer 3 interfaces.
Interfaces on all CE series switches from V100R005C10 can be switched to Layer 3
mode.
l Layer 3 interfaces consume chip forwarding resources, reducing the number of VLANs
that can be created on the switch. After interfaces are switched to Layer 3 mode, less
than 4063 VLANs can be created.
l The VLAN, VXLAN, carrier VLAN, main interface, and Eth-Trunk where card
interoperability mode is set to enhanced share system resources. If system resources are
insufficient, the configuration may fail.
l If internal forwarding resources are insufficient, you can still configure switching
between Layer 2 and Layer 3 modes. However, interfaces may not be properly switched
between Layer 2 and Layer 3 modes after the switch restarts.
l Eth-Trunk member interfaces do not support the configuration of switching between
Layer 2 and Layer 3 modes.
l Before switching an interface on the CE6855HI or CE7855EI to Layer 3 mode, run the
vlan reserved for main-interface startvlanid to endvlanid command to configure a
dedicated reserved VLAN for the Layer 3 main interface.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 21

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Port Isolation
l In V200R003C00 and later versions, the CE6880EI supports port isolation.
l Port isolation can be configured for interfaces on the same switch or on different
switches in a stack system. In versions earlier than V200R001C00, port isolation does
not take effect for Layer 2 packets that are encapsulated through TRILL and VXLAN
tunnels. In V200R001C00 and later versions, port isolation takes effect for Layer 2
packets that are encapsulated at the VXLAN service access side on the CE6870EI, and
for TRILL packets that are encapsulated at the access side. Port isolation for interfaces
on different switches in a stack system has been supported since V100R003C10.
l In V100R005C10, if 256, 512, or 1024 link aggregation groups are configured in an SVF
system consisting of fixed switches, port isolation cannot be configured on Eth-Trunks.
In V100R006C00 and later versions, if a leaf switch is a CE5810EI switch and 256, 512,
or 1024 link aggregation groups are configured in an SVF system consisting of fixed
switches, the CE5810EI leaf switch does not support the port isolation configuration on
Eth-Trunks. This limitation does not apply to other leaf switches.

Traffic Statistics
l By default, traffic statistics on an interface do not contain the interframe gap (IFG) or
preamble in versions earlier than V100R003C10. If a switch is upgraded to
V100R003C10 or a later version, traffic statistics on an interface contain the IFG and
preamble by default.
l For the CE6870EI, the following services are in descending order of priority: M-LAG
unidirectional isolation, MQC (traffic policing, traffic statistics, and packet filtering),
querying the outbound interface of packets with specified 5-tuple information, source
MAC address, and destination MAC address, local VLAN mirroring, sFlow, NetStream,
and statistics on the VLANIF interface. When the services are configured on an interface
in the outbound direction, the service with the highest priority takes effect. For example,
when both packet filtering and traffic statistics are configured on the VLANIF interface,
packet filtering takes effect.

Jumbo Frame
l An interface checks whether the length of a packet exceeds the maximum frame length
only in the inbound direction, but not in the outbound direction. After an interface
receives protocol packets such as VXLAN, TRILL, and VLAN packets, the chip
encapsulates packet headers containing certain bytes to the packets. As a result, the
length of the outgoing packets may exceed that of the incoming packets. Therefore, you
need to consider the packet header length when configuring the maximum frame length
allowed by an interface, preventing packets from being discarded because the packet
length exceeds the maximum frame length.
l Because switches running V100R005C00 and earlier versions do not support
configuration of the minimum length of a jumbo frame, pay attention to the following
points during version upgrade or downgrade. (value1 specifies the maximum frame
length allowed by an Ethernet interface, value2 specifies the minimum length of a jumbo
frame.)
– If value1 is configured in V100R005C00 or an earlier version and the system
software is upgraded to V100R005C10 or a later version, the configuration is
retained and the default minimum length of a jumbo frame is used.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 22

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

– If value1 and value2 are configured in V100R005C10 or a later version and the
system software is downgraded to V100R005C00 or an earlier version, the
configuration is lost.
– If only value1 is configured in V100R005C10 or a later version and the system
software is downgraded to V100R005C00 or an earlier version, the configuration is
saved because both the old and new versions support this parameter.

Others
l Single-fiber communication and internal loopback detection cannot be concurrently
configured on an interface.
l All 25GE interfaces on the CE8860EI are divided into groups, each of which contains
four interfaces (such as interfaces 1 to 4 and interfaces 5 to 8). In versions earlier than
V200R002C50, interfaces in a group must have the same medium attribute (copper or
fiber) configured. This limitation does not apply to V200R002C50 and later versions. If
the system software is downgraded from V200R002C50 or a later version to a version
earlier than V200R002C50 and interfaces in the same group have different medium
attributes, some interfaces may not go Up. Replace the media so that interfaces in the
same group have the same medium attribute. An interface works in fiber mode by
default. To use a copper attribute medium, configure the port mode copper command.
l A 40GE interface on a switch running a version earlier than V100R005C00 and a 40GE
interface on a switch running V100R005C00 or a later version cannot be connected
using a QSFP+ AOC. You are advised to upgrade the system software of the two
switches to the same version or replace the transmission medium.
l In V100R005C00 and later versions, if you replace the transmission medium on an
interface after entering the interface view, some features supported by the new
transmission medium may not be configured in the interface view. To solve this problem,
exit from the current interface view and run the interface interface-type interface-
number command to enter the interface view again. For example, a 40GE optical
interface supports the training function only when it connects to a high-speed cable. If
you enter the view of a 40GE interface with a 40GE optical module installed and replace
the optical module with a 40GE high-speed cable, you cannot configure the training
disable command in the current interface view. To configure the training disable
command, exit from the current 40GE interface view and enter the interface view again.
l A 5 m SFP+ high-speed cable cannot be used to connect 10GE optical interfaces
between the CE5850EI (running a version prior to V100R005C10) and CE5855EI
switches, and a 5 m 1-to-4 QSFP+ high-speed cable cannot be used to connect a 40GE
optical interface (split into four 10GE interfaces) and 10GE optical interfaces between
the CE5850EI (running a version prior to V100R005C10) and CE5855EI switches. You
can upgrade the system software of the CE5850EI to V100R005C10 or a later version, or
replace the transmission medium to connect the two switches.
– To connect 10GE optical interfaces between the two switches: use a 1 m, 3 m, 7 m,
or 10 m SFP+ high-speed cable, or use an AOC, or optical modules and optical
fibers.
– To connect a 40GE optical interface to four 10GE optical interfaces between the
two switches: use a 1 m or 3 m 1-to-4 QSFP+ high-speed cable, or use optical
modules.
l In V200R001C00 and earlier versions, interfaces on the CE8860EI may be unavailable
and the DEVM_1.3.6.1.4.1.2011.5.25.219.2.2.3 hwBoardFail alarm is generated in the
following situations.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 23

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

– Situation 1: The number of 25GE interfaces working at the rate of 10 Gbit/s using
the same chip reaches 20 and the switch restarts.
– Situation 2: The number of 25GE interfaces working at the rate of 10 Gbit/s using
the same chip reaches 8 and the card in another slot is replaced with a different type
of card.
In situation 1, some interfaces working at the rate of 10 Gbit/s will go Down. In situation
2, all interfaces on the newly replaced card go Down.
If the port mode 10g command is configured on a 25GE interface or a 100GE interface
is split into four 25GE interfaces and then has a 40GE medium installed, the 25GE
interfaces work at the rate of 10 Gbit/s.
Interface numbers corresponding to chips are related to card slots. For details, see Table
2-1.

Table 2-1 Mappings between chips and interfaces on the CE8860EI

Subcard Slot ID Interface Interface Interface Interface

Name Number Number Number Number
Correspon Correspon Correspon Correspon
ding to ding to ding to ding to
Chip 0 Chip 1 Chip 2 Chip 3

CE88- Slot 1 or 100GE 100GE 100GE 100GE

D8CQ slot 4 interfaces 1 interfaces 7 interfaces 3 interfaces 5
and 2 and 8 and 4 and 6

Slot 2 or 100GE 100GE 100GE 100GE

slot 3 interfaces 3 interfaces 5 interfaces 1 interfaces 7
and 4 and 6 and 2 and 8

CE88- Slot 1 or 40GE 40GE 40GE 40GE

D16Q slot 4 interfaces 1 interfaces interfaces 5 interfaces 9
(Problems to 4 13 to 16 to 8 to 12
related to Slot 2 or 40GE 40GE 40GE 40GE
this type slot 3 interfaces 5 interfaces 9 interfaces 1 interfaces
of card to 8 to 12 to 4 13 to 16
mainly
occur in
card
replaceme
nt
scenarios.
)

CE88- Slot 1 or 10GE 100GE 10GE 10GE

D24T2C slot 4 interfaces 1 interfaces 1 interfaces interfaces 9
Q to 8 and 2 17 to 24 to 16
(Problems
related to
this type
of card
mainly
occur in

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 24

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Subcard Slot ID Interface Interface Interface Interface

Name Number Number Number Number
Correspon Correspon Correspon Correspon
ding to ding to ding to ding to
Chip 0 Chip 1 Chip 2 Chip 3
scenarios Slot 2 or 10GE 10GE 10GE 100GE
where slot 3 interfaces interfaces 9 interfaces 1 interfaces 1
four 17 to 24 to 16 to 8 and 2
25GE
interfaces
split from
a 100GE
interface
are
configure
d to work
at the rate
of 10
Gbit/s.)

CE88- Slot 1 or 25GE 100GE 25GE 25GE

D24S2CQ slot 4 interfaces 1 interfaces 1 interfaces 9 interfaces
to 8 and 2 to 16 17 to 24

Slot 2 or 25GE 25GE 25GE 100GE

slot 3 interfaces 9 interfaces interfaces 1 interfaces 1
to 16 17 to 24 to 8 and 2

For example, a CE88-D8CQ card is installed in slot 1 and interfaces 1 and 2 are split and
have 40GE media installed, so eight interfaces work at the rate of 10 Gbit/s. A CE88-
D24S2CQ card is installed in slot 3 and the port mode 10g command is configured on
interfaces 1 to 8, so eight interfaces work at the rate of 10 Gbit/s. A CE88-D24S2CQ
card is installed in slot 4 and the port mode 10g command is configured on interfaces 1
to 8, so eight interfaces work at the rate of 10 Gbit/s. In this case, there are 24 interfaces
working at the rate of 10 Gbit/s using chip 0. Some of these interfaces will go Down, and
the DEVM_1.3.6.1.4.1.2011.5.25.219.2.2.3 hwBoardFail alarm is generated.
It is recommended that you upgrade the system software to V200R002C50 or a later
version, or restore the working rate of some 25GE interfaces using chip 0 to 25 Gbit/s,
configure these interfaces to work at the rate of 10 Gbit/s again, and then run the reset
slot slot-id [ card card-id ] command.

2.3 Default Configuration

This section describes the default configuration of Ethernet interfaces.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 25

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Table 2-2 Default configuration of GE electrical interfaces

Parameter Default Setting

Auto-negotiation mode Auto-negotiation (can be set to non-auto-negotiation)

Duplex mode Full-duplex

Rate l Auto-negotiation mode: negotiated by the local and

remote interfaces
l Non-auto-negotiation mode: 1 Gbit/s (configurable)

Maximum frame length 9216

(bytes)

MDI Auto

EEE Disabled
(EEE can be enabled only when an interface works in
auto-negotiation mode.)

Shut down or not No

Internal loopback Disabled

Flow control Disabled

Table 2-3 Default configuration of 10GE electrical interfaces

Parameter Default Setting

Auto-negotiation mode Auto-negotiation (can be set to non-auto-negotiation)

Duplex mode Full-duplex

Rate l Auto-negotiation mode: negotiated by the local and

remote interfaces
l Non-auto-negotiation mode: 10 Gbit/s (configurable)

Maximum frame length 9216

(bytes)

MDI Auto

EEE Disabled
(EEE can be enabled only when an interface works in
auto-negotiation mode.)

Shut down or not No

Internal loopback Disabled

Flow control Disabled

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 26

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Table 2-4 Default configuration of 10GE optical interfaces

Parameter Default Setting

Auto-negotiation mode l With an SFP-GE optical module installed: auto-

negotiation (can be set to non-auto-negotiation)
l With an SFP-GE copper module installed: auto-
negotiation (can be set to non-auto-negotiation)
l With an SFP+ optical module installed: non-auto-
negotiation (cannot be set to auto-negotiation)
l With a 10GE cable connected: non-auto-negotiation
(cannot be set to auto-negotiation)

Duplex mode Full-duplex

Rate l With an SFP-GE optical module installed: 1 Gbit/s

(not configurable)
l With an SFP-GE copper module installed, auto-
negotiation mode: negotiated by the local and remote
interfaces; non-auto-negotiation mode: 1 Gbit/s
(configurable)
l With an SFP+ optical module installed: 10 Gbit/s (not
configurable)
l With a 10GE cable connected: 10 Gbit/s (not
configurable)

Maximum frame length 9216

(bytes)

Shut down or not No

Internal loopback Disabled

Flow control Disabled

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 27

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Table 2-5 Default configuration of 25GE optical interfaces

Parameter Default Setting

Auto-negotiation mode l With an SFP+ optical module installed: non-auto-

negotiation (cannot be set to auto-negotiation)
l With an SFP+ cable connected: non-auto-
negotiation (cannot be set to auto-negotiation)
l With an SFP+ AOC connected: non-auto-
negotiation (cannot be set to auto-negotiation)
l With an SFP28 optical module installed: non-auto-
negotiation (cannot be set to auto-negotiation)
l With an SFP28 cable connected: non-auto-
negotiation (cannot be set to auto-negotiation)
l With an SFP28 AOC connected: non-auto-
negotiation (cannot be set to auto-negotiation)
NOTE
No SFP28 optical module can be installed on a 25GE
interface of the CE8860EI.

Duplex mode Full-duplex

Rate l With an SFP-GE copper module installed: 1 Gbit/s

(not configurable)
l With an SFP+ optical module installed: 10 Gbit/s
(not configurable)
l With an SFP+ cable module installed: 10 Gbit/s
(not configurable)
l With an SFP+ AOC module installed: 10 Gbit/s
(not configurable)
l With an SFP28 optical module installed: 25 Gbit/s
(not configurable)
l With an SFP28 cable module installed: 25 Gbit/s
(configurable)
l With an SFP28 AOC module installed: 25 Gbit/s
(not configurable)

Maximum frame length (bytes) 9216

Shut down or not No

Internal loopback Disabled

Flow control Disabled

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 28

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Table 2-6 Default configuration of 40GE optical interfaces

Parameter Default Setting

Auto-negotiation mode l With a QSFP+ optical module installed: non-auto-

negotiation (cannot be set to auto-negotiation)
l With a QSFP+ cable connected: auto-negotiation (can
be set to non-auto-negotiation)
l With a QSFP+ AOC connected: non-auto-negotiation
(cannot be set to auto-negotiation)

Duplex mode Full-duplex

Rate 40 Gbit/s (not configurable)

Maximum frame length 9216

(bytes)

Shut down or not No

Internal loopback Disabled

Flow control Disabled

Table 2-7 Default configuration of 100GE optical interfaces

Parameter Default Setting

Auto-negotiation mode l With a QSFP+ optical module installed: non-auto-

negotiation (cannot be set to auto-negotiation)
l With a QSFP+ AOC module installed: non-auto-
negotiation (cannot be set to auto-negotiation)
l With a QSFP28 optical module installed: non-
auto-negotiation (cannot be set to auto-
negotiation)
l With a QSFP28 copper module installed: auto-
negotiation (can be set to non-auto-negotiation)
l With a QSFP28 AOC module installed: non-auto-
negotiation (cannot be set to auto-negotiation)

Duplex mode Full-duplex

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 29

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Parameter Default Setting

Rate l With a QSFP+ optical module installed: 40 Gbit/s

(not configurable)
l With a QSFP+ AOC module installed: 40 Gbit/s
(not configurable)
l With a QSFP28 optical module installed: 100
Gbit/s (not configurable)
l With a QSFP28 copper module installed: 100
Gbit/s (configurable)
l With a QSFP28 AOC module installed: 100 Gbit/s
(not configurable)

Maximum frame length (bytes) 9216

Shut down or not No

Internal loopback Disabled

Flow control Disabled

NOTE

Only 10GE electrical interfaces on the CE6850-48T4Q-EI support half-duplex mode. Ethernet interfaces
on the other CE series switches work in full-duplex mode and do not support half-duplex mode.
The auto-negotiation capability of 10GE optical interfaces varies according to the optical modules
installed. Therefore, the auto-negotiation mode configured on the interface may be different from the
one that takes effect. You can run the display this interface command to check the auto-negotiation
mode that takes effect on the interface.

2.4 Configuring MEth Management Interface Attributes

Context
This section describes how to configure the MEth management interface attributes.

The management interface MEth0/0/0 is a special Ethernet interface used to log in to the
device to perform configuration and management. The MEth management interfaces are not
responsible for service transmission.

NOTE

For the detailed function configurations of management interface, see the CloudEngine
8800&7800&6800&5800 Series Switches Configuration Guide - Basic Configuration.

The CE6850HI, CE6855-48T6Q-HI, and CE6850U-HI switches have two combo ports: port 1
and port 2. Pay attention to the following when using the two combo ports:
l The two ports cannot be used together, and you must choose one of them to use.
l Before start of the switch, you can select MEth1 or MEth2 in the BIOS menu. MEth1 is
the default choice. For details, see Modify parameters.
l After registration of the switch succeeds:

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 30

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

– If both the management ports have a cable connected and are in Up state, port 1 acts
as the primary management port and port 2 becomes the backup automatically. The
management interface number displayed on the command line interface is
MEth0/0/0, regardless of which port is used.
– If cables are connected to the two combo ports after successful registration of the
switch, the port that is connected first is used as the primary management port.
– If port 1 fails, the system switches management traffic to port 2 automatically.
When port 1 recovers, management traffic cannot be switched back to port 1, unless
port 2 fails or the switch restarts. You can observe indicators on the combo ports to
determine which port is used currently. (The Link indicator of the combo port in use
is steady green. If data is being transmitted on this port, its ACT indicator is
blinking yellow. The indicators of the backup port are off.)
The combo port consists of an electrical port and an optical port. You can connect the
electrical or optical port to a configuration terminal or network management workstation to
set up the onsite or remote configuration environment. The electrical and optical ports are
multiplexed, and only one of them can work at a time.
The combo port automatically selects the working mode as follows:
l If the optical port has no optical module installed and the electrical port has no network
cable connected, the port type depends on which port is connected first. If the electrical
port is connected by a network cable first, the electrical port is used for data switching. If
the optical port has an optical module installed first, the optical port is used for data
switching.
l If the electrical port has a network cable connected and is in Up state, the electrical port
is still used for data switching when the optical port has an optical module installed.
l If the optical port has an optical module installed and is in Up state, the optical port is
still used for data switching when the electrical port has a network cable connected.
l If the optical port has an optical module and fiber installed and the electrical port has a
network cable connected, the optical port is used for data switching after the switch
restarts.
If the electrical interface works, you can configure the interface rate and negotiation mode. If
the electrical interface works, you cannot configure the interface rate and negotiation mode.
Pay attention to the following points:
l If you run the speed and negotiation disable commands on the working electrical
interface, and the management interface is switched from the electrical interface to an
optical interface, the configuration retains but does not take effect.
l If you run the speed and negotiation disable commands on the working optical
interface, the configuration exists but does not take effect. If the management interface is
switched from the optical interface to an electrical interface, the configuration takes
effect and the peer interface must use the same configuration.
The combo optical port uses a 100M or GE optical module and matching fibers. A 100M
optical module can be used only after the switch starts successfully. If a 10GE optical module
is installed, the interface can go Up, but the system displays an alarm message, indicating that
the interface does not support the optical module. If a GE copper module is installed and the
remote interface has a GE copper module installed, the local interface can go Up but does not
support rate configuration.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 31

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Procedure
1. Run:
system-view

The system view is displayed.

2. Run:
interface meth 0/0/0

The MEth management interface view is displayed.

3. Run:
negotiation disable

The MEth management interface is configured to work in non-auto negotiation mode.

By default, the MEth management interface works in auto-negotiation mode.
NOTE

In auto-negotiation mode, the rate cannot be configured for a management interface.

The negotiation modes must be the same on two ends of a link. If the negotiation modes are
different on the two ends, the two ends still cannot communicate with each other.
4. Run:
speed { 10 | 100 | 1000 }

The rate of the management interface is configured.

By default, the working rate of the management interface is 100 Mbit/s.
NOTE

If the optical interface of a combo interface on CE6850HI or CE6850U-HI works, the combo
interface supports only a GE optical module. The optical interface works at the rate of 1000 Mbit/s
and the remote interface must also work at the same rate.
5. Run:
commit

The configuration is committed.

2.5 Configuring a Port Group

Context
When the same configuration needs to be performed on multiple Ethernet interfaces, one-by-
one configuration on each interface may result in errors and a large amount of duplicated
workload.
The port group function can solve this problem. You can add the Ethernet interfaces to the
same port group. After you run the configuration command once in the port group view, the
configuration takes effect on all Ethernet interfaces in the port group, reducing your workload.
Two port group types are available:
l Temporary port group: To temporarily batch deliver the configuration to multiple
specified interfaces, you can configure a temporary port group. After the configuration is
batch delivered, the system will delete the temporary port group once you exit from the
port group view.
l Permanent port group: To batch deliver the configuration for multiple times, you can
configure a permanent port group. The port group and its port members still exist even if

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 32

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

you exit from the port group view. This facilities the next configuration batch
configuration. To delete a permanent port group, run the undo port-group { all | port-
group-name } command.

Procedure
l Configuring a temporary port group
a. Run:
system-view

The system view is displayed.

b. Run:
port-group group-member { interface-type interface-number1 [ to
interface-type interface-number2 ] } &<1-10>

A temporary port group is created and the temporary port group view is displayed.
c. Run:
commit

The configuration is committed.

l Configuring a permanent port group
a. Run:
system-view

The system view is displayed.

b. Run:
port-group port-group-name

A permanent port group is created and the permanent port group view is displayed.
c. Run:
group-member { interface-type interface-number1 [ to interface-type
interface-number2 ] } &<1-10>

The Ethernet interfaces are added to the permanent port group.

d. Run:
commit

The configuration is committed.

----End

Checking the Configuration

Run the display port-group [ all | port-group-name ] command to view information about
member interfaces in the permanent port group.

The port group function supports batch configuration, but you cannot check or save the port
group configuration. You can run the display this command in the member interface view to
check the current configuration of the member interface.

Follow-up Procedure
After entering the port group view, you can batch deliver the configurations based on the
following interface functions:
l 2.6.1 Switching Interfaces to Layer 3 Mode

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 33

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

l 2.6.2 Configuring Auto-Negotiation

l 2.6.3 Configuring the Interface Rate
l 2.6.4 Configuring Flow Control
l 2.6.5 Configuring the Delay in Reporting Physical Status Changes
l 2.6.7 Configuring an Interface to Transit to the Error-Down State When a Link
Flapping Occurs
l 2.6.8 Configuring an Interface to Transit to the CRC Error-Down State When the
Number of Received Error Packets Exceeds the Threshold
l 2.6.10 Configuring Traffic Statistics to Include the IFG and Preamble Statistics
l 2.6.9 Configuring the IFG
l 2.6.12 Setting the Jumbo Frame Length Allowed on an Interface
l 2.7.3 Configuring Unidirectional Single-Fiber Communication
l 2.9 Configuring Port Isolation
l 2.6.13 Configuring the Outbound/Inbound Bandwidth Usage Alarm Threshold
l 2.7.6 Pre-configuring the Transmission Medium Type on an Optical Interface
Both interface attribute commands and service commands can be configured in the port group
view. For details about service commands in the port group view, see the related
documentation.
The command executed in the port group view can take effect only when the member
interfaces support the function configuration. Therefore, the configuration that is batch
delivered may not take effect for all member interfaces. For example, the member interfaces
in a port group include optical and electrical interfaces. After you run the port transceiver-
power-low trigger error-down command in the port group view, the configuration only takes
effect on the optical interfaces.

2.6 Configuring Common Optical/Electrical Interface

Attributes
2.6.1 Switching Interfaces to Layer 3 Mode
Context
Based on the hardware structure of interface cards, the interfaces on some devices can only
function as Layer 2 or Layer 3 Ethernet interfaces, and some other interfaces have flexible
modes. In Layer 2 mode, an interface can be used as a Layer 2 Ethernet interface; in Layer 3
mode, the interface can be used as a Layer 3 Ethernet interface.
By default, an Ethernet interface on the device works in Layer 2 mode. To enable the Layer 3
functions on the interface, run the undo portswitch command on the interface.
When several interfaces need to be switched between Layer 2 and Layer 3 modes, run the
portswitch batch command in the system view to perform batch switching.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 34

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

NOTE

the CE6810LI supports switching between Layer 2 and Layer 3 modes and does not support sub-
interfaces.
Eth-Trunk member interfaces do not support the configuration of switching between Layer 2 and Layer
3 modes.

Procedure
l Switching a single Ethernet interface to Layer 3 mode in the Ethernet interface view
a. Run:
system-view

The system view is displayed.

b. Run:
interface interface-type interface-number

The Ethernet interface view is displayed.

c. For CE6855HI and CE7855EI, Run:
vlan reserved for main-interface startvlanid to endvlanid

Reserved VLANs are configured for a Layer 3 main interface.

d. Run:
undo portswitch

The interface is switched to Layer 3 mode.

By default, an Ethernet interface is in Layer 2 mode.

The mode switching function takes effect when the interface only has attribute
configurations (for example, shutdown and description configurations). If the
service configuration (for example, port link-type trunk configuration) exists on
the interface, you must clear the service configuration before running this
command.

NOTE

Like a VLAN, each interface switched to the Layer 3 mode will occupy the internal
forwarding resource of the device. Therefore, only less than 4063 VLAN can be created on
the device after the interfaces are switched to the Layer 3 mode.
You can still configure switching between Layer 2 and Layer 3 modes on an interface when
the device's internal forwarding resources exceed the specifications. However, an exception
may occur during switching between Layer 2 and Layer 3 modes after the device restarts.
e. (Option) Run:
shutdown network-layer

The protocol layer status of an interface is disabled.

By default, the protocol layer status of an interface is not disabled.

In scenarios where fault locating is implemented for optical modules or fibers, users
require that only the protocol status of an interface becomes Down but the physical
and link layer status remains unchanged. To meet this requirement, run the
shutdown network-layer command to disable the protocol status of the interface.
After faults are located at the physical and link layers, run the undo shutdown
network-layer command to restart the protocol status of the interface.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 35

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

NOTE

The shutdown network-layer and protocol up-delay-time commands cannot be configured

together.
f. Run:
commit

The configuration is committed.

l Switching Ethernet interfaces to Layer 3 mode in batches in the system view
a. Run:
system-view

The system view is displayed.

b. For CE6855HI and CE7855EI, Run:
vlan reserved for main-interface startvlanid to endvlanid

Reserved VLANs are configured for a Layer 3 main interface.

c. Run:
undo portswitch batch interface-type { interface-number1 [ to interface-
number2 ] } &<1-10>

The interfaces are batch switched to Layer 3 mode.

By default, an Ethernet interface is in Layer 2 mode.
d. Run:
commit

The configuration is committed.

----End

Checking the Configuration

Run the display interface [ interface-type [ interface-number ] ] command in any view or run
the display this interface command in the interface view to check the current interface status.
If the Switch Port field is displayed in the command output, the interface is a Layer 2
interface; if the Route Port field is displayed, the interface is a Layer 3 interface.

2.6.2 Configuring Auto-Negotiation

Context
There are a large number of devices with different transmission capabilities on networks. If
the devices at the two ends of a link cannot negotiate a proper data transmission capability,
they cannot communicate with each other. The auto-negotiation function allows interfaces at
both ends of a link to select the same operating parameters by exchanging information. In this
way, the transmission capabilities of the interfaces can reach the maximum supported by the
two ends.
The duplex mode, rate, and FEC of interfaces at both ends of a link are negotiated. If the
negotiation succeeds, the two interfaces use the same duplex mode, rate, and FEC mode. In
non-auto-negotiation mode, the operating parameters must be set manually.
Electrical interfaces on both ends of a link cannot negotiate to reduce the rate. For example,
when interfaces on both ends are GE electrical interfaces but network cables support only 100

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 36

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Mbit/s rate, the two interfaces cannot negotiate to a rate of 1000 Mbit/s rate and become
Down.

NOTE

For the auto-negotiation function supported by Ethernet interfaces, see Interface Query Tool.

Procedure
Step 1 Run:
system-view

The system view is displayed.

Step 2 Run:
interface interface-type interface-number

The interface view is displayed.

Step 3 Configure the auto-negotiation function.

l Run:
negotiation disable

The Ethernet interface is configured to work in non-auto-negotiation mode.

l Run:
undo negotiation disable

The Ethernet interface is configured to work in auto-negotiation mode.

For the default auto-negotiation mode of an Ethernet interface, see 2.3 Default
Configuration.

Step 4 Run:
commit

The configuration is committed.

----End

Checking the Configuration

Run the display interface [ interface-type [ interface-number ] ] command in any view or the
display this interface command in the interface view to check the current interface status.
For details, check the Negotiation field in the command output.

2.6.3 Configuring the Interface Rate

Context
You can configure the Ethernet interface rate in either auo-negotiation or non-auto negotiation
mode.

l In auo-negotiation mode, the interface rate is negotiated by the interfaces on two ends of
a link.
– If the negotiated interface rate does not comply with the actual requirement, you
can manually configure the rate.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 37

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

As shown in Figure 2-1, the network adapter rates on Server1, Server2, and
Server3 that form a server cluster are all 1000 Mbit/s, and the rate of GE1/0/4
connecting the server cluster to external networks is also 1000 Mbit/s. If the auto-
negotiation rate is not specified on the device, the rates negotiated by GE1/0/1,
GE1/0/2, and GE1/0/3 with their connected servers are all 1000 Mbit/s. When the
servers send data at the rate of 1000 Mbit/s concurrently, the outbound interface
GE1/0/4 may be blocked. In this case, you can configure the auto-negotiation rate to
100 Mbit/s for GE1/0/1, GE1/0/2, and GE1/0/3, preventing the outbound interface
from being blocked.

Figure 2-1 Networking diagram for configuring the interface auto-negotiation rate

Internet

Switch GE1/0/4

GE1/0/1 GE1/0/3
GE1/0/2

Server1 Server2 Server3

Service data flow

l When the peer device does not support auto-negotiation, you can configure the local
device to work in non-auto negotiation mode and configure the interface rate, ensuring
communication between the local and peer devices.
NOTE

l For the interface rate supported by Ethernet interfaces, see Interface Query Tool.
l When you configure the Ethernet interface rate, ensure that the same interface rate is configured for
the two ends.
l when SFP-GE electrical modules are installed on 10GE optical interfaces, the CE8860EI do not
support 10M/100M/1000M rates, and CE5810EI do not support 10M/100M rates.
l In V100R006C00, 10GE electrical interfaces on the CE8860EI do not support the speed and speed
auto commands and cannot work at the rate of 100 Mbit/s. They can negotiate with remote
interfaces to work at the rate of 10 Gbit/s or 1 Gbit/s. In V200R001C00 and later versions, 10GE
electrical interfaces on the CE8860EI support the speed { 10000 | 1000 } and speed auto { 10000 |
1000 } commands. They cannot work at the rate of 100 Mbit/s.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 38

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Procedure
l In the auto-negotiation mode
a. Run:
system-view

The system view is displayed.

b. Run:
interface interface-type interface-number

The Ethernet interface view is displayed.

c. Configuring interface rate
n GE electrical interface
Run:
speed auto { 10 | 100 | 1000 }*

The rate auto-negotiation function is configured for the GE electrical interface.

n 10GE electrical interface
Run:
speed auto { 100 | 1000 | 10000 }*

The rate auto-negotiation function is configured for the 10GE electrical

interface.
d. Run:
commit

The configuration is committed.

l In the non-auto negotiation mode
a. Run:
system-view

The system view is displayed.

b. Run:
interface interface-type interface-number

The Ethernet interface view is displayed.

c. (Optional) Run:
negotiation disable

The Ethernet interface is configured to work in non-auto-negotiation mode.

By default, GE and 10GE electrical interfaces work in auto-negotiation mode. If the
interface is a GE and 10GE electrical interface, perform this step. By default, 25GE
and 100GE optical interfaces work in non-auto-negotiation mode, and do not
support auto-negotiation mode. If the interface is a 25GE optical interface, skip this
step.
d. Configuring interface rate
n GE electrical interface
Run:
speed { 10 | 100 | 1000 }
The rate of the GE electrical interface is configured.
n 10GE electrical interface
Run:
speed { 100 | 1000 | 10000 }
The rate of the 10GE electrical interface is configured.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 39

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

n 25GE optical interface

Run:
port mode 10g

The 25GE optical interface is configured to work at the rate of 10 Gbit/s.

If a 10GE medium is installed on a 25GE optical interface, the interface goes
Up only after this command is run. If 40GE mediums are installed on four
25GE interfaces split from a 100GE interface, the four 25GE interfaces can
automatically work as four 10GE interfaces without configuration of this
command.
n 100GE QSFP28 optical interface
Run:
speed 40000
The rate of the 100GE optical interface is configured.
100GE QSFP28 optical interface supports speed 40000 command only after a
QSFP28 high-speed cable is installed.
e. Run:
commit

The configuration is committed.

By default, an Ethernet interface works at its maximum rate.

Checking the Configuration

Run the display interface [ interface-type [ interface-number ] ] command in any view or the
display this interface command in the interface view to check the current interface status.
For details, check the Speed field in the command output.

2.6.4 Configuring Flow Control

Context
Network congestion causes packet loss. Flow control is a method of avoiding packet loss. If
network congestion occurs on the local device after flow control is configured on both ends of
the link, the local device sends a message to the peer device, instructing the peer device to
reduce the packet sending rate. After receiving the message, the peer device stops sending
packets to the local device no matter whether its packet sending rate is high or low, which
prevents congestion.

As shown in Figure 2-2, SwitchA communicates with SwitchC through SwitchB. The flow
control function is implemented as follows:
1. 10GE1/0/1 on SwitchA is connected to 10GE1/0/2 on SwitchB, and the auto-negotiation
speed is 10000 Mbit/s. The interfaces send data packets to each other at the speed of
10000 Mbit/s.
2. The outbound interface GE1/0/1 on SwitchB only provides a maximum transmission
speed of 1000 Mbit/s. When congestion occurs during packet forwarding, SwitchB
caches the received packets. When the number of cached packets exceeds the forwarding
capability of GE1/0/1, packet loss may occur.
3. After flow control is configured on 10GE1/0/1 of SwitchA and 10GE1/0/2 of SwitchB,
10GE1/0/2 sends a Pause frame to 10GE1/0/1, instructing 10GE1/0/1 to stop sending
packets. After receiving the Pause frame, 10GE1/0/1 stops sending packets to 10GE1/0/2

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 40

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

within the period specified in the Pause frame. If congestion still occurs, 10GE1/0/2
continues to send Pause frames to 10GE1/0/1.
4. GE1/0/1 continues sending the cached data packets until congestion is eliminated.
5. After congestion is eliminated, SwitchB no longer sends Pause frames to SwitchA.
SwitchA then can continue to send data packets at the speed of 10000 Mbit/s.

Figure 2-2 Flow control

① Send data at the ② Congestion occurs when data is
speed of 10000 Mbit/s sent at the speed of 1000 Mbit/s

10GE1/0/1 10GE1/0/2 GE1/0/1 GE1/0/2

SwitchA ③ Send a Pause frame SwitchB SwitchC

to instruct the peer to
stop sending packets ④ Send the data cached on
SwitchB at the speed of 1000
Mbit/s until data transmission is
⑤ Continue to send data at complete
the speed of 10000 Mbit/s
Service data flow

Pause frame

You can enable flow control in both the inbound and outbound directions of an interface or
enable the function in either direction based on actual requirements. For example, if flow
control is enabled in the inbound direction of 10GE1/0/1 on SwitchA, the interface is
controlled by the remote interface and stops sending data packets after receiving a Pause
frame from the remote interface. If flow control is enabled in the outbound direction of
10GE1/0/1, the interface only sends a Pause frame to the remote interface, instructing the
remote interface to stop sending packets. If the inbound and outbound directions are not
specified, flow control is enabled in both directions.

When the peer device does not support auto-negotiation, you can configure flow control on
two ends of the link. When the devices on two ends of the link both support auto-negotiation,
you can configure flow control auto-negotiation on the two ends. The local device then
negotiates with the peer device to determine whether flow control is enabled based on the
network congestion status.

l Flow control and flow control auto-negotiation cannot be configured concurrently.

l Flow control and flow control auto-negotiation can take effect only when being
configured on both ends of a link.
l CE8860EI, CE7850EI, CE7855EI, CE6850HI, CE6850U-HI, CE6851HI, and CE6855HI
series switches do not support flow control when the packet forwarding mode is set to
cut-through. To enable flow control, run the assign forward mode store-and-forward
command to set the packet forwarding mode to store-and-forward.
NOTE

l For the flow control function supported by Ethernet interfaces, see Interface Query Tool.
l Flow control does not take effect for traffic on the following loopback interfaces:
l Loopback interface on the TRILL gateway
l Loopback interface of VXLAN services

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 41

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Procedure
l Configuring flow control
a. Run:
system-view

The system view is displayed.

b. Run:
interface interface-type interface-number

The interface view is displayed.

c. Run:
flow-control [ input | output ]

Flow control is configured.

By default, flow control is not configured on an Ethernet interface.

d. Run:
commit

The configuration is committed.

l Configuring flow control auto-negotiation

a. Run:
system-view

The system view is displayed.

b. Run:
interface interface-type interface-number

The interface view is displayed.

c. Run:
undo negotiation disable

The Ethernet interface is configured to work in auto-negotiation mode.

If the interface by default works in auto-negotiation mode and the working mode
cannot be modified, you do not need to perform this step.
d. Run:
flow-control negotiation

The flow control auto-negotiation function is configured on the interface.

By default, flow control auto-negotiation is not configured on the interface.

e. Run:
commit

The configuration is committed.

----End

2.6.5 Configuring the Delay in Reporting Physical Status Changes

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 42

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Context
An Ethernet interface can be physically Up and Down. When the physical interface status
changes, the system notifies the upper-layer protocol modules (such as routing and forwarding
modules) to guide packet sending and receiving. The system also generates traps and logs to
ask users whether to process the physical link.

After you configure the delay in reporting interface status changes, the system will not sense
the physical interface status switching within the delay. After the delay, the physical status
changes are reported to the system if the physical status is not recovered.

Procedure
Step 1 Run:
system-view

The system view is displayed.

Step 2 Run:
interface interface-type interface-number

The interface view is displayed.

Step 3 Run:
carrier { up-hold-time | down-hold-time } interval

The delay in reporting interface status changes is configured.

By default, the delay in reporting an Up event and that in reporting a Down event are both 0
ms.

The delay in reporting status changes depends on the network connection status.
l Set the delay to a larger value.
For example, the upper-layer protocol convergence time may be longer than the interface
status change time in some scenarios. If the interface status changes to Up when the
upper-layer protocol convergence is not complete, packet loss may occur. You can set the
delay to a larger value to prevent packet loss that occurs because the upper-layer protocol
convergence time is longer than the interface status change time.
l Set the delay to a smaller value.
For example, after the status of the interface on the master link group changes from
physically Up to Down, the system must immediately instruct the upper-layer service
forwarding protocol to send service packets from the standby interface on the backup
link. You must set a shorter interval to ensure real-time service switchover.

Step 4 Run:
commit

The configuration is committed.

----End

2.6.6 Configuring the Delay in Reporting an Interface Up Event

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 43

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Context
The physical status of an Ethernet interface can be Up or Down. When the interface physical
status changes, the system notifies upper-layer protocol modules (such as the routing and
forwarding modules) of the change to direct packet receiving and forwarding. The system also
automatically generates traps and logs to remind users to perform corresponding operations
on physical links.
After the device restarts or a card resets, the physical status of interfaces is Up. However, the
upper-layer protocol modules do not meet forwarding requirements. If an interface receives
service packets and sends the packets to the upper-layer protocol modules. The services
cannot run properly because the upper-layer protocol modules cannot process the service
packets. You can run the set up-delay command to set the delay after which the interface goes
Up. The interface then becomes physically Up after the upper-layer protocol modules meet
forwarding requirements, ensuring that packets can be processed.

Procedure
Step 1 Run:
system-view

The system view is displayed.

Step 2 Run:
interface interface-type interface-number

The interface view is displayed.

Step 3 Run:
set up-delay value

The delay after which an interface goes Up is set.

By default, the delay after which an interface goes Up is 0 seconds.
Step 4 Run:
commit

The configuration is committed.

----End

Checking the Configuration

Run the display interface [ interface-type [ interface-number ] ] command in any view or run
the display this interface command in the interface view within the delay to check the
current running status of the interface. For details, check the current state field in the
command output.

2.6.7 Configuring an Interface to Transit to the Error-Down State

When a Link Flapping Occurs
Context
The device records the status of an interface as Error-Down when it detects that a fault occurs.
The interface in Error-Down state cannot receive or send packets and the interface indicator is

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 44

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

off.Link flapping is one of the reasons that cause an interface to transit to the Error-Down
state. You can configure link flapping protection to enable the device to shut down interfaces
with frequent UP/Down transitions. This function prevents frequent network topology
changes from affecting services. For example, two links work in primary/backup mode. If the
interface of the primary link experiences frequent Up/Down transitions, services are switched
between the primary and backup links. Frequent service switchovers increase load of the
device and may result in service data loss. To solve this problem, you can configure link
flapping protection on the interface of the primary link to prevent frequent topology changes.
When the system detects frequent physical status changes on the interface of the primary link,
the system directly disables the interface to trigger a primary/backup link switchover. The
backup link then steadily transmits services. The link flapping protection function involves
the following parameters:
l Number of link flappings: One link flapping refers to one interface Up/Down transition.
l Link flapping period: It is a period during which the system counts the number of link
flappings.
If the number of link flappings on an interface reaches the threshold within a link flapping
period, the system disables the interface and records its status as ERROR DOWN(link-flap).
By default, after link flapping protection is enabled, an interface goes Error-Down if its status
changes five times within 10 seconds.

Procedure
Step 1 Run:
system-view

The system view is displayed.

Step 2 Run:
port link-flap trigger error-down

Link flapping protection is enabled on the interface.

By default, link flapping protection is enabled on an interface.

Step 3 Run:
interface interface-type interface-number

The interface view is displayed.

Step 4 (Optional) Run:

port link-flap { interval interval-value threshold threshold-value | interval
interval-value | threshold threshold-value }

The number of link flappings and the link flapping period are configured for the interface.

By default, the number of link flappings is 5 and the link flapping period is 10 seconds for an
interface.

Step 5 Run:
commit

The configuration is committed.

----End

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 45

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Checking the Configuration

Run the display error-down recovery [ interface interface-type interface-number ]
command in any view to check information about the interfaces in Error-Down state.

Follow-up Procedure
When interfaces are in Error-Down state, you are advised to first check whether the optical
modules and fiber on the two ends are securely installed on the optical interfaces or fail and
whether the network cables on the two ends are firmly connected to the electrical interfaces.
An interface in Error-Down state can be recovered using either of the following methods:
l Manual recovery (after an Error-Down event occurs):
If a few interfaces need to be recovered, run the shutdown and undo shutdown
commands in the interface view. Alternatively, run the restart command in the interface
view to restart the interfaces.
l Automatic recovery (before an Error-Down event occurs):
If a large number of interfaces need to be recovered, manual recovery is time consuming
and some interfaces may be omitted. To avoid this problem, you can run the error-down
auto-recovery cause link-flap interval command in the system view to enable
automatic interface recovery and set the recovery delay time. You can run the display
error-down recovery command to view information about automatic interface recovery.
NOTE

This method does not take effect on interfaces that are already in Error-Down state. It takes effect
only on interfaces that enter the Error-Down state after this configuration is complete.

2.6.8 Configuring an Interface to Transit to the CRC Error-Down

State When the Number of Received Error Packets Exceeds the
Threshold

Context
The device records the status of an interface as Error-Down when it detects that a fault occurs.
The interface in Error-Down state cannot receive or send packets and the interface indicator is
off.Receiving excessive CRC error packets is one of the reasons that cause an interface to
transit to the Error-Down state. When an Ethernet interface receives excessive CRC error
packets, faults such as packet loss occur. Because the interface is still in Up state, traffic is
still transmitted on the interface even if a backup link is configured. To avoid impact on
services, you can configure the interface to change to the Error-Down state when it receives
excessive CRC error packets. When the number of received CRC error packets on the
interface exceeds the upper alarm threshold, the system disables the interface and records the
interface status as ERROR DOWN(crc-statistics). Services are then switched to the backup
link immediately.

Procedure
Step 1 Run:
system-view

The system view is displayed.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 46

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Step 2 Run:
interface interface-type interface-number

The interface view is displayed.

Step 3 Run:
trap-threshold error-statistics threshold-value interval interval-value

The alarm threshold of CRC error packets and the alarm interval are configured.
By default, the alarm threshold of error packets is 3 and the alarm interval is 10 seconds.
Step 4 Run:
port crc-statistics trigger error-down

The interface is configured to transit to the Error-Down state when the number of received
CRC-error packets exceeds the threshold.
By default, an interface does not transit to the Error-Down state when the number of received
CRC-error packets exceeds the threshold.
Step 5 Run:
commit

The configuration is committed.

----End

Checking the Configuration

Run the display error-down recovery [ interface interface-type interface-number ]
command in any view to check information about the interfaces in Error-Down state.

Follow-up Procedure
If an interface is in Error-Down state, you are advised to find out the cause of the Error-Down
event first. If the two connected interfaces are optical interfaces, check whether optical
modules and fiber on the interfaces are securely installed or fail. If the two connected
interfaces are electrical interfaces, check whether the network cable is securely connected to
the interfaces or fails.
An interface in Error-Down state can be recovered using either of the following methods:
l Manual recovery (after an Error-Down event occurs):
If a few interfaces need to be recovered, run the shutdown and undo shutdown
commands in the interface view. Alternatively, run the restart command in the interface
view to restart the interfaces.
l Automatic recovery (before an Error-Down event occurs):
If a large number of interfaces need to be recovered, manual recovery is time consuming
and some interfaces may be omitted. To avoid this problem, you can run the error-down
auto-recovery cause crc-statistics interval command in the system view to enable
automatic interface recovery and set the recovery delay time. You can run the display
error-down recovery command to view information about automatic interface recovery.
NOTE

This method does not take effect on interfaces that are already in Error-Down state. It takes effect
only on interfaces that enter the Error-Down state after this configuration is complete.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 47

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

2.6.9 Configuring the IFG

Context
Inter-frame gap (IFG) is used to differentiate two data packets, as shown in Figure 2-3. You
can configure the IFG to improve data packet forwarding efficiency.

The packet forwarding rate, also called throughput, refers to the data forwarding capability on
an interface, in pps. The packet forwarding rate is calculated based on the number of 64-byte
data packets in a certain period. The payload of the preamble and IFG affects the packet
forwarding rate.

The default IFG is 12 bytes and is recommended. If you set the IFG to a small value, the
device may not have enough time to receive the next frame after receiving one data frame.
The packets then cannot be processed in real time, which results in packet loss.

Figure 2-3 IFG

8 bytes 64-1518 bytes 12 Bytes(variable) 8 bytes 64-1518 bytes
…... Data Data …...
Ethernet Ethernet
Preamble IFG Preamble
frame 1 frame 2

Procedure
Step 1 Run:
system-view

The system view is displayed.

Step 2 Run:
interface interface-type interface-number

The interface view is displayed.

Step 3 Run:
ifg ifg-value

The IFG is configured.

By default, the IFG is 12 bytes.

Step 4 Run:
commit

The configuration is committed.

----End

2.6.10 Configuring Traffic Statistics to Include the IFG and

Preamble Statistics

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 48

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Context
You can run the display interface command to check the running status and statistics of an
interface. The Last 300 seconds input rate or Last 300 seconds output rate field in the
command output indicates the traffic rate in the inbound or outbound direction of the interface
within the latest 300 seconds.

l If you want to obtain the total number of bytes passing through the interface within a
certain period, configure traffic statistics to include IFG and preamble statistics. The total
number includes the number of bytes in packets and the payload of the preamble and
inter-frame gap (IFG). In this case, Interface traffic statistics rate = (Original packet
length + IFG + Preamble) x Number of packets passing through the interface every
second.
l If you want to obtain the number of packet bytes passing through the interface within a
certain period, configure traffic statistics to exclude IFG and preamble statistics. The
number here excludes the payload of the preamble and IFG. In this case, Interface traffic
statistics rate = Original packet length x Number of packets passing through the interface
every second.
The inter-frame gap has a fixed value of 12 bytes and the preamble has a fixed value of 8
bytes. To adjust the inter-frame gap, run the ifg command.

Procedure
Step 1 Run:
system-view

The system view is displayed.

Step 2 Run:
interface interface-type interface-number

The interface view is displayed.

Step 3 Run:
set flow-statistics include-interframe

Interface traffic statistics include the IFG and preamble statistics.

By default, interface traffic statistics include the IFG and preamble statistics.

Step 4 Run:
commit

The configuration is committed.

----End

Checking the Configuration

Run the display interface [ interface-type [ interface-number ] ] command in any view or the
display this interface command in the interface view to check the current interface status.

2.6.11 Collecting Protocol Packet Statistics Based on Protocol

Types

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 49

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Context
After this function is enabled on interfaces or system, you can know how many packets of a
certain protocol are received on interfaces and whether all the protocol packets received on
interfaces are correct.

Procedure
l Configuring the system to collect protocol packet statistics based on protocol types
a. Run:
system-view

The system view is displayed.

b. Run:
ip host packet statistics protocol all

The system is configured to collect protocol packet statistics based on protocol

types.
By default, the system does not collect protocol packet statistics based on protocol
types.
c. Run:
commit

The configuration is committed.

l Configuring an interface to collect protocol packet statistics based on protocol types
a. Run:
system-view

The system view is displayed.

b. Run:
interface interface-type interface-number

The interface view is displayed.

c. Run:
ip host packet statistics protocol enable

The interface is configured to collect protocol packet statistics based on protocol

types.
By default, an interface does not collect protocol packet statistics based on protocol
types.
d. Run:
commit

The configuration is committed.

----End

Follow-up Procedure
l Run the display ip host packet statistics receive protocol interface interface-type
interface-number command in any view to check protocol packet statistics on interfaces.
l Run the display ip host packet statistics receive protocol command in any view to
check protocol packet statistics in the system.
l Run the display ip host packet statistics receive command in any view to check packet
statistics on the top 10 interfaces receiving the most packets.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 50

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

l Run the display ip host packet statistics receive protocol { arp | stp | lacp | lldp | isis |
icmp | ospf | pim | igmp | vrrp | snmp | dhcp | bgp | ldp | icmpv6 | ospfv3 | pimv6 |
igmpv6 | vrrpv6 | snmpv6 | dhcpv6 | bgp4plus | ldpv6 } command in any view to
check information about the top 10 interfaces receiving most specified protocol packets.

2.6.12 Setting the Jumbo Frame Length Allowed on an Interface

Context
Ethernet frames longer than 1518 bytes and VLAN frames longer than 1522 bytes are called
Jumbo frames.
When exchanging a large amount of data (for example, transmitting files), Ethernet interfaces
may receive Jumbo frames whose length exceeds that of common packets. If the length of the
received Jumbo frames exceeds the default data frame length that can be processed, the device
directly discards the Jumbo frames. You can set the Jumbo frame length allowed on an
interface.
After the Jumbo frame length allowed on an interface is adjusted, packet forwarding becomes
flexible. If multiple common Ethernet frames are used to transmit a data packet, redundant
information such as inter-packet gaps (IPGs) and headers is also transmitted. If Jumbo frames
are used to transmit a data packet of the same length, there are fewer frames without
unnecessary IPGs and headers to transmit, improving the bandwidth usage.
However, the Jumbo frame lengths allowed on interfaces cannot be all set to the maximum.
During transmission of a packet, each device that the packet passes though needs to support
forwarding of Jumbo frames; otherwise, the packet is discarded during forwarding. After an
interface receives protocol packets such as VXLAN, TRILL, and VLAN packets, the chip
encapsulates packet headers containing certain bytes to the packets. As a result, the length of
the outgoing packets may exceed that of the incoming packets. Therefore, you need to
consider the packet header length when configuring the maximum frame length allowed by an
interface, preventing packets from being discarded because the packet length exceeds the
maximum frame length.

Procedure
Step 1 Run:
system-view

The system view is displayed.

Step 2 Run:
interface interface-type interface-number

The interface view is displayed.

Step 3 Run:
jumboframe enable value1 [ value2 ]

The maximum length of a jumbo frame and the maximum length of a non-jumbo frame
allowed by an interface is set.
By default, the maximum frame length allowed by an interface is 9216 bytes and the
maximum length of a non-jumbo frame allowed by an interface is 1518 bytes.
The CE5810EI does not support parameter value2.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 51

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Step 4 Run:
commit

The configuration is committed.

----End

2.6.13 Configuring the Outbound/Inbound Bandwidth Usage

Alarm Threshold

Context
The bandwidth usage represents the load on a device. If the bandwidth usage exceeds the
threshold, bandwidth on the device is not sufficient for services and needs expansion. For
example, if the bandwidth usage exceeds 95%, an alarm is generated to indicate that
bandwidth resources are used up. Services may be interrupted before system expansion. You
can expand device capacity to prevent service interruption.

Procedure
Step 1 Run:
system-view

The system view is displayed.

Step 2 Run:
interface interface-type interface-number

The interface view is displayed.

Step 3 Run:
trap-threshold { input-rate | output-rate } bandwidth-in-use [ resume-rate resume-
threshold ]

The alarm threshold for the inbound/outbound bandwidth usage is configured for the
interface.

By default, the alarm thresholds of the outbound and inbound bandwidth usage are both 90.

NOTE

If the difference between the bandwidth-in-use and resume-threshold values is too small, alarms may be
frequently generated or cleared.

Step 4 Run:
commit

The configuration is committed.

----End

2.7 Configuring Attributes for an Optical Interface

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 52

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

2.7.1 Configuring Interface Split

Context
The interface split function allows a high-bandwidth physical interface on the device to be
configured as multiple independent low-bandwidth interfaces. A high-bandwidth interface on
the device can be configured as multiple low-bandwidth interfaces or directly used based on
the interface type on the remote device. The interface split function allows for flexible
networking and lowers hardware costs.
40GE and 100GE interfaces on CloudEngine 8800&7800&6800&5800 Series switches
support the interface split function. Table 2-8 shows interface split types and precautions. For
details about cables used on converted interfaces, see "Appearance and Structure" of the
corresponding model in the CloudEngine 8800&7800&6800&5800 Series Switches
Hardware Description.
When configuring interface split, pay attention to the following points:
l After configuring or canceling interface split on an interface, the interface will not exist
and the original configuration on the interface will be lost. If you run the rollback
configuration command to roll back the configuration to that before the interface split,
the original configuration on the interface will not be restored, and you need to manually
perform the configuration again. Therefore, exercise caution when deciding to roll back
the configuration.
l If a 40GE or 100GE interface has been added to a stack interface, the 40GE or 100GE
interface cannot be split.
l If a 40GE interface is not split and is connected to four 10GE interfaces on the remote
device using a 1-to-4 cable, the 40GE interface cannot go Up and the interface indicator
is off. The four 10GE interfaces on the remote device may go Up and the interface
indicators are steady on. However, the interfaces cannot work properly. After interface
split is correctly configured, the interfaces can work properly.
The interface view is not changed after an interface is split. For example, if 40GE x/y/n is
split into four 10GE interfaces, the view of the converted interfaces is still the 40GE interface
view, and the converted interfaces are numbered 40GE x/y/n:1, 40GE x/y/n:2, 40GE x/y/n:3,
and 40GE x/y/n:4.

NOTE

It is recommended that you use the Interface Split Query Tool when configuring the interface split
function. This tool provides details about interface split types supported by CE series switches, cables
used on converted interfaces, configuration procedure, and precautions.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 53

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Table 2-8 Interface split configuration on interfaces of CloudEngine

8800&7800&6800&5800 Series switches
Interface Interface Split Description
Type Type

40GE l A 40GE interface l If 40GE1/0/4 on a CE6850-48S4Q-EI is split

interface on the CE5850HI, into 10GE interfaces, super virtual fabric
CE5855EI, (SVF) cannot be configured on the
CE6800 series CE6850-48S4Q-EI. If SVF is configured on a
switches, and CE6850-48S4Q-EI, 40GE1/0/4 on the
CE7800 series CE6850-48S4Q-EI cannot be split into 10GE
switches can be interfaces.
split into four l 10GE interfaces split from 40GE optical
10GE interfaces. interfaces on CE series switches (excluding the
l A 40GE interface CE7850EI and CE7855EI) can be used for
on CE88-D16Q stack connections.
sub-card of the l On the CE6850HI, CE6851HI, CE6855HI,
CE8860EI can be CE6850U-HI, CE8800 series switches, and
split into two CE7800 series switches, converted 10GE or
10GE interfaces. 25GE interfaces do not support the cut-through
packet forwarding mode.
l In V100R005C10, when a CE6810-48S4Q-EI,
CE6810-48S4Q-LI, or CE6810-32T16S4Q-LI
functions as a leaf switch, its last 40GE
interface cannot be split. This limitation does
not apply to a CE7850-32Q-EI that functions
as a parent switch.
In V100R006C00 and later versions, by
default, when a CE6810-48S4Q-EI,
CE6810-48S4Q-LI, or CE6810-32T16S4Q-LI
functions as a leaf switch, its last 40GE
interface cannot be split. This limitation does
not apply to a CE7850-32Q-EI/CE7855-32Q-
EI that functions as a parent switch. If you
need to split the last 40GE interface, run the
device exclude leaf-type ce5810ei command
to prevent a CE5810EI from joining an SVF
system as a leaf switch.
l If a stack member switch with interface split
configuration on a 40GE interface is replaced
by a switch of the same model and same stack
ID, the original 40GE interface split
configuration takes effect on the new switch
and the service configuration on the converted
10GE interfaces is retained. The configuration
is automatically delivered to the new member
switch without manual intervention. However,
the new member switch takes a longer time to
start than other member switches.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 54

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Interface Interface Split Description

Type Type

100GE A 100GE interface on -

interface the CE8860EI or
CE6870EI is split
into:
l Four 25GE
interfaces when a
100GE optical
module is installed
l Four 10GE
interfaces when a
40GE optical
module is installed

Procedure
Step 1 Configure the interface split mode. (This configuration is only applicable to the CE7850EI
and CE7855EI.)
1. Run:
system-view

The system view is displayed.

2. Run:
port split mode { mode1 | mode2 | mode3 | mode4 } slot slot-id

The 40GE interface split mode is configured.

By default, the device uses mode 1 to split interfaces.

– If mode1, mode2, or mode3 is configured, the interface split function cannot be

configured on eight interfaces and can be configured on the other 24 interfaces.
Interface split can be configured on all or some of the 24 interfaces.
– If mode4 is configured, the interface split function cannot be configured on six
interfaces. The other 26 interfaces are automatically split into one hundred and four
10GE interfaces after the device restarts. The split configuration cannot be
canceled. You cannot perform the following split configuration steps.
3. Run:
commit

The configuration is committed.

4. Run:
quit

Return to the user view.

5. Run:
save

The current configuration is saved.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 55

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

6. Run:
reset slot slot-id

The device is restarted to make the interface split mode configuration take effect.

Step 2 Run:
system-view

The system view is displayed.

Step 3 Run:
port split dimension interface { interface-type interface-number1 [ to interface-
type interface-number2 ] } &<1-18> [ split-type split-type ]

Interface split is configured.

By default, an interface is not split.

In versions earlier than V100R003C00, converted interfaces are numbered using only the
interface sequence numbering rule. In V100R003C00 and later versions, converted interfaces
are numbered using only the dimension numbering rule.
l If the system software is upgraded from a version earlier than V100R003C00 to
V100R003C00 or a later version, the original converted interfaces still work properly. To
configure interface split again, you must run the port split refresh command to update
the interface numbering rule first.
l If the system software is downgraded from V100R003C00 or a later version to a version
earlier than V100R003C00, the interface split configuration will be lost.

NOTE

After the port split command is run, an interface is split, is in Offline state, and supports service
configurations. Converted interfaces go Up after the switch restarts and the configurations delivered when the
split interface is in Offline state take effect.

Step 4 Run:
commit

The configuration is committed.

Step 5 Run:
quit

Return to the user view.

Step 6 Run:
save

The current configuration is saved.

Step 7 Run:
reset slot slot-id

The device is restarted to make the numbering rule of converted interfaces take effect.

After configuring interface split or merge, save the configuration and restart the switch to
make the configuration take effect.

----End

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 56

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Checking the Configuration

Run the display port split [ slot slot-id ] command to view interface split information.

Related Topics
Interface Split Query Tool

2.7.2 Configuring Interface Aggregation

Context
The interface aggregation function allows multiple consecutive low-bandwidth physical
interfaces on the device to be aggregated into a high-bandwidth interface. With the interface
aggregation function, interfaces on a card can connect to various types of interfaces on the
remote device, allowing for flexible networking and lowering hardware costs.

Only four consecutive 10GE optical interfaces of the CE6810EI, CE6810LI, CE6850-48S6Q-
HI, CE6851HI, and CE5855-24T4S2Q-EI can be aggregated into a 40GE interface and the
type of the aggregated interface is 4x10GE. The number of the 4x10GE interface is based on
the number of the first 10GE interface that is aggregated. The indicator of the first 10GE
interface becomes the indicator of the 4x10GE interface, and is steady on or blinks. The
indicators of the other three 10GE interfaces are off. For example, 10GE1/0/1 to 10GE1/0/4
are aggregated into a 4x10GE interface. The number of the 4x10GE interface is 4x10GE1/0/1.
The indicator of 10GE1/0/1 functions as the indicator of 4x10GE1/0/1, and is steady on or
blinks. The indicators of 10GE1/0/2 to 10GE1/0/4 are off.

A 4x10GE interface can be connected to a remote 40GE interface using the following
methods:
l The 4x10GE interface can be directly connected to the remote 40GE interface using a
dedicated 1-to-4 cable. Both the local and remote interfaces do not need to have optical
modules installed. For details about the dedicated 1-to-4 cable, see High-Speed Cable in
the CloudEngine 8800&7800&6800&5800 Series Switches Hardware Description -
Cables.
l The 4x10GE interface can be connected to the remote 40GE interface using a 1-to-4
40GE optical module and a 1-to-4 optical fiber. The remote 40GE interface needs to
have a 40GE optical module with the same wavelength as the 10GE optical modules
installed. For details about the 1-to-4 optical fiber, see 40GE QSFP+ Optical Modules in
the CloudEngine 8800&7800&6800&5800 Series Switches Hardware Description -
Optical Module.
l If the 4x10GE interface is connected to a remote 4x10GE interface, the two 4x10GE
interfaces can have the same type of 10GE high-speed cable or Small Form-Factor
Pluggable Plus (SFP+) optical modules installed and be connected using four optical
fibers.

When 10GE interfaces are connected using a dedicated 1-to-4 cable or optical fiber, the cable
number sequence must be the same as the interface number sequence. For example, interfaces
10GE1/0/1 to 10GE1/0/4 are aggregated and connected using a 1-to-4 optical fiber, optical
fiber 1 must be connected to 10GE1/0/1 and optical fiber 2 must be connected to 10GE1/0/2.
The other two optical fibers must be connected similarly. Otherwise, the interfaces do not go
Up.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 57

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

NOTE

l After the interface aggregation configuration is committed, the aggregated interface takes effect
only after the switch is restarted. 10GE interfaces that are aggregated will be deleted and
configurations of these interfaces will be lost. Therefore, exercise caution when deciding to use the
interface aggregation function.
l An aggregated interface is in Up state only when all the 10GE interfaces that are aggregated are in
Up state and have the same type of optical modules installed. You can run the display interface
transceiver command in the view of the aggregated interface to check information about the
optical module on each 10GE interface that is aggregated.

Procedure
Step 1 Run:
system-view

The system view is displayed.

Step 2 Run:
port aggregate aggregate-type 4*10ge interface { interface-type interface-number
[ to interface-type interface-number ] } &<1-12>

The interface aggregation type and numbers of interfaces that are aggregated are configured.
By default, 10GE interfaces are not aggregated.
Step 3 Run:
commit

The configuration is committed.

----End

Checking the Configuration

Run the display port aggregate [ slot slot-id ] command in any view to check information
about aggregated interfaces.

Follow-up Procedure
l Restart the switch to make the aggregated interface take effect.
l Only 40GE interfaces aggregated from 10GE interfaces of the CE6810-48S-LI can be
used for stack connection.

2.7.3 Configuring Unidirectional Single-Fiber Communication

Context
During network management and maintenance, the administrator may need to send traffic
from users to a specified server for analysis, recording, and processing. If a server can receive
and send packets, there is a possibility that the server forwards user traffic to other devices,
causing a security risk. The unidirectional single-fiber communication function can address
this issue. A single fiber means that two optical modules are connected by only one fiber, and
unidirectional communication means that packets can be sent in only one direction. With this
function, a switch can only send but cannot receive packets, and an analysis server can only
receive but cannot send packets. The data security on the analysis server is ensured.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 58

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

An optical module provides a TX end and an RX end. Generally, two optical modules are
connected by two fibers. The TX and RX ends of one module are respectively connected to
the RX and TX ends of another module. A device transmits and receives packets through two
independent fibers. If the unidirectional single-fiber communication function is disabled, two
devices cannot communicate with each other through a single fiber. After this function is
configured, the devices can use only one fiber to communicate with each other.
As shown in Figure 2-4, SwitchA is connected to the upper-layer traffic distribution device
through 10GE2/0/1. The traffic sent from the traffic distribution device enters SwitchA
through 10GE2/0/1. SwitchA transmits packets through 10GE2/0/2, the analysis server
receives packets through the optical interface. After the unidirectional single-fiber function is
configured on interface 10GE2/0/2, you only need to connect the TX end of the optical
module on 10GE2/0/2 to the RX end of the optical module on the analysis server through one
fiber. Then the switch can transmit packets to the analysis server through a single fiber, and
the analysis server can receive packets through a single fiber. In addition, the TX end of the
optical module on the analysis server is not connected, so the analysis server cannot transmit
packets, ensuring data security on the server.

Figure 2-4 Networking diagram of unidirectional single-fiber communication

Traffic distribution
SwitchB device SwitchC

10GE2/0/1
SwitchA
TX RX
10GE2/0/2
Single
fiber

RX TX

Packet
analysis
Analyze packets
server
Optical module
Optical module TX end
Optical module RX end

Procedure
Step 1 Run:
system-view

The system view is displayed.

Step 2 Run:
interface interface-type interface-number

The interface view is displayed.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 59

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Step 3 Perform either of the following operations based on requirements: (The two commands
cannot be used together and the command configured later takes effect.)

l Run:
single-fiber enable

The unidirectional single-fiber communication function is configured.

By default, the unidirectional single-fiber communication function is disabled. After this
command is configured on an interface, the interface will be in Down state if no optical
module is installed on the interface or a single-fiber bidirectional optical module or high-
speed cable is installed.
l Run:
single-fiber rx

The single-fiber receiving function is configured.

By default, the single-fiber receiving function is disabled. After this command is
configured on an interface, the interface is in Up state only when both the local and
remote optical modules are installed and can properly receive optical signals. The
interface can only receive packets but cannot send packets. This configuration takes
effect only when the optical fiber is connected to the RX end of the local optical module.

Step 4 Run:
commit

The configuration is committed.

----End

2.7.4 Configuring the Training Function

Context
With the increase of transmission rate or frequency, attenuation of signal's high-frequency
components becomes increasingly severe. To guarantee transmission performance of signals,
it is necessary to compensate for signals, and commonly used compensation technologies are
pre-emphasis and balancing. The pre-emphasis technology increases high-frequency
components of signals at the transmitter end of transmission lines to compensate attenuation
during the transmission. However, the pre-emphasis technology increases crosstalk while
amplifying high-frequency components. To solve this problem, the balancing technology is
developed. The balancing technology is used at the receiver end of transmission lines to
function like a filter for filtering high-frequency crosstalk.

After the training function is enabled on an interface, the transmitter end exchanges frames
with the receiver end to automatically set the pre-emphasis and balancing parameters,
improving process efficiency of the two technologies. It should be noted that the negotiated
parameters for the training function are obtained based on the site environment. If the site
environment changes, for example, from high-temperature environment to low-temperature
environment, the parameters may be inaccurate. Therefore, error codes may occur when the
training function is enabled. The training function is optional in IEEE802.3 standards, and its
implementation on different types of products from various vendors may differ. When
connecting two devices, ensure that the training function must be enabled or disabled on both
ends simultaneously.

The following interfaces support the training disable configuration:

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 60

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

l A 40GE interface connecting to or pre-configured with a 40GE high-speed cable

l A QSFP28 interface connecting to or pre-configured with a 40GE high-speed cable
l A QSFP28 interface connecting to a 100GE high-speed cable and configured with the
speed 40000 command to work at the rate of 40 Gbit/s
40GE or QSFP28 interfaces on the following switches do not support the training disable
configuration: CE6855HI, CE6870EI and CE7855EI.

NOTE

l The training function must be enabled on interfaces at both ends of a link; otherwise, the interfaces
cannot work properly.
l The training function takes effect on an interface only when the interface works in auto-negotiation
mode. If you run the negotiation disable command to configure the interface to work in non-auto-
negotiation mode after the training function takes effect, the training function will become
ineffective.
l If a 40GE interface is configured as a physical member interface in a stack system, the training
function cannot be disabled. In V100R003C10 and earlier versions, if the training function is
disabled on a 40GE interface, the 40GE interface cannot be configured as a physical member
interface in a stack system. In V100R005 and later versions, if a 40GE interface is configured as a
physical member interface in a stack system, the training disable configuration will be
automatically deleted from the interface.
l When a training-disabled 40GE interface of a CE8860EI is connected to a 40GE interface of a
CE6851-48S6Q-HI, CE6850-48T6Q-HI or CE6855HI, the interfaces do not go Up. The interfaces
go Up only after the training function is enabled on the 40GE interface of the CE8860EI.
l When a 40GE interface on the CE88-D16Q subcard connects to a QSFP+ high-speed cable, you can
run the training disable command to disable the training function on a group of interfaces. Each
interface group contains two interfaces, such as interfaces 1 and 2, interfaces 3 and 4, and interfaces
15 and 16. The training function cannot be configured on one interface. This limitation does not
apply to a 100GE interface that connects to a QSFP+ high-speed cable on the CE8860EI. When a
40GE interface connects to a QSFP+ high-speed cable, pay attention to the following points:
l The training function can be enabled on a group of 40GE interfaces only when both interfaces
connect to QSFP+ high-speed cables and are not configured with interface split.
l To disable auto-negotiation on one of the 40GE interfaces, run the undo training disable
command to enable the training function on both interfaces, and run the negotiation disable
command to disable auto-negotiation. After auto-negotiation is disabled, the training disable
command cannot be configured.
l If the training disable command is configured on the two 40GE interfaces, and the QSFP+
high-speed cable on one interface is replaced with a 40GE optical module, the interface state
changes to DOWN(Transceiver type mismatch). The interface goes Up after the undo
training disable command is run.
l If the training disable command is configured on the 40GE interfaces, the pre-configured
medium type cannot be changed.
l The training disable command is mutually exclusive with the loopback internal, service
type tunnel, and service type trill unicast commands.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 61

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

NOTICE
In V100R001C00, V100R002C00, and V100R003C00SPC100, the training function is
disabled on an interface by default. Pay attention to the following points:
l If you upgrade a device from V100R001C00, V100R002C00, or V100R003C00SPC100
to a later version, the training function is enabled on an interface by default. To ensure
that the interface configuration remains unchanged, the device automatically generates
the training disable configuration on the interface to disable the training function. You
can install or preconfigure a transmission medium that supports the training function on
the interface, and run the undo training disable command to enable the training
function.
l If you roll back a device from V100R003C00SPC500 or a later version to
V100R001C00, V100R002C00, or V100R003C00SPC100, and the training function is
enabled on an interface and the interface is in Up state before the version rollback, the
interface cannot go Up after the version rollback.

Procedure
Step 1 Run:
system-view

The system view is displayed.

Step 2 Run:
interface interface-type interface-number

The interface view is displayed.

Step 3 Run:
training disable

The training function is disabled on the interface.

By default, the training function is enabled.
Step 4 Run:
commit

The configuration is committed.

----End

2.7.5 Configuring an Interface to Transit to Error-Down State

When the Receive Optical Power Is Low

Context
The device records the status of an interface as Error-Down when it detects that a fault occurs.
The interface in Error-Down state cannot receive or send packets and the interface indicator is
off.Low receive optical power is one of the reasons that cause an interface to transit to the
Error-Down state. Low receive optical power on an Ethernet optical interface may cause
problems such as packet loss. Because the interface is still in Up state, traffic is still

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 62

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

transmitted on the interface even if a backup link is configured. To avoid impact on services,
you can configure the interface to change to the Error-Down state when the receive optical
power is low. When the receive optical power on the interface falls below the lower alarm
threshold, the system disables the interface and records the interface status as ERROR
DOWN(transceiver-power-low). Services are then switched to the backup link immediately.

Procedure
Step 1 Run:
system-view

The system view is displayed.

Step 2 Run:
interface interface-type interface-number

The interface view is displayed.

Step 3 Run:
port transceiver-power-low trigger error-down

The interface is configured to transit to the Error-Down state when the receive optical power
is low.
By default, the Ethernet optical interface does not transit to the Error-Down state when the
receive optical power is low.
Step 4 Run:
commit

The configuration is committed.

----End

Checking the Configuration

Run the display error-down recovery [ interface interface-type interface-number ]
command in any view to check information about the interfaces in Error-Down state.

Follow-up Procedure
If an interface is in Error-Down state, you are advised to find out the cause first. For details,
see the possible causes and troubleshooting procedures.
An interface in Error-Down state can be recovered using either of the following methods:
l Manual recovery (after an Error-Down event occurs):
If a few interfaces need to be recovered, run the shutdown and undo shutdown
commands in the interface view. Alternatively, run the restart command in the interface
view to restart the interfaces.
l Automatic recovery (before an Error-Down event occurs):
If a large number of interfaces need to be recovered, manual recovery is time consuming
and some interfaces may be omitted. To avoid this problem, you can run the error-down
auto-recovery cause transceiver-power-low interval command in the system view to
enable automatic interface recovery and set the recovery delay time. You can run the
display error-down recovery command to view information about automatic interface
recovery.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 63

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

NOTE

This method does not take effect on interfaces that are already in Error-Down state. It takes effect
only on interfaces that enter the Error-Down state after this configuration is complete.

2.7.6 Pre-configuring the Transmission Medium Type on an

Optical Interface

Context
Some functions can be configured on an optical interface only after the interface connects to a
transmission medium (such as an optical module or copper module). Therefore, optical
interfaces must connect to transmission media before having these functions configured.
Sometimes the installation personnel and configuration personnel are different. The
configuration personnel can configure services only after the installation personnel install the
transmission media on interfaces. As a result, the service deployment time on interfaces is
lengthened. The configuration personnel can pre-configure the transmission medium type on
optical interfaces and configure functions on the interfaces. These functions take effect after
the installation personnel install the correct transmission media on the interfaces. In this way,
the configuration flexibility is improved and the service deployment time is shortened.

The transmission medium type pre-configuration function can be configured on an optical

interface only when the interface does not connect to a transmission medium or is shut down.
If the installed transmission medium type is the same as the pre-configured transmission
medium type, the interface has the condition to go Up and the configuration takes effect. If
the installed transmission medium type differs from the pre-configured transmission medium
type, the interface cannot go Up. To make the interface go Up, change the transmission
medium type or modify the conflict policy.

If the installed transmission medium type differs from the pre-configured transmission
medium type, and the two transmission media have incompatible configurations, the interface
status varies depending on the used transmission medium type as follows:
l If the installed transmission medium type is used, incompatible configurations (such as
auto-negotiation mode, rate, and duplex mode) on the interface are lost and the interface
goes Up.
l If the pre-configured transmission medium type is used, incompatible configurations on
the interface are saved and the interface goes Down. You can delete incompatible
configurations to make the interface go Up.

Procedure
Step 1 Run:
system-view

The system view is displayed.

Step 2 (Optional) Run:

device conflict-policy keep-config

The conflict policy is configured to be based on the pre-configured transmission medium

type.

By default, the conflict policy is based on the installed transmission medium type.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 64

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Step 3 Run:
interface interface-type interface-number

The interface view is displayed.

You can enter the corresponding interface view based on the transmission medium type to be
pre-configured.
Step 4 Run:
device transceiver transceiver-type

The transmission medium type is pre-configured on the interface.

By default, no transmission medium type is pre-configured on an interface.
Step 5 (Optional) Run:
undo shutdown

The interface is enabled.

Perform this step only when the optical interface is shut down.
Step 6 Run:
commit

The configuration is committed.

----End

Checking the Configuration

Run the display this command in the interface view to check the pre-configured transmission
medium.

Follow-up Procedure
If the installed transmission medium differs from the pre-configured transmission medium,
the interface status varies according to the configured conflict policy.
l If the conflict policy is based on the installed transmission medium type, the interface is
initialized based on the installed transmission medium and has the condition to go Up. In
addition, the configuration based on the pre-configured transmission medium type is
deleted.
l If the conflict policy is based on the pre-configured transmission medium type, the
configuration based on the pre-configured transmission medium type is retained on the
interface but the interface does not go Up. You need to install the pre-configured
transmission medium to make the interface have the condition to go Up and make the
configuration on the interface take effect. You can also make the interface have the
condition to go Up using the following methods, but the configuration based on the pre-
configured transmission medium type is deleted.
– Run the undo device conflict-policy keep-config command to change the conflict
policy to be based on the installed transmission medium type, and remove and
reinstall the transmission medium.
– Run the shutdown command to shut down the interface, run the undo device
transceiver command to delete the pre-configured transmission medium type, set
the transmission medium type to the installed transmission medium type, and run
the undo shutdown command.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 65

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

– Run the shutdown command to shut down the interface, run the undo device
transceiver command to delete the pre-configured transmission medium type, and
run the undo shutdown command. This method is not recommended because no
pre-configured transmission medium information exists on the interface.

NOTICE
l If the pre-configured transmission medium is a high-speed cable but an optical module is
installed, the optical module may be burned. Ensure that the installed transmission
medium is the same as the pre-configured transmission medium.
l If the installed optical module type is displayed as Unknown, the system cannot identify
the optical module. Only when the pre-configured transmission medium type and the
installed transmission medium type have the same bandwidth, the interface may go Up;
otherwise, the interface cannot go Up. To ensure that the interface works properly, you are
advised to use an optical module that is certified for Huawei Ethernet switches.

2.7.7 Configuring FEC

Context
NOTE

Only 25GE interfaces and QSFP28 100GE interfaces support this function.

Forward error correction (FEC) is a bit error correction technology that adds correction
information to data packets at the transmit end, and corrects bit errors generated during data
packet transmission at the receive end based on the correction information. FEC improves the
signal quality, but increases the signal transmission delay. You can enable or disable this
function based on actual requirements.
The switch supports three FEC modes: Base-R FEC, RS-FEC and None FEC.
l Base-R FEC: can be enabled or disabled only on 25GE interfaces (including 25GE
interfaces split from 100GE interfaces).
l RS-FEC: can be enabled or disabled on 100GE interfaces.
l None FEC: FEC is disabled.
FEC is a part of auto-negotiation on an interface. If auto-negotiation is enabled on an
interface, the local and remote interfaces negotiate to enable or disable FEC. If auto-
negotiation is disabled on an interface, the default FEC mode is used. You can configure RS-
FEC, Base-R FEC, or none FEC (disabling FEC) on an interface based on the interface's
support for FEC. If FEC is enabled at one end of a link, this function must also be enabled at
the other end of the link.
If an interface has been configured as a stack member interface, the FEC configuration
command cannot be run on the interface. Similarly, if the FEC configuration command has
been run on an interface, the interface cannot be configured as a stack member interface.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 66

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

NOTE

l The default FEC mode on an interface varies according to the device model and connection medium.
Interfaces at both ends of a link must work in the same FEC mode; otherwise, the interfaces do not
go Up. If their FEC modes are different, configure the same FEC mode on the interfaces when they
work in non-auto-negotiation mode. You can run the display interface command in any view or the
display this interface command in the interface view to check whether the FEC function is enabled
on an interface based on the Fec field in the command output.
l Disable auto-negotiation before changing the FEC mode on switches excluding the CE8860EI.

Procedure
l Enable the Base-R FEC function.
a. Run:
system-view

The system view is displayed.

b. Run:
interface interface-type interface-number

The interface view is displayed.

c. Run:
fec mode base-r

The Base-R FEC function is enabled.

By default, the Base-R FEC function is disabled on a 25GE interface.
d. Run:
commit

The configuration is committed.

l Enable the RS-FEC function.
a. Run:
system-view

The system view is displayed.

b. Run:
interface interface-type interface-number

The interface view is displayed.

c. Run:
fec mode rs

The RS-FEC function is enabled.

By default, the medium on a 100GE interface determines whether the RS-FEC
function is enabled on the interface.
d. Run:
commit

The configuration is committed.

l Disable the FEC function.
a. Run:
system-view

The system view is displayed.

b. Run:

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 67

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

interface interface-type interface-number

The interface view is displayed.

c. Run:
fec mode none

The FEC function is disabled.

NOTE

The fec mode none command disables the FEC function and the undo fec mode command
restores the default FEC mode. Therefore, if the FEC function is disabled on an interface
that has an optical module installed by default, you can run the fec mode none or undo fec
mode command to disable the FEC function on the interface.
d. Run:
commit

The configuration is committed.

----End

2.7.8 Configuring the Medium Attribute of a 25GE Interface

Context
NOTE

Only the CE8860EI supports this configuration.

Transmission media on interfaces are classified into copper media and fiber media. Copper
media include passive high-speed cables. Fiber media include optical fibers, active high-speed
cables, and active optical cables (AOCs).
The 24 SFP28 optical interfaces on a CE88-D24S2CQ card of the CE8860EI are divided into
six port groups, which contain interfaces 1-4, interfaces 5-8, interfaces 9-12, interfaces 13-16,
interfaces 17-20, and interfaces 21-24. The four interfaces in a group must have the same type
of media installed. For example, if interfaces 1-4 are configured to work in fiber mode, the
interfaces can only connect to optical fibers, active high-speed cables, or AOCs. If interface 2
connects to a passive high-speed cable, the interface does not go Up and is in
DOWN(Transceiver type mismatch) state.

Procedure
Step 1 Run:
system-view

The system view is displayed.

Step 2 Run:
interface 25ge interface-number

The 25GE interface view is displayed.

Step 3 (Optional) Run:
port mode 10g

The interface is configured to work at the rate of 10 Gbit/s.

By default, a 25GE interface works at the rate of 25 Gbit/s.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 68

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Step 4 Run:
port mode copper

The interface is configured to work at copper mode.

By default, a 25GE interface works in fiber mode. This configuration takes effect only for
25G and 10G media. The port mode copper and port mode ge commands are mutually
exclusive and cannot be configured together.
Step 5 Run:
commit

The configuration is committed.

----End

2.8 Configuring Attributes for an Electrical Interface

2.8.1 Setting the Duplex Mode

Context
There are the following two duplex modes:
l Half duplex mode: An interface in this mode only receives or sends data at a time within
a specified maximum transmission distance.
l Full duplex mode: An interface in this mode can receive and send data at the same time.
The maximum throughput in full duplex mode is double that in half duplex mode, and
the transmission distance is not limited.
If an interface is in auto-negotiation mode, it negotiates the duplex mode with the peer
interface. If the peer device does not support auto-negotiation, you can manually set the
duplex mode for the interface in non-auto-negotiation mode to ensure that the interface works
in the same duplex mode as the peer interface.
Only 10GE electrical interfaces on the CE6850-48T4Q-EI support the duplex mode
configuration.

Procedure
Step 1 Run:
system-view

The system view is displayed.

Step 2 Run:
interface 10ge interface-number

The 10GE electrical interface view is displayed.

Step 3 Run:
negotiation disable

The interface is set to work in non-auto-negotiation mode.

By default, a 10GE electrical interface works in auto-negotiation mode.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 69

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Step 4 Run:
speed 100

The working rate of the interface in non-auto-negotiation mode is set to 100 Mbit/s.
When a 10GE electrical interface works in non-auto-negotiation mode, the default working
rate of the interface is 10000 Mbit/s.
Step 5 Run:
duplex { full | half }

The duplex mode is set for the interface.

When a 10GE electrical interface works in non-auto-negotiation mode, the default duplex
mode of the interface is full duplex.
Step 6 Run:
commit

The configuration is committed.

----End

Checking the Configuration

Run the display interface [ interface-type [ interface-number ] ] command in any view or run
the display this interface command in the interface view to check the current running status
of the interface. For details, view the Duplex field in the command output.

2.9 Configuring Port Isolation

Context
To implement Layer 2 isolation between interfaces, you can add different interfaces to
different VLANs. This wastes VLAN resources. Port isolation can isolate interfaces in the
same VLAN. That is, you only need to add interfaces to a port isolation group to implement
Layer 2 isolation between these interfaces. Port isolation provides secure and flexible
networking schemes.
Figure 2-5 shows the port isolation method and application scenario. PC1, PC2, and PC3
belong to VLAN 10. After GE1/0/1 on PC1 and GE1/0/2 on PC2 are added to the port
isolation group, PC1 and PC2 cannot communicate with each other in VLAN 10, but they can
communicate with PC3.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 70

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Figure 2-5 Networking diagram of port isolation

Switch

GE1/0/1 GE1/0/3
Port isolation
group

GE1/0/2
X

PC1 PC2 PC3

10.10.10.1/24 10.10.10.2/24 10.10.10.3/24

VLAN 10

Send packets

Unidirectional port isolation can be configured in certain scenarios. When multiple hosts
connect to different interfaces of the same device, if a certain host with security risks sends
many broadcast packets to the other hosts, you can isolate the host from the other hosts
unidirectionally to prevent the other hosts from receiving packets from the host.

As shown in Figure 2-6, assume that PC4 with security risks sends many broadcast packets to
the other hosts. You can configure unidirectional isolation from GE1/0/5 and GE1/0/6 only on
GE1/0/4. In this way, the broadcast packets sent by PC4 cannot reach PC5 and PC6, but the
broadcast packets sent by PC5 and PC6 can reach PC4.

Figure 2-6 Networking diagram of port isolation

Switch

Configure
unidirectional GE1/0/4 GE1/0/6
isolation
GE 1/0/5

X
X

PC4 PC5 PC6

10.10.10.1/24 10.10.10.2/24 10.10.10.3/24

VLAN10

Send packets

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 71

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

NOTE

l CE6870EI only supports configure port isolation group, does not support unidirectional isolation .
l Only Layer 2 interfaces support port isolation.
l Port isolation takes effect only for the port isolation group members on the same device. This
function does not take effect on the interfaces of different devices.
l Port isolation can be configured for interfaces on the same switch or on different switches in a stack
system. In versions earlier than V200R001C00, port isolation does not take effect for Layer 2
packets that are encapsulated through TRILL and VXLAN tunnels. In V200R001C00 and later
versions, port isolation takes effect for Layer 2 packets that are encapsulated at the VXLAN service
access side on the CE6870EI, and for TRILL packets that are encapsulated at the access side. Port
isolation for interfaces on different switches in a stack system has been supported since
V100R003C10.
l In V100R005C10, if 256, 512, or 1024 link aggregation groups are configured in an SVF system
consisting of fixed switches, port isolation cannot be configured on Eth-Trunks. In V100R006C00
and later versions, if a leaf switch is a CE5810EI switch and 256, 512, or 1024 link aggregation
groups are configured in an SVF system consisting of fixed switches, the CE5810EI leaf switch does
not support the port isolation configuration on Eth-Trunks. This limitation does not apply to other
leaf switches.

Procedure
l Configure a port isolation group.
a. Run:
system-view

The system view is displayed.

b. Run:
interface interface-type interface-number

The Ethernet interface view is displayed.

c. Run:
port-isolate enable group group-id

Port isolation is enabled.

By default, port isolation is disabled.
d. Run:
commit

The configuration is committed.

l Configure unidirectional isolation on interfaces.
a. Run:
system-view

The system view is displayed.

b. Run:
interface interface-type interface-number

The Ethernet interface view is displayed.

c. Run:
am isolate { interface-type interface-number }&<1-8> or am isolate
interface-type interface-number1 [ to interface-number2 ]

Unidirectional port isolation is configured.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 72

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

By default, unidirectional port isolation is not configured.

NOTE

If interface A is isolated from interface B unidirectionally, packets sent from interface A

cannot reach interface B, but packets sent from interface B can reach interface A.
Interfaces in a port isolation group are isolated from each other, but interfaces in different
port isolation groups can communicate. To isolate interfaces in different port isolation
groups, configure unidirectional isolation between these interfaces.
d. Run:
commit

The configuration is committed.

----End

Checking the Configuration

Run the display port-isolate group { group-id | all } command in any view to check the
configuration of the interface isolation group.

2.10 Maintenance
You can detect the cable status, perform loopback tests on an Ethernet interface to check
whether the interface forwarding is normal, and clear traffic statistics on the Ethernet
interface.

2.10.1 Configuring VCT

Context
Virtual cable test (VCT) technology uses time domain reflectometry (TDR) to detect the cable
status. When a pulse is transmitted to the end of a cable or a failure point in the cable, some
pulse energies are reflected to the transmit end. The VCT algorithm measures the time spent
on transmitting pulses over cables, reaching the failure point, and returning the pulses. The
measured time is converted to the distance.
As shown in Figure 2-7, GE1/0/1 on SwitchA connects to GE1/0/2 on SwitchB with a
network cable. There is a failure point in the network cable. After VCT is configured on
GE1/0/1, the system generates pulse signals. After the pulse signals reach the failure point,
some energies are reflected to the transmit end. In this example, the distance between
SwitchA and the failure point is L, and the period from sending pulse signals to receiving
reflected pulse signals is T, the transmission rate of pulse signals in the cable is V. The
formula used to calculate the distance between the failure point and GE1/0/1 is as follows:
L = (V x T)/2

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 73

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Figure 2-7 VCT

SwitchA SwitchB

GE1/0/1 Failure point GE1/0/2

L
Pulse signal

VCT can detect the fault type of a network cable and identify failure points to help locate
network cable faults.

Procedure
Step 1 Run:
system-view

The system view is displayed.

Step 2 Run:
interface interface-type interface-number

The interface view is displayed.

Step 3 Run:
virtual-cable-test

The VCT function is configured.

NOTE

l Electrical interfaces and 10GE optical interfaces connected to an SFP-GE electrical interface module
support the setting.
l When 10GE optical interfaces use SFP-GE copper modules and the interfaces on the two ends of the
link are both Up, the test result may be not accurate and is only for reference.
l The test result is only for reference and may be inaccurate for cables from some vendors.
l Running the virtual-cable-test command may affect services on the interface in a short period of
time.
l Before configuring VCT, shut down the remote interface or remove the network cable between the
local and remote interfaces. Otherwise, signals from the remote interface may make the test result
inaccurate.

----End

Checking the Configuration

Run the display interface interface-type interface-number virtual-cable-test command to
check the latest cable test result on an Ethernet interface.

2.10.2 Configuring Internal Loopback Detection

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 74

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Context
When testing some special functions such as locating an Ethernet fault, you must enable
loopback detection on the Ethernet interface to check whether the interface is working
properly. When no fault occurs on the Ethernet interface, the physical interface status is
always Up after loopback detection is enabled. If a fault occurs on the Ethernet interface, the
interface status is Down.

NOTICE
l The loopback detection function (configured using the loopback command) affects other
functions; therefore, the interface or link cannot work properly. When the test is complete,
you must run the undo loopback command in real time to disable loopback. The original
configuration is restored after loopback detection is disabled.

Procedure
Step 1 Run:
system-view

The system view is displayed.

Step 2 Run:
interface interface-type interface-number

The interface view is displayed.

Step 3 Run:
loopback internal

Internal loopback detection is configured on the Ethernet interface.

By default, loopback detection is disabled on an Ethernet interface.

Step 4 Run:
commit

The configuration is committed.

----End

Checking the Configuration

Run the display interface [ interface-type [ interface-number ] ] command in any view or the
display this interface command in the interface view to check the current interface status.
For details, check the Loopback field in the command output.

2.10.3 Clearing Interface Statistics

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 75

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Context
To monitor the status of an interface or locate faults on the interface, you can collect traffic
statistics about the interface. Before collecting traffic statistics on an Ethernet interface within
a period, delete the existing traffic statistics on this interface.

NOTICE
The traffic statistics on an Ethernet interface cannot be restored after they are cleared.
Exercise caution before clearing the statistics.

Procedure
l Run the reset interface counters [ interface-type [ interface-number ] ] command to
clear the statistics on the specified interface.
l Run the reset interface [ interface-type interface-number ] virtual-cable-test command
to clear the cable test results of the interface.
l Run the reset ip host packet statistics receive protocol interface interface-type
interface-number command to clears statistics about protocol-specific packets received
on the interface.
l Run the reset ip host packet statistics receive command to clear statistics about packets
received on all interfaces.
----End

2.11 Configuration Examples

This section only provides configuration examples for single features. For details about multi-
feature configuration cases, feature-specific configuration cases, interconnection cases,
protocol or hardware replacement cases, and industry application cases, see the Typical
Configuration Cases.

2.11.1 Example for Configuring Interface Isolation

Networking Requirements
As show in Figure 2-8, Server1, Server2, and Server3 belong to VLAN 10. Server1 and
Server2 are not allowed to communicate with each other in VLAN10, but they are allowed to
communicate with Server3.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 76

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

Figure 2-8 Networking diagram of interface isolation configuration

Switch

10GE1/0/1 10GE1/0/3

Server1 10GE1/0/2
Server2 Server3
10.10.10.1/24 10.10.10.2/24 10.10.10.3/24

VLAN 10

Configuration Roadmap
The configuration roadmap is as follows:
1. Add the interfaces to the specified VLAN.
2. By default, interfaces are isolated at Layer 2 but can communicate at Layer 3. You can
add interfaces to an isolation group to implement Layer 2 isolation between these
interfaces.

Procedure
Step 1 Create VLAN 10 on Switch, and add interfaces to VLAN 10.
<HUAWEI> system-view
[~HUAWEI] sysname Switch
[*HUAWEI] commit
[~Switch] vlan 10
[*Switch-vlan10] quit
[*Switch] interface 10ge 1/0/1
[*Switch-10GE1/0/1] port default vlan 10
[*Switch-10GE1/0/1] quit
[*Switch] interface 10ge 1/0/2
[*Switch-10GE1/0/2] port default vlan 10
[*Switch-10GE1/0/2] quit
[*Switch] interface 10ge 1/0/3
[*Switch-10GE1/0/3] port default vlan 10
[*Switch-10GE1/0/3] quit
[*Switch] commit

Step 2 Configure interface isolation.

# Configure interface isolation for 10GE1/0/1.
[~Switch] interface 10ge 1/0/1
[~Switch-10GE1/0/1] port-isolate enable group 1
[*Switch-10GE1/0/1] quit
[*Switch] commit

# Configure interface isolation for 10GE1/0/2.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 77

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 2 Ethernet Interface Configuration

[~Switch] interface 10ge 1/0/2

[~Switch-10GE1/0/2] port-isolate enable group 1
[*Switch-10GE1/0/2] quit
[*Switch] commit

Step 3 Verify the configuration.

# Server1 and Server2 cannot communicate with each other.
# Server1 and Server3 can communicate with each other.
# Server2 and Server3 can communicate with each other.

----End

Configuration Files
Configuration file of Switch
#
sysname Switch
#
vlan batch 10
#
interface 10GE1/0/1
port default vlan 10
port-isolate enable group 1
#
interface 10GE1/0/2
port default vlan 10
port-isolate enable group 1
#
interface 10GE1/0/3
port default vlan 10
#
return

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 78

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 3 Logical Interface Configuration

3 Logical Interface Configuration

About This Chapter

3.1 Logical Interface Overview

3.2 Configuration Notes
This section provides the points of attention when configuring logical interface.
3.3 Configuring a Sub-interface
3.4 Configuring a Loopback Interface
3.5 Configuring a NULL Interface
3.6 Configuration Examples

3.1 Logical Interface Overview

This section describes logical interfaces supported by CloudEngine 8800&7800&6800&5800
Series.

Table 3-1 Logical interfaces

Interface Description Configuration Reference

Eth-Trunk An Eth-Trunk is formed by binding For details about how to configure

multiple Ethernet interfaces to provide an Eth-Trunk, see Link
more bandwidth and higher Aggregation Configuration in the
transmission reliability. CloudEngine
8800&7800&6800&5800 Series
Switches Configuration Guide -
Ethernet Switching.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 79

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 3 Logical Interface Configuration

Interface Description Configuration Reference

VLANIF A VLANIF interface has Layer 3 For details about how to configure
interface features and enables VLANs to a VLANIF interface, see VLAN
communicate after being assigned an IP Configuration in the CloudEngine
address. 8800&7800&6800&5800 Series
Switches Configuration Guide -
Ethernet Switching.

Layer 2 On VXLAN networks, service access For details about how to configure
sub- points are Layer 2 sub-interfaces. You a Layer 2 sub-interface, see
interface can configure different flow Configuring a Service Access
encapsulation modes on the Layer 2 Point in the CloudEngine
sub-interfaces, so that the sub-interfaces 8800&7800&6800&5800 Series
can process different data packets. Switches Configuration Guide -
DCN and Server Management.

Layer 3 A Layer 3 sub-interface is a virtual -

sub- interface configured on a main interface
interface to implement inter-VLAN
communication.
NOTE
l Before configuring a Layer 3 sub-
interface on a Layer 2 physical interface
or Layer 2 Eth-Trunk, run the undo
portswitch command to switch the
interface to a Layer 3 interface.
l The MAC address of a sub-interface is
the same as that of the main interface.
l A sub-interface can be assigned an IP
address.
l Unless otherwise stated, all sub-
interfaces mentioned in this chapter are
Layer 3 sub-interfaces.

Stack A stack interface is a logical interface For details about how to configure
interface used to implement the stacking a stack interface, see Stack
function. The stacking function Configuration in the CloudEngine
virtualizes multiple switches into a 8800&7800&6800&5800 Series
logical switch to provide massive data Switches Configuration Guide -
forwarding and improve network Device Management.
reliability in data centers.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 80

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 3 Logical Interface Configuration

Interface Description Configuration Reference

Loopback A loopback interface is always Up and -

interface can be configured with a 32-bit subnet
mask.
There is one unique loopback interface
called the InLoopBack0 interface,
which is automatically created at
startup. The InLoopback0 interface is a
special and fixed loopback interface. It
uses the fixed loopback address
127.0.0.1/8 to receive data packets
destined for the host where the
InLoopBack0 interface resides. The
loopback address of the InLoopBack0
interface is not advertised.

NULL A NULL interface used in route -

interface filtering because any network packets
sent to this interface are discarded.

Tunnel A tunnel interface has Layer 3 features, For details about how to configure
interface transmits packets, and identifies and a tunnel interface, see IPv6
processes packets transmitted over a Transition Technology
tunnel. Configuration in the CloudEngine
8800&7800&6800&5800 Series
Switches Configuration Guide - IP
Service and GRE Configuration
Guide in the CloudEngine
8800&7800&6800&5800 Series
Switches Configuration Guide -
VPN.

FCoE If Fibre Channel over Ethernet (FCoE) For details about how to configure
interface is configured, devices communicate and an FCoE interface, see FC and
transmit traffic through FCoE FCoE Configuration in the
interfaces. Therefore, you need to CloudEngine
configure FCoE logical interfaces on 8800&7800&6800&5800 Series
the devices. Switches Configuration Guide -
DCN and Server Management.

Fabric A Fabric interface is used as the logical For details about how to configure
interface interface of an SVF system to forward a Fabric interface, see SVF
SVF data or protocol packets. Configuration in the CloudEngine
8800&7800&6800&5800 Series
Switches Configuration Guide -
Virtualization Configuration.

NVE An NVE interface is the logical For details about how to configure
interface interface to establish VXLAN tunnels an NVE interface, see VXLAN
with other NVE devices. Configuration in the CloudEngine
8800&7800&6800&5800 Series
Switches Configuration Guide -
DCN and Server Management.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 81

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 3 Logical Interface Configuration

Interface Description Configuration Reference

VBDIF A VBDIF interface is the virtual For details about how to configure
interface interface based on a BD to support a VBDIF interface, see VXLAN
Layer 3 features and implement Configuration in the CloudEngine
communication between different BDs, 8800&7800&6800&5800 Series
between BD and non-BD networks, and Switches Configuration Guide -
between BDs and Layer 3 networks. DCN and Server Management.

FC An FC interface is used to connect to an For details about how to configure

interface FC switch, an FC SAN storage, or a an FC interface, see FC and FCoE
server with FC interfaces. Configuration in the CloudEngine
NOTE 8800&7800&6800&5800 Series
Only 10GE Ethernet interfaces on the Switches Configuration Guide -
CE6850U-HI support FC interfaces, and DCN and Server Management.
supported in V100R005C10 and later
versions.

Ethernet An Ethernet interface is used for For detailed configuration, see

communication between the VRP Open System Usage Guide in the
system and open system. CloudEngine
8800&7800&6800&5800 Series
Switches .

3.2 Configuration Notes

This section provides the points of attention when configuring logical interface.

Involved Network Elements

No other NEs are required.

License Support
An interface is a basic feature of a switch and is not under license control.

Version Support

Table 3-2 Minimum version supporting various types of logical interfaces

Seri Product Layer Layer 3 Sub- VLANIF Interface/ Fabric

es 2 Sub- interface Eth-Trunk/Stack Interface
interf Interface
ace

CE8 CE8860EI V100R V100R006C00 V100R006C00 Not

800 006C0 support
0

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 82

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 3 Logical Interface Configuration

Seri Product Layer Layer 3 Sub- VLANIF Interface/ Fabric

es 2 Sub- interface Eth-Trunk/Stack Interface
interf Interface
ace

CE8850EI V200R V200R002C50 V200R002C50 Not

002C5 support
0

CE7 CE7850-32Q-EI V100R V100R003C10 V100R003C00 V100R00

800 005C0 5C00
0

CE7855-32Q-EI V200R V200R001C00 V200R001C00 V200R00

001C0 1C00
0

CE6 CE6810-48S4Q- Not V100R005C00 V100R003C00 V100R00

800 EI support 5C00

CE6810-48S-LI/ Not Not support V100R003C10 Not

CE6810-48S4Q- support support
LI

CE6810-24S2Q- Not Not support V100R005C10 Not

LI/ support support
CE6810-32T16S4
Q-LI

CE6850-48S4Q- Not V100R005C00 V100R001C00 V100R00

EI support 3C00

CE6850-48T4Q- Not V100R005C00 V100R001C00 Not

EI support support

CE6850-48S6Q- V100R005C00 V100R00

HI 5C10

CE6850-48T6Q- V100R V100R005C10 V100R005C10 V200R00

HI 005C1 2C50
0

CE6851-48S6Q- V100R005C10 V100R00

HI 6C00

CE6850UHI V100R005C10

CE6855-48S6Q- V200R001C00
HI

CE6855-48T6Q- V200R001C00 Not

HI support

CE6860EI V200R V200R002C50 V200R002C50 Not

002C5 support
0

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 83

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 3 Logical Interface Configuration

Seri Product Layer Layer 3 Sub- VLANIF Interface/ Fabric

es 2 Sub- interface Eth-Trunk/Stack Interface
interf Interface
ace

CE6870-24S6CQ- V200R001C00 Not

EI/ support
CE6870-48S6CQ-
EI

CE6870-48T6CQ V200R002C50 Not

-EI support

CE6880EI V200R V200R002C50 V200R002C50 Not

002C5 support
0

CE5 CE5810EI Not V100R005C00 V100R002C00 Not

800 support support

CE5850-48T4S2 Not V100R005C00 V100R001C00 Not

Q-EI support support

CE5850-48T4S2 Not V100R005C00 V100R003C00 Not

Q-HI support support

CE5855EI Not V100R005C10 V100R005C10 Not

support support

S Produc Loopback Tunnel NVE VBDIF FCoE Ether

er t Interface/Null Interface Interfa Interf net
ie Interface ce ace Interf
s ace

C CE8860 V100R006C00
E EI
88
00 CE8850 V200R002C50
EI

C CE7850 V100R003C00 V100R003C V100R005C00 V100R

E -32Q-EI 00 006C0
78 0
00
CE7855 V200R001C00 V200R001C V200R0 V200R V200R V200R
-32Q-EI 00 01C00 001C00 001C0 001C0
0 0

C CE6810 V100R003C00 V100R003C Not Not V100R Not

E -48S4Q- 00 support support 005C0 suppor
68 EI 0 t
00

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 84

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 3 Logical Interface Configuration

S Produc Loopback Tunnel NVE VBDIF FCoE Ether

er t Interface/Null Interface Interfa Interf net
ie Interface ce ace Interf
s ace

CE6810 V100R003C10 Not support Not Not Not Not

-48S-LI/ support support suppor suppor
CE6810 t t
-48S4Q-
LI

CE6810 V100R005C10 Not support Not Not Not Not

-24S2Q- support support suppor suppor
LI/ t t
CE6810
-32T16S
4Q-LI

CE6850 V100R001C00 V100R001C Not Not V100R V100R

EI 00 support support 005C0 006C0
0 0

CE6850 V100R005C00 V100R

-48S6Q- 006C0
HI 0

CE6850 V100R005C10 V100R

-48T6Q- 006C0
HI/ 0
CE6851
-HI

CE6850 V100R005C10 V100R

U-HI 006C0
0

CE6855 V200R001C00
HI

CE6860 V200R002C50 Not

EI suppor
t

CE6870 V200R001C00 V200R001C V200R0 V200R V200R V200R

-24S6C 00 01C00 001C00 001C0 001C0
Q-EI/ 0 0
CE6870
-48S6C
Q-EI

CE6870 V200R002C50
-48T6C
Q-EI

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 85

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 3 Logical Interface Configuration

S Produc Loopback Tunnel NVE VBDIF FCoE Ether

er t Interface/Null Interface Interfa Interf net
ie Interface ce ace Interf
s ace

CE6880 V200R002C50 V200R002C V200R0 Not Not Not

EI 50 02C50 support suppor suppor
t t

C CE5810 V100R002C00 V100R002C Not Not Not Not

E EI 00 support support suppor suppor
58 t t
00
CE5850 V100R001C00 V100R001C Not Not Not Not
-48T4S2 00 support support suppor suppor
Q-EI t t

CE5850 V100R003C00 V100R003C Not Not Not Not

-48T4S2 00 support support suppor suppor
Q-HI t t

CE5850 V100R005C10 Not support Not Not Not Not

EI support support suppor suppor
t t

NOTE

Only the CE6850U-HI supports FC interfaces in V100R005C10 and later versions.

Feature Dependencies and Limitations

In V200R002C50 and earlier versions, when devices are dual-homed to gateways through M-
LAG, the M-LAG master and slave devices are configured with many sub-interfaces, and
optical interfaces on the M-LAG master and slave devices are connected to copper transceiver
modules, restarting the M-LAG master or slave device may cause packet loss for a long time.
In this case, you are advised to manually switch traffic to the other M-LAG device and
upgrade and restart it.

3.3 Configuring a Sub-interface

A sub-interface is a virtual interface configured on a physical interface to allow users belong
to different VLANs to communicate with each other.

NOTE

If a CE6810LI functions as a standalone switch or a leaf switch in an SVF system consisting of fixed
and modular switches, the CE6810LI does not support sub-interfaces. If a CE6810LI functions as a leaf
switch in an SVF system consisting of fixed switches, the CE6810LI supports sub-interfaces.
If the system resource mode is set to large ARP or super ARP using the system resource { large-arp |
super-arp } command, Layer 3 sub-interfaces configured on the CE5855EI do not take effect.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 86

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 3 Logical Interface Configuration

3.3.1 Configuring an Ethernet Sub-interface

Context
Users belong to different VLANs and are located on different network segments can
communicate with each other by configuring sub-interfaces.

As shown in Figure 3-1, PC1 and PC2 belong to VLAN 10 and VLAN 20 respectively and
they are located on the different network segments. PC1 and PC2 connect to the Layer 3
device SwitchB through the Layer 2 device SwitchA. If only one IP address is configured for
GE1/0/4 on SwitchB, PC1 and PC2 cannot communicate across VLANs. Although you can
configure a secondary IP address to solve this problem, the secondary IP address becomes
invalid after the primary IP address becomes invalid.

In this case, you can configure sub-interfaces on GE1/0/4. A sub-interface is independent

from a main interface. Although a sub-interface uses the same physical parameters as the
main interface, the IP address of the sub-interface is not affected by the change of the main
interface IP address. For example, you can create two sub-interfaces GE1/0/4.10 and
GE1/0/4.20 and specify IP addresses 10.10.10.10/24 and 10.20.20.20/24 for them. The two IP
addresses server as the gateway addresses for users in VLAN 10 and VLAN 20 respectively.
In this way, PC1 and PC2 on different network segments can communicate across VLANs.

Figure 3-1 Configuring an Ethernet sub-interface

Internet

SwitchB
GE1/0/4

GE1/0/4.10 GE1/0/4.20
10.10.10.10/24 10.20.20.20/24
VLAN 10 VLAN 20

GE1/0/3
SwitchA

GE1/0/1 GE1/0/2

PC1 PC2
10.10.10.1/24 10.20.20.2/24
VLAN 10 VLAN 20

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 87

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 3 Logical Interface Configuration

Pre-configuration Tasks
Before configuring an Ethernet sub-interface, complete the following tasks:

l Correctly connecting the physical interface of the sub-interface

l Configuring a Layer 3 main interface

Procedure
Step 1 Run:
system-view

The system view is displayed.

Step 2 Run:
interface interface-type interface-number.subinterface-number

The Ethernet sub-interface view is displayed.

subinterface-number specifies the number of a sub-interface after the Ethernet interface is

switched to Layer 3 mode.

Step 3 Run:
ip address ip-address { mask | mask-length } [ sub ]

The IP address of the Ethernet sub-interface is configured.

NOTE

Specify the sub parameter when you need to configure a secondary IP address for the Ethernet sub-
interface.

Step 4 Run:
dot1q termination vid vid

The sub-interface is configured to terminate single-tagged packets.

Step 5 Run:
commit

The configuration is committed.

----End

Checking the Configuration

Run the display interface [ interface-type [ interface-number [.subnumber ] ] ] command to
check the status of the specified Ethernet sub-interface.

3.3.2 Configuring an Eth-Trunk Sub-interface

Context
Sub-interfaces can be configured on a Layer 3 Eth-Trunk. When Layer 3 devices connect to
Layer 2 devices in different VLANs through the Layer 3 Eth-Trunk, sub-interfaces must be
configured on the Eth-Trunk to identify packets from different VLANs and to enable users in
different VLANs to communicate with each other.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 88

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 3 Logical Interface Configuration

Procedure
Step 1 Run:
system-view

The system view is displayed.

Step 2 Run:
interface eth-trunk trunk-id

An Eth-Trunk is created and the Eth-Trunk interface view is displayed.

Step 3 Run:
undo portswitch

A Layer 3 Eth-Trunk is configured.

Step 4 Run:
quit

The system view is displayed.

Step 5 Run:
interface eth-trunk trunk-id.subnumber

An Eth-Trunk sub-interface is created.

subnumber specifies the number of a sub-interface. The value ranges from 1 to 4094.
Step 6 Run:
ip address ip-address { mask | mask-length } [ sub ]

An IP address is configured for the sub-interface.

When configuring multiple IP addresses for an Eth-Trunk sub-interface, use the sub keyword
to indicate the IP addresses configured after the first one.
Step 7 Run:
dot1q termination vid vid

The Eth-trunk sub-interface is configured to terminate single-tagged packets.

Step 8 Run:
commit

The configuration is committed.

----End

3.4 Configuring a Loopback Interface

Context
The loopback interface has the following features:
l The loopback interface is always Up and has the loopback feature.
l The loopback interface can be configured with the mask of all 1s.
Based on the preceding features, the loopback interface has the following applications.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 89

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 3 Logical Interface Configuration

l The IP address of a loopback interface is specified as the source address of packets to

improve network reliability.
l When no Router ID is configured for dynamic routing protocols, the maximum IP
address of the loopback interface is configured as the router ID automatically.
l The loopback interface is configured as the source interface to send boarder gateway
protocol (BGP) packets to ensure that the BGP session functions correctly when the
physical interface is faulty.
l The loopback interface can be configured with the mask of 255.255.255.255 to save IP
address resources.

Pre-Configuration
Before configuring a loopback interface, complete the following task:
l Powering on the device and performing self-check

Procedure
Step 1 Run:
system-view

The system view is displayed.

Step 2 Run:
interface loopback loopback-number

A loopback interface is created and the loopback interface view is displayed.

You can create or delete loopback interface dynamically. After a loopback interface is created,
the link-layer protocol is always up unless the loopback interface is deleted.
Step 3 Run:
ip address ip-address { mask | mask-length }

The IP address of the loopback interface is configured.

Step 4 Run:
commit

Commit the configuration.

----End

Checking the Configuration

l Run the display interface loopback [ loopback-number ] command to check the status
of a loopback interface.

3.5 Configuring a NULL Interface

Context
A NULL0 interface is created automatically. The NULL0 interface is always Up and cannot
forward packets. Any packets sent to the NULL0 interface are discarded. If the next hop of a

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 90

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 3 Logical Interface Configuration

static route to a network segment is a null interface, all the data packets destined for this
network segment are discarded. Therefore, the packets that you want to filter out can be sent
to the NULL0 interface directly without configuring the access control list.
For example, run the following static route configuration command to discard packets sent to
the network segment of 192.168.0.0.
[~HUAWEI] ip route-static 192.168.0.0 255.255.255.0 NULL 0

Pre-Configuration
Before configuring a NULL interface, complete the following task:
l Powering on the device and performing self-check

Procedure
Step 1 Run:
system-view

The system view is displayed.

Step 2 Run:
interface null 0

The NULL interface view is displayed.

The NULL interface stays in the Up state. It cannot forward data packets. You cannot
configure an IP address for it or encapsulate it with protocols.

----End

Checking the Configuration

l Run the display interface null [ 0 ] command to check the status of a null interface.

3.6 Configuration Examples

This section only provides configuration examples for single features. For details about multi-
feature configuration cases, feature-specific configuration cases, interconnection cases,
protocol or hardware replacement cases, and industry application cases, see the Typical
Configuration Cases.

3.6.1 Example for Configuring Ethernet Sub-interfaces to

Implement Inter-VLAN Communication

Networking Requirements

As shown in Figure 3-2, PC1 and PC2 belong to VLAN 10 and VLAN 20 respectively and
are on different network segments. They connect to a Layer 3 switch SwitchB through a
Layer 2 switch SwitchA. It is required that physical interface resources be saved and PC1 and
PC2 communicate through a Layer 3 physical interface GE1/0/4 on SwitchB.

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 91

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 3 Logical Interface Configuration

Figure 3-2 Networking diagram for configuring Ethernet sub-interfaces to implement inter-
VLAN communication

Internet

SwitchB
GE1/0/4

GE1/0/4.10 GE1/0/4.20
10.10.10.10/24 10.20.20.20/24
VLAN 10 VLAN 20

GE1/0/3
SwitchA

GE1/0/1 GE1/0/2

PC1 PC2
10.10.10.1/24 10.20.20.2/24
VLAN 10 VLAN 20

Configuration Roadmap
The configuration roadmap is as follows:
1. Add interfaces on SwitchA to VLANs.
2. Switch a Layer 2 Ethernet interface on SwitchB to Layer 3 mode. Before creating sub-
interfaces on an Ethernet interface, switch the Ethernet interface to Layer 3 mode first.
3. Create sub-interfaces on SwitchB as gateway interfaces for network segments to
implement inter-VLAN communication.

Procedure
Step 1 On SwitchA, create VLAN 10 and VLAN 20, and add GE1/0/1 to VLAN 10, GE1/0/2 to
VLAN 20, and GE1/0/3 to VLAN 10 and VLAN 20.
<HUAWEI> system-view
[~HUAWEI] sysname SwitchA
[*HUAWEI] commit
[~SwitchA] vlan batch 10 20
[*SwitchA] commit
[~SwitchA] interface ge 1/0/1
[~SwitchA-GE1/0/1] port default vlan 10
[*SwitchA-GE1/0/1] quit
[*SwitchA] interface ge 1/0/2
[*SwitchA-GE1/0/2] port default vlan 20
[*SwitchA-GE1/0/2] quit

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 92

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 3 Logical Interface Configuration

[*SwitchA] interface ge 1/0/3

[*SwitchA-GE1/0/3] port link-type trunk
[*SwitchA-GE1/0/3] port trunk allow-pass vlan 10 20
[*SwitchA-GE1/0/3] quit
[*SwitchA] commit

Step 2 On SwitchB, switch GE1/0/4 to Layer 3 mode.

<HUAWEI> system-view
[~HUAWEI] sysname SwitchB
[*HUAWEI] commit
[~SwitchB] interface ge 1/0/4
[~SwitchB-GE1/0/4] undo portswitch
[*SwitchB-GE1/0/4] quit
[*SwitchB] commit

Step 3 On SwitchB, create sub-interfaces GE1/0/4.10 and GE1/0/4.20 and use them as the gateway
interfaces for network segments 10.10.10.0/24 and 10.20.20.0/24 respectively. Assign IP
addresses to the sub-interfaces and configure the sub-interfaces to terminate single-tagged
packets.
[~SwitchB] interface ge 1/0/4.10
[*SwitchB-GE1/0/4.10] ip address 10.10.10.10 24
[*SwitchB-GE1/0/4.10] dot1q termination vid 10
[*SwitchB-GE1/0/4.10] quit
[*SwitchB] interface ge 1/0/4.20
[*SwitchB-GE1/0/4.20] ip address 10.20.20.20 24
[*SwitchB-GE1/0/4.20] dot1q termination vid 20
[*SwitchB-GE1/0/4.20] quit
[*SwitchB-GE1/0/4] commit

Step 4 Verify the configuration.

# After the configuration is complete, perform ping operations on PC1 and PC2. PC1 and PC2
can communicate with each other.

----End

Configuration Files
l Configuration file of SwitchA
#
sysname SwitchA
#
vlan batch 10 20
#
interface GE1/0/1
port default vlan 10
#
interface GE1/0/2
port default vlan 20
#
interface 10GE1/0/3
port link-type trunk
port trunk allow-pass vlan 10 20
#
return

l Configuration file of SwitchB

#
sysname SwitchB
#
interface GE1/0/4
undo portswitch
#
interface GE1/0/4.10
ip address 10.10.10.10 255.255.255.0
dot1q termination vid 10

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 93

Copyright © Huawei Technologies Co., Ltd.
CloudEngine 8800&7800&6800&5800 Series Switches
Configuration Guide - Interface Management 3 Logical Interface Configuration

#
interface GE1/0/4.20
ip address 10.20.20.20 255.255.255.0
dot1q termination vid 20
#
return

Issue 05 (2017-05-08) Huawei Proprietary and Confidential 94

Copyright © Huawei Technologies Co., Ltd.