01-22 FAQs
01-22 FAQs
Series Switches
Troubleshooting Guide 22 FAQs
22 FAQs
NOTE
The configuration process provided in this document is basic configuration and is for reference only. For
the detailed configuration process and precautions, see the product configuration guide.
Unless otherwise specified, this document uses configuration for V100R003 as an example and the
configuration takes effect in two stages. Note the following in the two-stage mode:
l After completing a configuration, you must run the commit command to commit the
configuration. Otherwise, the configuration will not take effect.
l In V100R003 and later versions, if you do not run the commit command after modifying a
configuration, the system prompt ~ is changed to *, prompting you that the configuration is not
committed. The system prompt * will be restored to ~ after you run the commit command to
commit the configuration.
If you do not have the permission to download an upgrade guide, increase your user level.
Table 22-2 Dimensions of the CE8800, CE7800, CE6800, and CE5800 series switches
Device Model Dimensions (W x D x H)
A Huawei CE12800 and CE12800E cabinet can accommodate two CE12808 switches. To
install two CE12808 switches in the same cabinet, install a switch at the bottom and then
install the other on the top. Leave certain space above and below the two switches for cabling.
The power distribution unit (PDU) in the cabinet provides both 10 A PDU (C13) and 16 A
PDU (C19) sockets. If you want to connect a CE12800 or a CE12800E switch and another
small-sized device to the PDU simultaneously, properly plan the power cable connections to
ensure reliable power supply for the two devices. (Sockets on the PDU are divided into
different groups for proper power cable connections.) Ensure that the fault of one device does
not affect the other device.
A cabinet for the CE12800 and CE12800E series switches must meet the following
requirements:
l The cabinet/rack must provide reliable ground points for switch grounding. (The ground
cable delivered with a switch is 850 mm long.)
l The cabinet/rack must meet the requirements for front-to-back airflow. The cabinet must
meet the requirements for front-to-back airflow, and the front and rear doors must have a
porosity rate of more than 50%.
l If you use AC power cables delivered from Huawei, ensure that an AC power outlet is
available within 3000 mm (cable distance) away from the switch.
l The cabinet must be installed on concrete floor or ESD floor (placed on the floor directly
or fixed by expansion bolts). If necessary, use an earthquake-proof cabinet complying
with GR63 Zone4 or Zone3 standard.
In addition to the preceding requirements, different models of the CE series have their own
requirements for a cabinet/rack, as described in Table 22-3.
CE1 l The width of the cabinet/rack must be at least 600 mm (distance A in Figure
2804 22-1), and a cabinet/rack with a width of 800 mm is recommended. When a 600
mm wide cabinet/rack is used, do not install any power strip or power
distribution unit (PDU) within the CE12804 chassis height at the rear interior
sides of the cabinet/rack. Otherwise, the power strip or PDU will hinder chassis
installation and operations on fan modules.
l The depth of the cabinet/rack must be at least 1100 mm (distance B in Figure
22-1), and a cabinet/rack with a depth of 1200 mm is recommended. When a
1100 mm deep cabinet/rack is used, the front and rear doors must be single-
swing doors.
l The cabinet/rack must have sufficient vertical space for switch installation: at
least 12 U.
l When each cabinet accommodates only one switch, the minimum load bearing
capacity required for each cabinet is 178 kg.
l It is recommended that the distance between the front mounting rails and the
front door should be 150 mm to 180 mm (distance C in Figure 22-1), and the
distance between front and rear mounting rails should be 700 mm to 730 mm
(distance D in Figure 22-1).
Dev Requirement
ice
Mod
el
CE1 l The width of the cabinet/rack must be at least 600 mm (distance A in Figure
2808 22-1), and a cabinet/rack with a width of 800 mm is recommended. When a 600
mm wide cabinet/rack is used, do not install any power strip or PDU within the
CE12808 chassis height at the rear interior sides of the cabinet/rack. Otherwise,
the power strip or PDU will hinder chassis installation and operations on fan
modules.
l The depth of the cabinet/rack must be at least 1100 mm (distance B in Figure
22-1), and a cabinet/rack with a depth of 1200 mm is recommended. When a
1100 mm deep cabinet/rack is used, the front and rear doors must be single-
swing doors.
l The cabinet/rack must have sufficient vertical space for switch installation: at
least 18 U.
l When each cabinet accommodates only one switch, the minimum load bearing
capacity required for each cabinet is 266 kg.
l It is recommended that the distance between the front mounting rails and the
front door should be 150 mm to 180 mm (distance C in Figure 22-1), and the
distance between front and rear mounting rails should be 700 mm to 730 mm
(distance D in Figure 22-1).
CE1 l The width of the cabinet/rack must be at least 600 mm (distance A in Figure
2812 22-1), and a cabinet/rack with a width of 800 mm is recommended. When a 600
mm wide cabinet/rack is used, do not install any power strip or PDU within the
CE12812 chassis height at the rear interior sides of the cabinet/rack. Otherwise,
the power strip or PDU will hinder chassis installation and operations on fan
modules.
l The depth of the cabinet/rack must be at least 1100 mm (distance B in Figure
22-1), and a cabinet/rack with a depth of 1200 mm is recommended. When a
1100 mm deep cabinet/rack is used, the front and rear doors must be single-
swing doors.
l The cabinet/rack must have sufficient vertical space for switch installation: at
least 23 U.
l When each cabinet accommodates only one switch, the minimum load bearing
capacity required for each cabinet is 359 kg.
l It is recommended that the distance between the front mounting rails and the
front door should be 150 mm to 180 mm (distance C in Figure 22-1), and the
distance between front and rear mounting rails should be 700 mm to 730 mm
(distance D in Figure 22-1).
Dev Requirement
ice
Mod
el
CE1 l A cabinet/rack with a width of 800 mm (distance A in Figure 22-1) and a depth
2816 of 1200 mm (distance B in Figure 22-1) is recommended.
l The cabinet/rack must have sufficient vertical space for switch installation: at
least 37 U.
l When each cabinet accommodates only one switch, the minimum load bearing
capacity required for each cabinet is 521 kg.
l It is recommended that the distance between the front mounting rails and the
front door should be 150 mm to 180 mm (distance C in Figure 22-1), and the
distance between front and rear mounting rails should be 800 mm to 825 mm
(distance D in Figure 22-1).
CE1 l The width of the cabinet/rack must be at least 600 mm (distance A in Figure
2804 22-1), and a cabinet/rack with a width of 800 mm is recommended. When a 600
S mm wide cabinet/rack is used, do not install any power strip or PDU within the
CE12804S chassis height at the rear interior sides of the cabinet/rack.
Otherwise, the power strip or PDU will hinder chassis installation and
operations on fan modules.
l The depth of the cabinet/rack must be at least 800 mm (distance B in Figure
22-1), and a cabinet/rack with a depth of 1000 mm is recommended. When an
800 mm deep cabinet/rack is used, the front and rear doors must be single-
swing doors.
l The cabinet/rack must have sufficient vertical space for switch installation: at
least 9 U.
l When each cabinet accommodates only one switch, the minimum load bearing
capacity required for each cabinet is 132 kg.
l It is recommended that the distance between the front mounting rails and the
front door should be 150 mm to 170 mm (distance C in Figure 22-1), and the
distance between front and rear mounting rails should be at least 450 mm
(distance D in Figure 22-1). When Huawei expandable guide rails are used, the
distance between front and rear mounting rails should be 500 mm to 850 mm.
Dev Requirement
ice
Mod
el
CE1 l The width of the cabinet/rack must be at least 600 mm (distance A in Figure
2808 22-1), and a cabinet/rack with a width of 800 mm is recommended. When a 600
S mm wide cabinet/rack is used, do not install any power strip or PDU within the
CE12808S chassis height at the rear interior sides of the cabinet/rack.
Otherwise, the power strip or PDU will hinder chassis installation and
operations on fan modules.
l The depth of the cabinet/rack must be at least 800 mm (distance B in Figure
22-1), and a cabinet/rack with a depth of 1000 mm is recommended. When an
800 mm deep cabinet/rack is used, the front and rear doors must be single-
swing doors.
l The cabinet/rack must have sufficient vertical space for switch installation: at
least 17 U.
l When each cabinet accommodates only one switch, the minimum load bearing
capacity required for each cabinet is 216 kg.
l It is recommended that the distance between the front mounting rails and the
front door should be 150 mm to 170 mm (distance C in Figure 22-1), and the
distance between front and rear mounting rails should be at least 450 mm
(distance D in Figure 22-1). When Huawei expandable guide rails are used, the
distance between front and rear mounting rails should be 500 mm to 850 mm.
Dev Requirement
ice
Mod
el
Heat dissipation The CE8800, CE7800, CE6800, and CE5800 series switches use
front-to-back or back-to-front airflow design. The cabinet must
meet the requirement of the airflow.
Grounding The cabinet must have reliable ground points to ground the chassis.
Distance between When the CE8800, CE7800, CE6800, and CE5800 series switches
front and rear need to be installed in the same cabinet with a CE12800 switch,
mounting rails keep the distance between the front and rear mounting rails in the
range of 700 mm to 800 mm.
Cabling space On a switch, the side with power modules and fan modules is the
power supply side, and the side with service ports is the port side.
To ensure sufficient cabling space, reserve at least 130 mm of
vertical distance from the port side to the interior side of the
cabinet door, and at least 70 mm from the power supply side to the
interior side of the cabinet door.
CE-MPUA-S and CE-MPUB-S have the same management ports on the front panel. Figure
22-3 shows the management ports on the CE-MPUA-S as an example.
CE-MPUA-S
LINK LINK PWR FAN RUN/ALM
ACT
GE0 GE1 ACT GE0 GE1 ACT ETH CONSOLE RST STACK
RST
GE1 CLK1 RS485 ETH CONSOLE
The ETH management port can be connected to a network management workstation for
device management.
GE0 CE-MPUA
RUN/ALM
ACT
STACK
Attribute Description
Attribute Description
NOTE
l The console port and Mini USB/Micro USB port are multiplexed and share one internal serial port.
You can use the console port or Mini USB/Micro USB port as the serial port according to your
needs. When the Mini USB/Micro USB is activated, the console port cannot be used.
l When both the console port and Mini USB/Micro USB port have a cable connected, the Mini USB/
Micro USB port is used.
The console port can be connected to a configuration terminal for onsite system configuration.
GE0 CE-MPUA
RUN/ALM
ACT
STACK
Attribute Description
The Mini USB/Micro USB management port can be connected to a configuration terminal for
onsite system configuration, but the configuration terminal must have a USB serial port driver
installed. The Mini USB/Micro USB port is used as the serial port once a link is established
on the port.
GE0 CE-MPUA
RUN/ALM
ACT
STACK
Figure 22-5 Management ports on the CE8800, CE7800, CE6800, and CE5800 series
switches (except the CE6850HI, CE6855-48T6Q-HI, CE6856-48T6Q-HI, CE6850U-HI,
CE6875EI, CE8850-64CQ-EI, CE8860EI, CE8861EI, and CE8868EI)
Console management port
ACT
L/A
ID
ETH
PAC-150WA ~100-240V;50/60Hz,2.5A PWR1 FAN1 FAN2 PWR2 PAC-150WA ~100-240V;50/60Hz,2.5A
PWR1
FAN1
CONSOLE
SYS
MST
060A-F
FAN-
060A-F
FAN-
USB
ACT
FAN2
PWR2
ETH
Figure 22-7 shows the management ports on the CE8850-64CQ-EI, CE8860EI, CE8861EI,
and CE8868EI.
STAT
CE8860-4C-EI
FAN1
FAN2
FAN3
PWR1
1K2WA-F
整机条码
MAC地址
SLOT
180A-F
FAN-
180A-F
FAN-
1 2 3 4
1 2 ID STAT
PWR2
CONSOLE
SYS
MST
1K2WA-F
USB
ACT
STAT STAT
ETH
All management ports of the CE8800, CE7800, CE6800, and CE5800 series switches are
located at the front of the chassis (power supply side).
ETH (combo), console (RJ45), and Mini USB management ports of the CE8800, CE7800,
CE6800, and CE5800 series switches have the same attributes as those management ports on
the CE12800 series switches.
ETH management port (Combo)
The ETH management port (combo) 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.
l The electrical port uses a Category-5 or higher category network cable.
l The 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.
NOTE
NOTE
l When the altitude is between 1800 m and 4000 m, the highest operating temperature for the
CE12800&CE12800E series switches reduces by 1°C every time the altitude increases by 220 m.
l When the altitude is between 1800 m and 5000 m, the highest operating temperature for the
CE8800&7800&6800&5800 series switches reduces by 1°C every time the altitude increases by 220
m.
l The preceding temperature values refer to the ambient temperature, whereas the temperature shown
in the display command output is the PCB surface temperature near high-power chips in the chassis.
Whether the temperature is normal depends on whether it is within the temperature alarm threshold.
Therefore, if the displayed temperature of a switch does not exceed the alarm threshold, the switch is
operating under a normal temperature and has no risk of overtemperature.
NOTE
All temperature alarm thresholds are set by thermal design engineers based on device running status
monitoring results, and alarm thresholds of different device models vary slightly. The hardware alarm
threshold for a device is higher than its software alarm threshold.
NOTE
If the following conditions occur during installation of a CE12800 series switch, you can
remove the lower enclosure frame:
l If the chassis cannot be moved into the cabinet with the lower enclosure frame on the
chassis, remove the lower enclosure frame before placing the chassis into the cabinet.
NOTE
After 2/3 of the chassis is in the cabinet, install the lower enclosure frame back to the chassis (for
installing the air filter door), and then move the chassis completely in the cabinet.
l If the cabinet is not deep enough for the switch, determine whether the distance between
the front mounting rails and the front door meets the switch installation requirement. For
details, see 22.2.3 What Requirements Should Cabinets for the CE Series Switches
Meet?. If the front door cannot be closed after the switch is installed in the cabinet,
remove the lower enclosure frame from the chassis.
If lower enclosure frame is removed because the cabinet depth is small, the chassis
header and cable management frames cannot be installed either. Without cable
management frames, cables cannot be properly routed. This may affect normal service
operations. Therefore, use a qualified cabinet for the CE12800 series switches.
First check whether the input voltage for the switch is within the allowed range. If the input
voltage is normal, turn on the power switch of the power supply system, and then turn on the
power switch on the switch. (A CE12804S switch does not have a power switch.)
NOTE
After powering on a switch, check indicators on the switch to determine whether the switch is running
normally. For a CE12804/CE12808/CE12812/CE12816 switch, first check the indicators on the chassis
header. (The PWR, FAN, SFU, and MPU indicators should be steady green). For a CE12804S/
CE12808S/CE12804E/CE12808E/CE12816E switch, first check the indicators on the Main Control
Units (MPUs). (The PWR and FAN indicators should be steady green, and the RUN/ALM indicator
should be slow blinking green.) If the preceding indicators are in normal status, check indicators on
other modules.
Figure 22-8 shows the procedure for powering on a CE12800 switch for the first time.
Figure 22-8 Procedure for powering on a CE12800 or a CE12800E switch for the first time
Start
Locate and
rectify the fault
Is input voltage in No
normal range?
Yes
Locate and
rectify the fault
Yes
Yes
End
Power input C20 straight male Two-OT input C20 straight male
terminal connector terminal connector
The CE12800 AC, DC, and AC&high-voltage chassis with the same specifications can use
the same types of LPUs, MPUs, and CMUs (CE12804S and CE12808S have no independent
CMUs), SFUs, and fan modules.
CE12804 2 MPU slots, 2 SFU 3 fan module slots and a Figure 22-13
S chassis slots, 4 LPU slots, and 4 power distribution unit
power module slots
CE12808 2 MPU slots, 4 SFU 6 fan module slots and a Figure 22-14
S chassis slots, 8 LPU slots, and 8 power distribution unit
power module slots
NOTE
l In N+N backup mode, power module (PM) slots in the chassis are divided into area A and area B,
and power modules in the two areas work as a hot backup for each other.
l MPU: Main Processing Unit
l SFU: Switch Fabric Unit
l CMU: Centralized Monitoring Unit
l LPU: Line Processing Unit
l PM: power module
l FAN: fan module
Slot7-CMU Slot8-CMU
FAN7 FAN8 FAN9
Slot6-MPU
FAN5 FAN6
Slot5-MPU
Slot11-CMU Slot12-CMU
FAN11 FAN12 FAN13
Slot10-MPU
FAN9 FAN10
Slot9-MPU
Slot8-LPU
FAN7 FAN8
Slot7-LPU
Slot6-LPU Slot Slot Slot Slot Slot Slot
FAN5 13 14 15 16 17 18 FAN6
Slot5-LPU
S S S S S S
Slot4-LPU
FAN3
FAN1 F F F F F F FAN4
Slot3-LPU U U U U U U
Slot2-LPU
FAN1 FAN2
Slot1-LPU
Slot15-CMU Slot16-CMU
FAN15 FAN16 FAN17
Slot14-MPU
FAN13 FAN14
Slot13-MPU
Slot12-LPU
FAN11 FAN12
Slot11-LPU
Slot10-LPU
FAN9 FAN10
Slot9-LPU
Slot4-LPU
FAN3
FAN1 FAN4
Slot3-LPU
Slot2-LPU
FAN1 FAN2
Slot1-LPU
Slot19-CMU Slot20-CMU
Slot18-MPU
FAN17 FAN18
Slot17-MPU
Slot16-LPU
FAN15 FAN16
Slot15-LPU
Slot14-LPU
FAN13 FAN14
Slot13-LPU
Slot12-LPU
FAN11 FAN12
Slot11-LPU
Slot10-LPU Slot Slot Slot Slot Slot Slot
FAN9 FAN10
Slot9-LPU 21 22 23 24 25 26
Slot8-LPU S S S S S S
FAN7 F F F F F F FAN8
Slot7-LPU U U U U U U
Slot6-LPU
FAN5 FAN6
Slot5-LPU
Slot4-LPU
FAN3 FAN4
Slot3-LPU
Slot2-LPU
FAN1 FAN2
Slot1-LPU
Slot5-MPU Slot6-MPU
Slot4-LPU
Slot3-LPU
Slot8-SFU FAN1 FAN2 FAN3
Slot7-SFU
Slot2-LPU
Slot1-LPU
Slot9-MPU Slot10-MPU
Slot8-LPU
Slot7-LPU
Slot6-LPU FAN4 FAN5 FAN6
Slot5-LPU
Slot14-SFU
Slot13-SFU
Slot12-SFU
Slot11-SFU
Slot4-LPU FAN1 FAN2 FAN3
Slot3-LPU
Slot2-LPU
Slot1-LPU
Area A Area B
Slot8-LPU
Slot7-LPU
Slot6-LPU
Slot5-LPU Slot Slot Slot Slot Slot Slot
Slot4-LPU 11 12 13 14 15 16
FAN4 FAN5
Slot3-LPU S S S S S S
F F F F F F
Slot2-LPU U U U U U U
Slot1-LPU
Slot9-MPU Slot10-MPU
PM PM PM PM PM PM PM PM
1 2 3 4 5 6 7 8 FAN1 FAN2 FAN3
Area A Area B
Slot16-LPU
Slot15-LPU
Slot14-LPU
Slot13-LPU
FAN8 FAN9
Slot12-LPU
Slot11-LPU
Slot10-LPU
Slot Slot Slot Slot Slot Slot
Slot9-LPU 19 20 21 22 23 24
Slot8-LPU S S S S S S
F F F F F F
Slot7-LPU U U U U U U
Slot6-LPU
Slot5-LPU
Slot4-LPU FAN6 FAN7
Slot3-LPU
Slot2-LPU
Slot1-LPU
Slot17-MPU Slot18-MPU
PM PM PM PM PM PM PM PM PM PM
1 2 3 4 5 6 7 8 9 10 FAN1 FAN2 FAN3 FAN4 FAN5
Area A Area B
Figure 22-18 Front panel of the CE8800, CE7800, CE6800, and CE5800 series switches
(except the CE6850HI, CE6855-48T6Q-HI, CE6856-48T6Q-HI, CE6850U-HI, CE6875EI,
CE8860EI, CE8861EI, and CE8868EI)
1 3 4 2
Figure 22-19 shows the slots on the front panel of the CE6850HI, CE6855-48T6Q-HI,
CE6856-48T6Q-HI, CE6850U-HI, and CE6875EI switches.
PWR1
FAN1
CONSOLE
SYS
MST
060A-F
FAN-
060A-F
FAN-
USB
ACT
FAN2
PWR2
ETH
Figure 22-20 shows the slots on the front panel of the CE8860, CE8861, and CE8868 switch.
Figure 22-20 Front panel of the CE8860EI, CE8861EI, and CE8868EI switch
Fan module slot Fan module slot Power module slot Power module slot
STAT
CE8860-4C-EI
FAN1
FAN2
FAN3
PWR1
1K2WA-F
整机条码
MAC地址
SLOT
180A-F
FAN-
180A-F
FAN-
1 2 3 4
1 2 ID STAT
PWR2
CONSOLE
SYS
MST
1K2WA-F
USB
ACT
STAT STAT
ETH
Figure 22-21 shows the slots on the front panel of the CE8850-64CQ-EI switch.
STAT
CE8850-64CQ-EI
FAN1
FAN2
FAN3
PWR1
1K2WA-F
整机条码
MAC地址
SLOT
180A-F
FAN-
180A-F
FAN-
180A-F
FAN-
1 2 3 4
1 2 ID STAT
PWR2
CONSOLE
SYS
MST
1K2WA-F
USB
ACT
NOTE
– N+0 mode (N
≤ 4): 1500 W
x 4 = 6000 W
x 8 = 12000
W
– N+0 mode (N
≤ 12): 1500
W x 12 =
18000 W
– N+1 backup
(N ≤ 19):
1500 W x 19
= 28500 W
– N+0 mode (N
≤ 20): 1500
W x 20 =
30000 W
NOTE
N+N backupc: If the actual chassis power consumption exceeds the maximum output power provided by
power modules in the N+N backup mode, this backup mode cannot be used. The N+1 backup mode is
recommended in this case.
If the AC power modules use a 110 V power source, follow these suggestions:
1. Use double live wires. In this case, the output power of a power module is 2700 W/3000 W.
2. If only one live wire is used, the output power of a power module is 1200 W/1500 W. In this case, it
is recommended that you use more power modules or change the power module backup mode to
increase the output power.
Power modules in a CE12800 or a CE12800E series switch can work in N+N backup, N+1
backup, or N+0 mode. To ensure power reliability, it is recommended that you use the N+N
backup mode and connect the power modules to two power supply systems. In this way, the
switch will not be powered off or restart if one power supply system fails.
NOTE
N is the number of power modules configured based on actual power consumption of a chassis. The
maximum power capacity of the system is the total maximum output power of N power modules.
l N+N backup: N power modules supply power for the chassis, and N power modules work as backup
to implement power redundancy.
l N+1 backup: N power modules supply power for the chassis, and one power module works as
backup to implement power redundancy.
l N+0 (no backup): N power modules supply power for the chassis, and no power module works as
backup.
The CE12800 and CE12800E series switches do not support the PoE function.
Power module configuration modes for the CE12800 and CE12800E series switches
l In N+N backup mode, power module (PM) slots in the chassis are divided into area A
and area B, and power modules in the two areas work as a hot backup for each other, as
shown in Slot layout in the CE12804 chassis, Slot layout in the CE12808 chassis, Slot
layout in the CE12812 chassis, Slot layout in the CE12816 chassis, Slot layout in the
CE12804S chassis, Slot layout in the CE12808S chassis, Slot layout in the
CE12804E chassis, Slot layout in the CE12808E chassis, and Slot layout in the
CE12804E chassis. To use the N+N backup mode, install N power modules in the PM
slots in area A, and the other N power modules in the PM slots in area B. For example,
to use the 4+4 backup mode in a CE12812 chassis, install four power modules in any
four slots among PM1, PM2, PM5, PM6, PM9, and PM10, and the other four power
modules in any four slots among PM3, PM4, PM7, PM8, PM11, and PM12.
l In N+1 backup or N+0 mode, power module slots are not divided into two areas for
backup, and N+1 or N power modules can be installed in any PM slots. However, you
are advised to install (N+1)/2 or N/2 power modules in area A and the other power
modules in area B. (Here, area division used in the N+N backup mode is mentioned only
for convenience of description. Actually, area A and area B do not back up each other.) If
N+1 or N is an odd number, the redundant one can be installed in either area A or area B.
– For example, to use the 6+0 mode in a CE12812 chassis, you are advised to install
three power modules in any three slots among PM1, PM2, PM5, PM6, PM9, and
PM10, and the other three power modules in any three slots among PM3, PM4,
PM7, PM8, PM11, and PM12.
– To use the 6+1 backup mode in a CE12812 chassis, you are advised to install four
power modules in any four slots among PM1, PM2, PM5, PM6, PM9, and PM10,
and the other three power modules in any three slots among PM3, PM4, PM7, PM8,
PM11, and PM12. Alternatively, install three power modules in any three slots
among PM1, PM2, PM5, PM6, PM9, and PM10, and the other four power modules
in any four slots among PM3, PM4, PM7, PM8, PM11, and PM12.
CE8800&7800&6800&5800
The CE8800, CE7800, CE6800, and CE5800 series switches have two power module slots, in
which pluggable power modules can be installed to provide power to the chassis. A chassis
can use one or two power modules. Double power modules can provide higher reliability.
The CE8800, CE7800, CE6800, and CE5800 series switches support double power modules
working in 1+1 backup mode.
l When both power modules are working normally, each of them provides half of the
power required for the chassis.
l When one power module fails, the other one provides all power required for the chassis.
NOTE
l If only one power module is installed in a switch, the switch generates power module absence
alarms for the idle power slots.
l Failure of a single power module does not affect operations of the switch but will degrade the
reliability of the switch.
All power modules are hot swappable. The CE8800, CE7800, CE6800, and CE5800 series
switches do not support the PoE function.
NOTE
Power modules in a chassis must have the same power and same heat dissipation method.
NOTE
Power modules of the CE12800&CE12800E series switches are hot swappable. After you remove a
power module from the chassis, wait at least 30 seconds before installing it back into the chassis.
l AC&high-voltage DC power modules that receive AC power inputs can be used together
with AC power modules in a chassis.
l In other cases, AC, DC, and AC&high-voltage DC power modules cannot be used together
in a chassis.
PDC-2200 A A A A A A
WA maximum maximum maximum maximum maximum maximum
of 4 power of 8 power of 12 of 20 of 4 power of 8 power
modules modules power power modules modules
supported supported modules modules supported supported
in a DC in a DC supported supported in a DC in a DC
chassis, chassis, in a DC in a DC chassis, chassis,
not not chassis, chassis, not not
supported supported not not supported supported
in an AC in an AC supported supported in an in an
or or in an AC in an AC AC&high- AC&high-
AC&high- AC&high- or or voltage voltage
voltage voltage AC&high- AC&high- DC DC
DC DC voltage voltage chassis chassis
chassis chassis DC DC NOTE NOTE
NOTE NOTE chassis chassis This This
This This NOTE NOTE power power
power power This This module is module is
module is module is power power supported supported
supported supported module is module is in in
in in supported supported V100R00 V100R00
V100R00 V100R00 in in 5C00 and 5C00 and
2C00 and 2C00 and V100R00 V100R00 later later
later later 2C00 and 3C00 and versions. versions.
versions. versions. later later
versions. versions.
PHD-3000 A A A A A A
WA maximum maximum maximum maximum maximum maximum
of 4 power of 8 power of 12 of 20 of 4 power of 8 power
modules modules power power modules modules
supported supported modules modules supported supported
in an AC in an AC supported supported in an in an
chassis or chassis or in an AC in an AC AC&high- AC&high-
an an chassis or chassis or voltage voltage
AC&high- AC&high- an an DC DC
voltage voltage AC&high- AC&high- chassis, chassis,
DC DC voltage voltage not not
chassis, chassis, DC DC supported supported
not not chassis, chassis, in a DC in a DC
supported supported not not chassis chassis
in a DC in a DC supported supported NOTE NOTE
chassis chassis in a DC in a DC This This
NOTE NOTE chassis chassis power power
This This module is module is
NOTE NOTE
power power supported supported
This This
module is module is in in
power power
supported supported V100R00 V100R00
module is module is
in in 5C00 and 5C00 and
supported supported
V100R00 V100R00 later later
in in
5C00 and 5C00 and versions. versions.
V100R00 V100R00
later later 5C00 and 5C00 and
versions. versions. later later
versions. versions.
Table 22-7 lists the CE12804E/CE12808E/CE12816E switch product and version support for
various power modules.
Table 22-7 CE12800E series switch product and version support for power modules
Power Module CE12804E Chassis CE12808E Chassis CE12816E Chassis
Name
CE12800, CE12800E
The AC power cables applicable to the CE12800, CE12800E series switches depend on the
type of power sockets used in your equipment room. Two types of AC power cables are
available to suit different power sockets: PDU power cable and country-specific power cable.
l PDU power cables are delivered with a power distribution unit (PDU) with C20 straight
sockets. A PDU power cable has a C19 straight female connector at one end and a C20
straight male connector at the other end, as shown in Figure 22-22.
l Country-specific power cables are delivered in compliance with standards of the
destination country or region. For example, AC power cables used in China have PI
angle male connectors, as shown in Figure 22-23.
If PDUs with C20 straight sockets are available in a cabinet, use PDU power cables
complying with local standards.
If there is no PDU with C20 straight sockets in a cabinet, use power cables specific to the
country or region where the switch is used.
Connection
l The C19 straight female connector is connected to a socket on the power distribution unit
at the rear of the CE12800 chassis.
l The country-specific connector or C20 straight male connector at the other end is
connected to an external power outlet.
l 150 W/350 W/600 W AC power modules: The AC power sockets or directly connected
circuit breakers must have a current rating of 10 A.
l 350 W DC power modules: The directly connected circuit breakers must have a current
rating of 16 A.
l 600 W/1200 W AC&240 V DC power modules: The AC power sockets or directly
connected circuit breakers must have a current rating of 10 A.
l 600 W/1200 W high-voltage DC power modules: The directly connected circuit breakers
must have a current rating of 10 A.
NOTE
When 110 V power input is provided to a 1200 W AC&240 V DC power module, the AC power socket
or directly connected circuit breaker must have a current rating of 16 A.
If a switch uses two power modules for 1+1 power redundancy, the superior circuit breaker for the
power distribution box or PDU to which the power modules are connected must have a current rating of
25 A.
The AC power cables delivered with a switch must comply with the standards used in the delivery
destination.
CE8800&7800&6800&5800
Some power modules for the CE8800, CE7800, CE6800, and CE5800 switches have a power
switch, as listed in Table 22-8.
Table 22-8 Power modules for the CE8800, CE7800, CE6800, and CE5800 series switches
PAC-150WA Yes
ES0W2PSA015 Yes
0
PAC-350WA-B Yes
PAC-350WA-F Yes
PDC-350WA-B No
PDC-350WA-F No
PAC-600WA-B Yes
PAC-600WA-F Yes
PAC-600WB-B No
PAC-600WB-F No
PHD-600WA-B No
PHD-600WA-F No
PAC-1K2WA-B No
PAC-1K2WA-F No
PHD-1K2WA-B No
PHD-1K2WA-F No
PDC-1K2WA-B No
PDC-1K2WA-F No
Table 22-10 lists the indicators on a 2200 W DC power module and describes the meaning of
each indicator state.
NOTE
When AC power supply is used and the input voltage is lower than or equal to 175 V AC, the rated
output power reduces to 1500 W.
If the CE8800, CE7800, CE6800, and CE5800 switches use power modules that do not match
their product models, the switches generate an alarm and may be powered off due to power
shortage.
When the power module and fan module use forcible heat dissipation, they must use the same
airflow method. For example, if the power module with back-to-front airflow is used, the fan
module with back-to-front airflow must be used. Type-B power modules can only be used
with type-B fan modules and cannot be used with type-F fan modules.
Table 22-16 Applicable power modules, fan modules, and version support
Series Model Power Module Fan Module
1200 W high-voltage DC
power modules:
l PHD-1K2WA-B
l PHD-1K2WA-F
1200 W DC power
modules:
l PDC-1K2WA-B
l PDC-1K2WA-F
l A PM is a power module used to supply power to the equipment. Figure 22-24 shows
the appearance of a PM.
l A PEM is the power entry module on the chassis, an iron frame that provides power
inputs to PMs and provides other functions such as wave filtering. Generally, a PEM is
included in the chassis sales part. Figure 22-25 shows the appearance of a PEM.
The Hardware Configuration Tool helps you quickly calculate the number of power
modules required in a CE12800 or CE12800E switch. The calculator is also integrated in the
UniSTAR system.
You can also use the following formula to calculate the number of power modules required:
Minimum number of power modules required = Sum of maximum power consumption of all
working modules in the switch/Maximum power of a single power module (integral part of
the result) + 1
For example, a CE12808 switch uses 2700 W AC power modules and is configured with two
CE-CMUAs, five CE-SFU08As, two CE-MPUAs, and two CE-L48XS-EC line cards. Based
on the power consumption values of the cards provided in the "Cards" section of the
Hardware Description and the power consumption of the chassis (including fans) and power
modules, the sum of maximum power consumption of all modules can be obtained: 2519 W.
Therefore, a minimum of one 2700 W power module is required.
To ensure power reliability of a CE12800 switch, it is recommended that you use the N+N
backup mode with double power supply systems so that the failure of a single power supply
system will not cause power failure or restart of the switch. In the preceding example, the 1+1
backup mode is recommended.
22.3.13 What Are the Differences Among N+0, N+1, and N+N
Backup Modes of Power Modules?
In the N+1 backup mode, N power modules provide power supply, and one power module
acts as a backup. If one power module fails, the redundant power module provides power
supply to ensure normal operations of the switch.
In the N+N backup mode, N power modules provide power supply, and N power modules act
as backups. Even if all the N primary power modules fail, the N redundant power modules
can still ensure normal operations of the switch.
In the N+0 mode, N power modules provide power supply, and no redundant power module is
available. If one power module fails and the total output of the other power modules is smaller
than the actual power consumption of the switch, the switch cannot run normally.
The 2+2 power module configuration is supported as long as the output power of the power
modules is sufficient for the switch.
All power cables are hot swappable if they are connected to matching power modules.
CE12804/CE12808/CE12812/CE12816/CE12804S/CE12808S
NOTE
l The CE12804/CE12808/CE12812/CE12816 chassis use the same heat dissipation mode. The
CE12812 chassis is used as an example here.
l The CE12804S/CE12808S chassis use the same heat dissipation mode. The CE12808S chassis is
used as an example here.
l If a CE12800 chassis is not fully loaded with cards, cover vacant slots with filler panels to ensure
efficient heat dissipation and EMC compliance.
Figure 22-26 shows the airflow for heat dissipation of SFUs and power modules in a
CE12812 chassis.
l Cold air flows into the chassis from the air intake between the LPU cage and power
modules and reaches the bottom of the SFU cage. After cold air flows upwards through
the SFU cage, it turns into hot air and is exhausted from the chassis by the fan modules
above SFUs.
l Cold air flows into the power modules from air holes on power module panels and is
blown through the power module area by built-in fans of the power modules. Cold air
then becomes hot air, passes through the backplane, and goes out of the chassis from air
holes on the power distribution unit.
Figure 22-26 Airflow for heat dissipation of SFUs and power modules (side view)
Backplane
CMU FAN
Hot air
MPU
LPU
SFU
LPU
Front Rear
LPU
Cold air
Cold air
Backplane
Hot air
Figure 22-27 shows the airflow for heat dissipation of MPUs, CMUs, and LPUs in a
CE12812 chassis. Cold air flows into the chassis through air holes on these cards, and goes
through the cards to dissipate the heat. After that, cold air turns into hot air, bypasses the
backplane, and is finally exhausted from the chassis by the fan modules at two sides of the
SFUs.
Figure 22-27 Airflow for heat dissipation of MPUs, CMUs, and LPUs (top view)
Backplane
MPU、CMU、LPU
Cold air
Cold air
Front
Hot air
Figure 22-28 shows the airflow for heat dissipation of MPUs, SFUs, LPUs, and power
modules in a CE12808S chassis.
l Cold air flows into the chassis through air holes on panels of the MPUs, SFUs, LPUs,
and goes through these cards to dissipate the heat. After that, cold air turns into hot air,
goes through the backplane, and is finally exhausted from the chassis by the fan modules
at the rear of the chassis.
l Cold air flows into the power modules from air holes on power module panels and is
blown through the power module area by built-in fans of the power modules. Cold air
then becomes hot air, passes through the backplane, and goes out of the chassis from air
holes on the power distribution unit.
Figure 22-28 Airflow for heat dissipation of MPUs, SFUs, LPUs, and power modules (side
view)
Backplane
FAN
Cold air LPU Hot air
Cold air
CE12804E/CE12808E/CE12816E
NOTE
l The CE12804E has a heat dissipation mode different from other chassises.
l The CE12808E/CE12816E chassis use the same heat dissipation mode. The CE12808E chassis is
used as an example here.
l If a CE12800 chassis is not fully loaded with cards, cover vacant slots with filler panels to ensure
efficient heat dissipation and EMC compliance.
The SFUs in the CE12804E chassis are cooled in the following way:
1. Cold air flows into the chassis from the SFU air intake vents below the chassis header at
the front of the chassis, and then arrives at the top of the SFUs.
2. Cold air goes down through the SFU cage and becomes hot air, which is exhausted out of
the chassis by the two fan modules at the bottom left and bottom right.
Figure 22-29 Airflow for heat dissipation of the SFUs (side view)
Cold air
LPU
SFU
Front Rear
MPU
Cold air
Hot air
The MPUs and LPUs in the CE12804E chassis are cooled in the following way:
1. Front panels of the MPUs and LPUs have air holes. Cold air flows from these air holes
to the MPUs and LPUs.
2. Cold air turns into hot air and goes through two sides of the SFUs in the middle. Finally,
hot air is exhausted from the chassis by the fan modules at both sides of the SFU cage.
Figure 22-30 shows the airflow for heat dissipation of the MPUs and LPUs.
Figure 22-30 Airflow for heat dissipation of the MPUs and LPUs (top view)
Hot air Rear Hot air
MPU、LPU
Cold air
Cold air
Front
Hot air
Figure 22-31 Airflow for heat dissipation of the power modules (side view)
LPU
SFU
Front Rear
MPU
FAN
Cold air PM
Hot air
Cold air
Hot air
The SFUs in the CE12808E chassis are cooled in the following way:
1. Cold air flows into the chassis from the SFU air intake vents below the chassis header at
the front of the chassis, and then arrives at the top of the SFUs.
2. Cold air goes down through the SFU cage and becomes hot air, which is exhausted out of
the chassis by the three fan modules below the SFUs.
Figure 22-32 shows the airflow for heat dissipation of SFUs.
Figure 22-32 Airflow for heat dissipation of the SFUs (side view)
Cold air
LPU
SFU
Front Rear
LPU
MPU
Cold air
Hot air
The MPUs and LPUs in the CE12808E chassis are cooled in the following way:
1. Front panels of the MPUs and LPUs have air holes. Cold air flows from these air holes
to the MPUs and LPUs.
2. Cold air turns into hot air and goes through two sides of the backplane in the middle.
Finally, hot air is exhausted from the chassis by the fan modules at both sides of the SFU
cage.
Figure 22-33 shows the airflow for heat dissipation of the MPUs and LPUs.
Figure 22-33 Airflow for heat dissipation of the MPUs and LPUs (top view)
Hot air Rear Hot air
MPU、LPU
Cold air
Cold air
Front
Hot air
Figure 22-34 Airflow for heat dissipation of the power modules (side view)
LPU
SFU
Front Rear
LPU
MPU
FAN
Cold air PM
Hot air
Cold air
Hot air
CE8800&7800&6800&5800
The CE8800&7800&6800&5800 series switches are cooled through front-to-back or back-to-
front airflow, depending on the airflow directions of the fan modules and power modules used
in the chassis. Each fan module and power module has an airflow flag.
l Front-to-back airflow: Power modules and fan modules with front-to-back airflow are
identified by a or flag. Air flows into the chassis from the power
supply side and is exhausted from the port side, as shown in Figure 22-35 (using a
CE5800 chassis as an example).
l Back-to-front airflow: Power modules and fan modules with back-to-front airflow are
identified by a or flag. Air flows into the chassis from the port side
and is exhausted from the power supply side, as shown in Figure 22-36 (using a CE5800
chassis as an example).
NOTE
l The CE5850EI and CE5810EI series switches support 1+1 redundancy of fan modules. A CE5850EI
or CE5810EI switch can work normally with a single fan module, but the device reliability degrades
in this case.
l The CE5850HI, CE5855EI, CE5880EI, CE6800, and CE7800 switches have two power modules,
each of which consists of two fans. The four fans in a switch work in 3+1 redundancy mode.
l Power modules using forced air cooling and fan modules in a switch must have the same airflow
direction. For example, power modules with back-to-front airflow must be used with fan modules
with back-to-front airflow.
l When a switch uses fanless 150 W AC power modules, it can use fan modules with either of the
airflow directions based on the actual conditions.
CONSOLE CE5850-48T4S2Q-EI
STATUS STATUS STATUS STATUS
SYS
MST
ACT
L/A
ID
ETH
PAC-150WA ~100-240V;50/60Hz,2.5A FAN-40EA-F PWR1 FAN1 FAN2 PWR2 FAN-40EA-F PAC-150WA ~100-240V;50/60Hz,2.5A
1 2
3 4
1 2
48
46 47
44 45
42 43
40 41
38 39
37
36
34 35
32 33
30 31
28 29
26 27
25
24
22 23
20 21
18 19
16 17
14 15
13
12
10 11
8 9
6 7
4 5
2 3
1
Cold air
Hot air
CONSOLE CE5850-48T4S2Q-EI
STATUS STATUS STATUS STATUS
SYS
MST
ACT
L/A
ID
ETH
PAC-150WA ~100-240V;50/60Hz,2.5A FAN-40EA-B PWR1 FAN1 FAN2 PWR2 FAN-40EA-B PAC-150WA ~100-240V;50/60Hz,2.5A
1 2
3 4
1 2
48
46 47
44 45
42 43
40 41
38 39
37
36
34 35
32 33
30 31
28 29
26 27
25
24
22 23
20 21
18 19
16 17
14 15
13
12
10 11
8 9
6 7
4 5
2 3
1
SYS
MST
STAT
SPEED
Cold air
STACK
4S2Q-EI
CE5850-48T
MODE/ID
Hot air
Airflow of a switch cannot be changed through software, but you can determine the airflow
by selecting specific power modules and fan modules.
l Power modules with different airflow directions or power values cannot be used in the
same switch.
l Fan modules with different airflow directions cannot be used in the same switch.
l An alarm will be generated if a switch does not use any power module and fan module
combination listed in Table 22-17.
Table 22-17 lists the combinations of power modules and fan modules allowed by
CE8800&7800&6800&5800 series switches.
1. The cards in a zone overheat, and fans in this zone must run at full speed to lower card
temperature.
2. Some cards in a zone have all temperature sensors failed and card temperature cannot be
obtained.
3. Management channels in a zone are interrupted, so the system cannot obtain card
temperature in this zone.
4. The management channels of some fans are interrupted, so the speed of these fans cannot
be adjusted based on real-time card temperature.
5. If any of the preceding situations occurs in a zone, fans in the adjacent zones with
running cards also run at full speed.
6. No Centralized Monitoring Module (CMU) is installed in the switch or the CMUs in the
switch have failed.
7. Fans also run at full speed when you set the fan speed to the highest using the set fan-
speed command.
You can see alarm messages on the network management system in situations 1 to 5.
1. There is only one fan module in the switch. In this case, the switch generates fan module
absence alarms for idle fan slots, and fans in the present fan module work at full speed.
2. Power modules and fan modules in the switch use different airflow designs. In this case,
the system reports alarms about inconsistent airflow directions of the power modules and
fan modules. Inconsistent airflow directions between power modules, fan modules, or
between power modules and fan modules degrade system reliability when the switch
needs to work in a high-temperature environment for a long time, and shorten service life
of the switch.
3. The management channels of some fans are interrupted, so the speed of these fans cannot
be adjusted based on real-time equipment temperature.
4. Fans also run at full speed when you set the fan speed to the highest using the set fan-
speed command.
Possible Causes
l The fan module is not installed properly in the fan slot.
l Fan blades are blocked or stop rotating because of heavy dust.
l The fan module is faulty.
Troubleshooting Procedure
1. Pull out the fan module and reinstall it to ensure that it is securely installed in the slot.
(Fan modules are hot swappable.)
2. Remove the fan module from the chassis and check whether the fan blades are blocked
or whether there is too much dust on the fan blades.
– If fan blades are blocked by an obstacle, remove the obstacle.
– If the fan blades have heavy dust, use a brush to clean the fan blades.
3. If the problem persists, install another fan module of the same model into the fan slot. If
the new fan module can work normally, the original fan module has failed and needs to
be replaced.
MPU RUN/ALM: Yellow Steady on: The card is in power-off state. (For
running example, the card has been forcibly powered
SFU status off using the power off command or is about
CMU indicator to start.)
l Main Processing Units (MPUs) and Centralized Monitoring Units (CMUs) support 1:1
hot standby.
l Switch Fabric Units (SFUs) support N+M hot standby. To remove an SFU, hold down
the OFL button (at the bottom of the enlarged image) for at least 6s until the OFL
indicator turns red (at the top of the enlarged image). Then remove the SFU from the
chassis. Packets may be lost if you remove the SFU before the OFL indicator turns red.
CE-SFU04
RUN/ALM
OFL
OFL
CE-CMUA
RUN/ALM
ACT
ETH port
It is a management port and supports rates of 10 Mbit/s, 100 Mbit/s, and 1000 Mbit/s. This
port is reserved for future use.
RS485 port
Ports RS485 0 and RS485 1 are reserved for future function extension.
I/O port
Ports ALM IN and ALM OUT are reserved for future function extension.
The typical power consumption of a CE-L36LQ-EG card is 605 W, and its maximum power
consumption is 986 W.
You can obtain hardware specifications of the CE series switches using the Hardware Query
Tool.
Figure 22-38 Naming conventions of the CE12800 and CE12800E series switches
CE128 04 S
A B C
Table 22-20 describes the naming conventions of the CE12800 and CE12800E series
switches.
Table 22-20 Naming conventions of the CE12800 and CE12800E series switches
Identifier Meaning Description
A CloudEngine series data Stands for the CE12800 series data center core
center core switches switches.
B Number of LPU slots l 04: The chassis has 4 line card slots.
l 08: The chassis has 8 line card slots.
l 12: The chassis has 12 line card slots.
l 16: The chassis has 16 line card slots.
Cards
Figure 22-39 shows the naming conventions of the CE12800's line cards.
CE - L 24 L Q - EC1
A B C D E F
Table 22-21 describes the naming conventions of the CE12800's line cards.
Figure 22-40 shows the naming conventions of the CE12800E's line cards.
Table 22-22 describes the naming conventions of the CE12800E's line cards.
Figure 22-41 Naming conventions of the CE8800, CE7800, CE6800, and CE5800 series
switches
CE6850U-48S6Q-EI
A B C D E F
Table 22-23 describes the naming conventions of the CE8800, CE7800, CE6800, and
CE5800 series switches.
Table 22-23 Naming conventions of the CE8800, CE7800, CE6800, and CE5800 series
switches
Ide Meaning
ntifi
er
C Special function flag. This flag is not present if the product does not provide special
functions.
U: model supporting Fiber Channel (FC) interfaces
Table 22-25 lists the active optical cables (AOCs) supported by the CE series switches.
The CE series switches must use high-speed cables that are certified for Huawei Ethernet switches.
High-speed copper cables that are not certified for Huawei Ethernet switches cannot ensure transmission
reliability and may affect service stability. Huawei is not liable for any problem caused by the use of
high-speed copper cables that are not certified for Huawei Ethernet switches and will not fix such
problems.
Figure 22-42 Appearance of SFP+ to SFP+ and SFP28 to SFP28 high-speed copper cable
Figure 22-43 shows the appearance of QSFP+ to QSFP+ and QSFP28 to QSFP28 high-speed
copper cable.
Figure 22-43 Appearance of QSFP+ to QSFP+ and QSFP28 to QSFP28 high-speed copper
cable
Figure 22-44 shows the appearance of QSFP+ to 4*SFP+ and QSFP28 to 4*SFP28 high-
speed copper cable.
Figure 22-44 Appearance of QSFP+ to 4*SFP+ and QSFP28 to 4*SFP28 high-speed copper
cable
Figure 22-46 shows the structure of SFP+ to SFP+ and SFP28 to SFP28 high-speed copper
cable.
Figure 22-46 Structure of SFP+ to SFP+ and SFP28 to SFP28 high-speed copper cable
X1 X2
Figure 22-47 shows the structure of QSFP+ to QSFP+ and QSFP28 to QSFP28 high-speed
copper cable.
Figure 22-47 Structure of QSFP+ to QSFP+ and QSFP28 to QSFP28 high-speed copper
cable
X1 X2
Amphenol
Amphenol
Figure 22-48 shows the structure of QSFP+ to 4*SFP+ and QSFP28 to 4*SFP28 high-speed
copper cable.
Figure 22-48 Structure of QSFP+ to 4*SFP+ and QSFP28 to 4*SFP28 high-speed copper
cable
X2
X1 C
Figure 22-49 shows the structure of a CXP to CXP high-speed copper cable.
If 1-to-4 optical jumpers are routed to optical distribution frames (ODFs) and connected to
four 10GE multimode optical fibers each to transmit data over a certain distance, pay attention
to the following points:
l Determine the fiber connector type (LC/FC) supported by the ODF and select
appropriate optical jumpers for the supported connector type.
l Optical jumpers on the switches at both ends must be connected to the ODFs at the same
sequence.
l Each pair of TX and RX optical fibers between the ODFs must be cross-connected to
ensure normal transmission between the optical modules at both ends.
Fiber connection
sequence between
two ends
1-to-4 optical CE12804 chassis 1A 1A
CE12804 chassis 1-to-4 optical
jumper CMU CMU 1B 1B CMU CMU jumper
MPU MPU
MPU 2A 2A MPU
LPU 2B 2B LPU
LPU LPU
LPU 3A 3A LPU
LPU 3B 3B LPU
4A 4A
PM1 PM2 PM3 PM4 PM1 PM2 PM3 PM4
4B 4B
NOTE
ODFs rarely support MOP optical jumpers. If the two switches are not far from each other, it is
recommended that they be directly connected using MPO-MPO optical fibers.
MPO fibers are used for 40G and 100G optical modules. An MPO fiber consists of multiple
multi-mode fiber strands, and each multi-mode fiber strand provides one laser transmission
channel. Some fiber suppliers produce 8-strand MPO optical fibers, while some suppliers
produce 12-strand or 24-strand MPO fibers.
l A 40G optical module uses four channels to transmit laser and four channels to receive
laser. That is, a total of eight channels are required for a 40G optical module. 8-strand
and 12-strand MPO fibers use the same definition of fiber channels. Therefore, they are
equivalent in functionality when connecting to 40G optical modules.
l When 100G optical modules are used, choose MPO fibers according to the following
principles:
– Choose 24-strand fibers for CXP, CFP, and CFP2 modules.
– Choose 8-strand or 12-strand fibers for QSFP28 modules.
l The MPO-MPO, MPO-2*MPO, and MPO-3*MPO fibers have similar appearances except for the
number of MPO connectors at the other end (1, 2, and 3 respectively).
l The MPO-4*DLC, MPO-8*DLC, MPO-10*DLC, and MPO-12*DLC fibers have similar
appearances except for the number of DLC connectors at the other end (4, 8, 10, and 12 pairs
respectively).
Figure 22-50 shows the structure of an MPO-MPO fiber used for a 40GE optical module, and
Table 22-27 lists its pin assignments.
Figure 22-50 Structure of an MPO-MPO fiber used for a 40GE optical module
X1 X2
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
10 10
11 11
12 12
Table 22-27 Pin assignments of an MPO-MPO fiber used for a 40GE optical module
X1 Pin X2 Pin
1 12
2 11
3 10
4 9
9 4
10 3
11 2
12 1
Figure 22-51 shows the structure of an MPO-MPO fiber used for a 100GE optical module,
and Table 22-28 lists its pin assignments.
Figure 22-51 Structure of an MPO-MPO fiber used for a 100GE optical module
X1 X2
1 13 1 13
2 X1 14 2 X2 14
3 15 3 15
4 16 4 16
5 17 5 17
6 18 6 18
7 19 7 19
8 20 8 20
9 21 9 21
10 22 10 22
11 23 11 23
12 24 12 24
Table 22-28 Pin assignments of an MPO-MPO fiber used for a 100GE optical module
X1 Pin X2 Pin X1 Pin X2 Pin
1 24 13 12
2 23 14 11
3 22 15 10
4 21 16 9
5 20 17 8
6 19 18 7
7 18 19 6
8 17 20 5
9 16 21 4
10 15 22 3
11 14 23 2
12 13 24 1
Figure 22-52 and Figure 22-53 show the structures of MPO-4*DLC and MPO-8*FC fibers.
Their ping assignments are the same, as listed in Table 22-29.
4B
1 4A
2
3
4
5
6
7
8
9
10
11
12
2B
1
4A
2
3
4
5
6
7
8
9
10
11
12
1 1B
2 2B
3 3B
4 4B
9 4A
10 3A
11 2A
12 1A
Figure 22-54 shows the structure of an MPO-2*MPO fiber, and Table 22-30 lists its pin
assignments.
X1
X3
13 1 1
X1 X2/X3
14 2 2
15 3 3
16 4 4
17 5 5
18 6 6
19 7 7
20 8 8
21 9 9
22 10 10
23 11 11
24 12 12
2 12 NA
3 11 NA
4 10 NA
5 9 NA
7 NA 12
8 NA 11
9 NA 10
10 NA 9
14 1 NA
15 2 NA
16 3 NA
17 4 NA
19 NA 1
20 NA 2
21 NA 3
22 NA 4
Figure 22-55 shows the structure of an MPO-3*MPO fiber, and Table 22-31 lists its pin
assignments.
X2
X1 X3
X4
13 1 1
X1 X2/X3/X4
14 2 2
15 3 3
16 4 4
17 5 5
18 6 6
19 7 7
20 8 8
21 9 9
22 10 10
23 11 11
24 12 12
1 NA NA 9
2 12 NA NA
3 11 NA NA
4 10 NA NA
5 9 NA NA
6 NA 12 NA
7 NA 11 NA
8 NA 10 NA
9 NA 9 NA
10 NA NA 12
11 NA NA 11
12 NA NA 10
13 NA NA 4
14 1 NA NA
15 2 NA NA
16 3 NA NA
17 4 NA NA
18 NA 1 NA
19 NA 2 NA
20 NA 3 NA
21 NA 4 NA
22 NA NA 1
23 NA NA 2
24 NA NA 3
Figure 22-56 shows the structure of an MPO-8*DLC fiber, and Table 22-32 lists its pin
assignments.
8B
1 13 8A
2 14
3 15
4 16
5 17
6 18
7 19
8 20
9 21
10 22
11 23
12 24
2 1A 14 1B
3 2A 15 2B
4 3A 16 3B
5 4A 17 4B
7 5A 19 5B
8 6A 20 6B
9 7A 21 7B
10 8A 22 8B
Figure 22-57 shows the structure of an MPO-10*DLC fiber, and Table 22-33 lists its pin
assignments.
NOTE
If a 100GE interface is split into eight 10GE interfaces, pins 5A/5B and 10A/10B of the MPO-10*DLC
fiber become unusable. The first to fourth 10GE ports use the pins 1A/1B, 2A/2B, 3A/3B, and 4A/4B
respectively; the fifth 10GE interface uses pins 6A/6B; the sixth to eighth 10GE interfaces use pins
7A/7B, 8A/8B, and 9A/9B respectively.
10B
1 13 10A
2 14
3 15
4 16
5 17
6 18
7 19
8 20
9 21
10 22
11 23
12 24
2 1A 14 1B
3 2A 15 2B
4 3A 16 3B
5 4A 17 4B
6 5A 18 5B
7 6A 19 6B
8 7A 20 7B
9 8A 21 8B
10 9A 22 9B
11 10A 23 10B
Figure 22-58 shows the structure of an MPO-12*DLC fiber, and Table 22-34 lists its pin
assignments.
12B
1 13 12A
2 14
3 15
4 16
5 17
6 18
7 19
8 20
9 21
10 22
11 23
12 24
1 1A 13 1B
2 2A 14 2B
3 3A 15 3B
4 4A 16 4B
5 5A 17 5B
6 6A 18 6B
7 7A 19 7B
8 8A 20 8B
9 9A 21 9B
10 10A 22 10B
11 11A 23 11B
12 12A 24 12B
The figures provided here are for reference only and may differ from the actually delivered cables.
AC Power Cable
The AC power cables applicable to the CE12800&CE12800E series switches depend on the
type of power sockets used in your equipment room. Two types of AC power cables are
available to suit different power sockets: PDU power cable and country-specific power cable.
l PDU power cables are delivered with a power distribution unit (PDU) with C20 straight
sockets. A PDU power cable has a C19 straight female connector at one end and a C20
straight male connector at the other end, as shown in Figure 22-59.
l Country-specific power cables are delivered in compliance with standards of the
destination country or region. For example, AC power cables used in China have PI
angle male connectors, as shown in Figure 22-60.
If PDUs with C20 straight sockets are available in a cabinet, use PDU power cables
complying with local standards.
If there is no PDU with C20 straight sockets in a cabinet, use power cables specific to the
country or region where the switch is used.
Connection
An AC power cable transmits AC power to a chassis and allows maximum current of 16 A. It
is connected in the following way:
l The C19 straight female connector is connected to a socket on the power distribution unit
at the rear of the CE12800&CE12800E chassis.
l The country-specific connector or C20 straight male connector at the other end is
connected to an external power outlet.
DC Power Cable
DC power cables for a CE12800 or CE12800E switch include a -48 V power cable and an
RTN ground cable. DC power cables with the following cross-sectional areas are available: 16
mm2, 25 mm2, and 35 mm2.
You need to cut the delivered DC power cables into appropriate lengths according to actual
situations in your site and make DC power cables onsite.
Figure 22-61 shows the appearance of a DC power cable.
Connection
A DC power cable transmits DC power to a chassis and allows maximum current of 62 A. It
is connected in the following way:
l The JG2 M6 terminal is connected to power terminals on the power distribution unit at
the rear of CE12800&CE12800E chassis.
l The OT M8 terminal is connected to an external power source.
Connection
A high-voltage DC PDU power cable transmits 240 V high-voltage DC power to a chassis
and allows maximum current of 18.5 A. It is connected in the following way:
l The C19 straight female connector is connected to a socket on the power distribution unit
at the rear of the CE12800 chassis.
l The C20 straight male connector is connected as follows:
– If a high-voltage DC PDU is used, directly connect the C20 straight male connector
to a socket on the high-voltage DC PDU.
AC Power Cable
NOTE
The AC power cables delivered with a switch must comply with the standards used in the delivery
destination. This section uses the AC power cables used in China as an example.
PI straight male
C13 straight connector
female connector
Figure 22-65 shows a C13 straight female to C14 straight male AC power cable.
Figure 22-65 C13 straight female to C14 straight male AC power cable
Connection
An AC power cable transmits power to an AC power module in the chassis. It is connected in
the following way:
l The C13 straight female connector is connected to the power socket on the AC power
module.
l The PI/C14 male connector is connected to an external power source.
When a 600/1200 W AC&240 V DC power module uses 240 V DC power input, it must be
connected to a power supply device using a C13 straight female to C14 straight male AC
power cable. This power cable is connected as follows:
l The C13 straight female connector is connected to the power socket on the power
module.
l The C14 straight male connector is connected to a high-voltage DC PDU. If a DC power
distribution box is used, make OT or cord end terminals for the cable. Cut the C14
straight male connector off and crimp OT or cord end terminals on the bare wires.
Connect the blue wire to a positive terminal on the DC power distribution box, the
brown wire to a negative terminal, and the yellow-green wire to a protection ground.
DC Power Cable
DC power cables consist of the power cable for a 350 W DC power module and the power
cable for a 1200 W DC power module.
Figure 22-66 shows the appearance of the power cable for a 350 W DC power module.
Figure 22-66 Appearance of the power cable for a 350 W DC power module
Figure 22-67 shows the structure of the power cable for a 350 W DC power module.
Figure 22-67 Structure of the power cable for a 350 W DC power module
Figure 22-68 shows the appearance of the power cable for a 1200 W DC power module.
Figure 22-68 Appearance of the power cable for a 1200 W DC power module
Figure 22-69 shows the structure of the power cable for a 1200 W DC power module.
Figure 22-69 Structure of the power cable for a 1200 W DC power module
RTN
+
-
X1
NEG
Connection
Bare wires
High-voltage DC
straight female
connector
Connection
A 380 V high-voltage DC power cable has a high-voltage DC straight female connector at one
end and bare wires at the other end. It is used to connect a 600/1200 W high-voltage DC
power module to a power supply device:
l The high-voltage DC straight female connector is connected to the power socket on the
power module.
l The bare wires are connected to a 380 V high-voltage DC power distribution frame or
power distribution box. Crimp OT or cord end terminals on the bare wires, and then
connect the brown wire to a positive terminal, the blue wire to a negative terminal, and
the yellow-green wire to a protection ground.
A power module must use a power cable matching its current rating. Using an unmatched
power cable may cause a failure of the power module.
CE12800&CE12800E
To determine the optical modules and cables supported by a card of the CE12800&CE12800E
series switches, see the "Indicators and Ports" section of the card model under "Cards" in the
Hardware Description.
CE8800&7800&6800&5800
To determine the optical modules and cables supported by a switch, see the "Appearance and
Structure" section of the specific chassis model under Chassis > Chassis Models in the
CloudEngine 8800&7800&6800&5800 Series Switches Hardware Description.
AC power cables need to be selected based on the type of power sockets used in your
equipment room. Two types of AC power cables are available to suit different power sockets:
PDU power cable and country-specific power cable.
l PDU power cables are delivered with a power distribution unit (PDU) with C20 straight
female sockets. A PDU power cable has a C19 straight female connector at one end and
a C20 straight male connector at the other end. The number of power cables delivered
with a switch is the same as the number of power modules.
l Country-specific power cables are delivered in compliance with standards of the
destination country or region. For example, AC power cables used in China have PI
angle male connectors. The number of power cables delivered with a switch is the same
as the number of power modules.
The power entry module (PEM) for an AC chassis provides C20 sockets for C19 straight
female connectors of AC power cables. The PEM for a DC chassis provides double DC power
terminals for each power module.
Optical fibers are classified into multimode and single-mode optical fibers:
l Multimode optical fibers can be used with multimode optical modules to implement low-
cost, short-distance transmission. Multimode optical fibers are classified into OM1,
OM2, OM3, and OM4 types. OM1 and OM2 optical fibers are usually orange, whereas
OM3 and OM4 optical fibers are usually light green. For specifications of various optical
fibers, see the text printed on optical fibers.
l Single-mode optical fibers, usually G.652 fibers, can be used with single-mode optical
modules to implement long-distance transmission. Single-mode optical fibers are yellow.
Table 22-37 lists the interface standards supported by the CE series switches.
CE6810LI 16 CE6810-48S-LI,
l CE6810-48S-LI CE6810-48S4Q-LI:
V100R003C10 and later
l CE6810-48S4Q-LI versions
l CE6810-24S2Q-LI CE6810-24S2Q-LI,
l CE6810-32T16S4Q-LI CE6810-32T16S4Q-LI:
V100R005C10 and later
versions
CE6850HI/CE6850U-HI 16 CE6850-48S6Q-HI:
l CE6850-48S6Q-HI V100R005C00 and later
versions
l CE6850-48T6Q-HI
CE6850-48T6Q-HI,
l CE6850U-24S2Q-HI CE6850U-24S2Q-HI,
l CE6850U-48S6Q-HI CE6850U-48S6Q-HI:
V100R005C10 and later
versions
CE6870EI 9 CE6870-24S6CQ-EI,
l CE6870-24S6CQ-EI CE6870-48S6CQ-EI:
V200R001C00 and later
l CE6870-48S6CQ-EI versions
l CE6870-48T6CQ-EI CE6870-48T6CQ-EI:
V200R002C50 and later
versions
CE8850EI 16 CE8850-32CQ-EI:
l CE8850-32CQ-EI V200R002C50 and later
versions
l CE8850-64CQ-EI
CE8850-64CQ-EI:
V200R005C00 and later
versions
Type of ports used l 10GE optical All ports except GE ports can be used for stack
for stack connections ports connections.
l 10GE
electrical ports
l 40GE optical
ports
l 100GE optical
ports
Stack Port Types of CE8800, CE7800, CE6800, and CE5800 Series Switches
Table 22-40 Stack port requirements of CE8800, CE7800, CE6800, and CE5800 series
switches
Item Requirement Remarks
Type of ports l 10GE electrical All ports except GE ports can be used for stack
used for stack ports connections.
connections l 10GE optical
ports
l 25GE optical
ports
l 40GE optical
ports
l 100GE optical
ports
When setting up a stack, ensure that optical modules and cables on ports used for stack
connections are properly installed and these ports are Up.
NOTE
l 10GE optical ports can be used for stack connections only when they have 10GE optical modules
installed. They cannot be used for stack connections when having GE optical modules or GE copper
modules installed.
l SIP ports on the CE12800 and CE12800E can use network cables or LC fibers. When LC fibers are used,
the SIP ports must have GE optical modules installed.
l For CE8800, CE7800, CE6800, and CE5800, in V100R006C00, 40GE optical modules of the
QSFP-40G-SR-BD model cannot be used to set up a stack. This limitation does not apply to
V200R001C00 and later versions.
The following example describes basic stack configurations. For details, see the related product configuration
manual.
Attribute Configuration:
---------------------------------------------------------------
MemberID Domain Priority Mode Enable
Oper(Conf) Oper(Conf) Oper(Conf) Oper(Conf) Oper
---------------------------------------------------------------
Stack-Port Configuration:
------------------------------------------------------------------------------
--
Stack-Port Member Ports
------------------------------------------------------------------------------
--
Stack-Port1/1 10GE1/1/0/1 10GE1/1/0/2
------------------------------------------------------------------------------
--
In the command output, the Oper(Conf) field indicates the current configuration and
next startup configuration.
– Check whether all member switches use the same stack domain ID. All member
switches must use the same stack domain ID. Otherwise, they cannot set up a stack.
If member switches use different stack domain IDs, run the stack member
{ member-id | all } domain domain-id command to change their stack domain IDs
to the same.
– (Applicable only to the CE12800 and CE12800E) Check whether all member
switches use the same stack connection mode. All member switches must use the
same stack connection mode. Otherwise, they cannot set up a stack. MB indicates
the default MPU connection mode, and LC indicates the LPU connection mode.
If member switches use different stack connection modes, run the stack member
{ member-id | all } link-type { mainboard-direct | linecard-direct } command to
change their stack connection modes to the same.
– (Applicable only to the CE12800 and CE12800E) Check whether the stacking
function has been enabled on member switches. Member switches must have the
stacking function enabled to set up a stack. Enable indicates that the stacking
function is enabled, and Disable indicates that the stacking function is disabled.
If member switches have the stacking function disabled, run the stack enable
command to enable this function.
– Check whether member switches have offline configuration. The configuration
marked with an asterisk (*) is offline configuration. If offline configuration exists,
you must delete it. This is because offline configuration may case a stack
configuration conflict, which will cause a failure to set up a stack.
3. Check whether stack connections are correct.
Check whether stack connections are consistent with the plan and configuration.
4. Check whether ports used for stack connections are Up.
Run the display interface brief command to check whether the ports used for stack
connections are physically up, including stack member ports and SIP ports (applicable
only to the CE12800 and CE12800E). If these ports are physically Down, check whether
optical modules and fibers are faulty.
In V100R005C00 and later versions, you can also run the display stack link-state last-
down-reason command to check the reason why the stack link protocol becomes Down.
The reason includes incorrect configuration or cable connection.
5. Check stack failure event information.
Run the display stack troubleshooting command to check whether stack failure events
occur. This command can record some failures occurring during a stack setup, including
configuration errors and connection errors. Troubleshoot these failures according to
failure event description.
The following example describes how to load licenses in a stack of CE8800, CE7800,
CE6800, and CE5800 series switches.
l If there are multiple license files:
Compress multiple .dat license files into a .zip file, upload and load the .zip file to the
stack master.
<HUAWEI> license active license_file.zip
NOTE
Before a master/standby switchover is performed in a stack of CE12800 and CE12800E switches, ensure that
the master switch has two MPUs installed.
22.7.15 Can Four 10GE Ports of a CE5800 Provide Two Ports for
Stack Connections and the Other Two as Service Ports?
CE5810EI, CE5855EI and CE5850HI support this application but CE5850EI does not. When
10GE ports of a CE5850EI are used for stack connections, all the four 10GE ports must be
configured as physical member ports and cannot function as service ports again. This
restriction does not apply to CE5810EI, CE5855EI and CE5850HI.
SIP ports are located on MPUs. Each MPU has two SIP ports, as shown in Figure 22-71. A
SIP port is a combo port consisting of a GE electrical port and a GE optical port. It starts to
work immediately after a cable is connected and does not require any configuration. By
default, the working mode of a combo port depends on whether the electrical port or optical
port has a cable connected first. If the electrical and optical ports are connected at the same
time, the optical port works.
Figure 22-72 shows the recommended SIP port connections when each stack member switch
has two MPUs.
MPU A MPU A
MPU B MPU B
SIP port
NOTE
It is recommended that MPUs of the master and standby chassis should be directly connected through four
cables to ensure reliability. Although a stack can still operate when the MPUs are directly connected through
one cable, reliability is low. Therefore, one cable is not recommended.
l LPU connection mode: integrates management links and forwarding links. In this mode,
management links and forwarding links are connected through service ports on LPUs
without requiring SIP ports on MPUs.
SIP port
Service port
The MPU connection mode is recommended because it separates management links and
forwarding links and so ensures stack reliability.
Chain topology
Stack-port1/1 Stack-port2/2
Stack-port2/1 Stack-port3/1
SwitchA SwitchB SwitchC
Ring topology
SwitchA
Stack-port1/1 Stack-port1/2
Stack-port2/1
Stack-port3/2
Stack-port2/2 Stack-port3/1
SwitchB SwitchC
Member switches require at least one member port to set up a stack. To improve reliability, it
is recommended that two or more member ports be used to set up a stack.
22.7.28 Can 10GE Ports Split from a 40GE Port Be Used to Set Up a
Stack?
On all switch models except the CE7850EI and CE7855EI, 10GE ports split from a 40GE
port can be used to set up a stack.
e. Run the display stack command to check whether SwitchA is the master switch. If
so, perform an active/standby switchover in the stack. If not, go to the next step.
<HUAWEI> display switchover state //Check whether the system meets
switchover requirements.
Switchover State : Ready //You can perform an active/
standby switchover only when the Switchover State field displays Ready.
Switchover Policy : Board Switchover
MainBoard : 1
SlaveBoard : 2
<HUAWEI> system-view
[~HUAWEI] slave switchover enable //Enable the active/standby
switchover function.
[*HUAWEI] commit
[~HUAWEI] slave switchover //Perform an active/standby switchover.
Warning: This operation will switch the slave board to the master board.
Continue? [Y/N]: y
After cables are connected, run the display stack topology command to check
whether the stack connection topology is changed to the ring topology.
b. After the stack connection topology changes to ring topology, replace the switch
according to Replace one member switch in a stack of two member switches.
c. To restore the stack connection topology to chain topology after the replacement,
remove the stack cables connected in step 1.
Before restarting a member switch, save the stack configuration to ensure that the restarted
switch can join the stack normally.
To improve reliability, you are advised to add at least two physical member ports to each stack
port.
It is recommended that upstream and downstream devices connect to a stack using inter-
chassis link aggregation to ensure link reliability.
SwitchA SwitchB
10GE1/0/1 10GE2/0/1
Eth-Trunk 1
10GE1/0/1 10GE1/0/2
SwitchC
iStack Link
Common Link
Eth-Trunk
#
stack member 1 domain 10
stack member 1 priority 150
#
stack member 2 domain 10
stack member 2 priority 120
#
3. Check the stack configuration after the stack restarts, finding that the stack priority of the
switch with stack member ID 2 is the same as that of the switch with stack member ID 1
but not the configured one.
<HUAWEI> display stack configuration
Oper : Operation
Conf : Configuration
* : Offline configuration
Isolated Port : The port is in stack mode, but does not belong to any Stack-
Port
Attribute Configuration:
-----------------------------------------
MemberID Domain Priority
Oper(Conf) Oper(Conf) Oper(Conf)
-----------------------------------------
1(1) 10(10) 150(150)
2(2) 10(10) 150(120)
-----------------------------------------
......
To disable local traffic preferential forwarding, run the local-preference disable command in
the Eth-Trunk interface view.
<HUAWEI> system-view
[~HUAWEI] interface eth-trunk 10
[*HUAWEI-Eth-Trunk10] undo local-preference disable
10GE
Initial connection
1
10GE
Connected into a
ring topology
40GE
2
If the stack has been set up in a ring topology, remove 10GE ports from the logical stack port,
remove their cables, add 40GE ports in the logical stack port, and then connect the 40GE
ports using cables.
l System restart
a. Upload the new system software to the master switch and copy it to all the other
member switches. If a member switch has multiple MPUs, copy the new system
software to all MPUs.
b. Run the startup system-software system-file all command to specify the name of
the system software to use at the next startup.
c. Run the reboot command to restart the stack.
l Fast upgrade
a. Upload the new system software to the master switch and copy it to all the other
member switches. If a member switch has multiple MPUs, copy the new system
software to all MPUs.
b. Run the startup system-software system-file all command to specify the name of
the system software to use at the next startup.
c. Run the system-view command to enter the system view.
d. Run the stack command to enter the stack management view.
e. Run the stack upgrade fast command to start a fast upgrade.
After performing a fast upgrade, you can run the display stack upgrade status
command to check the upgrade status.
2. Clear the configuration file for next start and restart the switch.
<HUAWEI> reset saved-configuration
The configuration will be erased to reconfigure.Continue? [Y/N]: y
<HUAWEI> reboot
Warning: The current configuration will be saved to the next startup saved-
configuration file. Continue? [Y/N]: n
Warning: The system will reboot. Continue? [Y/N]: y
l When the stacking function is enabled on a switch, run the stack member member-id
renumber new-member-id [ inherit-config ] command to change its stack member ID.
<HUAWEI> system-view
[~HUAWEI] stack
[~HUAWEI-stack] stack member 1 renumber 2
Warning: The device will use the configuration of member ID 2 after the
device resets. Continue? [Y/N]: y
l If the stacking function has been enabled on the switch, run the stack member
{ member-id | all } domain domain-id command to change the stack domain ID.
<HUAWEI> system-view
[~HUAWEI] stack
[~HUAWEI-stack] stack member 1 domain 10
Changing the Stack Domain ID of the CE8800, CE7800, CE6800, and CE5800
Run the stack member { member-id | all } domain domain-id command to change the stack
domain ID.
<HUAWEI> system-view
[~HUAWEI] stack
[~HUAWEI-stack] stack member 1 domain 10
It is recommended that switches use the same system software version before they set up a
stack.
In V100R003C10 and later versions, stacking of CE12800 and CE12800E switches does not
require that two member switches use the same number of stack cards.
1. Connect SIP ports on MPUs to meet cable connection requirements of the MPU
connection mode.
2. Run the stack member all link-type mainboard-direct command to change the stack
connection mode to MPU connection.
<HUAWEI> system-view
[~HUAWEI] stack
[~HUAWEI-stack] stack member all link-type mainboard-direct
[*HUAWEI-stack] commit
3. Save the configuration and then restart the stack to make the configuration take effect.
2. Delete the stack member port configuration from ports to restore the ports as ordinary
service ports. Two methods are available:
– In the stack management view, run the undo port mode stack interface interface-
type { interface-number1 [ to interface-number2 ] } &<1-32> command.
<HUAWEI> system-view
[~HUAWEI] stack
[~HUAWEI-stack] undo port mode stack interface 10ge 1/0/1
– In the interface view, run the undo port mode stack command.
10GE1/0/1 10GE2/0/1
DAD Link
Stack Link
Relay
10GE1/0/1 10GE1/0/2
Eth-Trunk 1
10GE1/0/5 10GE2/0/5
DAD Link
Stack Link
# Configure the DAD relay function on the Eth-Trunk interface of the relay agent.
<HUAWEI> system-view
[~HUAWEI] sysname SwitchC
[*HUAWEI] commit
[~SwitchC] interface eth-trunk 1
[*SwitchC-Eth-Trunk1] trunkport 10ge 1/0/1
[*SwitchC-Eth-Trunk1] trunkport 10ge 1/0/2
[*SwitchC-Eth-Trunk1] dual-active proxy //Configure the relay function
on the Eth-Trunk interface of the relay agent.
[*SwitchC-Eth-Trunk1] commit
Management
network
MEth0/0/0 MEth0/0/0
DAD link
Stack link
l Stack port (supported only when the stack is set up by CE12800 and CE12800E switches
in MPU connection mode)
l Management port
SwitchA
Eth-Trunk
SwitchB SwitchC
Eth-Trunk
SwitchD
l If the Eth-Trunk is configured in manual mode, no loop will occur on SwitchA and
SwitchD after the stack split. The traffic forwarding paths are the same as those before
the stack split, for example, SwitchA -> SwitchB -> SwitchD and SwitchA -> SwitchC -
> SwitchD.
l When the Eth-Trunk is configured in static LACP mode, the traffic forwarding paths are
the same before and after the stack split if the standby switch still uses the system MAC
address. If the standby switch uses its own MAC address after the stack split, traffic can
only be forwarded through one port and no loop will occur. This is because, in the static
LACP mode, a physical member port in the Eth-Trunk goes Down and only one physical
member port can go Up.
Table 22-41 Device models and versions used to set up SVF systems of fixed switches
CE7850-32Q-EI l CE6810-48S4Q
-EI
CE6810-48S4Q-EI l CE5810-48T4S
-EI
l CE5810-24T4S
-EI
l CE5850-48T4S
2Q-EI
CE7850-32Q-EI l CE5850-48T4S
2Q-EI
l CE5855-48T4S
2Q-EI
l CE5855-24T4S
2Q-EI
l CE6810-48S4Q
-EI
l CE6810-48S-LI
l CE6810-48S4Q
-LI
l CE6810-24S2Q
-LI
l CE6810-32T16
S4Q-LI
l CE6850-48S6Q
-HI
l CE6850-48T6Q
-HI
l CE6851-48S6Q
-HI
l CE6850-48S6Q l CE5810-48T4S
-HI -EI
l CE6850U-48S6 l CE5810-24T4S
Q-HI -EI
l CE6850U-24S2 l CE6810-48S4Q
Q-HI -EI
l CE6810-48S-LI
l CE6810-48S4Q
-LI
l CE6810-24S2Q
-LI
l CE6810-32T16
S4Q-LI
CE6851-48S6Q-HI l CE5855-48T4S
2Q-EI
l CE5855-24T4S
2Q-EI
l CE6810-48S-LI
l CE6810-48S4Q
-LI
l CE6810-24S2Q
-LI
l CE6810-32T16
S4Q-LI
l CE7850-32Q- l CE5850-48T4S
EI 2Q-EI
l CE7855-32Q- l CE5855-48T4S
EI 2Q-EI
l CE5855-24T4S
2Q-EI
l CE6810-48S4Q
-EI
l CE6810-48S-LI
l CE6810-48S4Q
-LI
l CE6810-24S2Q
-LI
l CE6810-32T16
S4Q-LI
l CE6850-48S6Q
-HI
l CE6850-48T6Q
-HI
l CE6851-48S6Q
-HI
l CE6850U-48S6
Q-HI
l CE6850U-24S2
Q-HI
NOTE
The
CE6850U-48S6Q-
HI/
CE6850U-24S2Q-
HI can function as a
leaf switch from
V200R002C50.
l CE6850-48S6Q l CE5810-48T4S
-HI -EI
l CE6850U-48S6 l CE5810-24T4S
Q-HI -EI
l CE6850U-24S2 l CE6810-48S4Q
Q-HI -EI
l CE6850-48T6 l CE6810-48S-LI
Q-HI l CE6810-48S4Q
NOTE -LI
The l CE6810-24S2Q
CE6850-48T6Q-HI
-LI
can function as a
parent switch from l CE6810-32T16
V200R002C50. S4Q-LI
l CE6851-48S6Q l CE5855-48T4S
-HI 2Q-EI
l CE6855-48S6Q l CE5855-24T4S
-HI 2Q-EI
l CE6856-48S6Q l CE6810-48S-LI
-HI l CE6810-48S4Q
NOTE -LI
The
CE6856-48S6Q-HI l CE6810-24S2Q
can function as a -LI
parent switch from l CE6810-32T16
V200R002C50.
S4Q-LI
– If there is only one license file, run the license active file-name command once.
3. Run the display license command to check whether a license file has been loaded.
Centralized forwarding Leaf and parent switches Leaf switches need to send
have the same entry all traffic to the parent
specifications; therefore, switch for forwarding,
low-cost leaf switches can leading to a high latency.
be deployed.
1. Run the command in the switch mode stack member member-id command in the stack
management view to set the next startup mode of the leaf switch to the stack mode.
2. Run the reset slot slot-id command in the user view to restart the leaf switch.
If a leaf switch has been removed from an SVF system, perform the following steps to restore
the switch to the stack mode:
1. Log in to the switch through the console port.
2. During the startup process, press Ctrl+B to enter the BIOS menu. The prompt
information is "Press CTRL+B to enter BIOS menu or CTRL+E to boot DFX".
3. In the BIOS menu, choose 6. Modify stack parameters > 2. Modify stack
configuration > 2. Stack mode > 3. Return > 1. Continue to boot to set the working
mode of the switch to stack mode, and then choose to continue the startup
22.9 Virtualization - VS
CE12800E (equipped with ED-E, EG-E, and EGA-E series cards) switches support VS in
V200R002C50 and later versions, and each switch can have a maximum of 16 VSs (including
the Admin-VS) configured.
The CE12800E does not support the VS function after FD-X series cards are installed.
VS1
LPU VS2
To check which ports on an LPU belong to the same chip, run the display device port-
map [ slot slot-id ] command in the user or system view.
<HUAWEI> display device port-map slot 1
-------------------------------------------------------------------------
Slot Type Engine Interface
-------------------------------------------------------------------------
1 CE-L24XS-EA 0 10GE1/0/0 - 10GE1/0/23
-------------------------------------------------------------------------
For example, if 24 ports on a CE-L24XS-EA belong to the same forwarding chip, all
these ports will be assigned to VSs in group mode.
l Any ports on an LPU can be assigned to VSs in port mode according to network
requirements, as shown in Figure 22-82.
VS1
LPU VS2
The following example describes basic stack configurations. For details, see the related product configuration
manual.
<HUAWEI> system-view
[~HUAWEI] admin
[~HUAWEI-admin] virtual-system vs1 //Create a VS named vs1.
[*HUAWEI-admin-vs:vs1] port-mode group //Set the VS mode to group mode. Two
VS modes are available: group mode and port mode.
[*HUAWEI-admin-vs:vs1] assign interface 10GE 1/0/0 //Assign a physical port
to VS.
Warning: All configurations of the interfaces will be deleted. All interfaces of
the same group will be assigned. Continue? [Y/N]: y
[*HUAWEI-admin-vs:vs1] quit
[*HUAWEI-admin] commit
[~HUAWEI-admin] display virtual-system
---------------------------------------------
Name Status
---------------------------------------------
Admin-VS running
vs1 running
---------------------------------------------
If the status of a VS is running, the VS is created.
Hardware management Hardware operations, such as device restart, card reset, and e-
label backup, can be performed only in the admin-VS and are not
supported in non-admin-VSs. VS information such as CPU and
memory usage can be viewed in non-admin-VSs.
Agile Controller-DCN Only the admin-VS can connect to the Agile Controller-DCN.
Feature Support in a VS
MPLS, GRE, L3VPN, l Among all VSs in port mode, MPLS, GRE, L3VPN, VLL,
VLL, PWE3, VPLS, PWE3, VPLS, EVN and multicast functions can only be
EVN and multicast configured in the admin-VS.
l MPLS, GRE, L3VPN, VLL, PWE3, VPLS, EVN and
multicast functions can be configured in any VS in group
mode.
M-LAG l Among all VSs in port mode, M-LAG can only be configured
in the admin-VS.
l M-LAG can be configured in any VS in group mode.
Feature Support in a VS
TCAM ACL l All VSs in group mode support TCAM ACL customization.
customization l In earlier versions of V100R006C00, among VSs in port
mode, only the admin-VS supports TCAM ACL
customization. In V100R006C00 and later versions, the VS in
port mode and TCAM ACL customization cannot be
configured simultaneously.
22.9.16 Can VSs in Group Mode and VSs in Port Mode Co-exist?
VSs in group mode and VSs in port mode can be configured on the same switch.
In V100R005C00 and later versions, after some ports in a group have been configured as
stack member ports, other ports in this group can be assigned to a VS in group mode.
For CE7800 series switches, run the display device manufacture-info command.
<HUAWEI> display device manufacture-info
-------------------------------------------------------------------------
---
Slot Card Type Serial-number Manu-date
-------------------------------------------------------------------------
---
1 -- CE7850-32Q-EI 210235925010E9000010 2014-09-19
FAN1 FAN-40HA-F 210235909610E9000038 2014-09-11
FAN2 FAN-40HA-F 210235909610E9000024 2014-09-11
PWR1 PAC-600WA-F 2102310PMJD0E9000037 2014-09-09
PWR2 PAC-600WA-F 2102310PMJD0E9000010 2014-09-08
-------------------------------------------------------------------------
---
NOTE
For details about the CE series switch license, including license introduction, common operations, and
application procedure, see the CloudEngine Series Switches License Use Guide.
NOTE
l The FCF feature is controlled by two kinds of licenses: CE-LIC-FCF-ALL and CE-LIC-FCF-PORT.
For CE12800 and CE12800E series switches, the license CE-LIC-FCF-ALL supports FCF
configuration on all the interfaces. The license CE-LIC-FCF-PORT supports FCF configuration on a
total of 48 interfaces of the same card or different cards. You can load multiple CE-LIC-FCF-PORT
licenses to meet capacity expansion requirements. For CE8800, CE7800, and CE6800 series
switches, the license CE-LIC-FCF-PORT supports FCF configuration on a total of 16 interfaces.
l After a license is loaded, you can configure the Telemetry function only after the device is restarted
in the V200R003C00 version, and you can configure functions corresponding to other features
without the need of restarting the device.
l For the CE8868EI, after one of the following licenses is loaded, you need to run the active card-
license command to enable the corresponding license in the specified subcard slot. The CE8868EI
has four subcard slots. You can purchase licenses based on the number of required subcard slots.
l 25GE rate upgrade license: CE-LIC-25G01
l 40GE port license: CE-LIC-40G01
l 100GE port license: CE-LIC-100G01
l For the CE6880EI, the 1588v2 function has been controlled by the license since V200R005C10. If
the switch is upgraded from V200R005C00 to V200R005C10 or a later version, the 1588v2 function
is not affected, but the 1588v2 function configuration cannot be added or modified. You are advised
to apply for and load a 1588v2 license after the upgrade.
Table 22-44 License control items supported by the CE12800 series switches
Table 22-45 License control items supported by the CE12800E series switches
License Model Product Before After License
Control Item Model and License Activation
Minimum Activation
Version
Table 22-46 License control items supported by the CE8800 series switches
License Model Product Model Before After License
Control Item and Minimum License Activation
Version Activation
Table 22-47 License control items supported by the CE7800 series switches
License Model Product Model Before After License
Control Item and Minimum License Activation
Version Activation
Table 22-48 Functional license control items supported by the CE6800 series switches
License Model Product Model Before After License
Control Item and Minimum License Activation
Version Activation
FCF for all CE68-LIC- CE6800 series This function This function
ports of the FCFAL switches does not take takes effect and
CE6800: CE- excluding the effect and can be
LIC-FCF-ALL CE6810LI and cannot be configured.
CE6880EI: configured.
V100R005C00
Table 22-49 Functional license control items supported by the CE5800 series switches
License Model Product Model Before After License
Control Item and Minimum License Activation
Version Activation
The following example describes how to load licenses in a stack of CE8800, CE7800,
CE6800, and CE5800 series switches.
l If there are multiple license files:
Compress multiple .dat license files into a .zip file, upload and load the .zip file to the
stack master.
<HUAWEI> license active license_file.zip
NOTE
A license authorization certificate is in electronic PDF format or printed format and made together with
the sales contract. The printed certificates are delivered with the products. The electronic certificates in
PDF format are sent to the customers' email boxes by the license management system.
l If you do not have a license authorization certificate, send an email that contains the
Huawei contract number of the purchased device to [email protected] to seek
technical support.
NOTE
If you do not have the Huawei contract number, contact the dealer.
Master board license state: Demo. The license for the current configuration will
expire in 12 day(s).
Apply for authentic license before the current license expires.
After you purchase a license, you can obtain the license file (*.dat file) based on the license
authorization code (LAC) and device ESN using the following methods:
l Visit the website app.huawei.com/isdp, click Password Activation under License
Activation, enter the activation code and ESN as prompted to complete license
activation, and then download the license file.
l Send the LAC and device ESN to [email protected] to obtain the license file.
Run the following command in any view to view a license revocation code:
l CE12800 and CE12800E series switches: Run the display license revoke-ticket
[ chassis chassis-number ] command.
l CE8800, CE7800, CE6800, and CE5800 series switches: Run the display license
revoke-ticket [ slot slot-id ] command.
chassis-number indicates a stack chassis ID, and slot-id indicates a switch slot ID.
<HUAWEI> display license revoke-ticket
Info: The revoke ticket is:
LIC20121103006100:27C1B773ED11D9F877855CDAEE74ABFE60E07126.
Version Support
The following table lists the models and minimum software versions supporting NQA.
CE8868EI/CE8861EI V200R005C10
CE8860EI V100R006C00
CE8850-32CQ-EI V200R002C50
CE8850-64CQ-EI V200R005C00
CE7850EI V100R003C00
CE7855EI V200R001C00
CE6810EI V100R003C00
CE6850EI V100R001C00
CE6850-48S6Q-HI V100R005C00
CE6850-48T6Q-HI/CE6850U-HI/ V100R005C10
CE6851HI
CE6855HI V200R001C00
CE6870-24S6CQ-EI/CE6870-48S6CQ-EI V200R001C00
CE6870-48T6CQ-EI/CE6860EI/CE6856HI/ V200R002C50
CE6880EI
CE6857EI V200R005C10
CE6875EI V200R003C00
CE6865EI V200R005C00
CE5880EI V200R005C10
CE5810EI V100R002C00
CE5850EI V100R001C00
CE5850HI V100R003C00
CE5855EI V100R005C10
CE12804/CE12808/CE12812 V100R001C00
CE12816 V100R003C00
CE12804S/CE12808S V100R005C00
CE12800E V200R002C50
NOTE
The CE6810LI does not support IPv4 or IPv6 Layer 3 forwarding. After the IPv4 or IPv6 function is enabled
on an interface of the CE6810LI, the configured IPv4 or IPv6 address can only be used to manage the switch.
License Support
CE12800 and CE12800E series switches:
The NQA IPv6 function can be used only after the IPv6 feature is enabled, whereas the IPv6
feature is under license control. By default, the IPv6 feature is disabled on new purchased
CE12800 and CE12800E series switche. To use the IPv6 feature, apply for and purchase the
license from the equipment supplier.
NQA is a basic feature of CE8800, CE7800, CE6800, and CE5800 Series Switches and is not
under license control.
Run the following commands in the user view to revoke a license, change it to the trial state,
and obtain a revocation code.
l On CE12800 and CE12800E series switches, run the license revoke [ chassis chassis-
number ] command.
l On CE8800, CE7800, CE6800, and CE5800 series switches, run the license revoke
[ slot slot-id ] command.
chassis-number specifies a stacked chassis ID, and slot-id specifies a switch's slot ID.
<HUAWEI> license revoke
Warning: The license will switch to trial state. Continue? [Y/N]: y
MainBoard:
Info: Succeeded in revoking the license. The revoke ticket is
LIC201411261KSC50:87CE09A70A7401C7D0E1853B7931E3FA755AC88D.
A license is bound to the backplane ESN of a switch, that is, the chassis ESN.
<HUAWEI> display sn
Chassis 1:
Equipment SN(ESN): 210211377210E2000079
License ESN: 210211377210E2000079
----------------------------------------------------------------------------
Slot Sub Type SN P/N
----------------------------------------------------------------------------
-- -- BackPlane 210211377210E2000079 02113772
1 -- CE-L48XS-EA 021KNW10D1000065 03021KNW
2 -- CE-L24XS-EF 022GGE10D8000020 03022GGE
5 -- CE-L36CQ-FD 023CLS10G4000122 03023CLS
6 -- CE-L48GT-EA 022BEU10D7000114 03022BEU
7 -- CE-L48XS-ED -- --
9 -- CE-L24XS-ED -- --
11 -- CE-L48XS-EA 021KNW10D2000002 03021KNW
13 -- CE-MPUA 021NUD10D1000120 03021NUD
14 -- CE-MPUA 021NUD10E1000495 03021NUD
15 -- CE-CMUA 021NMV10E1000648 03021NMV
16 -- CE-CMUA 021NMV10E1000632 03021NMV
---- More ----
Networking Description
As shown in Figure 22-84, the device functions as an SSH server and has reachable routes to
PC1 and PC2, and the IP address of the management interface on the SSH server is
10.137.217.203. Two login users client001 and client002 need to be configured on the SSH
server. PC1 uses client001 to log in to the SSH server and undergoes a password
authentication; PC2 uses client002 to log in to the SSH server and undergoes an RSA
authentication.
Figure 22-84 Networking diagram for logging in to the device using STelnet
10.137.217.10/24 10.137.217.203/24
PC1
PC2
Precautions
Take the following precautions when configuring STelnet-based login:
l STelnet V1 has security vulnerabilities. You are advised to log in to the device using
STelnet V2.
l Before configuring STelnet-based login, install the SSH server login software on PC1,
and install the key pair generation software, public key conversion software, and SSH
server login software on PC2. This example uses the third-party software PuTTY as the
SSH server login software.
Procedure
1. Generate a local key pair on the SSH server.
<HUAWEI> system-view
[~HUAWEI] sysname SSH Server
[*HUAWEI] commit
[~SSH Server] rsa local-key-pair create //Generate local RSA host and
server key pairs.
The key name will be: SSH Server_Host
The range of public key size is (512 ~ 2048).
NOTE: Key pair generation will take a short while.
Input the bits in the modulus [default = 2048] : 2048 //You are advised to
set the size of the key pairs to 2048 to improve device security. In
V200R001C00 and later versions, the switch supports only 2048-bit key pairs.
You do not need to enter the value.
[*SSH Server] commit
# Create an SSH user named client001 and configure the password authentication mode
for the user.
[*SSH Server] aaa
[*SSH Server-aaa] local-user client001 password irreversible-cipher
Huawei@123 //Configure the local user name and password.
[*SSH Server-aaa] local-user client001 level 3 //Set the user level to 3.
[*SSH Server-aaa] local-user client001 service-type ssh //Set the service
type for the local user to SSH.
[*SSH Server-aaa] quit
[*SSH Server] ssh user client001 authentication-type password //Set the
authentication mode for client001 to password authentication.
# Create another SSH user named client002 and configure related information for the
user.
a. Create an SSH user named client002 and configure the RSA password
authentication mode for the user.
[*SSH Server] ssh user client002 authentication-type rsa //Set the
authentication mode for client002 to RSA authentication.
[*SSH Server] ssh authorization-type default root
b. Run puttygen.exe on PC2 to generate the public and private key files.
Select SSH2 RSA and click Generate. By moving the cursor in the blank area to
generated the key.
After the key is generated, click Save public key to save the key as the key.pub
file.
Click Save private key. In the PuTTYgen Warning dialog box that is displayed,
click Yes. The private key is saved as the private.ppk file.
c. Run sshkey.exe on the client. Convert the generated public key to the character
string required for the device.
Open the key.pub file.
Click Convert(C). The original and converted public keys are displayed.
d. On the SSH server, enter the RSA public key generated on PC2.
[*SSH Server] rsa peer-public-key rsakey001 //Enter the RSA public key
view.
[*SSH Server-rsa-public-key] public-key-code begin //Enter the RSA
public key editing view.
[*SSH Server-rsa-public-key-rsa-key-code] 30820108 02820101 00DD8904
1A5E30AA 976F384B 5DB366A7 //Edit the public key.
[*SSH Server-rsa-public-key-rsa-key-code] 048C0E79 06EC6B08 8BB9567D
75914B5B 4EA7B2E5 1938D118
[*SSH Server-rsa-public-key-rsa-key-code] 4B863A38 BA7E0F0D BE5C5AE4
CA55B192 B531AC48 B07D21E3
[*SSH Server-rsa-public-key-rsa-key-code] 62E3F2A5 8C04C443 CF51CF51
136B5B9E 812AB1B7 1250EB24
[*SSH Server-rsa-public-key-rsa-key-code] A4AE5083 A1DB18EC E2395C9B
B806E8F0 0BE24FB5 16958784
[*SSH Server-rsa-public-key-rsa-key-code] 403B617F 8AAAB1F8 C6DE8C3C
F09E4D23 7D1C17BF 4AAF09C4
[*SSH Server-rsa-public-key-rsa-key-code] 74C083AF 17CD3075 3396B322
32C57FF0 B1991971 02F1033B
[*SSH Server-rsa-public-key-rsa-key-code] 81AA6D47 44520F23 685FAF72
04BA4B6E 615EF224 14E64E2A
[*SSH Server-rsa-public-key-rsa-key-code] 331EEB7F 188D9805 96DBFD30
0C947A5A BA879DC4 F848B769
[*SSH Server-rsa-public-key-rsa-key-code] 513C35CD B52B2917 02B77693
F79910EE 5287F252 977F985E
e. On the SSH server, bind the RSA public key to the client002 user.
[*SSH Server] ssh user client002 assign rsa-key rsakey001
[*SSH Server] commit
4. Configure the STelnet service type for the client001 and client002 users.
[~SSH Server] ssh user client001 service-type stelnet
[*SSH Server] ssh user client002 service-type stelnet
[*SSH Server] commit
Verification
After the configuration is complete, use PuTTY to log in to the SSH server from the PC.
d(The output information may vary according to version. Therefore, the output information on
your device may be different from that provided in this example.)
l Log in to the SSH server as the client001 user from PC1 in password authentication
mode.
# Log in to the device using PuTTY, enter the device's IP address, and select the SSH
protocol.
Figure 22-90 Logging in to the SSH server using PuTTY in password authentication
mode
# Click Open. On the displayed page, enter the user name client001 and password
Huawei@123, then press Enter to log in to the SSH server.
login as: client001
Sent username "client001"
[email protected]'s password:
Info: The max number of VTY users is 21, the number of current VTY users
online is 2, and total number of terminal users online is 2.
The current login time is 2012-08-04 20:09:11+00:00.
First login successfully.
<SSH Server>
l Log in to the SSH server as the client002 user from PC2 in RSA authentication mode.
# Log in to the device using PuTTY, enter the device's IP address, and select the SSH
protocol.
# Choose Connection > SSH in the navigation tree. The page shown in Figure 22-92 is
displayed. Select 2 under Preferred SSH protocol version.
# Choose Connection > SSH > Auth in the navigation tree. The page shown in Figure
22-93 is displayed. Select the private.ppk file corresponding to the public key
configured on the server.
# Click Open. Enter the user name client002 at the prompt, and press Enter. You have
logged in to the SSH server.
login as: client002
Authenticating with public key "rsa-key"
Info: The max number of VTY users is 5, the number of current VTY users
online is 2, and total number of terminal users online is 2.
The current login time is 2012-08-06 04:30:23+00:00.
First login successfully.
<SSH Server>
2. The switch functions as the SSH server and its SSH version is later than that of the SSH
client. The SSH login fails because the SSH versions are different.
Solution:
– In V100R002C00, run the ssh server compatible-ssh1x disable command in the
system view to configure the SSH server to be compatible with earlier SSH
versions. By default, compatibility with earlier SSH versions is enabled.
– In V100R003C00 and later versions, run the ssh server compatible-ssh1x enable
command in the system view to configure the SSH server to be compatible with
earlier SSH versions. By default, compatibility with earlier SSH versions is
disabled.
3. The switch functions as an SSH client and connects to the SSH server for the first time.
The authentication fails because the SSH server's public key is not configured on the
switch.
Solution: Run the ssh client first-time enable command in the system view to enable
first-time authentication on the SSH client. By default, the function is disabled. Access
the SSH server and save the SSH server's public key on the SSH client. The switch uses
the saved public key to authenticate the SSH server next time it connects to the server.
4. The SSH service type is not configured. By default, no service type is configured.
Solution: Run the ssh user user-name service-type { stelnet | all } command in the
system view to set the SSH service type to STelnet or all.
If the fault persists, contact technical support personnel.
Table 22-51 Mapping between authentication modes and IP address or user name locking
Login Mode Authentication Mode Authentication Mode
Causing IP Address Causing User Name
Locking Locking
l After the IP address of a client is locked, wait for 5 minutes until the system
automatically unlocks the IP address, and attempt to log in to the switch. Alternatively,
use another account to log in to the switch and perform the following operation to unlock
the IP address in advance.
– Telnet login
– STelnet login
# Unlock the IP address 10.1.2.3.
<HUAWEI> activate ssh server ip-block ip-address 10.1.2.3
l If the current user name is locked, use another user name to log in to the switch, and run
the activate aaa local-user user-name command in the user view to unlock the user
name.
# Unlock the user name admin1234.
<HUAWEI> activate aaa local-user admin1234
If you log in to a CE switch for the first time, you need to connect your terminal and the
switch through a serial cable. CE12800E allow you to use a Micro USB port for fist login,
and CE12800, CE8860EI, CE8861EI, CE8868EI, CE8850-64CQ-EI, CE6875EI, CE6850HI,
CE6856-48T6Q-HI, CE6855-48T6Q-HI, and CE6850U-HI also allow you to use a mini USB
port for first login. That is, you can connect your terminal and the switch through a Mini
USB/Micro USB cable for first login.
Configuration rollback is similar to the system restore function of the Windows operating
system. A configuration rollback point is generated each time the commit command is
executed. You can run the rollback configuration last number-of-commits command in the
user view to roll back the current configuration to a previous configuration rollback point.
number-of-commits specifies the configuration rollback point.
For example, run the following commands to create VLANIF 2 and configure an IP address
for VLANIF 2.
<HUAWEI> system-view
[~HUAWEI] vlan batch 2
[*HUAWEI] interface vlanif 2
[*HUAWEI-Vlanif2] ip address 192.168.0.1 255.255.255.0
[*HUAWEI-Vlanif2] commit
[~HUAWEI-Vlanif2] quit
[~HUAWEI] quit
Run the rollback configuration last 1 command in the user view to roll back the
configuration to the previous configuration rollback point.
<HUAWEI> rollback configuration last 1
Warning: This operation will revert configuration changes to the previous status.
Continue? [Y/N]: y
Run the display configuration commit changes last 1 command in any view to check the
configuration changes at configuration rollback point 1.
NOTE
The command output provided here is used for reference only. The output information on your device
may differ from that provided in this example.
NOTE
After the configuration is complete, you can connect to the device from your PC through FTP.
Assume that the management IP address of the FTP server is 10.1.1.1. You can perform the
following steps on your PC to connect to the device (The messages may vary according to the
operating system of the PC, and the following messages are for reference only):
C:\Documents and Settings\Administrator> ftp 10.1.1.1
Connected to 10.1.1.1.
220 FTP service ready.
User (10.136.23.5:(none)): admin1234 //Enter the user name admin1234.
331 Password required for admin1234.
Password: //Enter the password Helloworld@6789. The password will not be
displayed on the terminal.
230 User logged in.
ftp> binary
200 Type set to I.
ftp> put devicesoft.cc //Upload the system file devicesoft.cc to the device.
200 Port command okay.
150 Opening BINARY mode data connection for /devicesoft.cc
226 Transfer complete.
ftp: 4342 bytes sent in 223.23Seconds 560.79Kbytes/sec.
ftp> get vrpcfg.cfg //Back up the configuration file vrpcfg.cfg to the PC.
200 Port command okay.
150 Opening BINARY mode data connection for /vrpcfg.cfg
226 Transfer complete.
ftp: 4342 bytes received in 0.04Seconds 120.6Kbytes/sec.
NOTE
After specifying the configuration file for next startup, do not save the current configuration when
using the reboot command to restart the device. When the following message is displayed, enter
N:
Warning: The current configuration will be saved to the next startup saved-configuration file.
Continue? [Y/N]: N
Alternatively, run the copy source-filename startup destination-filename [ slot slot-id |
all ] command to copy the configuration file and set the file copy as the configuration
file for next startup. This command has been supported since V200R005C00.
NOTE
CE12800 and CE12800E series switches do not support slot slot-id and all parameters.
l To specify patch.pat as the patch file for next startup, run (all indicates all cards or stack
member switches):
<HUAWEI> startup patch patch.pat all
l The system software package used by the CE12800E series switches is different from
that used by the CE12800 series switches. (The file name extension of the system
software package is .cc.)
The system software package of V200R002C50SPC800 is used as an example.
– The system software package used by the CE12800E series switches is named as
CE12800E-V200R002C50SPC800.cc.
– The system software package used by the CE12800 series switches is named as
CE12800-V200R002C50SPC800.cc.
NOTE
The system software of the CE12800E equipped with the ED-E, EG-E, or EGA-E series cards is
different from that of the CE12800E equipped with the FD-X series cards. For example, the name
of the system software is CE12800E-V200R005C00.cc when the ED-E, EG-E, or EGA-E series
cards are installed, and the name of the system software is CE12800E-X-V200R005C10.cc when
the FD-X series cards are installed. Supported configurations differ depending on the system
software. Before replacing the system software, verify that the current configurations match the
system software of the device.
l CE8800, CE7800, CE6800, and CE5800 series switches of EI/HI/LI models have
corresponding system software packages that are named based on the switch models.
For example:
– CE7850EI-V100R003C00SPC600.cc
– CE6810EI-V100R003C00SPC600.cc
– CE6850EI-V100R003C00SPC600.cc
– CE5810EI-V100R003C00SPC600.cc
– CE5850EI-V100R003C00SPC600.cc
– CE5850HI-V100R003C00SPC600.cc
Some models are further classified into the optical and electrical models. For example,
for CE6870EI series switches, the system software package used by the
CE6870-48T6CQ-EI is different from that used by other CE6870EI switches.
– The system software package used by the CE6870-48T6CQ-EI is named as
CE6870T-EI-V200R002C50SPC800.cc.
– The system software package used by other CE6870EI switches is named as
CE6870S-EI-V200R002C50SPC800.cc.
l Symptom 3: The switch restarts repeatedly, the BIOS menu is displayed, but the switch
fails to register.
Troubleshooting method:
a. Check whether the system software package is damaged. If the system displays
error messages indicating that the size of the system software package is incorrect
or CRC check fails, the system software package is damaged. You are advised to
use the BIOS menu to upload the correct system software package again.
b. Check whether the MPU successfully obtains the system software package. For
details, see How Do I Handle an MPU's Failure to Obtain the System Software
Package During a Startup.
c. Check whether the memory is normal. Restart the switch, and press CTRL+T
within 3 seconds when "Press CTRL+T for full memory test" is displayed to enter
the BIOS memory test menu to test the system memory. If the test result shows that
the memory is abnormal, contact technical support personnel.
Starting ...
If this fault occurs, perform the following steps to locate the fault (save operation records):
Step 1: Enter the BIOS menu and view the system flash directory to check whether the switch
contains the system software package (with the file name extension .cc).
l If so, go to step 2.
l If not, go to step 4.
Take system software package CE12800-V100R002C00SPC200.cc as an example. The
operation procedure is as follows:
Press CTRL+B to enter BIOS menu: //Press CTRL+B to enter the BIOS menu.
1. Continue to boot
In the file system submenu, select 3. Return to return to the BIOS menu.
Step 2: Check and modify system startup parameters, and specify the system software
package in the flash directory as the next startup file.
1. Check the current startup file. The operation procedure is as follows:
BIOS Menu (Version: 221)
1. Continue to boot
2. Update from serial interface
3. Update from ethernet interface
4. Modify startup parameters
5. File system
6. Modify BIOS password
7. Modify console password
8. Reboot
Startup Submenu
Startup Submenu
To prevent information loss, power off the MPU after collecting the preceding
information.
2. Set the connected interface and communication parameters, as shown in Figure 22-95.
Select the serial interface that is connected to the switch on your PC. For example, on a
PC running a Windows OS, you can view interface information in Device Manager.
Communication parameter settings on the terminal emulation software must be the same
as the default values on the switch, which are: 9600 bit/s baud rate, 8 data bits, 1 stop bit,
no parity check, and no flow control.
NOTE
By default, no flow control mode is configured on the switch. Because RTS/CTS is selected in the
software by default, you need to deselect RTS/CTS; otherwise, you cannot enter commands.
NOTE
A PC may have more than one serial interface. Select the interface to which the console cable is
connected.
If you modify the serial interface communication parameters on the switch, modify the communication
parameters on the PC to ensure that they use the same communication parameter settings, and then set
up the connection again.
Preparations
Obtain the system software package (.cc file) and patch file (.pat file).
l To obtain a system software package (.cc file), log in to Huawei official website, choose
a CE series switch, click the Downloads tab page, and select the software version.
l The method of obtaining a patch file (.pat file) is the same as that of obtaining a system
software package.
Upgrade Process
1. Upload the system software package and patch file to the switch. For details on how to
upload a file, see 22.13.1 How Do I Enable the FTP Server on the Device?.
2. Configure the system software package for next startup. The following uses a CE12800
as an example, if the CE12800 has two MPUs, copy the system software package to the
standby MPU and specify the next startup file for the standby MPU.
<HUAWEI> copy CE12800-V100R006C00SPC300.cc slave#flash:
<HUAWEI> startup system-software CE12800-V100R006C00SPC300.cc all
NOTE
The CE switch does not need to be restarted. If a fault has occurred, see the precaution in
patch release notes to determine whether to restart the switch or card for fault rectification.
You can run the display patch-information verbose command in any view to check detailed
information about a patch file.
****************************************************************************
* Information about patch errors is as follows: *
****************************************************************************
SlotId CurrentVersion
----------------------------------------------------------------------------
No patch error occurs on any board
Board Info :
----------------------------------------------------------------------------------
------------
SlotId ProcId State PatchType Valid PatchEffectiveTime
PatchFileName
----------------------------------------------------------------------------------
------------
1 1049 Running C YES 2017-11-14 14:02:09.297
HP000012.pat
1 1049 Running C YES 2017-11-14 14:02:09.308
HP000028.pat
----------------------------------------------------------------------------------
------------
Total = 2
22.15.4 What Are the Differences Between the reboot and reboot
fast Commands?
When the reboot command is executed, the system displays a message asking you whether to
save the configuration before restart.
When the reboot fast command is executed, the system restarts quickly without displaying
any message and the unsaved configuration is lost.
Exercise caution when you use the command. You are advised to use it under the guidance of
technical personnel.
l Restart the switch with no configuration using the BIOS/BootLoader menu. The
following uses the CE6850EI as an example.
Enter the BIOS main menu, and select 4 to access the startup submenu. Then select 2 to
modify startup configurations.
BIOS Menu (Version: 281)
1. Continue to boot
2. Update from serial interface
3. Update from ethernet interface
4. Modify startup parameters
5. File system
6. Modify BIOS password
7. Modify console password
8. Restore factory defaults
9. Reboot
Startup Submenu
configuration
NOTE
By default, link flapping protection is enabled on CE series switches in V100R002C00 with the patch
V100R002SPH006 loaded as well as V100R003C00 and later versions.
You can run the display interface command to view the interface status.
If an interface is in ERROR DOWN(link-flap) state, you are advised to perform the following
operations:
1. Check historical alarms. If many Up/Down alarms of the interface were generated for a
long time, the interface status is unstable. You are advised to check whether the optical
modules and optical fibers on the local and remote interfaces are normal.
2. Restore the interface to Up state. By default, an interface in ERROR DOWN state does
not automatically go Up. Run the shutdown and undo shutdown commands to
manually enable the interface. In V200R002C50 and later versions, you can remove and
reinstall the transmission medium on an optical interface to manually restore the
interface to Up state. If an interface is not in Error-Down state, run the error-down
auto-recovery cause link-flap interval interval-value command to configure the
interface to automatically go Up when it is in ERROR DOWN(link-flap) state.
To disable the device from detecting Up/Down events on an interface, you can perform either
of the following operations:
l Disable link flapping detection on the interface. To disable link flapping detection on an
interface, enter the interface view first.
– In V100R002C00 and earlier versions, run the undo port link-flap trigger error-
down command to disable link flapping detection.
– In V100R002C00 with the patch V100R002SPH006 loaded as well as
V100R003C00 and later versions, run the port link-flap threshold 0 command to
disable link flapping detection.
NOTE
If link flapping detection is disabled on an interface, the system does not detect faults caused by
link flapping on the interface in real time. Exercise caution when deciding to perform this
operation.
l Adjust the link flapping detection threshold. If the network adapter of a server
disconnects intermittently during server startup, the device may incorrectly determine the
interface status. You can run the port link-flap { interval interval-value threshold
threshold-value | interval interval-value | threshold threshold-value } command to
increase the link flapping detection threshold.
For more Error-Down troubleshooting methods, see CloudEngine Series Switches Error-
Down Mechanism.
l 850 nm: multimode optical modules. They are often used for short-distance transmission.
l 1310 nm: mostly single-mode optical modules and a few multimode optical modules.
They are often used for medium- or long-distance transmission.
l 1550 nm: single-mode optical modules. They are often used for long-distance
transmission.
You can also obtain optical module information in the following ways:
GE and 10GE optical modules look the same. You can identify them by the interface type and speed on
the labels attached to the optical modules:
l eSFP or SFP identifies a GE optical module. SFP+ or XFP identifies a 10GE optical module.
l A GE optical module has GE or 1.25G on its label, whereas a 10GE optical module has 10G on its
label.
l The transmit power of a long-distance optical module is often larger than its overload
power. Therefore, when using such optical modules, select optical fibers of an
appropriate length to ensure that the actual receive power is smaller than the overload
power. If the optical fibers connected to a long-distance optical module are too short, use
an optical attenuator to reduce the receive power on the remote optical module.
Otherwise, the remote optical module may be burnt.
l It is recommended that idle optical modules be covered with dust-proof plugs to keep the
optical modules clean. An unclean optical module may degrade the quality of optical
signals or even cause link failures or error codes on the link.
l An optical module usually has Rx/Tx flags or in/out arrowheads (▼/▲) to help you
identify the transmitter and receiver. The transmitter on one end must be connected to
receiver on the other end. Otherwise, the link between two ends cannot go Up.
To avoid injury to your eyes, do not look into optical modules on a device without eye
protection, whether they are long-distance or short-distance optical modules.
Optical modules work with optical fibers to transmit optical signals. An optical module works
in full-duplex mode with a pair of optical fibers, one for transmission and one for receiving,
as shown in the following figure.
Interface 1 Interface 2
Optical fiber
Rx Tx
Optical module Optical module
Tx Rx
Optical fiber
Both optical modules and fibers support two modes: single mode and multimode. The
following table describes mapping between optical modules and fibers.
Table 22-52 describes usage scenarios of high-speed cables and cable connections in these
scenarios.
QSFP+ to QSFP+ l Scenario 1: used for 40GE optical port connection or stacking
high-speed cable between CE12800&CE12800E series switches or
CE8800&7800&6800&5800 series switches
l Scenarios 2: used for 40GE connection between
CE12800&CE12800E and CE8800&7800&6800&5800
switches
Both ends connect to a 40GE optical port.
QSFP+ to 4*SFP+ When a 40GE optical port is split into four 10GE optical ports:
high-speed cable l Scenario 1: used for 10GE optical port connection or stacking
between CE12800&CE12800E series switches or
CE8800&7800&6800&5800 series switches
l Scenarios 2: used for 10GE connection between
CE12800&CE12800E and CE8800&7800&6800&5800
switches
One end connects to the 40GE optical port, and the other end
connects to four 10GE optical ports.
Optical Fiber
Table 22-53 describes usage scenarios of optical fibers and fiber connections in these
scenarios.
NOTE
Table 22-53 describes connections of pigtails at two ends of the optical fibers. Generally, optical ports of
two switches are not directly connected by optical fibers. An optical distribution frame (ODF) is usually
deployed between the two switches.
MPO-4*DLC fiber When a 40GE optical port is split into four 10GE optical ports:
MPO-8*FC fiber l Scenario 1: used for 10GE optical port connection or stacking
between CE12800&CE12800E series switches or
CE8800&7800&6800&5800 series switches
l Scenarios 2: used for 10GE connection between
CE12800&CE12800E and CE8800&7800&6800&5800
switches
One end connects to the 40GE optical port using a QSFP+ optical
module, and the other end connects to four 10GE optical ports using
SFP+ optical modules.
AOC cable
Figure 22-103 shows the appearance of a QSFP+ to QSFP+ AOC cable.
Table 22-54 describes usage scenarios of AOC cables and cable connections in these
scenarios.
QSFP+ to QSFP+ l Scenario 1: used for 40GE optical port connection or stacking
AOC cable between CE12800&CE12800E series switches or
CE8800&7800&6800&5800 series switches
l Scenarios 2: used for 40GE connection between
CE12800&CE12800E and CE8800&7800&6800&5800
switches
Both ends connect to a 40GE optical port.
QSFP+ to 4*SFP+ When a 40GE optical port is split into four 10GE optical ports:
AOC cable l Scenario 1: used for 10GE optical port connection or stacking
between CE12800&CE12800E series switches or
CE8800&7800&6800&5800 series switches
l Scenarios 2: used for 10GE connection between
CE12800&CE12800E and CE8800&7800&6800&5800
switches
One end connects to the 40GE optical port, and the other end
connects to four 10GE optical ports.
NOTE
To know which optical modules are certified for Huawei Ethernet switches, see "Optical Modules" in the
Hardware Description. All the optical modules listed in this section have been certified for Huawei
Ethernet switches. You can also contact Huawei Technical Assistance Center (TAC) to certify your
optical modules.
[Board Properties]
BoardType=PLRXPLSCS4322N
BarCode=CB45UF1A2
Item=
Description=10300Mb/sec-850nm-LC-80m(0.05mm)-30m(0.0625mm)-300m(OM3)
Manufactured=2011-11-07
VendorName=JDSU
IssueNumber=
CLEICode=
BOM=
......
l If the device runs V100R003 or a later version, you can check whether an optical module
is certified for Huawei Ethernet switches to determine whether the optical module is
officially delivered by Huawei. If not, an alarm is generated. Therefore, you can check
whether an alarm is generated to determine whether an optical module is officially
delivered by Huawei.
The following alarm is generated when an optical module that is not certified for Huawei
Ethernet switches is detected. You can determine the interface where such an optical
module resides using EntPhysicalName.
1.3.6.1.4.1.2011.5.25.219.2.4.9 hwOpticalUnAuthorized
The optical module was not certified by HUAWEI.
(EntPhysicalIndex=[EntPhysicalIndex], EntPhysicalName=[EntPhysicalName],
EntityTrapFaultID=[EntityTrapFaultID], Reason=[Reason])
NOTE
When an interface detects an optical module that is not certified for Huawei Ethernet switches, an alarm
is generated. The generation of alarms is controlled by the alarm function. By default, the alarm
function is enabled. You can run the undo transceiver non-certified-alarm disable command in the
interface view to enable the alarm function.
22.18.8 Why Are Alarms About Optical Modules That Are Not
Certified for Huawei Ethernet Switches Generated When Optical
Modules Having the Huawei Logo Are Used?
Vendor information of optical modules early delivered from Huawei may not be recorded.
Therefore, alarms about optical modules that are not certified for Huawei Ethernet switches
are generated. You can run the transceiver non-certified-alarm disable command to disable
the alarm function for these optical modules.
<HUAWEI> system-view
[~HUAWEI] transceiver non-certified-alarm disable
[*HUAWEI] commit
When the receive or transmit power of an optical module exceeds the upper threshold or falls
below the lower threshold, the optical module may fail to work properly.
[Board Properties]
BoardType=AFBR-703SDZ
BarCode=AD1148A084M
Item=
In the command output, the BarCode field indicates the serial number.
You can run the display interface [ interface-type interface-number | slot slot-id ]
transceiver [ verbose ] command to check information about optical modules.
Generally, the label of an optical module provides the following information: manufacturer,
serial number, speed, center wavelength, mode (single-mode/multimode), and transmission
distance.
l There are FE, GE, 10GE, 25GE, 40GE, and 100GE optical modules to suit various
interface speeds, as well as those with special speeds, including 2G, 4G, 8G, and 16G
optical modules.
l Center wavelengths of optical modules include:
– 850 nm: multimode optical modules. They are often used for short-distance
transmission.
– 1310 nm: mostly single-mode optical modules and a few multimode optical
modules. They are often used for medium- or long-distance transmission.
– 1550 nm: single-mode optical modules. They are often used for long-distance
transmission.
l Depending on core diameters and features, optical fibers are classified into single-mode
and multimode fibers. Generally, multimode fibers have large core diameters and severe
dispersion, so they transmit optical signals over short distances. Single-mode fibers have
small dispersion and can transmit optical signals over long distances.
l Optical signals of different wavelengths can travel different distances. Transmission
distances of optical modules are affected by attenuation and dispersion of optical signals
during transmission. Generally, a distance of less than 2 km is considered a short
transmission distance, a distance of 10 km to 20 km is considered a medium transmission
distance, and a distance beyond 20 km is considered a long transmission distance. The
optical modules used on CE series switches support a transmission distance of up to 100
km.
The CE series switches support CFP, CFP2, CXP, and QSFP28 form factors for 100GE
interfaces.
Two 100GE optical interfaces can interconnect successfully when the following two
conditions are met:
l The two interfaces comply with the same standard. For example, a 100GBASE-SR10
interface with the CFP form factor can interconnect with a 100GBASE-SR10 interface
with the CXP form factor.
l Interface parameters (wavelength and physical interface attributes) are the same. For
example, a 100GBASE-SR10 interface with the CFP form factor and a 100GBASE-
eSR10 interface with the CFP2 form factor can interconnect because they both have 100
Gbit/s speed and 850 nm wavelength.
BIDI optical modules must be used in pairs. For example, an SFP-GE-LX-SM1310-BIDI module must
be used with an SFP-GE-LX-SM1490-BIDI module.
l Huawei CFP and CFP2 100G optical modules can interoperate with Cisco CPAK 100G
optical modules.
l Among QSFP28 100G optical modules, only the ones with 10 km transmission distance
can interoperate with Cisco CPAK 100G optical modules.
l Among CXP 100G optical modules, only the 100GBASE-SR10 modules can
interoperate with Cisco CPAK 100G optical modules.
A 40GE interface on the following cards can be split into two 10GE interfaces:
l A CE88-D16Q card
l A 40GE interface on a CE-L48XS-FDA card can be split into two 10GE interfaces. The
two 40GE interfaces on the CE-L48XS-FDA card must be split simultaneously so that
converted 10GE interfaces can work properly.
A 40GE or 100GE interface on the following models and cards can be split into four 10GE
interfaces:
l CE switch models excluding the preceding models
l A QSFP28 interface (or a 100GE interface converted from a 40GE interface) that
connects to a 40GE optical module on the CE8860EI, CE8861EI, CE8868EI, CE8850EI,
CE6870EI, CE6875EI, CE6860EI, CE-L12CQ-FD card, CE-L36CQ-FD card, CE-
L36CQ-FG card, CE-L36CQ-FD1 card, CE-L36CQ-SD card, CE-L16CQ-FD card, FD-
X series cards, CEL16CQEG-E card, CEL24XSEGA-E card, CEL48XSEGA-E card,
CE-L36LQ-FD card, and CE-L24LQ-FD card
1. Configure the interface split mode. (Only the CE8850-64CQ-EI, CE7850EI and CE7855EI
support this step.)
<HUAWEI> system-view
[~HUAWEI] port split mode mode2 slot 1 //Configure the interface split mode.
Warning: This operation may cause the split configurations on the board to be
lost. This mode will take effect after board is reset.
Continue? [Y/N]: y
[*HUAWEI] commit
[~HUAWEI] quit
<HUAWEI> save
<HUAWEI> reset slot 1 //Restart the switch to make the configuration take effect.
NOTE
For device models that support dynamic interface split, see Configuration - Interface Management
Configuration Guide - Ethernet Interface Configuration - Configuring Attributes for an Optical Interface
- Configuring Interface Split.
3. Create an Eth-Trunk.
<HUAWEI> system-view
[~HUAWEI] interface eth-trunk 2 //Create an Eth-Trunk.
[*HUAWEI-Eth-Trunk2] trunkport 40ge 1/0/1:1 to 1/0/1:2 //Add 40GE1/0/1:1 and
40GE1/0/1:2 to Eth-Trunk 2.
[*HUAWEI-Eth-Trunk2] commit
Figure 22-106 Mapping between 10GE ports and sequence number indicators
1 23
CE-L12LQ-EF
1
RUN/ALM
2
3
4
40GE
0 1 2 3 4 5 6 7 8 9 10 11 Breakout
CE-L12LQ-EF
1
RUN/ALM
1
2 2
3 3
4 4
40GE
Breakout
Figure 22-107 Mapping between 10GE ports and sequence number indicators
1 2
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
SYS
MST
STAT
SPEED
STACK
MODE/ID 40GE 1 2 3 4
CE7850-32Q-EI Breakout
29 30 31 32
40GE 1 2 3 4
Breakout
1
2
3
4
CE-L08CC-EC l 3*40GE
l 12*10GE
CE-L12CF-EG l 2*40GE
l 8*10GE
NOTE
If a 100GE interface of the CE-L12CF-
EG card connects to a CFP2-100G-
LR4 optical module, the 100GE
interface cannot be split into eight
10GE interfaces.
CE8860EI, – 4*25GE
CE8861EI, For 100GEx/1/2 of the
CE8868EI CE88_D24S2CQ and
CE88_D24S2CQ_U, 100GEx/1/1
of the CE88_D24T2CQ,
100GEx/1/8 of the CE88_D8CQ,
40GEx/1/13 and 40GEx/1/14 of the
CE88_D16Q on the CE8861EI and
CE8868EI, the interface split
function and the IOAM function are
mutually exclusive and cannot be
configured together. When the
interface split function is canceled
and the IOAM function is then
configured, the IOAM function
takes effect only after the device is
restarted.
CE8850EI – 4*25GE
CE6870EI – 4*25GE
and
CE6875EI
CE6860EI – 4*25GE
and For 100GE1/0/3 on the CE6865EI,
CE6865EI the interface split function and the
IOAM function are mutually
exclusive and cannot be configured
together.
on different switches are numbered in the following ways. CE5800, CE7800 and
CE8800 series switches do not support this numbering rule.
– On CE6800 series switches, a 10GE interface converted from 40GE x/y/n is
numbered 10GE x/y/[48 + 4(n - 1) + z]. z indicates the interface location and the
value of z ranges from 1 to 4.
– On CE12800 series switches, a 10GE interface converted from 40GE x/y/0/n is
numbered 10GE x/y/0/(4n + z). z indicates the interface location and the value of z
ranges from 0 to 3.
A 40GE interface on a CE88-D16Q card can be split into two 10GE interfaces. For example,
two 10GE interfaces converted from 40GE 1/1/1 are numbered: 40GE 1/1/1:1 and 40GE
1/1/1:2.
Numbering rules after 100GE interfaces are split
For example, a 100GE interface is numbered 100GE x/y/0/n and is split. Take a 100GE
interface on a CE-L04CF-EF card as an example. The converted interfaces are numbered
using the following two rules:
l If the dimension numbering rule is used, the 100GE interface supports the following split
types:
– If the 100GE interface is split into ten 10GE interfaces, a converted 10GE interface
is numbered 100GE x/y/0/n:z. z indicates the interface location and the value of z
ranges from 1 to 10.
– If the 100GE interface is split into eight 10GE interfaces, a converted 10GE
interface is numbered 100GE x/y/0/n:z. z indicates the interface location and the
value of z ranges from 1 to 8.
– If the 100GE interface is configured as two 40GE interfaces, a converted 40GE
interface is numbered 100GE x/y/0/n:z. z indicates the interface location and the
value of z is 1 or 2.
– If the 100GE interface is configured as one 40GE interface, the converted 40GE
interface is numbered 100GE x/y/0/n:1.
l If the interface sequence numbering rule is used, the 100GE interface supports the
following split types:
– If the 100GE interface is split into ten 10GE interfaces, a converted 10GE interface
is numbered 10GE x/y/0/(10n + z). z indicates the interface location and the value
of z ranges from 0 to 9.
– If the 100GE interface is split into eight 10GE interfaces, a converted 10GE
interface is numbered 10GE x/y/0/(10n + z). z indicates the interface location and
the value of z ranges from 0 to 7.
– If the 100GE interface is configured as two 40GE interfaces, a converted 40GE
interface is numbered 40GE x/y/0/(2n + z). z indicates the interface location and the
value of z is 0 or 1.
– If the 100GE interface is configured as one 40GE interface, the 40GE interface is
numbered 40GE x/y/0/2n.
<HUAWEI> system-view
[~HUAWEI] interface eth-trunk 2
[*HUAWEI-Eth-Trunk2] trunkport 40ge 1/0/1:1
[*HUAWEI-Eth-Trunk2] commit
Figure 22-108 Connecting a split 40GE interface to the remote 10GE interfaces
10GE
10GE
C
Ampheno
l
40GE 10GE
A
High-speed cable
10GE
Fiber connection
sequence between
two ends
1-to-4 optical CE12804 chassis 1A 1A
CE12804 chassis 1-to-4 optical
jumper CMU CMU 1B 1B CMU CMU jumper
MPU MPU
MPU 2A 2A MPU
LPU 2B 2B LPU
LPU LPU
LPU 3A 3A LPU
LPU 3B 3B LPU
4A 4A
PM1 PM2 PM3 PM4 PM1 PM2 PM3 PM4
4B 4B
You can run the display interface [ interface-type [ interface-number ] ] command in any
view to check the Speed and Duplex fields of the local and remote interfaces to determine
whether the two interfaces use the same rate and duplex mode.
You can use Interface Query 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.
To use the interface query tool for CE series switches, visit http://support.huawei.com/
onlinetoolsweb/DC/en/interface_query.html.
In the preceding command output, the current state field displays Down, indicating that the
interface cannot work properly.
Common Causes
The possible causes are as follows:
l The interface is manually shut down.
l The cable or optical fiber is faulty.
l The optical module is faulty.
l The local and remote interfaces fail to negotiate parameters including the duplex mode
and rate.
Troubleshooting Procedure
1. Check whether the interface has been manually shut down using the shutdown
command. If so, run the undo shutdown command in the interface view to enable the
interface.
2. Connect the interface to another interface. If the interface is still Down, replace the
optical module and optical fiber or cable on the interface to check whether the interface,
optical module, or optical fiber/cable has failed.
3. Check whether the negotiate parameters including the duplex mode and rate.of local and
remote interfaces are same. If not, change them to same.
4. If the fault persists, see "Interface Troubleshooting" in the Troubleshooting - Hardware.
Troubleshooting Procedure
After determining the interface, run the display interface brief command, and view the
bandwidth usage of incoming and outgoing traffic on the interface in the last 300 seconds in
the InUti/OutUti field. Perform the following operations based on the actual situation.
l If burst traffic exceeds the upper alarm threshold instantly and does not affect services,
you are advised to run the trap-threshold { input-rate | output-rate } bandwidth-in-use
[ resume-rate resume-threshold ] command in the interface view to increase the upper
alarm threshold.
l If the traffic rate on an interface remains high for a long time, you are advised to expand
the link bandwidth. Determine the link bandwidth based on the traffic volume on the
network to ensure that bandwidth can be fully used and packet loss will not occur.
l If you consider that the alarm severity is too high and want to mask this alarm, run the
undo snmp-agent trap enable feature-name devm [ trap-name trap-name ] command
in the system view to mask this type of alarms on all the interfaces. After the alarms are
masked, the device still record alarm messages, but does not display them.
NOTE
For types of single-fiber directional optical modules, see 22.18.17 Do the CE Series Switches Support
BIDI Optical Modules?. Optical modules except for single-fiber directional optical modules support
single-fiber unidirectional connections.
NOTE
l Before running the undo portswitch command in the interface view to switch an interface to Layer
3 mode, delete all the non-default configurations on the interface.
l Before running the portswitch command in the interface view to switch an interface from Layer 3
mode to Layer 2 mode, delete all the Layer 3 service configurations on the interface.
l Before configuring the undo portswitch and undo portswitch batch commands on the CE12800E
that has the FD-X series cards installed, CE6865EI, CE6857EI, CE6855HI, CE6856HI, CE7855EI,
CE8861EI, and CE8868EI, run the vlan reserved for main-interface startvlanid to endvlanid
command in the system view to configure a dedicated reserved VLAN for Layer 3 main interfaces.
l The CE6810LI does not support IPv4 or IPv6 Layer 3 forwarding. After the IPv4 or IPv6 function is
enabled on an interface of the CE6810LI, the configured IPv4 or IPv6 address can only be used to
manage the switch.
MTU
Issue 09 (2019-03-05) Copyright © Huawei Technologies Co., Ltd. 620
IP packet Network layer
Destination Source
MAC MAC
Type IP packet FCS Data link layer
CloudEngine 12800, 12800E, 8800, 7800, 6800, and 5800
Series Switches
Troubleshooting Guide 22 FAQs
NOTE
If connected optical interfaces do not go Up, locate the faults by referring to Two Connected Optical
Interfaces Cannot Go Up.
Switch interfaces discard packets when traffic is overloaded or burst traffic occurs.
Buffer threshold parameters of an interface include the reserved buffer threshold, threshold
for triggering flow control, threshold for stopping flow control, and headroom buffer size.
For CE switches running V200R003C00 and later versions, you can run the flow-control
[ input | output ] buffer { reserved reserved-value | xon xon-value | xoff xoff-value | hdrm
hdrm-value } command in the interface view to modify values of the buffer threshold
parameters of an interface.
The following tables list default values and value ranges of different buffer threshold
parameters on interfaces of different models of devices.
Table 22-59 Value ranges and default values of the xoff and xon parameters in a standalone
scenario
Card reserved xoff xoff- xon xon-value hdrm hdrm-
reserved-value value value
FD-X series Value range: 1 Value range: 1 Value range: 1 Value range: 1
cards to 131071 to 131071 to 131071 to 4095
Default value: 7 Default value: Default value: Default value:
1024 912 256
Table 22-60 Value ranges and default values of threshold parameters on CE8800, CE7800,
CE6800, and CE5800 series switches
Product reserved xoff xoff- xon xon-value hdrm hdrm-
reserved-value value (Threshold value
(Reserved (Threshold for Stopping (Headroom
Buffer for Triggering Flow Control) Buffer Size)
Threshold) Flow Control)
CE5850EI and Value range: 1 Value range: 1 Value range: 1 Value range: 1
CE5855EI to 32767 to 32767 to 4095 to 4095
Default value: 7 Default value: Default value: Default value:
16383 on a 3840 100 on a 40GE
40GE interface interface and 16
and 4095 on on other
other interfaces interfaces
CE7850EI and Value range: 1 Value range: 1 Value range: 1 Value range: 1
CE7855EI to 131071 to 131071 to 131071 to 4095
Default value: 7 Default value: Default value: Default value:
1024 912 256
After loopback detection is enabled on an interface of a CE switch, a test packet sent from the
interface is sent back to the interface inside the system. This function can only be used to
check whether the interface functions properly, that is, this function cannot be used to check
packets sent and received on this interface or check whether packet loss occurs on the link.
By default, the CE-L36LQ-FD card has 36 40GE interfaces. The 40GE optical interfaces
numbered 4n and 4n+1 (n is an integer that ranges from 0 to 8) can be switched into 100GE
interfaces using the port mode 100ge command. If a 40GE interface is switched into a
100GE interface, the interface can work at the rates of 40 Gbit/s and 100 Gbit/s, and other
40GE interfaces that are not switched can work properly.
Run the display mac-address command in any view to check the mapping between the MAC
address of the downstream device and interface.
------------------------------------------------------------------------------
-
Total items on slot 1: 1
Original-Port: indicates the interface that learns the MAC address first. Move-Ports
indicates the interface that learns the MAC address later. There may be multiple values
of Move-Ports.
2. Check whether a loop occurs.
– Run the display interface brief | include up command to check traffic on the
interface. Run this command multiple times to check traffic.
When the switch is deployed on a network where loops occur, the response speed of
the switch is slow. Check traffic on all interfaces in Up state. The values of InUti
(inbound bandwidth usage) and OutUti (outbound bandwidth usage) increase
gradually on the interface where a loop occurs, and even approximate 100%,
exceeding service traffic.
When multiple network adapters connect to a server, add the interface of the switch connected to the
server to the Eth-Trunk or disable MAC address flapping detection in a VLAN.
4. Collect information and seek technical support.
NOTE
When MAC address flapping occurs, see An Alarm About MAC Address Flapping Is Generated on
a CE Switch.
– When 256 is specified, the switch allows a maximum of 256 LAGs and each LAG
supports a maximum of 8 member interfaces.
– When 512 is specified, the switch allows a maximum of 512 LAGs and each LAG
supports a maximum of 4 member interfaces.
– When 1024 is specified, the switch allows a maximum of 1024 LAGs and each
LAG supports a maximum of 2 member interfaces.
CE8800, CE7800, CE6800, and CE5800 Series Switches (Excluding CE6870EI and
CE6875EI)
Table 22-61 describes LAG specifications supported by other models of CE8800, CE7800,
CE6800, and CE5800 series switches except the CE5810EI, CE5880EI and CE6880EI. The
CE5810EI allows a maximum of 128 LAGs and each LAG supports a maximum of 8 member
interfaces. The CE5880EI and CE6880EI allow a maximum of 1024 LAGs and each LAG
supports a maximum of 64 member interfaces.
Table 22-61 LAG specifications supported by other models of CE8800, CE7800, CE6800,
and CE5800 series switches
Version Configurable Specification Default Value
NOTE
After the assign forward eth-trunk mode command is used to change the number of LAGs supported
by the switch, restart the switch to make the configuration take effect.
CE6810LI of V100R005C00:
l When 24 is specified, the switch allows a maximum of 24 LAGs and each LAG supports
a maximum of 32 member interfaces.
l When 48 is specified, the switch allows a maximum of 48 LAGs and each LAG supports
a maximum of 16 member interfaces.
l When 96 is specified, the switch allows a maximum of 96 LAGs and each LAG supports
a maximum of 8 member interfaces.
On other models:
l When 64 is specified, the switch allows a maximum of 64 LAGs and each LAG supports
a maximum of 32 member interfaces.
l When 128 is specified, the switch allows a maximum of 128 LAGs and each LAG
supports a maximum of 16 member interfaces.
V100R005C10 and later versions:
l When 64 is specified, the switch allows a maximum of 64 LAGs and each LAG supports
a maximum of 32 member interfaces.
l When 128 is specified, the switch allows a maximum of 128 LAGs and each LAG
supports a maximum of 16 member interfaces.
l When 256 is specified, the switch allows a maximum of 256 LAGs and each LAG
supports a maximum of 8 member interfaces.
l When 512 is specified, the switch allows a maximum of 512 LAGs and each LAG
supports a maximum of 4 member interfaces.
l When 1024 is specified, the switch allows a maximum of 1024 LAGs and each LAG
supports a maximum of 2 member interfaces.
Manual load balancing mode This mode is used when the remote All versions
device does not support LACP.
LACP mode Static LACP This mode is used when the remote All versions
mode device supports LACP.
NOTE
When the number of Eth-Trunk member interfaces is nth power of 2, traffic can evenly load
balanced. For example, when an Eth-Trunk has 2, 4, or 8 member interfaces(divided by 16),
the Eth-Trunk can obtain an integral value of timeslots and the load balancing is even. If the
number of Eth-Trunk member interfaces is not nth power of 2 (for example, 3), one interface
can send packets six times and the other two interfaces can send packets five times. In this
case, the load balancing is uneven.
It is recommended that the number of Eth-Trunk member interfaces is nth power of 2 to
achieve even load balancing.
The load balancing mode is only valid for the outbound interface of traffic. In multi-level load balancing
scenarios, if traffic of the inbound interface is uneven, change the load balancing mode of the outbound
interface on the uplink device.
In the scenario where Eth-Trunk and ECMP are deployed and traffic is unevenly load balanced, configure
different hash algorithms for the Eth-Trunk and ECMP and change the value of universal-id as needed.
NOTE
l Uplink and downlink interfaces cannot be located on the CE-L24XS-EA, CE-L48XS-EA, CE-
L48GT-EA, CE-L48GT-EC, CE-L48GS-EA, or CE-L48GS-EC card.
l CE12800 of V100R003C10 and later versions supports traffic replication, Only the CE6870EI and
CE6875EI of CE8800, CE7800, CE6800, and CE5800 series switches support traffic replication,
other devices do not support this function.
l All the interfaces that join the downlink Eth-Trunk replicate traffic on uplink interfaces.
l The switch can replicate traffic of up to 24 uplink interfaces to downlink interfaces, and traffic on
one uplink interface can be replicated to up to 5 downlink interfaces.
l This function is invalid for the Eth-Trunk that contains GE member interfaces.
Figure 22-111 Recommended deployment when multiple Eth-Trunk member interfaces are
deployed on different cards
0 2 4 ...
Slot 1 ...
1 3 5 ...
0 2 4 ...
Slot 2 ...
1 3 5 ...
Run the mode { lacp-static | lacp-dynamic | manual [ load-balance ] } command in the Eth-
Trunk interface view to configure the LACP mode of the Eth-Trunk on the CE switch.
When a CE switch connects to a third-party device, the LACP modes must be consistent.
Mainstream network devices that connect to CE switches include devices from Cisco, Juniper,
and H3C. The following table describes the mapping between aggregation modes.
Table 22-63 Mapping between link aggregation modes on a Cisco device and a Huawei
device
Table 22-64 Mapping between link aggregation modes on a Juniper device and a Huawei
device
LACP lacp-static
Table 22-65 Mapping between link aggregation modes on a H3C device and a Huawei device
LACP lacp-static
Table 22-66 Interworking mode of a Huawei CE series switch and a Linux server
Network Adapter Switch Description
Binding Mode Interworking Mode
Table 22-67 Interworking mode of a Huawei CE series switch and a Windows server
Network Adapter Switch Interworking Description
Binding Mode Mode
Switch fault Connect two switches. Network adapters of a server use SFT, so
tolerance (SFT) two switches are used.
A server generally has multiple network adapters. You can configure network adapter
aggregation to improve bandwidth and reliability. This document describes how to connect
CE series switches to servers using network adapter groups consisting of mainstream network
adapters, including Intel(R) Server Adapter I340, HP Ethernet 1Gb 4-port 331T, and
Mellanox ConnectX Ethernet Adapter.
l Intel(R) Server Adapter I340 Network Adapter Group
Table 22-68 describes the five modes supported by Intel(R) Server Adapter I340
network adapter group.
Table 22-68 Modes supported by Intel(R) Server Adapter I340 network adapter group
Server Mode Network Adapter Description
Group Mode
Load balancing IEEE 802.3ad dynamic Each group supports two to eight
mode link aggregation network adapters working at the same
rate and in full-duplex mode.
Table 22-69 Modes supported by HP Ethernet 1Gb 4-port 331T network adapter group
Server Network Adapter Group Description
Mode Mode
Table 22-70 Modes supported by Mellanox ConnectX Ethernet Adapter network adapter
group
l Trunk interface
<HUAWEI> system-view
[~HUAWEI] vlan batch 10 to 30
[*HUAWEI] interface 10ge 1/0/1
[*HUAWEI-10GE1/0/1] port link-type trunk //Set the link type to trunk.
[*HUAWEI-10GE1/0/1] port trunk allow-pass vlan 10 to 30 //Set the range of
VLANs allowed by the trunk interface to 10-30.
[*HUAWEI-10GE1/0/1] quit
[*HUAWEI] commit
l Hybrid interface
<HUAWEI> system-view
[~HUAWEI] vlan batch 3 to 9
[*HUAWEI] interface 10ge 1/0/1
[*HUAWEI-10GE1/0/1] port link-type hybrid //Set the link type to hybrid.
[*HUAWEI-10GE1/0/1] port hybrid untagged vlan 3 to 5 //Set the range of
VLANs allowed by the hybrid interface in untagged mode to 3-5.
[*HUAWEI-10GE1/0/1] port hybrid tagged vlan 6 to 9 //Set the range of VLANs
allowed by the hybrid interface in tagged mode to 6-9.
[*HUAWEI-10GE1/0/1] quit
[*HUAWEI] commit
CE switches support 4063 VLANs. The VLAN ID ranges from 1 to 4094. VLAN 1 exists by
default, and VLANs 4064 to 4094 are default reserved VLANs. You can run the vlan
reserved vlan-id command in the system view to adjust the range of reserved VLANs.
You can run the vlan vlan-id or vlan batch { vlan-id1 [ to vlan-id2 ] } &<1-10> command in
the system view to create other VLANs except reserved VLANs.
The VLANIF interface can go Up only when there is the physical interface in Up state in the
VLAN corresponding to the VLANIF interface.
Check the MIB object with the OID of 1.3.6.1.4.1.2011.5.25.42.3.1.3.4.1 through the SNMP
agent.
In V100R005C00 and earlier versions, only one VLANIF interface can be created on the
CE6810LI. The VLANIF interface is used for in-band management only, and is not used for
data forwarding.
NOTE
The CE6810LI does not support IPv4 or IPv6 Layer 3 forwarding. After the IPv4 or IPv6 function is
enabled on an interface of the CE6810LI, the configured IPv4 or IPv6 address can only be used to
manage the switch.
For the CE12800 and CE12800E series, CE5880EI, CE6870EI, CE6875EI, and CE6880EI, a
maximum of 256 sub-VLANs can be added to the same super-VLAN.
For CE8800, CE7800, CE6800, and CE5800 series switches excluding the CE5880EI,
CE6870EI, CE6875EI, and CE6880EI:
l In earlier versions of V200R002C50, a maximum of 32 sub-VLANs can be added to the
same super-VLAN.
l In V200R002C50 and later versions, a maximum of 256 sub-VLANs can be added to the
same super-VLAN.
Selective QinQ can be implemented based on the VLAN ID or MQC. The CE5880EI and
CE6880EI switches do not support VLAN ID-based selective QinQ. The CE6870EI and
CE6875EI switch do not support MQC-based selective QinQ. Other models support both.
CE12800E that has the ED-E/EG-E/EGA-E series cards installed, CE5880EI and CE6880EI
switches do not support GVRP. Other CE switches support GVRP starting from
V200R001C00.
If a specified port of a CE switch directly connects to a server but not another switch, you can
configure the port as an edge port. Then the port does not participate in spanning tree
calculation, which shortens the network topology convergence time and improves network
stability.
Normally, an edge port will not receive BPDUs. If a switch is attacked by forged BPDUs, the
edge port will receive the BPDUs. The switch then sets the edge port as a non-edge port and
recalculates the spanning tree, resulting in network flapping. You can enable BPDU protection
to prevent the attacks.
Configuration
When a CE switch connects to a server, configure the port directly connects to the server as an
edge port and configure BPDU protection simultaneously. The RSTP mode is used as the
example:
<HUAWEI> system-view
[~HUAWEI] stp mode rstp //Configure the device to work in RSTP mode.
[*HUAWEI] interface 10ge 1/0/1
[*HUAWEI-10GE1/0/1] stp edged-port enable //Configure the port as the edge port.
[*HUAWEI-10GE1/0/1] quit
[*HUAWEI] stp bpdu-protection //Enable BPDU protection.
[*HUAWEI] commit
Run the stp [ process process-id ] [ instance instance-id ] cost cost command in the Eth-
Trunk interface view to set the path cost of the Eth-Trunk to a fixed value. Then member
interface change does not affect the path cost of the Eth-Trunk, and the network topology is
not affected.
l When the peer-link fails but the heartbeat status is normal, only the interface where the
m-lag unpaired-port suspend command is configured enters the Error-Down state in
addition to the M-LAG interface in Error-Down state.
l The m-lag unpaired-port suspend command cannot be configured on the M-LAG
interface. It is recommended that this command be configured on both the M-LAG
master and backup devices.
In Figure 22-112, HostA connects to SwitchA through 10GE1/0/1, and the server directly
connects to 10GE1/0/2 on SwitchA. The server needs to monitor packets that carry 802.1p
priority 6 and are sent from HostA.
<HUAWEI> system-view
[~HUAWEI] sysname SwitchA
[*HUAWEI] commit
[~SwitchA] observe-port 1 interface 10ge 1/0/2 //Configure an observing port.
[*SwitchA] commit
[~SwitchA] traffic classifier c1 //Configure a traffic classifier.
[*SwitchA-classifier-c1] if-match 8021p 6 //Define a matching rule in the
traffic classifier.
[*SwitchA-classifier-c1] commit
[~SwitchA-classifier-c1] quit
[~SwitchA] traffic behavior b1 //Configure a traffic behavior.
[*SwitchA-behavior-b1] mirroring observe-port 1 //Mirror traffic that
matches the rule to the observing port.
[*SwitchA-behavior-b1] commit
[~SwitchA-behavior-b1] quit
[~SwitchA] traffic policy p1 //Configure a traffic policy.
[*SwitchA-trafficpolicy-p1] classifier c1 behavior b1 //Bind the traffic
classifier and traffic behavior to the traffic policy.
[*SwitchA-trafficpolicy-p1] commit
[~SwitchA-trafficpolicy-p1] quit
[~SwitchA] interface 10ge 1/0/1
[~SwitchA-10GE1/0/1] traffic-policy p1 inbound //Apply the traffic policy to
the inbound direction of 10GE1/0/1.
[*SwitchA-10GE1/0/1] commit
most eight observing ports at a time, while other switch models can use at most four
observing ports at a time. On a CE6880EI or CE5880EI switch, each observing port can
be used to copy both incoming and outgoing traffic. On a switch except CE6880EI and
CE5880EI, the maximum numbers of available observing ports for incoming and
outgoing traffic are calculated separated. When the same observing port is specified for
both incoming and outgoing traffic, the remaining number of observing ports for
incoming and outgoing traffic is 2 not 3.
l A CE6870EI or CE6875EI switch also supports observing port groups and can have at
most 128 observing port groups configured, but can use at most 8 observing port groups
at a time. Each observing port group contains at most 64 member ports. An observing
port group is a group of observing ports. Packets copied to an observing port group will
be copied to all its member ports.
Figure 22-113 Mirrored packet forwarding in a Layer 2 remote port mirroring scenario
Internet
VLAN 20
SwitchB
SwitchA
VLAN 10
VLAN 20
Server
HostA HostB
Mirrored port
Remote observing port
Original packets
Mirrored packets
In most Layer 2 remote mirroring scenarios, intermediate devices will not learn destination
MAC addresses of mirrored packets in the corresponding VLANs. This means that mirrored
packets can be forwarded to the monitoring devices in broadcast mode. In the following
scenarios, however, Layer 2 remote port mirroring will fail because destination MAC
addresses of mirrored packets can be learned in the corresponding VLANs.
l The original service traffic and mirrored traffic are forwarded in the same VLAN.
Figure 22-114 Layer 2 remote port mirroring fails when the original service traffic and
mirrored traffic are forwarded in the same VLAN
MacS
DNS Server
VLAN 10
SwitchB’s MAC address table MacS MacS
MacA MacA
VLAN 10 VLAN 10
Outbound 10GE1/0/3 VLAN 10
MAC VLAN
interface 10GE1/0/2
SwitchB VLAN 10
MacA VLAN 10 10GE1/0/1
MacS VLAN 10 10GE1/0/3 10GE1/0/1 Server
VLAN 10
MacS MacS
MacA MacA
VLAN 10 VLAN 10
VLAN 10
SwitchA
MacS VLAN 10
MacA
VLAN 10
R&D
HostA HostB
MacA MacB
Mirrored port
Remote observing port
Original packets
Mirrored packets
To resolve this problem, use different VLANs to forward the original service traffic and
mirrored traffic, as shown in Figure 22-115. If the original service traffic and mirrored
traffic need to be forwarded in the same VLAN, run the mac-address learning disable
command on the intermediate devices to disable MAC address learning in this VLAN.
However, disabling MAC address learning will cause a waste of link bandwidth.
Figure 22-115 Original service traffic and mirrored traffic are forwarded in different
VLANs
MacS
DNS Server
VLAN 20
MacS MacS
SwitchB’s MAC address table MacA MacA
VLAN 20 VLAN 10
10GE1/0/3 VLAN 20
Outbound
MAC VLAN 10GE1/0/2
interface
SwitchB VLAN 10
MacA VLAN 20 10GE1/0/1
MacS VLAN 20 10GE1/0/3 10GE1/0/1 Server
VLAN 20
VLAN 10
MacS MacS
MacA MacA
VLAN 20 VLAN 10
VLAN 20
SwitchA
VLAN 20
MacS
MacA
VLAN 20
R&D
HostA HostB
MacA MacB
Mirrored port
Remote observing port
Original packets
Mirrored packets
l Mirrored packets from different source MAC addresses are forwarded in the same
VLAN.
Figure 22-116 Layer 2 remote port mirroring fails when mirrored packets from different
source MAC addresses are forwarded in the same VLAN
Server
VLAN 10
MacD MacA
SwitchA MacA MacD SwitchC
VLAN 10 VLAN 10
VLAN 10 VLAN 10
10GE1/0/2 10GE1/0/1
SwitchB
Outbound
MAC VLAN
interface Marketing
R&D
MacA VLAN 10 10GE1/0/2
MacD VLAN 10 10GE1/0/1
SwitchB’s MAC address table
Mirrored port
Remote observing port
Mirrored packets from SwitchA
Mirrored packets from SwitchC
VLAN 10
SwitchA SwitchC
VLAN 10 VLAN 10
SwitchB
MAC address learning is disabled in VLAN 10
R&D Marketing
Mirrored port
Remote observing port
Mirrored packets from SwitchA
Mirrored packets from SwitchC
VLAN 10
SwitchA SwitchC
VLAN 10 VLAN 10
SwitchB
R&D Marketing
Mirrored port
Remote observing port
Mirrored packets from SwitchA
Mirrored packets from SwitchC
VLAN 20
VLAN 10
MacD MacA
SwitchA MacA MacD SwitchC
VLAN 10 VLAN 20
VLAN 10 VLAN 20
10GE1/0/2 10GE1/0/1
SwitchB
Outbound
MAC VLAN
interface Marketing
R&D
MacA VLAN 10 10GE1/0/2
MacD VLAN 20 10GE1/0/1
Mirrored port
Remote observing port
Mirrored packets from SwitchA
Mirrored packets from SwitchC
NOTE
The alias alias-name parameter is only supported in V100R003C00 and later versions. To facilitate
identification of community names, set alias for community names. The alias are stored in the configuration
file as plain text.
Configuring SNMPv2c
Set the SNMP version to v2c and read/write community name to Private123, and configure
access control.
<HUAWEI> system-view
[~HUAWEI] acl 2001
[*HUAWEI-acl4-basic-2001] rule permit source 192.168.1.0 0.0.0.255 //Allow only
the NMS on network segment 192.168.1.0 to access the switch.
[*HUAWEI-acl4-basic-2001] commit
[~HUAWEI-acl4-basic-2001] quit
[~HUAWEI] snmp-agent mib-view included alliso iso //Set the MIB view name to
alliso and accessed view includes iso.
[*HUAWEI] snmp-agent sys-info version v2c //The SNMP version used by the switch
must be the same as the SNMP version used by the NMS; otherwise, the switch
cannot communicate with the NMS.
[*HUAWEI] snmp-agent community write Private123 mib-view alliso acl 2001
[*HUAWEI] snmp-agent target-host trap address udp-domain 10.1.1.2 params
securityname adminNMS1234 v2c
//Set the IP address of trap host to 10.1.1.2, security name to
adminNMS1234, and trap version to v2c.
[*HUAWEI] commit
After the preceding configuration is complete, the NMS can connect to the switch using the
configured read/write community name.
Configuring SNMPv3
The security levels of the trap host, user, and user group are in descending order.
The security levels include:
l privacy: authentication and encryption
l authentication: authentication and no encryption
l none: no authentication and no encryption
If the user group is at the privacy level, the user and trap host must be at the privacy level. If
the user group is at the authentication level, the user and trap host must be at the privacy or
authentication level.
Set the user group name to huawei_group and security level to privacy, and configure access
control.
<HUAWEI> system-view
[~HUAWEI] acl 2001
[*HUAWEI-acl4-basic-2001] rule permit source 192.168.1.0 0.0.0.255 //Allow only
the NMS on network segment 192.168.1.0 to access the switch.
[*HUAWEI-acl4-basic-2001] commit
[~HUAWEI-acl4-basic-2001] quit
[~HUAWEI] snmp-agent mib-view included alliso iso //Set the MIB view name to
alliso and accessed view includes iso.
[*HUAWEI] snmp-agent sys-info version v3 //The SNMP version used by the switch
must be the same as the SNMP version used by the NMS; otherwise, the switch
cannot communicate with the NMS.
[*HUAWEI] snmp-agent group v3 huawei_group privacy write-view alliso acl 2001
[*HUAWEI] snmp-agent usm-user v3 huawei_user group huawei_group //Set the SNMPv3
user name to huawei_user and add the user to user group huawei_group.
[*HUAWEI] snmp-agent usm-user v3 huawei_user authentication-mode sha
Please configure the authentication password (8-255)
Enter Password: //Enter the authentication password.
Confirm Password: //Confirm the authentication password.
[*HUAWEI] snmp-agent usm-user v3 huawei_user privacy-mode aes256
Please configure the privacy password (8-255)
Enter Password: //Enter the encryption password.
Confirm Password: //Confirm the encryption password.
[*HUAWEI] snmp-agent target-host trap address udp-domain 10.1.1.2 params
securityname huawei_user v3 privacy
//Set the IP address of trap host to 10.1.1.2, security name to
huawei_user, and trap version to v3.
[*HUAWEI] commit
After the preceding configuration is complete, the NMS can connect to the switch using the
configured user name, authentication password, and encryption password.
Packet is too large The size of an SNMP Run the snmp-agent packet
packet sent by the NMS max-size command in the
exceeds the threshold set system view to increase the
on the device. packet size threshold.
ACL denied The IP address used by Run the display acl {acl-
the NMS to send SNMP number | name acl-name | all }
request packets is denied command in any view to check
by an ACL. ACL configuration.
If the IP address used by the
NMS to send requests is denied
by the ACL, run the rule
command in the corresponding
ACL view to allow the NMS to
access the device.
If the NMS is on a VPN but no
VPN instance name is bound to
the ACL rule, run the rule
command in the corresponding
ACL view to bind the ACL rule
to the VPN instance.
UsmUser Name is incorrect The SNMPv3 USM user l If the device uses an
names or AAA local user SNMPv3 USM user name
names configured on the to communicate with the
NMS and device are NMS, run the display
different. snmp-agent usm-user
command in any view to
check whether the SNMPv3
USM user name is the same
as that configured on the
NMS. If not, modify the
SNMPv3 USM user name
on the NMS or device to
make them consistent.
l If the device uses an
SNMPv3 AAA local user
name to communicate with
the NMS, run the display
snmp-agent local-user
command in any view to
check whether the SNMPv3
AAA local user name is the
same as that configured on
the NMS. If not, modify the
SNMPv3 AAA user name
on the NMS or device to
make them consistent.
Wrong Protocol Parameter The security level of the Run the display snmp-agent
SNMPv3 user is lower group and display snmp-agent
than that of the SNMPv3 usm-user commands in any
user group. view to check the security
levels of the SNMPv3 user
group and user. If the security
level of the SNMPv3 user is
lower than that of the user
group, run the snmp-agent
usm-user v3 user-name
authentication-mode { md5 |
sha } command in the system
view to set an authentication
password for the SNMPv3 user
and the snmp-agent usm-user
v3 user-name privacy-mode
{ 3des168 | aes128 | aes192 |
aes256 | des56 } command in
the system view to set an
encryption password for the
SNMPv3 user.
The SNMPv3 security levels in
descending order are as
follows:
l Level 1: privacy
(authentication and
encryption)
l Level 2: authentication
(only authentication)
l Level 3: noauthentication
(no authentication, no
encryption)
The security level of the
SNMPv3 user cannot be lower
than the security level of the
user group; otherwise, the
device cannot communicate
with the NMS.
Context name is incorrect The contextname on the Set the contextname on the
NMS is incorrect. NMS to be empty.
Wrong Security level for The SNMPv3 USM user Run the display snmp-agent
USM User security level on the usm-user command in any
NMS is higher than that view to check whether an
on the device. authentication or encryption
method is configured for the
SNMPv3 USM user.
l If no authentication or
encryption method is
configured for SNMPv3
USM user on the device,
this indicates that the
security level of the
SNMPv3 USM user on the
device is noauthentication.
However, the security level
of the SNMPv3 USM user
on the NMS is
authentication. Perform the
following operations:
– Run the snmp-agent
usm-user v3 user-name
authentication-mode
{ md5 | sha } command
in the system view to set
the SNMPv3 USM
authentication password
to be the same as that on
the NMS.
– Change the security
level of SNMPv3 USM
user on the NMS to be
the same as that on the
device.
l If an authentication method
is configured for the
SNMPv3 USM user on the
device, this indicates that
the security level of
SNMPv3 USM user on the
device is authentication.
However, the security level
of the SNMPv3 USM user
on the NMS is privacy.
Perform the following
operations:
– Run the snmp-agent
usm-user v3 user-name
privacy-mode
{ 3des168 | aes128 |
aes192 | aes256 |
des56 } command in the
system view to set the
SNMPv3 USM
encryption password to
be the same as that on
the NMS.
– Change the security
level of SNMPv3 USM
user on the NMS to be
the same as that on the
device.
Pipeline is full The NMS has sent too Reduce the number of SNMP
many SNMP request request packets sent by the
packets to congest the NMS.
channel.
# Configure the MIB view named alliso and the accessed view includes iso.
[~HUAWEI] snmp-agent mib-view included alliso iso
[*HUAWEI] commit
# Allow some NMSs using the specified SNMP community name to access only the objects
in the MIB view alliso on the switch.
[~HUAWEI] snmp-agent community write private_user mib-view alliso acl 2001
[*HUAWEI] commit
# Allow some NMSs using the user names in the specified SNMPv3 user group to access only
the objects in the MIB view alliso on the switch.
[~HUAWEI] snmp-agent group v3 huawei_group privacy write-view alliso acl 2001
[*HUAWEI] commit
# Allow some NMSs using the user names in an SNMPv3 user group to manage the switch.
[~HUAWEI] snmp-agent usm-user v3 huawei_user acl 2001
[*HUAWEI] commit
notation. The first binary digit has a fixed value 1. Therefore, the engine ID in
hexadecimal format is 800007DB.
l The device information is configured manually. You are advised to use the IP address or
MAC address of the switch as device information to uniquely identify the switch.
You can query basic information about each member device in the stack to learn which device
is the standby device. hwStackMemberInformationTable provides information about basic
attributes of a stack member device, including the stack priority, role, and device type. In this
table, hwStackRole indicates the stack role. As shown in the following figure, hwStackRole.
2[integer]standby[2] indicates that the ID of the standby device in the stack is 2. In
hwStackRole.2, 2 indicates the stack ID.
To query information, for example, the fan status, about the standby device in a stack, you can
query the hwEntityFanState object. As shown in the following figure, hwEntityFanState.
2.2[integer]normal[1] indicates the fan status of the standby device. In hwEntityFanState.2.2,
the first 2 indicates the stack ID, and the second 2 indicates the fan ID.
SNMPv3, but the notify view name is not configured on the device. To resolve the problem,
perform the following steps:
1. Run the display snmp-agent statistics command to check SNMP packet statistics. If 0
Trap PDU sent is displayed, the device does not send any traps. In this case, go to step
2. If the value of Trap-PDU sent is not 0, the device has sent traps. In this case, contact
technical support personnel.
<HUAWEI> display snmp-agent statistics
3158 Messages delivered to the SNMP entity
0 Messages which were for an unsupported version
0 Messages which used an SNMP community name not known
0 Messages which represented an illegal operation for the community supplied
0 ASN.1 or BER errors in the process of decoding
3152 Messages passed from the SNMP entity
0 SNMP PDUs which had badValue error status
0 SNMP PDUs which had genErr error status
0 SNMP PDUs which had noSuchName error status
0 SNMP PDUs which had tooBig error status
3135 MIB objects retrieved successfully
0 MIB objects altered successfully
0 GetRequest PDU accepted and processed
3158 GetNextRequest PDU accepted and processed
0 GetBulkRequest PDU accepted and processed
3152 GetResponse PDU sent
0 SetRequest PDU accepted and processed
0 Trap PDU sent
0 Inform PDU sent
0 Inform PDU received with no acknowledgement
0 Inform PDU received with acknowledgement
CE8868EI/CE8861EI V200R005C10
CE8860EI V100R006C00
CE8850-32CQ-EI V200R002C50
CE8850-64CQ-EI V200R005C00
CE7850EI V100R003C00
CE7855EI V200R001C00
CE6810EI V100R003C00
CE6810-48S4Q-LI/CE6810-48S-LI V100R003C10
CE6810-32T16S4Q-LI/CE6810-24S2Q-LI V100R005C10
CE6850EI V100R002C00
CE6850-48S6Q-HI V100R005C00
CE6850-48T6Q-HI/CE6850U-HI/ V100R005C10
CE6851HI
CE6855HI V200R001C00
CE6870-24S6CQ-EI/CE6870-48S6CQ-EI V200R001C00
CE6860EI/CE6856HI/CE6870-48T6CQ-EI/ V200R002C50
CE6880-24S4Q2CQ-EI/
CE6880-48S4Q2CQ-EI/
CE6880-48T4Q2CQ-EI
CE6857EI V200R005C10
CE6875EI V200R003C00
CE6865EI V200R005C00
CE5880EI V200R005C10
CE5810EI V100R002C00
CE5850EI V100R002C00
CE5850HI V100R003C00
CE5855EI V100R006C00
CE12804/CE12808/CE12812 V100R002C00
CE12816 V100R003C00
CE12804S/CE12808S V100R005C00
CE12800E V200R002C50
obtained through SNMP is the same as the length of the interface index carried in a
NetStream packet.
To view the mapping between 32-bit and 16-bit interface indexes, run the display netstream
map command in the diagnostic view. InterfaceName indicates the interface name, IfnetIndex
indicates the 32-bit interface index, and NSIndex indicates the 16-bit interface index.
<HUAWEI> system-view
[~HUAWEI] diagnose
[~HUAWEI-diagnose] display netstream map
Netstream index information in map:
----------------------------------------------------------------------
InterfaceName IfnetIndex NSIndex
----------------------------------------------------------------------
MEth6/0/0 1494 1
Sip6/0/0 1331 2
Sip6/0/1 1332 3
NULL0 4 4
InLoopBack0 5 5
MEth0/0/0 1333 6
40GE2/0/0 1334 7
40GE2/0/1 1335 8
40GE2/0/2 1336 9
40GE2/0/3 1337 10
40GE2/0/4 1338 11
40GE2/0/5 1339 12
40GE2/0/6 1340 13
40GE2/0/7 1341 14
40GE2/0/8 1342 15
40GE2/0/9 1343 16
40GE2/0/10 1344 17
40GE2/0/11 1345 18
# Set the NetStream sampling rate. Set the sampling rate in both the inbound and outbound
directions of 10GE1/0/1 to 8192.
<HUAWEI> system-view
[~HUAWEI] interface 10ge 1/0/1
[~HUAWEI-10GE1/0/1] netstream sampler random-packets 8192 inbound
[*HUAWEI-10GE1/0/1] netstream sampler random-packets 8192 outbound
[*HUAWEI-10GE1/0/1] quit
# Set the NetStream flow aging mode and aging time. Set the inactive aging time to 100
seconds and enable FIN- and RST-based aging.
[*HUAWEI] netstream timeout ip inactive 100
[*HUAWEI] netstream timeout ip tcp-session
# Set the source IP address, destination IP address, and destination port number of the
exported packets carrying original flow statistics.
[*HUAWEI] netstream export ip source 10.1.2.1
[*HUAWEI] netstream export ip host 10.1.2.2 6000
# Enable NetStream statistics collection on 10GE1/0/1 for incoming and outgoing packets.
[*HUAWEI] interface 10ge 1/0/1
[*HUAWEI-10GE1/0/1] netstream outbound ip
[*HUAWEI-10GE1/0/1] netstream inbound ip
[*HUAWEI-10GE1/0/1] quit
[*HUAWEI] commit
NetStream can be deployed at the access, sFlow can be deployed at the access,
aggregation, or core layer based on service aggregation, or core layer based on service
requirements to learn about the service requirements to analyze traffic transmission
types on the network, network traffic on the data center network in real time. With
characteristics, and network running status. sFlow, users can quickly detect abnormal
With NetStream, users can detect network flows and locate the source of attack traffic.
structure defects or performance bottlenecks
in a timely manner.
The cache is required for NetStream and its The cache is unnecessary. The ASIC chip
capacity affects NetStream performance. directly delivers received packets to the
The cache increases costs. CPU, which then exports the data to the
sFlow collector.
This method is flexible. Users can This method is inflexible, and no template
customize a template to obtain key can be customized.
information in network traffic.
Table 22-75 Description of the display lldp neighbor brief command output
Item Description
LLDP-MIB: configures LLDP, queries statistics on sent and received LLDPDUs, and queries
local and remote device information. In addition, this object can configure the traps sent to the
NMS when the specific events happen. The root object is 1.0.8802.1.1.2.
For details about LLDP MIB objects, see the MIB reference.
All IPv6 functions on CE12800 series switches are controlled by license, so sFlow (IPv6)
requires license.
All IPv6 functions on CE12800E series switches are controlled by license, so sFlow (IPv6)
requires license.
CE12800E does not support sFlow when configured with ED-E/EG-E/EGA-E series cards.
CE8868EI/CE8861EI V200R005C10
CE8860EI V100R006C00
CE8850-32CQ-EI V200R002C50
CE8850-64CQ-EI V200R005C00
CE7850EI V100R003C00
CE7855EI V200R001C00
CE6810EI V100R003C00
CE6810-48S4Q-LI/CE6810-48S-LI V100R003C10
CE6810-32T16S4Q-LI/CE6810-24S2Q-LI V100R005C10
CE6850EI V100R002C00
CE6850-48S6Q-HI V100R005C00
CE6850-48T6Q-HI/CE6850U-HI/ V100R005C10
CE6851HI
CE6855HI V200R001C00
CE6870-24S6CQ-EI/CE6870-48S6CQ-EI V200R001C00
CE6860EI/CE6856HI/CE6870-48T6CQ-EI V200R002C50
CE6857EI V200R005C10
CE6875EI V200R003C00
CE6865EI V200R005C00
CE5810EI V100R002C00
CE5850EI V100R002C00
CE5850HI V100R003C00
CE5855EI V100R006C00
CE12804/CE12808/CE12812 V100R002C00
CE12816 V100R003C00
CE12804S/CE12808S V100R005C00
In the command, rate rate indicates the denominator of the flow sampling rate. The actual
sampling rate is 1/rate. For example, the default rate is 8192, indicating that one packet is
sampled every 8192 packets.
The LSP ping test and LSP trace test are supported only on the CE12800, CE8860EI,
CE8850EI, CE7855EI, CE7850EI, CE6875EI, CE6870EI, CE6865EI, CE6860EI, CE6857EI,
CE6856HI, CE6855HI, CE6851HI, CE6850HI, and CE6850U-HI.
l The frequency parameter indicates the interval at which an NQA test instance is
automatically performed. It is configured using the frequency interval command.
l The interval parameter indicates the interval at which probe packets are sent in an NQA
test instance. It is configured using the interval { milliseconds interval | seconds
interval } command.
l The probeCount parameter indicates the number of probes to be sent each time for an
NQA test instance. It is configured using the probe-count number command.
Except for the ICMP jitter and UDP jitter tests, the number of probes is equal to the
number of probe packets. For the ICMP jitter and UDP jitter tests, the number of probe
packets is calculated as follows: probeCount x jitter-packetnum.
l The timeout parameter indicates the timeout period of a probe for an NQA test instance.
It is configured using the timeout time command.
l The jitter-packetnum parameter indicates the number of packets sent in each probe for
an NQA test instance. It is configured using the jitter-packetnum number command.
The frequency value varies according to the test instance type. If the frequency value does not
meet the frequency verification formula, the test result may be "no result".
Table 22-77 Description of the display openflow flows standard command output
Item Description
Item Description
Table 22-78 Products and minimum version supporting the OpenFlow Agent
Product Minimum Version Required
CE8861EI/CE8868EI V200R005C10
CE8860EI V100R006C00
CE8850-32CQ-EI V200R002C50
CE8850-64CQ-EI V200R005C00
CE7850EI V100R003C10
CE7855EI V200R001C00
CE6850-48S6Q-HI V100R005C00
CE6850-48T6Q-HI/CE6850U-HI/ V100R005C10
CE6851HI
CE6855HI V200R001C00
CE6870-24S6CQ-EI/CE6870-48S6CQ-EI V200R001C00
CE6870-48T6CQ-EI/CE6860EI/CE6856HI/ V200R002C50
CE6880EI
CE6857EI V200R005C10
CE6875EI V200R003C00
CE6865EI V200R005C00
CE5880EI V200R005C10
CE12804/CE12808/CE12812/CE12816 V100R003C10
CE12804S/CE12808S V100R005C00
CE12800E V200R002C50
After the tracert 10.1.2.2 command is run on switch A, by default, switch A displays the
following information:
<SwitchA> tracert 10.1.2.2
traceroute to 10.1.2.2(10.1.2.2), max hops: 30 ,packet length: 40,press CTRL_C to
break
1 10.1.1.2 10 ms 4 ms 4 ms
2 10.1.2.2 6 ms 4 ms 3 ms
If you do not want to expose switch B's IP address on other devices (displayed as * on switch
A), run the icmp name ttl-exceeded send disable command in the system view of switch B
to disable switch B from sending ICMP TTL timeout packets. Then, when the tracert
10.1.2.2 command is run on switch A, the following information is displayed:
<SwitchA> tracert 10.1.2.2
traceroute to 10.1.2.2(10.1.2.2), max hops: 30 ,packet length: 40,press CTRL_C to
break
1 * * *
2 10.1.2.2 6 ms 4 ms 3 ms
10GE1/0/1 10GE1/0/2
10.1.1.1/24 10.1.2.1/24
10GE1/0/1 10GE1/0/1
SwitchA 10.1.1.2/24 SwitchB 10.1.2.2/24 SwitchC
3. Insufficient memory space. When a switch receives a large number of packets, the switch
does not have sufficient memory space to process these packets, resulting in slow packet
processing speed and long network latency.
You can run the ping command to test network latency. The test results are only for reference
and cannot be used as an absolute value of network latency measurement. No reference value
is available for determining whether the ping latency is normal because requirement for
network latency varies depending on network status. Other measurement methods such as
network quality analysis (NQA) are also required to accurately measure network latency. Pay
attention to the following points when analyzing a ping latency:
1. When a switch forwards packets through the hardware at a high speed, network latency
is short. For example, ping a PC connected to the switch. When packets need to be
processed by the CPU, network latency is long. For example, ping a gateway.
Though network latency is long when the switch pings the gateway, packets are normally
forwarded because the packets are processed by the underlying chip rather than the CPU.
You can run undo icmp echo-reply fast disable command to enable the fast ICMP reply
function on the switch to shorten network latency when the switch pings the gateway.
After the function is enabled, the switch quickly responds to received Echo Request
packets destined for its own IP address. The CPU of the LPU directly responds to the
received ICMP packets, improving the processing speed of ICMP packets and shortening
network latency.
2. The processing priority of ICMP packets has been minimized to prevent impacts of
common ping attacks on the switch, so that ICMP packets are the last to be transmitted
and processed. Therefore, the network latency is long.
Ping Implementation
The following is a sample of ping operation procedure.
In Figure 22-120, PC1 attempts to ping PC2 (11.1.1.2).
3. When receiving the ARP reply packet, PC1 obtains the MAC address of the next hop.
Then PC1 encapsulates the ICMP Echo Request packet into an Ethernet frame and sends
the Ethernet frame to the switch.
Before sending the ARP request packet to the switch, PC1 fills the mapping between its
own IP address and MAC address into the packet. When receiving the ARP request
packet from PC1, the switch fills the address mapping of PC1 into the local ARP cache.
This improves efficiency of subsequent communication between the switch and PC1 and
reduces communication data.
4. After receiving the ICMP Echo Request packet, the switch removes the Ethernet frame
header, and sends the packet to the IP layer. The IP layer finds that the destination
address 11.1.1.2/24 is not a local address, so it searches the routing table and re-
encapsulates the packet.
The switch does not know the destination MAC address (MAC address matching
11.1.1.2/24), so the switch sends a broadcast ARP request packet.
5. PC2 receives the ARP request packet and finds that the destination address of the packet
is a local address, so PC2 returns a unicast ARP reply packet that contains the MAC
address matching 11.1.1.2/24.
In addition, PC2 records the mapping between the IP address and MAC address of the
switch's port b into the local ARP cache.
6. The switch obtains the MAC address of PC2 from the ARP reply packet, encapsulates an
Ethernet frame header into the packet, and sends the packet to PC2.
7. After receiving the packet, PC2 removes the Ethernet frame header. When finding that
the packet is the ICMP request packet, PC2 sends an ICMP reply packet to PC1. In this
case, the source IP address becomes IP address 11.1.1.2/24 of PC2 and the destination IP
address becomes 10.1.1.1/24 of PC1.
Since the source and destination IP addresses are located on different network segments,
PC2 searches the FIB table for the next hop. The next hop is the switch's port b
(11.1.1.1/24). As mentioned in preceding steps, PC2 has recorded the address mapping
of the switch's port b in the ARP cache, so PC2 does not need to send an ARP request
packet to the switch. Instead, PC2 obtains the MAC address matching 11.1.1.1/24 from
its local ARP cache, encapsulates the MAC address into the ICMP reply packet, and
sends the packet to the switch.
Similarly, the switch does not need to send an ARP request packet to PC1. The switch
directly encapsulates PC1's MAC address found in the ARP cache into the ICMP reply
packet. Then the switch sends the packet to PC1.
8. After receiving the packet, PC1 decapsulates the Ethernet frame header and IP header.
Finally, it learns that the packet is an ICMP reply packet and the ping operation
succeeds.
The NMS or network performance monitoring software calculates the real-time network
bandwidth at an interval of seconds to minutes. In such an interval, the network traffic seems
to be stable, as shown in Figure 22-121. However, at lower levels of granularity, for example,
milliseconds, the traffic rate exhibits a sawtooth pattern, as shown in Figure 22-122. Many
microbursts occur in the actual traffic.
Common Misunderstandings
Common misunderstandings on microbursts include:
1. Why cannot microbursts be presented in the value of Input peak rate on port statistics
of a switch?
On a switch, the packet rate is the quotient of the total number of packets in a specific
period of time divided by the period. For CE series switches, the values of Input peak
rate and Last 300 seconds input rate are calculated based on a period of 300 seconds
by default. The value of Input peak rate is the highest average rate in all historical
statistics periods. The interval for collecting statistics on a port is configurable, but the
minimum interval is 10 seconds.
Impact
When a microburst exceeds the forwarding capability of the switch, the switch buffers the
burst data for later transmission. If the switch does not have sufficient buffer space, the excess
data can only be discarded, causing congestion and packet loss.
1MB Buffer
Figure 22-123 shows a typical millisecond-level microburst scenario. Assume that Port1 and
Port2 respectively send 5 MB data to Port3 at a line rate of 10 Gbit/s. The total transmission
rate is 20 Gbit/s. Port3 supports only a rate of 10 Gbit/s, which is a half of the total
transmission rate. It sends only 5 MB data out and buffers the other 5 MB data for
transmission later. However, the switch has only 1 MB buffer space. Therefore, 4 MB data is
discarded due to insufficient buffer space. Without considering overhead data such as the
inter-frame gap, preamble, frame checksum, and packet header, the microburst duration is 4
ms (5 MB/10 Gbit/s).
Typically, a switch is equipped with a buffer space of 1 to 20 MB. In the preceding scenario
where two ports transmit data at a rate of 10 Gbit/s, the maximum microburst duration is
shorter than 16 ms (20 MB/10 Gbit/s). On an actual network, traffic may be originated from
multiple ports to one port. In this scenario, more buffer space is consumed, and severer
congestion and packet loss occur upon microbursts.
Detection Method
You can use the packet capture tools and Wireshark to detect microbursts.
On the Wireshark, open a file of captured packets and choose Statistics > I/O Graph to view
the traffic graph, as shown in Figure 22-124.
To view the millisecond-level burst traffic in the IO graph, change the unit for the y-axis to
Bits and the interval to 1 ms, as shown in Figure 22-125.
Countermeasures
You can take the following countermeasures to mitigate microbursts:
1. For problems such as severe bursts, intensive buffer utilization, poor performance on
lossy lines, and large delay and jitter in traditional TCP congestion control mechanisms,
use common improvement technologies in the industry to minimize the possibility of
microbursts.
2. During network service traffic planning, avoid scenarios with an excessively high
convergence ratio (traffic is transmitted from multiple ports to one port), and expand the
capacity in a timely manner for ports with severe bursts to eliminate burst bottlenecks.
NOTE
Each Layer 3 interface can have a maximum of 255 secondary IP addresses.
Switch
10GE1/0/1
VLANIF100
172.16.1.1/24
172.16.2.1/24 sub
LSW
On an Ethernet interface, run the undo portswitch command to switch the interface mode to
Layer 3 first, and then you can configure IP addresses for it.
NOTE
The CE6810LI does not support IPv4 or IPv6 Layer 3 forwarding. After the IPv4 or IPv6 function is
enabled on an interface of the CE6810LI, the configured IPv4 or IPv6 address can only be used to
manage the switch.
EXP: Expire-
time
10.1.1.1 306b-2079-2205 I
Vlanif101
– Before the switch receives ARP reply packets, IP packets matching the temporary
ARP entries are discarded and do not trigger ARP Miss messages.
– After the switch receives ARP reply packets, it generates correct ARP entries to
replace the temporary ARP entries.
l If the aging time of the temporary ARP entries expires, the switch clears the temporary
ARP entries.
------------------------------------------------------------------------------
Total:6 Dynamic:2 Static:0 Interface:2 OpenFlow:2
......
......
......
......
l local
In Figure 22-128, the local mode indicates that ARP information is identified based on
the key composed of the IP address, logical interface index, and physical interface index.
ARP information is stored only on the chip where the outbound interface is located, that
is, ARP information stored on each chip is different.
In local mode, the maximum number of ARP resources of the device is relevant to the
chip where the outbound interface is located.
– If all outbound interfaces corresponding to ARP information are located on the
same chip, the maximum number of ARP resources of the device is that of the chip.
– If all outbound interfaces corresponding to ARP information are located on
different chips, the maximum number of ARP resources of the device is the sum of
ARP resources of all chips.
......
......
......
......
......
ARP N ARP N ARP N ARP N ARP N
l extend
In Figure 22-129, the extend mode indicates that ARP information is identified based on
the key composed of the highest 44 bits of a MAC address. ARP information is stored on
all chips based on the same resource index, that is, ARP information stored on each chip
is the same. The difference between extend and global modes is that ARP information
corresponding to contiguous MAC addresses is aggregated in extend mode. That is,
ARP information with the same highest 44 bits of a MAC address corresponds to the
same ARP resource.
In local mode, the maximum number of ARP resources of the device depends on
whether MAC addresses are contiguous.
– If MAC addresses corresponding to ARP information is noncontiguous, the
maximum number of ARP resources of the device refers to that of the chip with
minimum specifications among all chips.
– If MAC addresses corresponding to ARP information is contiguous, the maximum
number of ARP resources of the device refers to that of the chip with minimum
specifications among all chips multiplied by 16.
......
......
......
......
ARP ARP ARP ARP ARP
N-16~N N-16~N N-16~N N-16~N N-16~N
NOTE
l The CE5880EI and CE6880EI use the extend mode, which cannot be changed.
l Only the CE12804, CE12808, CE12812, CE12816, CE12804S, and CE12808S support the local
mode for ARP resource allocation. This mode has high requirements for network deployment.
Exercise caution when you use this mode. To use this mode, contact technical support personnel.
Depending on the number of remaining ARP forwarding resources on the device, this alarm
can be generated in either of the following situations, as indicated by different message
digests:
l If the message digest is hwBoardResWarningThresholdExceed_active, the number of
remaining ARP forwarding resources on the device is insufficient. The number of used
forwarding resources has exceeded the alarm threshold 75%.
Dec 6 2017 04:30:43-08:00 DC08-CR03 %%01FEI/4/
hwBoardResWarningThresholdExceed_active(l):CID=0x807f042c-
alarmID=0x095e2029;The number of forwarding resources reaches the alarm
threshold. (Slot = 1, Threshold = 75, Reason = 174, Description : The number
of ARP exceeded the warning threshold.)
If the alarm The number of ARP exceeded the warning threshold is generated on the
device, use either of the following methods to clear the alarm:
l Increase the number of ARP forwarding resources supported by the device.
– For the CE12804, CE12808, CE12812, CE12816, CE12804S, CE12808S,
CE6870EI, and CE6875EI, if MAC addresses corresponding to ARP entries are
contiguous, run the arp resource-mode extend command in the system view to set
the ARP resource allocation mode to extend.
– For the CE12800E equipped with CEL24XSED-E, CEL48XSED-E, CEL24LQED-
E, CEL36LQED-E, or CEL36CQFD-X cards, CE5855EI, CE6850HI, CE6850U-
HI, CE6851HI, CE6855HI, CE6856HI, CE6857EI, CE6860EI, CE6865EI,
CE7850EI, CE7855EI, CE8850EI, CE8861EI, CE8868EI, and CE8860EI, run the
system resource large-arp command in the system view to set the system resource
mode to large-arp.
l Reduce the usage of ARP forwarding resources on the device.
Properly plan network deployment, reduce the number of terminals connected to the
device, and prevent all Layer 3 packets from being forwarded by the device. These
measures help reduce the number of ARP entries learned by the device.
CE8850-32CQ-EI V200R002C50
CE8850-64CQ-EI V200R005C00
CE8860EI V100R006C00
CE8861EI V200R005C10
CE8868EI V200R005C10
CE7850EI V100R003C00
CE7855EI V200R001C00
CE6810EI V100R003C00
CE6850EI V100R001C00
CE6850-48S6Q-HI V100R005C00
CE6850-48T6Q-HI/CE6850U-HI/ V100R005C10
CE6851HI
CE6855HI V200R001C00
CE6856HI V200R002C50
CE6857EI V200R005C10
CE6860EI V200R002C50
CE6865EI V200R005C00
CE6870-24S6CQ-EI/CE6870-48S6CQ-EI V200R001C00
CE6870-48T6CQ-EI V200R002C50
CE6875-48S4CQ-EI V200R003C00
CE6880EI V200R002C50
CE5810EI V100R002C00
CE5850EI V100R001C00
CE5850HI V100R003C00
CE5855EI V100R005C10
CE5880EI V200R005C10
CE12804/CE12808/CE12812 V100R001C00
CE12816 V100R003C00
CE12804S/CE12808S V100R005C00
CE12800E V200R002C50
To prevent a DHCP server from allocating an IP address in use to another client, run the dhcp
server ping { packet number | timeout milliseconds } * command in the system view of the
DHCP server to set the maximum number of ping packets to be sent and the maximum
response time of a ping packet.
If this command is run, the DHCP server sends a ping packet to check whether an IP address
is in use before allocating the IP address to a client. If there is no response to the ping packet
within the maximum response time specified by milliseconds, the DHCP server continues to
send ping packets until the number of sent ping packets reaches the maximum number
specified by number. If there is still no response, the DHCP server considers that no device on
the local network segment uses the IP address, and allocates the IP address to the client.
The time for the DHCP server to detect whether an IP address is in use should not be too
long; otherwise, DHCP clients cannot obtain IP addresses. It is recommended that the
configured total detection time (maximum number number of sent ping packets x maximum
response time milliseconds of a ping packet) be less than 8 seconds.
No. CE series switches support system-based DHCP. After a DHCP address pool is
configured in the system view of a switch as a DHCP server, the switch can allocate IP
addresses in the address pool to DHCP clients connected to all interfaces of the switch.
If the ping command is run on a device, the device sends ICMP packets to check the network
connectivity. You can run commands on a Huawei device to configure the device to response
or not response to an ICMP packet whose destination address is a broadcast address.
Therefore, broadcast addresses can be successfully pinged or cannot be pinged on Huawei
devices, meeting customer's different requirements.
l Run the icmp broadcast-address echo disable command to disable the device to
response to an ICMP packet whose destination address is a broadcast address.
l Run the undo icmp broadcast-address echo disable command to enable the device to
response to an ICMP packet whose destination address is a broadcast address.
By default, the device responses to an ICMP packet whose destination address is a broadcast
address.
l The CE6810LI does not support IPv6 Layer 3 forwarding. After the IPv6 function is
enabled on an interface of the CE6810LI, the configured IPv6 address can only be used
to manage the switch.
l In versions earlier than V200R002C50, the CE5855EI does not support IPv6. However,
interfaces on a CE5855EI provide the IPv6 capability when the switch functions as a leaf
switch in an SVF system and the SVF forwarding mode is set to centralized or hybrid.
l Table 22-80 lists CE series switches supporting IPv6 and the minimum version required.
CE12800 and 12800E series switches support IPv6 only after a license is loaded.
CE8850-32CQ-EI V200R002C50
CE8850-64CQ-EI V200R005C00
CE8860EI V100R006C00
CE8861EI V200R005C10
CE8868EI V200R005C10
CE7850EI V100R003C00
CE7855EI V200R001C00
CE6810EI V100R003C00
CE6850EI V100R002C00
CE6850-48S6Q-HI V100R005C00
CE6850-48T6Q-HI/CE6850U-HI/ V100R005C10
CE6851HI
CE6855HI V200R001C00
CE6856HI V200R002C50
CE6857EI V200R005C10
CE6860EI V200R002C50
CE6865EI V200R005C00
CE6870-24S6CQ-EI/CE6870-48S6CQ-EI V200R001C00
CE6870-48T6CQ-EI V200R002C50
CE6875-48S4CQ-EI V200R003C00
CE6880EI V200R002C50
CE5810EI V100R002C00
CE5850EI V100R002C00
CE5850HI V100R003C00
CE5855EI V200R002C50
CE5880EI V200R005C10
CE12804/CE12808/CE12812 V100R002C00
CE12816 V100R003C00
CE12804S/CE12808S V100R005C00
CE12800E V200R002C50
In V100R002 and later versions, the possible reasons for IPv6 address conflict on an interface
are as follows:
l The peer interface directly connected to the local interface has been configured with the
same IPv6 address as that of the local interface.
l After the inloop detection function is enabled on an Ethernet interface, the link local
IPv6 address status of the interface is DUPLICATE, and the global unicast IPv6 address
status of the interface is TENTATIVE (the address is a tentative address and does not
pass the duplicate address detection). You can run the undo loopback command in the
Ethernet interface view to disable the loopback detection mode on the interface.
l Loops exist on the network.
NOTE
By default, the device performs duplicate address detection after an interface is configured with an IPv6
address. You can run the display ipv6 interface interface-type interface-number command in any view
to check whether the IPv6 address of the interface conflicts with other IPv6 addresses.
In V100R002 and later versions, the possible causes for failure in learning ND entries are as
follows:
l The physical status or protocol status of the interface is Down. You can run the display
ipv6 interface interface-type interface-number command in any view to check the
physical status and protocol status of the interface.
l The STP state of the interface is Discarding. You can run the display stp brief command
in any view to check the STP state of the interface.
l The number of ND entries or IPv6 routes exceeds the limit.
– You can run the display ipv6 neighbors command in any view to check
information about ND entries.
– You can run the display ipv6 routing-table statistics command in any view to
check statistics information about IPv6 routes.
l In versions earlier than V200R002C50, CE series switches can only function as DHCPv6
relay agents. In V200R002C50 and later versions, CE series switches can function as
DHCPv6 relay agents and DHCPv6 servers.
l Table 22-81 lists CE series switches supporting DHCPv6 and the minimum version
required.
CE8850-32CQ-EI V200R002C50
CE8850-64CQ-EI V200R005C00
CE8860EI V100R006C00
CE8861EI V200R005C10
CE8868EI V200R005C10
CE7850EI V100R003C00
CE7855EI V200R001C00
CE6810EI V100R003C00
CE6850EI V100R003C00
CE6850-48S6Q-HI V100R005C00
CE6850-48T6Q-HI/CE6850U-HI/ V100R005C10
CE6851HI
CE6855HI V200R001C00
CE6856HI V200R002C50
CE6857EI V200R005C10
CE6860EI V200R002C50
CE6865EI V200R005C00
CE6870-24S6CQ-EI/CE6870-48S6CQ-EI V200R001C00
CE6870-48T6CQ-EI V200R002C50
CE6875-48S4CQ-EI V200R003C00
CE6880EI V200R002C50
CE5810EI V100R003C00
CE5850EI/CE5850HI V100R003C00
CE5855EI V200R002C50
CE5880EI V200R005C10
CE12804/CE12808/CE12812/CE12816 V100R003C00
CE12804S/CE12808S V100R005C00
CE12800E V200R002C50
22.47 Security
SN:21021141 SN:2102
7810DC000242 7810DC00
Y ET1BS12708S0 FAN7 FAN8 FAN9 Y ET1BS12
FAN5 FAN6
FAN3 No dragglng or lifting with fan handles. No dragglng or lifting with fan handles. FAN4
⾷↘⭞伄ᣀᢁᩢᣢૂᤌ䇴༽Ⱦ ⾷↘⭞伄ᣀᢁᩢᣢૂᤌ䇴༽Ⱦ
FAN1 FAN2
9 10 11 12 13 14
PEM
Figure 22-131 CE8800, CE7800, CE6800, and CE5800's serial number label
STATUS STATUS CONSOLE CE6850-48S4Q-EI STATUS STATUS
SYS
MST
ACT
L/A
ID
ETH
PWR1 FAN1 FAN2 PWR2
Serialᒿࡇਧḷㆮ
Number Label
[BackPlane_1]
/$[ArchivesInfo Version]
/$ArchivesInfoVersion=3.0
[Board Properties]
BoardType=CE12804-AC
BarCode=2102113774P0E4000003
Item=02113774
... ...
BarCode in BackPlane_1 in the command output indicates the serial number.
The display device elabel command can also display serial numbers of components
on the device, including LPUs, power modules, fan modules, and optical modules.
– Run the display sn [ all ] command to view the serial number of the device. This
command applies to V100R005C00 and later versions of CE12800 and CE12800E
and CE8800, CE7800, CE6800, and CE5800 series switches.
In the following example, the display sn [ all ] command is executed on a CE12800
switch.
<HUAWEI> display sn
Chassis 1:
Equipment SN(ESN): 2102113774P0E4000003
License ESN: 2102113774P0E4000003
... ...
This command can also display the serial numbers of fan modules, power modules,
and optical modules of the device.
In the command output, the Online field indicates the component online status, the Register
field indicates the component registration status, and the Alarm field indicates the component
operating status.
In the command output, the Average power consumption field indicates the average device
power, the Total Power Allocated field indicates the consumed device power, and the Total
Power Available field indicates the remaining device power.
In the command output, the Status field indicates the temperature status, the Current field
indicates the current temperature, the Major field indicates the upper threshold for a major
temperature alarm, and the Fatal field indicates the upper threshold for a fatal temperature
alarm.
[BackPlane_1]
/$[ArchivesInfo Version]
/$ArchivesInfoVersion=3.0
[Board Properties]
BoardType=CE12804-AC
BarCode=2102113774P0E4000003
Item=02113774
... ...
In the command output, the ActualPower field indicates the current power, while the
RatedPower field indicates the rated power.
22.48.7 What Are the OIDs for Obtaining CPU Usage and Memory
Usage?
Table 22-82 lists OIDs for obtaining CPU usage and memory usage.
Table 22-82 OIDs for obtaining CPU usage and memory usage
Object OID Description
22.48.9 How Can I Quickly Locate the Cause of High CPU Usage
and Collect Relevant Information?
When CPU usage of a device is high, you can perform the following steps to quickly locate
the cause and collect information. Then you can report the information to technical support
personnel.
1. Run the display cpu command to view the services having high CPU usage.
<HUAWEI> display cpu
CPU utilization statistics at 2016-02-02 02:07:22 366 ms
System CPU Using Percentage : 6%
CPU utilization for five seconds: 6%, one minute: 6%, five minutes: 6%.
Max CPU Usage : 38%
Max CPU Usage Stat. Time : 2016-02-01 12:29:26 821 ms
State: Non-overload
Overload threshold: 90%, Overload clear threshold: 75%, Duration: 480s
---------------------------
ServiceName UseRate
---------------------------
SYSTEM 6%
AAA 0%
ARP 0%
CMF 0%
......
2. Run the display system service service-name command in the diagnostic view to check
the names of components having high CPU usage.
<HUAWEI> system-view
[~HUAWEI] diagnose
[~HUAWEI-diagnose] display system service SYSTEM
Service : SYSTEM
------------------------------------------------------------------------------
--
Component CID CGID Process HaState CpuUsage
MemUsage
------------------------------------------------------------------------------
--
DFS 0x81DE271D 0x1DE271B 10003 PRIMARY 0%
4006312
TELC 0x8091271F 0x91271D 10001 PRIMARY 0%
19000
SSHC 0x80922720 0x92271E 10001 PRIMARY 0%
41208
SSHS 0x80932723 0x932721 10001 PRIMARY 0%
48844
3. Run the display system thread process process-id command in the diagnostic view
repeatedly to check information about the processes having high CPU usage.
<HUAWEI> system-view
[~HUAWEI] diagnose
[~HUAWEI-diagnose] display system thread process 10003
Info: Operating, please wait for a moment......done.
------------------------------------------------------------------------------
-
Process ID Thread ID Thread Type BindComp BindCpu BindFlag
Usage
------------------------------------------------------------------------------
-
10003 1210525456 main thread Bind all Free
0%
10003 1227637888 DefSch0200 Free all Free
0%
10003 1227506816 IPC0000 Free all Free
0%
10003 1227768960 DefSch0300 Free all Free
0%
10003 1237476480 DefSch0100 Free all Free
0%
10003 1237607552 DefSch0101 Free all Free
0%
10003 1226847360 DMS_PIPE_RECV_TASK Free all Free
0%
10003 1218028672 TICK Free all Free
0%
10003 1226716288 DMS_TIPC_SEND Free all Free
0%
10003 1217766528 BOX_Out Free all Free
0%
10003 1217897600 VCLK Free 0 Free
0%
10003 1237738624 CliGetThreadCpu Free all Free
0%
------------------------------------------------------------------------------
-
Total = 12
4. Run the display system thread callstack process process-id thread-id command in the
diagnostic view to check process call stack information.
<HUAWEI> system-view
[~HUAWEI] diagnose
[~HUAWEI-diagnose] display system thread callstack process 10003 1210525456
Thread 1210525456 (Thread MainThread):
#00 libc.so.6(epoll_wait)
#01 location(Frame_FdMainThread)
#02 location(Frame_Main)
#03 location(main)
#04 libc.so.6()
#05 libc.so.6()
To stop the display of diagnostic information, log in to the device through Telnet and then run
the kill user-interface console command to terminate the Console interface connection.
<HUAWEI> kill user-interface console 0
To view ZTP execution information, log in through the Console port of the active MPU.
If the device is running the ZTP process, you can log in to the device through the console port
and enter CTRL+C to terminate the ZTP process.
10GE1/0/1
SwitchA
10GE1/0/2
In the command output, Total Power reserved for MPU,SFU,CMU and FAN Board(s)
indicates the power reserved for MPUs, SFUs, CMUs, and fan modules. MPUs, SFUs,
CMUs, and fan modules are mandatory components that ensure the normal device operation.
The reserved power is reserved only for the mandatory components. When the corresponding
component is not installed, the system reserves the rated power for it. After the component is
installed to the device, the system reduces the rated power from the reserved power, but does
not reduce the rated power from the available power. That is, when the MPUs, SFUs, CMUs,
and fan modules are installed to the device, their corresponding power values are reduced
from the reserved power of the device, without affecting the available power of the device.
If the collected log information is saved in the logfile, you can run the following command to
enter the logfile and view log files that are generated when the fault occurs (for example,
March 10, 2018), in order to export files.
<HUAWEI> cd logfile/ //Enter the logfile.
<HUAWEI> dir //View files in the logfile.
Directory of flash:/
logfile/
diaglog_6_20180309223152.log.zip
3 -rw- 577,576 Mar 10 2018 05:39:09
diaglog_6_20180310053908.log.zip
4 lrw- 42 Mar 10 2018 06:20:34 diagnostic_information.zip ->
flash:/diagnostic_information.zip //This file is a link file. -> indicates
that the former one the link file of the latter one, and the operating system
diagnosis information can be exported in both directories.
5 -rw- 7,392,575 Mar 10 2018 19:22:18 log.log
6 -rw- 171,657 Mar 08 2018 13:44:38
log_6_20180308134437.log.zip
7 -rw- 71,056 Mar 09 2018 02:44:38
log_6_20180309024438.log.zip
8 -rw- 73,260 Mar 09 2018 15:33:09
log_6_20180309153309.log.zip
9 -rw- 71,613 Mar 10 2018 04:21:34
log_6_20180310042134.log.zip
NOTE
To collect logs of the standby MPU, you can run the cd slave#flash command to enter the directory of the
standby MPU, then run the cd logfile/ command to enter the logfile, and run the dir to view the files in the
logfile.
When the log files are generated, you can export the files from the device using FTP, SFTP, or
SCP. For details, see Local File Management.
NOTE
You can also run the display logbuffer and display trapbuffer commands to view the log and trap
information on the device, and save the information in diagnostic files on a disk. For details, see Diagnostic
File Obtaining Guide.
After the undo terminal monitor command is executed, all prompt information will not
be displayed, including alarm, log, and debugging information.
l Enable prompt information.
<HUAWEI> terminal monitor //Enable a terminal to display prompt
information.
<HUAWEI> terminal alarm //Enable a terminal to display alarm
information.
<HUAWEI> terminal logging //Enable a terminal to display log
information.
<HUAWEI> terminal debugging //Enable a terminal to display debugging
information.
Module Module name Name of the module by which logs are output.
User logs are saved in the log.log file in flash:/logfile, and diagnostic logs are saved in the
diag.log file in flash:/logfile. When the log file size reaches 8 MB, the system compresses the
log file into a log_slot-id_time.log.zip or diaglog_slot-id_time.log.zip file. slot-id and time
indicate the slot ID and log file generation time.
The storage space for saving user logs and diagnostic logs is limited. To view the storage
space and usage, run the display logfile storage-usage command. The log storage space
cannot be adjusted currently. When the remaining storage space of a type of log is less than
20% of the total space, the system deletes the earliest compressed log file to ensure that the
latest logs can be recorded.
When log files occupy more than 80% of the whole storage space or the number of log files
exceeds the upper threshold, the system deletes the earliest log file to ensure that the latest
logs can be recorded.
In versions earlier than V100R006C00, VSs and the Admin-VS share the flash storage space.
The log storage space of VSs is not fixed and depends on the upper-level directory size. In
V100R006C00 and later versions, VSs can be assigned independent flash storage space. In
this situation, the log storage space of VSs depends on the storage space of these VSs.
<HUAWEI> system-view
[~HUAWEI] info-center loghost 192.168.2.2
[*HUAWEI] commit
Unicast The unicast client/server mode is used on a You only need to configure
client/ higher stratum on a synchronization subnet. the client. The server only
server In this mode, the IP address of the server needs to be configured with
mode needs to be obtained in advance. an NTP primary clock.
Note that the client can be
synchronized to the server but
the server cannot be
synchronized to the client.
Symmetri The symmetric peer mode is used on a lower You only need to configure
c peer stratum on the synchronization subnet. In this the symmetric active peer.
mode mode, a symmetric active peer and a The symmetric passive peer
symmetric passive peer can be synchronized does not need to be
with each other. configured with an NTP
command.
In symmetric peer mode, a
symmetric peer of a higher
stratum is synchronized to a
symmetric peer of a lower
stratum.
Multicast The multicast mode applies to the high-speed Relevant commands need to
mode network that has multiple clients and does not be run on the server and the
require high precision. In a typical scenario, client.
one or more clock servers on the network Note that the client can be
periodically send multicast packets to clients, synchronized to the server but
and the clients synchronize time based on the the server cannot be
multicast packets. synchronized to the client.
Manycast The manycast mode applies to the scenario Relevant commands need to
mode where servers are scattered on a network. The be run on the server and the
client can discover and synchronize to the client.
closest manycast server. The manycast mode Note that the client can be
applies to the scenario where the servers are synchronized to the server but
not stable and clients on the entire network do the server cannot be
not need to be configured again due to a synchronized to the client.
change of the server.
NOTE
After the local clock of the CE switch is configured as the NTP clock source, other devices on the network
synchronize time with the CE switch. Exercise caution when performing this configuration to prevent clock
errors on network devices. It is recommended that a high-precision clock source (such as the GPS) be used if
the user network requires high clock precision.
# Configure the NTP master clock, enable NTP authentication, and enable the NTP server
function.
<HUAWEI> system-view
[~HUAWEI] ntp refclock-master 1
[*HUAWEI] ntp authentication enable
[*HUAWEI] ntp authentication-keyid 42 authentication-mode hmac-sha256 Hello@123456
[*HUAWEI] ntp trusted authentication-keyid 42
[*HUAWEI] undo ntp server disable
[*HUAWEI] commit
22.51 Reliability
No.
The VRID on an interface must be unique, but different interfaces can be bound to VRRP
groups with the same VRID. If the number of VRIDs is insufficient, bind different interfaces
to VRRP groups with the same VRID.
If the VRRP backup does not receive VRRP packets from the VRRP master within the
timeout interval, the VRRP backup considers that the VRRP master does not work properly.
Then the VRRP backup becomes the master.
The timeout interval of VRRP packets cannot be configured directly on a CE series switch.
The calculation formula is as follows:
(3*Advertisement_Interval) + Skew_time
Where,
Advertisement_Interval specifies the interval for sending VRRP packets. The default value is
1s.
The value of Skew_Time is as follows:
Skew_Time = (256 - Priority)/256
The value of Skew_Time ranges from 0 to 1.
You can run the vrrp vrid timer advertise command to change the interval for sending
VRRP packets to adjust the timeout interval of VRRP packets.
After the configuration is complete, run the display vrrp verbose command. The
TimerConfig field displays the configured interval for sending VRRP packets.
<HUAWEI> display vrrp verbose
Vlanif100 | Virtual Router 1
State : Master
Virtual IP : 10.1.1.100
Master IP : 10.1.1.2
Send VRRP Packet To Subvlan : all
PriorityRun : 120
PriorityConfig : 120
MasterPriority : 120
Preempt : YES Delay Time : 20 s Remain : --
TimerRun : 2 s
TimerConfig : 2 s
Auth Type : MD5 Auth Key : ******
Virtual MAC : 0000-5e00-0101
Check TTL : YES
Config Type : Normal
Track BFD : atob Priority Reduced :20
BFD-session State : UP
Create Time : 2014-10-07 15:43:42
Last Change Time : 2014-10-07 15:44:03
The device configured with a single-hop BFD session cannot interwork with the device
configured with a multi-hop BFD session.
When a BFD session is configured on a CE series switch and the outbound interface of the
BFD session is specified, the BFD session is a single-hop BFD session. Otherwise, the BFD
session is a multi-hop BFD session. Devices of a BFD session must be configured with the
same BFD session type (single-hop or multi-hop); otherwise, the BFD session cannot be
established.
CE12804/CE12808/CE12812 V100R002C00
CE12816 V100R003C00
CE12804S/CE12808S V100R005C00
CE12800E V200R002C50
CE8868EI V200R005C10
CE8861EI V200R005C10
CE8860EI V100R006C00
CE8850-32CQ-EI V200R002C50
CE8850-64CQ-EI V200R005C00
CE7850EI V100R003C00
CE7855EI V200R001C00
CE6850EI V100R002C00
CE6810EI V100R003C00
CE6810-48S4Q-LI/CE6810-48S-LI V100R003C10
CE6810-32T16S4Q-LI/CE6810-24S2Q-LI V100R005C10
CE6850-48S6Q-HI V100R005C00
CE6850-48T6Q-HI V100R005C10
CE6850U-HI V100R005C10
CE6851HI V100R005C10
CE6855HI V200R001C00
CE6870-24S6CQ-EI/CE6870-48S6CQ-EI V200R001C00
CE6870-48T6CQ-EI V200R002C50
CE6856HI V200R002C50
CE6860EI V200R002C50
CE6880EI V200R002C50
CE6875EI V200R003C00
CE6865EI V200R005C00
CE6857EI V200R005C10
CE5880EI V200R005C10
CE5850EI V100R002C00
CE5810EI V100R002C00
CE5850HI V100R003C00
CE5855EI V100R005C10
Run the shutdown command in the interface view to shut down the interface, and then run
the undo erps ring command to delete the interface from the ERPS ring.
22.52 TRILL
NOTE
CE8861EI, CE8868EI, CE6880EI, CE6865EI, CE6857EI, CE5880EI, CE5810EI and CE6810LI does not
support TRILL.
In V100R003C10 and later versions, you can configure association between TRILL and
MSTP for transparent transmission of BPDUs on a TRILL network without changing the root
bridge.
22.53 VXLAN
The VXLAN function of the CE8800, CE7800, CE6800, and CE5800 is controlled by the
license. By default, the VXLAN function is disabled on a newly purchased device. To use the
VXLAN function, apply for and purchase the license from the equipment supplier.
You can run the mac-address mac-address command on the VBDIF interface to configure a
MAC address for the VBDIF interface. The MAC address ranges on different device models
are different.
l The CE12800E (equipped with ED-E, EG-E, and EGA-E series cards) supports 500 any
unicast MAC addresses.
l The CE12800E (equipped with FD-X series cards) supports 1000 any unicast MAC
addresses.
l When CE12800 works in non-enhanced mode, the following situation will occur:
– If the switch is equipped with the EA series card, CE-L48GT-EC card, or CE-
L48GS-EC card, the MAC address supported by the switch is in the range of
0000-5e00-0100 to 0000-5e00-0107, and a maximum of eight MAC addresses are
allowed.
– If the switch is equipped with other cards but not the preceding cards, the MAC
address supported by the switch is in the range of 0000-5e00-0100 to
0000-5e00-0107 and a maximum of eight MAC addresses are allowed.
In enhanced mode, the MAC address supported by the switch is in the range of
0000-5e00-0100 to 0000-5e00-01ff and a maximum of 16 MAC addresses are allowed.
l The CE8800, CE7800, CE6800, and CE5800 series switches running a version earlier
than V200R002C50 support MAC addresses in the range of 0000-5e00-0100 to
0000-5e00-01ff. Among these models, the CE6870EI/CE6875EI supports a maximum of
16 MAC addresses, while the other models support a maximum of 256 MAC addresses.
In V200R002C50 and later versions, the CE6870EI/CE6875EI supports a maximum of
16 MAC addresses in the range of 0000-5e00-0100 to 0000-5e00-01ff. The CE6865EI/
CE6857EI/CE8861EI/CE8868EI supports a maximum of 1000 any unicast MAC
addresses, while the other models support a maximum of 500 any unicast MAC
addresses.
22.54 FCoE
CE12816 V100R003C00
CE12804S/CE12808S V100R005C00
CE8850-32CQ-EI V200R002C50
CE8850-64CQ-EI V200R005C00
CE8868EI V200R005C10
CE8861EI V200R005C10
CE7855EI V200R001C00
CE7855EI V200R001C00
CE6810EI V100R003C00
CE6810-48S4Q-LI/ V100R003C10
CE6810-48S-LI
CE6810-32T16S4Q- V100R005C10
LI/CE6810-24S2Q-
LI
CE6850-48S6Q-HI V100R005C00
CE6850-48T6Q-HI V100R005C10
CE6850U-HI V100R005C10
CE6851HI V100R005C10
CE6855HI V200R001C00
CE6870-24S6CQ- V200R001C00
EI/
CE6870-48S6CQ-EI
CE6870-48T6CQ-EI V200R002C50
CE6856HI V200R002C50
CE6860EI V200R002C50
CE6875EI V200R003C00
CE6865EI V200R005C00
CE6857EI V200R005C10
CE6810EI V100R005C00
CE6850HI V100R005C00
CE6850U-HI V100R005C10
CE6851HI V100R005C10
CE6855HI V200R001C00
CE6856HI V200R002C50
CE6860EI V200R002C50
CE6865EI V200R005C00
CE6857EI V200R005C10
CE6870-48T6CQ-EI V200R002C50
CE6875EI V200R003C00
The FCF and NPV functions of the CE8800&7800&6800 are controlled by the license. By
default, the FCF and NPV functions are disabled on a newly purchased device. To use the
FCF and NPV functions, apply for and purchase the license from the equipment supplier.
There are two types of licenses for the FCF function:
l CE-LIC-FCF-ALL
The FCF function can be configured on all interfaces.
l CE-LIC-FCF-PORT
The FCF function can be configured on a maximum of 48 interfaces of the same card or
different cards of the CE12800.
The FCF function can be configured on a maximum of 16 interfaces of the
CE8800&7800&6800.
Multiple licenses of CE-LIC-FCF-PORT can be loaded on the device to meet expansion
requirements.
The FSB function is a basic function of the switch, and can be used without a license.
Y indicates compatible, N indicates incompatible, N/A indicates that the function is not
supported or there is no such a scenario. The hyphen (-) indicates that compatibility is not
ensured.
NOTE
l Interworking compatibility is not ensured in the scenarios that are not listed in the following table.
l Only the CE8861EI, CE8860EI and CE6850U-HI support Fiber Channel (FC) interfaces.
The Cisco
switch used as
the NPV
switch does
not respond to
PLOGI
packets with
the FC_ID of
the NPV
switch as the
destination
FC_ID.
When both the
FCF and NPV
device are
used, it is
recommended
that zone
members on
the FCF be
added to the
zone based on
the WWN. Do
not use the
interface
mode.
are used,
the dual-
homing
networking
is
recommen
ded. In
single-
homing
networking
, when the
FCoE-
enabled
device
connects to
the Qlogic
network
adapter,
the link
recovery
period is
long.
after the
network
adapter is
disabled.
SNS5 FC N/ N/ N/ N/ Y -
120 FCF A A A A
SNS2 FC N/ N/ N/ N/ Y -
124 FCF A A A A
SNS2 FC N/ N/ N/ N/ Y -
248 FCF A A A A
FC N/ N/ N/ N/ Y - When a server
FCF A A A A and a storage
device connect
to the N5K FCF,
the server
cannot detect
the storage
device.
FC N/ N/ N/ N/ - - None
FCF A A A A
FC N/ N/ N/ N/ - - None
NPV A A A A
8000 FC N/ N/ N/ N/ Y - configured on
FCF A A A A the Brocade
8000 FCF. In
this case, the
server cannot
detect the
storage device.
DCX FC N/ N/ N/ N/ - Y DCX8510 8B
8510 FCF A A A A (Fifth
generation)
Multiple physical interfaces cannot join an FCoE interface. One physical interface
corresponds to one FCoE interface.
FCoE registration information can be queried through the display fcoe name-server
command.
Table 22-94 Description of the display fcoe name-server brief command output
Item Description
Table 22-95 Description of the display fcoe name-server interface command output
Item Description
22.55 DCB
2octets Ethertype
Time(0)
16octets Time(n)
Time(7)
4octets CRC
Destination address Destination MAC address, which has a fixed value of 01-80-
c2-00-00-01.
Time(0)-Time(7) Backpressure timer. The unit is the time at which an interface sends
512 bits. If Time(n) is 0, backpressure is canceled.
Pad(transmit as Reserved.
zero) The value of this field is 0 during PFC frame transmission.
22.56 QoS
For CE-L48XS-FDA cards, CE12800 series switches do not support queue-based traffic
statistics on multicast packets. They support only queue-based traffic statistics on unicast
packets. To check the queue-based traffic statistics on unicast packets on an interface,
run the display qos queue statistics interface interface-type interface-number command
in any view.
For cards other than CE-L48XS-FDA, queue-based traffic statistics on multicast packets
can be collected. Traffic statistics on multicast packets can also be collected
independently. To check the queue-based traffic statistics on multicast packets on an
interface, run the display qos queue statistics multicast interface interface-type
interface-number command in any view.
l CE12800E equipped with ED-E/EG-E/EGA-E series cards
The CE12800E equipped with ED-E/EG-E/EGA-E series cards supports queue-based
traffic statistics on multicast packets, which are included in queue-based traffic statistics
and cannot be collected separately. To check the queue-based traffic statistics on an
interface, run the display qos queue statistics interface interface-type interface-number
command in any view.
l CE12800E equipped with FD-X series cards
The CE12800E equipped with FD-X series cards supports queue-based traffic statistics
on multicast packets. To check the queue-based traffic statistics on multicast packets on
an interface, run the display qos queue statistics multicast interface interface-type
interface-number command in any view.
[*HUAWEI-10GE1/0/1] quit
[*HUAWEI] commit
0 times matched indicates the number of times software ACL rules match services.
l Packets of FTP, TFTP, Telnet, SNMP, HTTP, routing, and multicast are sent to the CPU
and match software ACL rules. You can use the display acl all command to check the
number of times packets match ACL rules.
l When ACL-based MQC and ACL-based simplified traffic policy are configured on a
CE12800 and 12800E, packets are forwarded through the hardware and match hardware
ACL rules. The display acl hardware statistics command is used to check statistics on
packets matching hardware ACL rules. In addition, you must first use the statistics
enable command in the traffic behavior view to enable traffic statistics and the traffic-
statistics command in the interface, system, VLAN, or QoS group view to enable traffic
statistics of simplified traffic policy.
Only the CE12800E equipped with FD-X series cards, CE6850HI, CE6850U-HI, CE6851HI,
CE6855HI, CE6856HI, CE6857EI, CE6860EI, CE6865EI, CE7850EI, CE7855EI, CE8850EI,
CE8861EI, CE8868EI, and CE8860EI support this mode.
Networking Requirements
You want to apply a traffic policy to allow only TCP packets in ESTABLISHED state to be
sent from VLANIF 30 to VLANIF 150 and discard other packets sent in this direction.
Procedure
1. Configure the traffic classifiers c1, c2, and c3.
<HUAWEI> system-view
[~HUAWEI] acl 3000
[*HUAWEI-acl4-advance-3000] rule permit tcp tcp-flag established // Match
TCP packets in ESTABLISHED state.
[*HUAWEI-acl4-advance-3000] quit
[*HUAWEI] traffic classifier c1
22.57 IP Multicast
CE12804/CE12808/CE12812 V100R001C00
CE12816 V100R003C00
CE12804S/CE12808S V100R005C00
CE12800E V200R002C50
CE8861EI/CE8868EI V200R005C10
CE8860EI V100R006C00
CE8850-32CQ-EI V200R002C50
CE8850-64CQ-EI V200R005C00
CE7850EI V100R003C00
CE7855EI V200R001C00
CE6810EI V100R003C00
CE6850EI V100R001C00
CE6850-48S6Q-HI V100R005C00
CE6850-48T6Q-HI/CE6850U-HI/ V100R005C10
CE6851HI
CE6855HI V200R001C00
CE6856HI V200R002C50
CE6857EI V200R005C10
CE6860EI V200R002C50
CE6865EI V200R005C00
CE6870-24S6CQ-EI/CE6870-48S6CQ-EI V200R001C00
CE6870-48T6CQ-EI V200R002C50
CE6875EI V200R003C00
CE6880EI V200R002C50
CE5810EI V100R002C00
CE5850EI V100R001C00
CE5850HI V100R003C00
CE5855EI V100R006C00
CE5880EI V200R005C10
CE12804/CE12808/CE12812 V100R002C00
CE12816 V100R003C00
CE12804S/CE12808S V100R005C00
CE8861EI/CE8868EI V200R005C10
CE8860EI V100R006C00
CE8850-32CQ-EI V200R002C50
CE8850-64CQ-EI V200R005C00
CE7850EI V100R003C00
CE7855EI V200R001C00
CE6810EI V100R003C00
CE6850EI V100R002C00
CE6850-48S6Q-HI V100R005C00
CE6850-48T6Q-HI/CE6850U-HI/ V100R005C10
CE6851HI
CE6855HI V200R001C00
CE6856HI V200R002C50
CE6857EI V200R005C10
CE6860EI V200R002C50
CE6865EI V200R005C00
CE6870-24S6CQ-EI/CE6870-48S6CQ-EI V200R001C00
CE6870-48T6CQ-EI V200R002C50
CE6875EI V200R003C00
CE5810EI V100R002C00
CE5850EI V100R002C00
CE5850HI V100R003C00
CE12804/CE12808/CE12812 V100R001C00
CE12816 V100R003C00
CE12804S/CE12808S V100R005C00
CE12800E V200R002C50
CE8861EI/CE8868EI V200R005C10
CE8860EI V100R006C00
CE8850-32CQ-EI V200R002C50
CE8850-64CQ-EI V200R005C00
CE7850EI V100R003C00
CE7855EI V200R001C00
CE6810EI V100R003C00
CE6850EI V100R001C00
CE6850-48S6Q-HI V100R005C00
CE6850-48T6Q-HI/CE6850U-HI/ V100R005C10
CE6851HI
CE6855HI V200R001C00
CE6856HI V200R002C50
CE6857EI V200R005C10
CE6860EI V200R002C50
CE6865EI V200R005C00
CE6870-24S6CQ-EI/CE6870-48S6CQ-EI V200R001C00
CE6870-48T6CQ-EI V200R002C50
CE6875EI V200R003C00
CE6880EI V200R002C50
CE5810EI V100R002C00
CE5850EI V100R001C00
CE5850HI V100R003C00
CE5855EI V100R006C00
CE5880EI V200R005C10
CE12804/CE12808/CE12812 V100R002C00
CE12816 V100R003C00
CE12804S/CE12808S V100R005C00
CE8861EI/CE8868EI V200R005C10
CE8860EI V100R006C00
CE8850-32CQ-EI V200R002C50
CE8850-64CQ-EI V200R005C00
CE7850EI V100R003C00
CE7855EI V200R001C00
CE6810EI V100R003C00
CE6850EI V100R002C00
CE6850-48S6Q-HI V100R005C00
CE6850-48T6Q-HI/CE6850U-HI/ V100R005C10
CE6851HI
CE6855HI V200R001C00
CE6856HI V200R002C50
CE6857EI V200R005C10
CE6860EI V200R002C50
CE6865EI V200R005C00
CE6870-24S6CQ-EI/CE6870-48S6CQ-EI V200R001C00
CE6870-48T6CQ-EI V200R002C50
CE6875EI V200R003C00
CE5810EI/CE5850EI V100R002C00
CE5850HI V100R003C00
CE5855EI V100R006C00
CE12804/CE12808/CE12812/CE12816/ V100R006C00
CE12804S/CE12808S
CE12800E V200R002C50
CE8861EI/CE8868EI V200R005C10
CE8860EI V100R006C00
CE8850-32CQ-EI V200R002C50
CE8850-64CQ-EI V200R005C00
CE7850EI V100R006C00
CE7855EI V200R001C00
CE6810EI/CE6850EI/CE6850HI/ V100R006C00
CE6850U-HI/CE6851HI
CE6855HI V200R001C00
CE6856HI V200R002C50
CE6857EI V200R005C10
CE6860EI V200R002C50
CE6865EI V200R005C00
CE6870-24S6CQ-EI/CE6870-48S6CQ-EI V200R001C00
CE6870-48T6CQ-EI V200R002C50
CE6875EI V200R003C00
CE6880EI V200R002C50
CE5810EI/CE5850EI/CE5855EI/CE5850HI V100R006C00
CE5880EI V200R005C10
IPv6 PIM
CE12804/CE12808/CE12812 V100R002C00
CE12816 V100R003C00
CE12804S/CE12808S V100R005C00
CE8861EI/CE8868EI V200R005C10
CE8860EI V100R006C00
CE8850-32CQ-EI V200R002C50
CE8850-64CQ-EI V200R005C00
CE7850EI V100R003C00
CE7855EI V200R001C00
CE6810EI V100R003C00
CE6850EI V100R002C00
CE6850-48S6Q-HI V100R005C00
CE6850-48T6Q-HI/CE6850U-HI/ V100R005C10
CE6851HI
CE6855HI V200R001C00
CE6856HI V200R002C50
CE6857EI V200R005C10
CE6860EI V200R002C50
CE6865EI V200R005C00
CE6870-24S6CQ-EI/CE6870-48S6CQ-EI V200R001C00
CE6870-48T6CQ-EI V200R002C50
CE6875EI V200R003C00
CE5810EI V100R002C00
CE5850EI V100R002C00
CE5850HI V100R003C00
CE12804/CE12808/CE12812/CE12816 V100R003C00
CE12804S/CE12808S V100R005C00
CE8861EI/CE8868EI V200R005C10
CE8860EI V100R006C00
CE8850-32CQ-EI V200R002C50
CE8850-64CQ-EI V200R005C00
CE7850EI V100R003C00
CE7855EI V200R001C00
CE6810EI/CE6850EI V100R003C00
CE6850-48S6Q-HI V100R005C00
CE6850-48T6Q-HI/CE6850U-HI/ V100R005C10
CE6851HI
CE6855HI V200R001C00
CE6856HI V200R002C50
CE6857EI V200R005C10
CE6860EI V200R002C50
CE6865EI V200R005C00
CE6870-24S6CQ-EI/CE6870-48S6CQ-EI V200R001C00
CE6870-48T6CQ-EI V200R002C50
CE6875EI V200R003C00
CE5810EI/CE5850EI/CE5850HI V100R003C00
CE12804/CE12808/CE12812 V100R001C00
CE12816 V100R003C00
CE12804S/CE12808S V100R005C00
CE12800E V200R002C50
CE8861EI/CE8868EI V200R005C10
CE8860EI V100R006C00
CE8850-32CQ-EI V200R002C50
CE8850-64CQ-EI V200R005C00
CE7850EI V100R003C00
CE7855EI V200R001C00
CE6810EI V100R003C00
CE6850EI V100R001C00
CE6850-48S6Q-HI V100R005C00
CE6850-48T6Q-HI/CE6850U-HI/ V100R005C10
CE6851HI
CE6855HI V200R001C00
CE6856HI V200R002C50
CE6857EI V200R005C10
CE6860EI V200R002C50
CE6865EI V200R005C00
CE6870-24S6CQ-EI/CE6870-48S6CQ-EI V200R001C00
CE6870-48T6CQ-EI V200R002C50
CE6875EI V200R003C00
CE6880EI V200R002C50
CE5810EI V100R002C00
CE5850EI V100R001C00
CE5850HI V100R003C00
CE5855EI V100R006C00
CE5880EI V200R005C10
Multicast VPN
CE12804/CE12808/CE12812/CE12816/ V100R005C10
CE12804S/CE12808S
CE12804/CE12808/CE12812 V100R001C00
CE12816 V100R003C00
CE12804S/CE12808S V100R005C00
CE12800E V200R002C50
CE8861EI/CE8868EI V200R005C10
CE8860EI V100R006C00
CE8850-32CQ-EI V200R002C50
CE8850-64CQ-EI V200R005C00
CE7850EI V100R003C00
CE7855EI V200R001C00
CE6810EI V100R003C00
CE6850EI V100R001C00
CE6850-48S6Q-HI V100R005C00
CE6850-48T6Q-HI/CE6850U-HI/ V100R005C10
CE6851HI
CE6855HI V200R001C00
CE6856HI V200R002C50
CE6857EI V200R005C10
CE6860EI V200R002C50
CE6865EI V200R005C00
CE6870-24S6CQ-EI/CE6870-48S6CQ-EI V200R001C00
CE6870-48T6CQ-EI V200R002C50
CE6875EI V200R003C00
CE6880EI V200R002C50
CE5810EI V100R002C00
CE5850EI V100R001C00
CE5850HI V100R003C00
CE5855EI V100R006C00
CE5880EI V200R005C10
CE12804/CE12808/CE12812 V100R002C00
CE12816 V100R003C00
CE12804S/CE12808S V100R005C00
CE8861EI/CE8868EI V200R005C10
CE8860EI V100R006C00
CE8850-32CQ-EI V200R002C50
CE8850-64CQ-EI V200R005C00
CE7850EI V100R003C00
CE7855EI V200R001C00
CE6810EI V100R003C00
CE6850EI V100R002C00
CE6850-48S6Q-HI V100R005C00
CE6850-48T6Q-HI/CE6850U-HI/ V100R005C10
CE6851HI
CE6855HI V200R001C00
CE6856HI V200R002C50
CE6857EI V200R005C10
CE6860EI V200R002C50
CE6865EI V200R005C00
CE6870-24S6CQ-EI/CE6870-48S6CQ-EI V200R001C00
CE6870-48T6CQ-EI V200R002C50
CE6875EI V200R003C00
CE5810EI V100R002C00
CE5850EI V100R002C00
CE5850HI V100R003C00
CE12804/CE12808/CE12812 V100R001C00
CE12816 V100R003C00
CE12804S/CE12808S V100R005C00
CE12800E V200R002C50
CE8861EI/CE8868EI V200R005C10
CE8860EI V100R006C00
CE8850-32CQ-EI V200R002C50
CE8850-64CQ-EI V200R005C00
CE7850EI V100R003C00
CE7855EI V200R001C00
CE6810-48S4Q-LI/CE6810-48S-LI V100R003C10
CE6810-32T16S4Q-LI/CE6810-24S2Q-LI V100R005C10
CE6810EI V100R003C00
CE6850EI V100R001C00
CE6850-48S6Q-HI V100R005C00
CE6850-48T6Q-HI/CE6850U-HI/ V100R005C10
CE6851HI
CE6855HI V200R001C00
CE6856HI V200R002C50
CE6857EI V200R005C10
CE6860EI V200R002C50
CE6865EI V200R005C00
CE6870-24S6CQ-EI/CE6870-48S6CQ-EI V200R001C00
CE6870-48T6CQ-EI V200R002C50
CE6875EI V200R003C00
CE6880EI V200R002C50
CE5810EI V100R002C00
CE5850EI V100R001C00
CE5850HI V100R003C00
CE5855EI V100R005C10
CE5880EI V200R005C10
CE12804/CE12808/CE12812/CE12816/ V200R002C50
CE12804S/CE12808S
CE8861EI/CE8868EI V200R005C10
CE8860EI V200R002C50
CE8850-32CQ-EI V200R002C50
CE8850-64CQ-EI V200R005C00
CE7850EI/CE7855EI V200R002C50
CE6810LI/CE6810EI/CE6850EI/ V200R002C50
CE6850HI/CE6850U-HI/CE6851HI/
CE6855HI/CE6856HI/CE6860EI/
CE6870-24S6CQ-EI/CE6870-48S6CQ-EI
CE6857EI V200R005C10
CE6865EI V200R005C00
CE6870-48T6CQ-EI V200R002C50
CE6875EI V200R003C00
CE5810EI/CE5850EI/CE5850HI V200R002C50
CE12804/CE12808/CE12812 V100R001C00
CE12816 V100R003C00
CE12804S/CE12808S V100R005C00
CE12800E V200R002C50
CE8861EI/CE8868EI V200R005C10
CE8860EI V100R006C00
CE8850-32CQ-EI V200R002C50
CE8850-64CQ-EI V200R005C00
CE7850EI V100R003C00
CE7855EI V200R001C00
CE6810-48S4Q-LI/CE6810-48S-LI V100R003C10
CE6810-32T16S4Q-LI/CE6810-24S2Q-LI V100R005C10
CE6810EI V100R003C00
CE6850EI V100R001C00
CE6850-48S6Q-HI V100R005C00
CE6850-48T6Q-HI/CE6850U-HI/ V100R005C10
CE6851HI
CE6855HI V200R001C00
CE6856HI V200R002C50
CE6857EI V200R005C10
CE6860EI V200R002C50
CE6865EI V200R005C00
CE6870-24S6CQ-EI/CE6870-48S6CQ-EI V200R001C00
CE6870-48T6CQ-EI V200R002C50
CE6875EI V200R003C00
CE6880EI V200R002C50
CE5810EI V100R002C00
CE5850EI V100R001C00
CE5850HI V100R003C00
CE5855EI V100R005C10
CE5880EI V200R005C10
Multicast VLAN
CE12804/CE12808/CE12812 V100R002C00
CE12816 V100R003C00
CE12804S/CE12808S V100R005C00
CE12800E V200R002C50
CE8861EI/CE8868EI V200R005C10
CE8860EI V100R006C00
CE8850-32CQ-EI V200R002C50
CE8850-64CQ-EI V200R005C00
CE7850EI V100R003C00
CE7855EI V200R001C00
CE6810EI V100R003C00
CE6850EI V100R002C00
CE6850-48S6Q-HI V100R005C00
CE6850-48T6Q-HI/CE6850U-HI/ V100R005C10
CE6851HI
CE6855HI V200R001C00
CE6856HI V200R002C50
CE6857EI V200R005C10
CE6860EI V200R002C50
CE6865EI V200R005C00
CE6870-24S6CQ-EI/CE6870-48S6CQ-EI V200R001C00
CE6870-48T6CQ-EI V200R002C50
CE6875EI V200R003C00
CE6880EI V200R002C50
CE5810EI V100R002C00
CE5850EI V100R002C00
CE5850HI V100R003C00
CE5855EI V100R005C10
CE5880EI V200R005C10
NOTE
If you do not specify an ACL or specify an ACL with no rules configured in the preceding commands, only
one static RP can be configured, and the static RP serves all the multicast groups. If you specify an ACL with
rules configured, the configured static RP serves only the multicast groups that match the ACL.
If both static and dynamic RPs are configured, which type of RP takes effect depends on the preferred
keyword in the preceding commands. If the preferred keyword is not specified, the dynamic RP takes
precedence over the static RP. If this keyword is specified, the static RP takes precedence over the dynamic
RP.
If the bidir keyword is not specified, the configured static RP serves only the Protocol Independent Multicast
Sparse Mode (PIM-SM) network. If this keyword is specified, the configured static RP serves only the
Bidirectional PIM (Bidir-PIM) network.
To enable static RPs to function normally, perform the same static-rp configuration on all the devices in a
PIM domain.
# On an IPv4 PIM-SM network, specify different static RPs for group addresses in different
ranges and ensure that the static RPs are preferred.
<HUAWEI> system-view
[~HUAWEI] acl number 2001
[*HUAWEI-acl4-basic-2001] rule permit source 225.1.0.0 0.0.255.255
[*HUAWEI-acl4-basic-2001] quit
[*HUAWEI] acl number 2002
[*HUAWEI-acl4-basic-2002] rule permit source 226.1.0.0 0.0.255.255
[*HUAWEI-acl4-basic-2002] quit
[*HUAWEI] multicast routing-enable
[*HUAWEI] pim
[*HUAWEI-pim] static-rp 10.10.10.1 2001 preferred //Specify 10.10.10.1 as a
static RP and configure it to serve only groups in the range of 225.1.0.0/16.
[*HUAWEI-pim] static-rp 10.10.20.1 2002 preferred //Specify 10.10.20.1 as a
static RP and configure it to serve only groups in the range of 226.1.0.0/16.
[*HUAWEI-pim] quit
[*HUAWEI] commit
# On an IPv6 Bidir-PIM network, specify different static RPs for group addresses in different
ranges and ensure that the static RPs are preferred.
<HUAWEI> system-view
[~HUAWEI] acl ipv6 name rp1
[*HUAWEI-acl6-advance-rp1] rule permit ipv6 destination ff03::101 128
[*HUAWEI-acl6-advance-rp1] quit
[*HUAWEI] acl ipv6 name rp2
[*HUAWEI-acl6-advance-rp2] rule permit ipv6 destination ff04::101 128
[*HUAWEI-acl6-advance-rp2] quit
[*HUAWEI] multicast ipv6 routing-enable
[*HUAWEI] pim ipv6
[*HUAWEI-pim6] bidir-pim //Enable Bidir-PIM.
[*HUAWEI-pim6] static-rp fc00:10::1 acl6-name rp1 preferred bidir //Specify
fc00:10::1 as a static RP and configure it to serve only group ff03::101.
[*HUAWEI-pim6] static-rp fc00:20::1 acl6-name rp2 preferred bidir //Specify
fc00:20::1 as a static RP and configure it to serve only group ff04::101.
[*HUAWEI-pim6] quit
[*HUAWEI] commit
By default, the aging time of a multicast traffic triggered entry in a VLAN is 210 seconds.
You can set a proper aging time for (S, G) or (*, G) entries according to the number of
multicast entries used on your network. If a large number of multicast entries are used on your
network, a too short aging time will make the multicast forwarding table incomplete.
However, if the aging time is too long, invalid entries will be retained for a long time, wasting
system resources. The following table lists the recommended aging time values for different
quantities of multicast entries.
Number of Entries Recommended Aging Time
The following example sets the aging time of multicast traffic triggered entries to 1000
seconds in VLAN 100.
<HUAWEI> system-view
[~HUAWEI] igmp snooping enable
[*HUAWEI] vlan 100
[*HUAWEI-vlan100] igmp snooping enable
[*HUAWEI-vlan100] multicast layer-2 source-lifetime 1000
[*HUAWEI-vlan100] commit
In the command output, Group Address indicates the IP address of the group that users have
joined, and Last Reporter indicates the last host that has sent an IGMP Report message for
the group.
22.58 MPLS
The MPLS function of the CE12800 is controlled by the license. By default, the MPLS
function is disabled on a newly purchased device. To use the MPLS function, apply for and
purchase the license from the equipment supplier.
You can also run the lsp-trigger { all | host | ip-prefix ip-prefix-name | none } command to
modify an LSP setup triggering policy based on actual needs.
The CE8800, CE7800, CE6800, and CE5800 support MPLS LDP and MPLS QoS.
The following tables describe the products and minimum version supporting MPLS features.
Table 22-112 Products and minimum version supporting MPLS LDP and MPLS QoS
Product Minimum Version Required
CE12804/CE12808/CE12812 V100R001C00
CE12816 V100R003C00
CE12804S/CE12808S V100R005C00
CE6850-48S6Q-HI V100R005C00
CE6850-48T6Q-HI/CE6850U-HI/ V100R005C10
CE6851HI
CE6855HI V200R001C00
CE6856HI V200R002C50
CE6857EI V200R005C10
CE6860EI V200R002C50
CE6865EI V200R005C00
CE6870-24S6CQ-EI/CE6870-48S6CQ-EI V200R001C00
CE6870-48T6CQ-EI V200R002C50
CE6875-48S4CQ-EI V200R003C00
CE7850EI V100R003C10
CE7855EI V200R001C00
CE8850-32CQ-EI V200R002C50
CE8850-64CQ-EI V200R005C00
CE8860EI V100R006C00
CE8861EI V200R005C10
CE8868EI V200R005C10
CE12804/CE12808/CE12812/CE12816 V100R003C00
CE12804S/CE12808S V100R005C00
CE12804/CE12808/CE12812/CE12816/ V200R002C50
CE12804S/CE12808S
22.59 VPN
The following tables describe the products and minimum version supporting VPN features.
CE12804/CE12808/CE12812/CE12816 V100R003C00
CE12804S/CE12808S V100R005C00
CE8868EI V200R005C10
CE8861EI V200R005C10
CE8860EI V100R006C00
CE8850-32CQ-EI V200R002C50
CE8850-64CQ-EI V200R005C00
CE7850EI V100R003C00
CE7855EI V200R001C00
CE6850EI/CE6810EI V100R003C00
CE6850HI V100R005C00
CE6850U-HI/CE6851HI V100R005C10
CE6855HI V200R001C00
CE6870-24S6CQ-EI/CE6870-48S6CQ-EI V200R001C00
CE6870-48T6CQ-EI V200R002C50
CE6856HI V200R002C50
CE6860EI V200R002C50
CE6875EI V200R003C00
CE6865EI V200R005C00
CE6857EI V200R005C10
CE5850EI/CE5810EI/CE5850HI V100R003C00
Table 22-117 Products and minimum version supporting BGP/MPLS IP VPN (CE12800)
Product Minimum Version Required
CE12804/CE12808/CE12812 V100R001C00
CE12816 V100R003C00
CE12804S/CE12808S V100R005C00
CE8868EI V200R005C10
CE8861EI V200R005C10
CE8860EI V200R002C50
CE8850-32CQ-EI V200R002C50
CE8850-64CQ-EI V200R005C00
CE7850EI V200R002C50
CE7855EI V200R002C50
CE6870EI V200R002C50
CE6860EI V200R002C50
CE6856HI V200R002C50
CE6855HI V200R002C50
CE6851HI V200R002C50
CE6850HI V200R002C50
CE6850U-HI V200R002C50
CE6875EI V200R003C00
CE6865EI V200R005C00
CE6857EI V200R005C10
Table 22-119 Products and minimum version supporting BGP/MPLS IPv6 VPN (CE12800)
CE12804/CE12808/CE12812 V100R003C00
V100R002C00 supports only IPv6 MCE.
CE12816 V100R003C00
CE12804S/CE12808S V100R005C00
Table 22-120 Products and minimum version supporting BGP/MPLS IPv6 VPN
(CE8800&7800&6800&5800)
CE8868EI V200R005C10
CE8861EI V200R005C10
CE8860EI V200R005C00
CE8850-32CQ-EI V200R005C00
CE8850-64CQ-EI V200R005C00
CE7850EI V200R005C00
CE7855EI V200R005C00
CE6870EI V200R005C00
CE6860EI V200R005C00
CE6856HI V200R005C00
CE6855HI V200R005C00
CE6851HI V200R005C00
CE6850HI V200R005C00
CE6850U-HI V200R005C00
CE6875EI V200R005C00
CE6865EI V200R005C00
CE6857EI V200R005C10
CE12800E V200R002C50
CE8868EI V200R005C10
CE8861EI V200R005C10
CE8860EI V100R006C00
CE8850-32CQ-EI V200R002C50
CE8850-64CQ-EI V200R005C00
CE7850EI V100R003C00
CE7855EI V200R001C00
CE6850EI V100R001C00
CE6850HI V100R005C00
CE6810EI V100R003C00
CE6850U-HI/CE6851HI V100R005C10
CE6855HI V200R001C00
CE6870-24S6CQ-EI/CE6870-48S6CQ-EI V200R001C00
CE6870-48T6CQ-EI V200R002C50
CE6856HI V200R002C50
CE6860EI V200R002C50
CE6880EI V200R002C50
CE6875EI V200R003C00
CE6865EI V200R005C00
CE6857EI V200R005C10
CE5880EI V200R005C10
CE5850EI V100R001C00
CE5810EI V100R002C00
CE5850HI V100R003C00
CE5855EI V100R006C00
CE8868EI V200R005C10
CE8861EI V200R005C10
CE8860EI V100R006C00
CE8850-32CQ-EI V200R002C50
CE8850-64CQ-EI V200R005C00
CE7850EI V100R003C00
CE7855EI V200R001C00
CE6850EI V100R002C00
CE6850HI V100R005C00
CE6810EI V100R003C00
CE6850U-HI/CE6851HI V100R005C10
CE6855HI V200R001C00
CE6870-24S6CQ-EI/CE6870-48S6CQ-EI V200R001C00
CE6870-48T6CQ-EI V200R002C50
CE6856HI V200R002C50
CE6860EI V200R002C50
CE6875EI V200R003C00
CE6865EI V200R005C00
CE6857EI V200R005C10
CE5850EI/CE5810EI V100R002C00
CE5850HI V100R003C00
CE12804/CE12808/CE12812/CE12816 V100R003C00
CE12804S/CE12808S V100R005C00
CE12804/CE12808/CE12812/CE12816 V100R003C00
CE12804S/CE12808S V100R005C00
CE12804/CE12808/CE12812 V100R002C00
CE12816 V100R003C00
CE12804S/CE12808S V100R005C00
CE6850-48S6Q-HI V200R005C00
CE6850-48T6Q-HI/CE6850U-HI/ V200R005C00
CE6851HI
CE6855HI V200R005C00
CE6856HI V200R005C00
CE6860EI V200R005C00
CE6865EI V200R005C00
CE6857EI V200R005C10
CE6870-24S6CQ-EI/CE6870-48S6CQ-EI V200R005C00
CE6870-48T6CQ-EI V200R005C00
CE6875-48S4CQ-EI V200R005C00
CE7850EI V200R005C00
CE7855EI V200R005C00
CE8850-32CQ-EI V200R005C00
CE8850-64CQ-EI V200R005C00
CE8860EI V200R005C00
CE8868EI V200R005C10
CE8861EI V200R005C10
l You can use the VPLS feature only after the MPLS function is enabled. MPLS is a basic
feature of switches and is not under license control.
A VPN target, also called the route target (RT), is a BGP extension community attribute.
BGP/MPLS IP VPN uses VPN targets to control VPN routes advertisement.
A VPN instance is associated with one or more VPN target attributes. VPN target attributes
are classified into the following types:
l Export target: After a PE device learns IPv4 routes from directly connected sites, it
converts the routes to VPN-IPv4 routes and sets the export target attribute for those
routes. The export target attribute is advertised with the routes as a BGP extended
community attribute.
l Import target: After a PE device receives VPN-IPv4 routes from other PE devices, it
checks the export target attribute of the routes. If the export target is the same as the
import target of a VPN instance on the local PE device, the local PE device adds the
route to the VPN routing table.
BGP/MPLS IP VPN uses VPN targets to control advertisement and receiving of VPN routes
between sites. VPN export targets are independent of import targets. An export target and an
import target can be configured with multiple values to implement flexible VPN access
control and VPN networking.
For example, if the import target of a VPN instance contains 100:1, 200:1, and 300:1, any
route with the export target of 100:1, 200:1, or 300:1 is added to the routing table of the VPN
instance.
A CE series switch can interwork with a Cisco switch through GRE, and they support the
Keepalive function of a GRE tunnel. The configuration example is as follows (basic
configurations including routing are not provided):
1. Check whether the route for establishing a tunnel interface is valid. If the route is invalid,
troubleshoot the route problem.
2. For the tunnel interface configured with the Keepalive function, check the traffic and
CPU usage of the tunnel interface on the Cisco switch. Traffic on the tunnel interface of
an old Cisco switch is forwarded by the CPU. If there is heavy traffic, Keepalive packets
cannot be forwarded normally. As a result, the tunnel interface becomes Down.
SW8-Cisco#show processes cpu | exclude 0.00%
CPU utilization for five seconds: 69%/29%; one minute: 71%; five minutes:
77%
The following table lists products and minimum version supporting GRE.
CE12804/CE12808/CE12812/CE12816 V100R003C00
CE12804S/CE12808S V100R005C00
CE8868EI V200R005C10
CE8861EI V200R005C10
CE8860EI V100R006C00
CE7850EI V100R003C00
CE7855EI V200R001C00
CE6850EI/CE6810EI V100R003C00
CE6850HI V100R005C00
CE6850U-HI/CE6851HI V100R005C10
CE6855HI V200R001C00
CE6870-24S6CQ-EI/CE6870-48S6CQ-EI V200R001C00
CE6870-48T6CQ-EI V200R002C50
CE6856HI V200R002C50
CE6860EI V200R002C50
CE6875EI V200R003C00
CE6865EI V200R005C00
CE6857EI V200R005C10
CE5850EI/CE5810EI/CE5850HI V100R003C00
NOTE
NSR is enabled on the device by default and does not need to be configured.
Support for graceful restart (GR) of CE series switches varies according to the routing
protocols:
l The Routing Information Protocol (RIP) does not support GR.
l The Open Shortest Path First (OSPF), Intermediate System to Intermediate System (IS-
IS), and Border Gateway Protocol (BGP) do not support the GR restarter function
(replaced by NSR), but support the GR helper function.
Table 22-129 Impacts of GR and NSR on the network upon an active/standby switchover
Active/Standby Active/Standby Active/Standby
Switchover Without GR Switchover with GR Switchover with NSR
or NSR
Packets are lost during No packet is lost during No packet is lost during
forwarding and services are forwarding and services are forwarding and services are
interrupted. not affected. not affected.
The entire network detects Except the neighbors of The entire network does not
route changes and route devices that perform the detect route changes.
flapping occurs for a short active/standby switchover,
period of time. other routers do not detect
route changes.
l You can run the load-balance profile profile-name command to create a global load
balancing profile and enter the load balancing profile view. Parameter profile-name
specifies a profile name and does not need to be invoked.
By default, there is a load balancing profile named default. The CE12800 supports only
one load balancing profile.
l You can run the display load-balance profile profile-name command to view detailed
information about a specified load balancing profile. If you run the load-balance profile
profile-name command to specify a profile name different from that of the existing load
balancing profile, the profile is renamed and the configuration in the profile remains
unchanged.
Direct 0
OSPF 10
IS-IS 15
Static 60
RIP 100
IBGP 255
EBGP 255
NOTE
Except the CE6810LI, other CE series switches in V100R006C00 and later versions allow users to
manually configure the preference of direct routes. In addition, the preference of each static route varies.
During route selection, the device first compares the external preferences of routes. When the
same external preference is set for different routing protocols, the device selects the optimal
route based on the internal preference. Internal preferences of routing protocols cannot be
manually configured. Table 22-131 lists the internal preferences of routing protocols.
Direct 0
OSPF 10
IS-IS Level-1 15
IS-IS Level-2 18
Static 60
RIP 100
IBGP 200
EBGP 20