- Table of Contents
- Related Documents
-
| Title | Size | Download |
|---|---|---|
| 01-Text | 19.42 MB |
Preventing electrostatic discharge
Grounding methods to prevent electrostatic discharge
Server models and chassis view
Front 8LFF UniBay drive backplane (PBP-8LFF-UniBay-A-2U-G6)
Rear 4SFF UniBay drive backplane (PCA-BP-4SFF-4UniBay-2U-G6)
Front 12LFF UniBay drive backplane (PCA-BP-12LFF-UniBay-2U-G6)
Front 8SFF UniBay drive backplane (PCA-BP-8UniBay-2U-G6)
Component installation guidelines
Storage controllers and power fail safeguard modules
Installing or removing the server
Preparing for the installation
Airflow direction of the server
Temperature and humidity requirements
Equipment room height requirements
Installing or removing the server
(Optional) Installing cable management brackets
Connecting a mouse, keyboard, and monitor
Removing the server from a rack
Powering on and powering off the server
Configuring basic BIOS settings
Installing the operating system and drivers
Replacing a liquid-cooled module
Replacing the server management module
Replacing riser cards and PCIe modules
Replacing a storage controller and a power fail safeguard module
Replacing a standard PCIe network adapter
Replacing the OCP network adapter
Replacing a SATA M.2 SSD and the front M.2 SSD expander module
Installing the LCD smart management module
Replacing the LCD smart management module
Installing and setting up a TCM or TPM
TPM/TCM installation and setup flowchart
Installing the TPM or TCM module
Configuring encryption in the operating system
8LFF Unibay drives at the front
12LFF drives at the front + rear 4SFF UniBay drives at the rear
8SFF Unibay at the front +8SFF Unibay + 8SFF Unibay drives
Connecting cables for the OCP adapter
Connecting the supercapacitor cable
Connecting cables for the front M.2 SSD expander module
Connecting cables for riser cards
Connecting the LCD smart management module cable
Connecting an inlet temperature sensor cable
Connecting cables for the mounting bracket
Monitoring the temperature and humidity of the equipment room
Safety
Safety information
Before operating the server, carefully review the following safety information:
General operating safety
· Only H3C authorized or professional server engineers are allowed to install, service, repair, operate, or upgrade the server.
· Place the server on a clean, stable table or floor for servicing.
· Make sure all cables are correctly connected before you power on the server.
· To ensure good ventilation and proper airflow, follow these guidelines:
¡ Do not block the ventilation openings in the server chassis.
¡ Install blanks if the following module slots are empty:
- Drive bays.
- Fan bays.
- PCIe slots.
- Power supply slots.
¡ To avoid thermal damage to the server, do not operate the server for long periods in any of the following conditions:
- Access panel open or uninstalled.
- Air baffles uninstalled.
- PCIe slots, drive bays, fan bays, or power supply slots empty.
¡ Minimize the time of removing the access panel when maintaining hot-pluggable components.
· To avoid being burnt, allow the server and its internal modules to cool before touching them.
· To prevent server damage from insufficient heat dissipation, do not block the server vents. When you stack the server and other devices vertically in a cabinet, leave a minimum vertical gap of 2 mm (0.08 in) between two devices.
Electrical safety
|
WARNING! If you put the server in standby mode with the power on/standby button on the front panel, the power supplies continue to supply power to some circuits in the server. To remove all power for servicing safety, you must first press the button, wait for the system LED to turn off, and then remove the power cords from the server. |
· To avoid bodily injury or damage to the server, always use the power cords that came with the server.
· Do not use the power cords that came with the server for any other devices.
· Power off the server when installing or removing any components that are not hot swappable.
Battery safety
The server's system board contains a system battery, which is designed with a lifespan of 3 to 5 years.
If the server no longer automatically displays the correct date and time, replace the battery. When you replace the battery, follow these safety guidelines:
· Do not attempt to recharge the battery.
· Do not expose the battery to a temperature higher than 60°C (140°F).
· Do not disassemble, crush, puncture, short external contacts, or dispose of the battery in fire or water.
· Dispose of the battery at a designated facility. Do not throw the battery away together with other wastes.
Rack mounting recommendations
|
|
NOTE: To protect the server from unstable power or power outage, use uninterrupted power supplies (UPSs) to provide power for the server. This power supply protects server hardware from damage caused by power surges and voltage spikes, and can ensure continuous operation during power failures. |
To avoid bodily injury or damage to the equipment, follow these guidelines when you rack mount a server:
· Mount the server in a standard 19-inch rack.
· Make sure the leveling jacks are extended to the floor and the full weight of the rack rests on the leveling jacks.
· Couple the racks together in multi-rack installations.
· Load the rack from the bottom to the top, with the heaviest hardware unit at the bottom of the rack. Install devices in a rack from the bottom to the top of the rack, and install the heaviest device first.
· Get help to lift and stabilize the server during installation or removal, especially when the server is not fastened to the rails. As a best practice, a minimum of two people are required to safely load or unload a rack. A third person might be required to help align the server if the server is installed higher than chest level.
· For rack stability, make sure only one unit is extended at a time. A rack might get unstable if more than one server unit is extended.
· Make sure the rack is stable when you operate a server in the rack.
· To maintain correct airflow and avoid thermal damage to the server, use blank panels to fill empty rack units.
ESD prevention
Preventing electrostatic discharge
Electrostatic charges that build up on people and tools might damage or shorten the lifespan of the system board and electrostatic-sensitive components.
To prevent electrostatic damage, follow these guidelines:
· Transport or store the server with the components in antistatic bags.
· Keep the electrostatic-sensitive components in separate antistatic bags until they arrive at an ESD-protected area.
· Place the components on a grounded surface before removing them from their antistatic bags.
· Avoid touching pins, leads, or circuitry.
Grounding methods to prevent electrostatic discharge
The following are grounding methods that you can use to prevent electrostatic discharge:
· Wear an ESD wrist strap and make sure it makes good skin contact and is reliably grounded. Keep the wristband close to the skin and make sure it can flexibly stretch.
· Take adequate personal grounding measures, including wearing antistatic clothing and static dissipative shoes.
· Use conductive field service tools.
· Use a portable field service kit with a folding static-dissipating work mat.
Dust residue statement
To ensure product quality, H3C servers undergo strict reliability testing before leaving the factory. During the test process, high-speed fans might cause minor dust accumulation on the air baffle and fan surfaces. This is normal and does not affect product performance or reliability.
Safety sign conventions
To avoid bodily injury or damage to the server or its components, make sure you are familiar with the safety signs on the server chassis or its components.
|
Sign |
Description |
Warning |
|
Circuit or electricity hazards are present. Only H3C authorized or professional server engineers are allowed to service, repair, or upgrade the server. |
To avoid bodily injury, do not open any components with the field-servicing forbidden sign in any circumstances. unless you have authorization to do so. |
|
|
Electrical hazards are present. Field servicing or repair is not allowed. Do not open this part under any circumstances. |
To avoid bodily injury, do not open any components with the field-servicing forbidden sign in any circumstances. |
|
|
The RJ-45 ports on the server cannot be used for network connections. |
To avoid electrical shocks, fire, or damage to the equipment, do not connect an RJ-45 port to a telephone. |
|
|
The surface or component might be hot and present burn hazards. To avoid being burnt, allow hot surfaces or components to cool before touching them. |
To avoid being burnt, allow hot surfaces or components to cool before touching them. |
|
|
The server or component is heavy and requires more than one people to carry or move. |
To avoid bodily injury or damage to hardware, do not move a heavy component alone. In addition, observe local occupational health and safety requirements and guidelines for manual material handling. |
|
|
The server is powered by multiple power supplies. |
To avoid bodily injury from electrical shocks, make sure you disconnect all power supplies if you are performing offline servicing. |
For more information about safety, see H3C Indoor Device Operating Environment Requirements.
Identifying the server
|
|
NOTE: · This document provides general product information. The information in this document might differ from your product if it contains custom configuration options or features. · The model name of a hardware option in this document might differ slightly from its model name label. For example, the DDR5-6400B-32G-2Rx8-R memory model represents memory module labels including UN-DDR5-6400B-32G-2Rx8-R and UN-DDR5-6400B-32G-2Rx8-R-2-S, which have different suffixes. · The figures in this document are for illustration only. |
Server models and chassis view
The H3C UniServer R4900 G7 server (hereinafter referred to as R4900 G7 server) is an H3C self-developed 2U dual-processor rack server based on Intel’s new-generation Birch Stream-AP platform. It is widely applicable to general-purpose computing scenarios, and is specifically optimized for high-performance computing, artificial intelligence, and cloud desktop workloads.
It is suitable for typical applications across industries such as Internet, telecom operators, enterprises, and government organizations, featuring high computing performance, large storage capacity, low power consumption, strong scalability, high reliability, and ease of management and deployment. Figure 1 shows the appearance of the server.
The servers are available in the models listed in Table 2.
Table 2 R4900 G7 server models
|
Model |
Drive configuration |
|
LFF |
Supports up to 12LFF drives at the front + 4SFF drives at the rear |
|
SFF |
Supports up to 24SFF drives at the front |
Server specifications
Technical specifications
Table 3 Technical specifications
|
Item |
Description |
|
Processor |
· Supports up to two Intel Birch Stream-AP processors. · Supports up to 500W power consumption per processor · For more information, see Compatibility Matrix by Server Model |
|
Memory |
· Supports up to 24 DDR5 memory slots · Supports RDIMM and MRDIMM ¡ RDIMM: With 1DPC, supports a maximum memory speed of 6400 Mt/s ¡ MRDIMM: With 1DPC, supports a maximum memory speed of 8800 MT/s |
|
Storage controllers |
· High-performance storage controller · NVMe VROC module |
|
Integrated graphics card |
Graphics card chip integrated in the BMC management chip Chip model AST2600 Supported maximum resolution ratio: 1920 × 1200@60Hz (32bpp) Where, · Resolution ratio: ¡ 1920 × 1200: 1920 horizontal pixels and 1200 vertical pixels ¡ 60 Hz: Screen refresh rate of 60 times per second ¡ 32Bpp: Color bits. Higher the color bits, more colors that can be displayed · When both the front and rear VGA connectors are used, only the monitor connected to the front VGA connector is available |
|
Network connectors |
· 1 × 1Gb/s onboard HDM dedicated network port · Support 1 × OCP 3.0 network adapter slot · OCP 3.0 network adapters support NCSI functions |
|
· 5 × USB connectors ¡ 3 × USB 3.0 connectors (one on the right mounting bracket and two on the rear panel) ¡ 2 × USB 2.0 connectors (one on the system board, and one on the left mounting bracket only when the mounting bracket is used) · 1 × RJ-45 HDM dedicated network port (on the rear panel) · 2 × VGA connectors (one on the rear panel and one on the left mounting bracket only when the mounting bracket is used) · 1 × serial port · 1 × HDM dedicated management port (on the front panel, available only when the mounting bracket is used) |
|
|
Expansion slots |
Supports up to 10 standard PCIe 5.0 slots |
|
Supports external USB optical drive |
|
|
Power supply |
Supports 2 × hot-swappable power supplies with 1+1 redundancy |
|
Certification |
CCC, UL, CE |
Physical specifications
Table 4 Technical specifications
|
Category |
Item |
Specifications |
|
Physical specifications |
Dimensions (H × W × D) |
· Without a security bezel: ¡ 87.5 × 445.4 × 850 mm (3.44 × 17.54 × 33.46 in) · With a security bezel: ¡ 87.5 × 445.4 × 878 mm (3.44 × 17.54 × 34.57 in) |
|
Maximum weight (fully equipped) |
· 46.5kg (102.51 lb) |
|
|
Environment specifications |
Temperature |
Operating temperature: 5°C to 45°C (41°F to 113°F) The maximum temperature varies by hardware option presence. For more information, see the Appendix A. |
|
Storage temperature: –40°C to 70°C (–40°F to 158°F) |
||
|
· Operating humidity: 8% to 90%, noncondensing · Storage humidity: 5% to 95%, noncondensing |
||
|
· Operating altitude: –60 m to +3000 m (–196.85 ft to +9842.52 ft) · The allowed maximum temperature decreases by 0.33°C (32.59°F) as the altitude increases by 100 m (328.08 ft) from 900 m (2952.76 ft) · Storage altitude: –60 m to +5000 m (–196.85 ft to +16404.20 ft) |
Components
Table 5 Descriptions of R4900 G7 server components
|
No. |
Name |
Description |
|
1 |
Access panel |
N/A |
|
2 |
System battery |
Powers the system clock to ensure that the system date and time are correct. |
|
3 |
Standard PCIe network adapter |
Installed in a standard PCIe slot to provide network ports. |
|
4 |
Storage controller |
Provides RAID capability to SAS/SATA drives, including RAID configuration and RAID scale-up. It supports online upgrade of the controller firmware and remote configuration. |
|
5 |
GPU module |
Provides image processing and artificial intelligence computing services for the server. |
|
6 |
NVMe VROC module |
Used to activate and achieve the RAID capabilities of NVMe drives in conjunction with the VMD technology. |
|
7 |
Riser card |
Provides PCIe slots for installing PCIe modules to the server. |
|
8 |
Server management module |
Provides HDM out-of-band management features. |
|
9 |
DIMM |
Used to temporarily store computational data from the processors and data exchanged with external storage devices such as drives. |
|
10 |
Rear drive backplane |
Provides power to drives installed at the server rear and offers a data transmission channel. This document installs the 4SFF drive backplane at the server rear. |
|
11 |
Rear drive cage |
Installed at the server rear to accommodate drives. |
|
12 |
IO board |
Provides I/O connectors. |
|
13 |
OCP network adapter |
Network adapter installed onto the OCP network adapter connector on the system board. |
|
14 |
Riser card blank |
Installed on an empty PCIe riser connector to ensure server good ventilation. |
|
15 |
Power supply module |
Supplies power to the server. Power supply module supports hot swapping and 1+1 redundancy. |
|
16 |
Chassis |
Accommodate all components. |
|
17 |
Mounting bracket |
Attach the server to the rack. The right ear is integrated with the front I/O component, and the left ear is integrated with a VGA connector, an HDM dedicated management connector, and a USB connector 2.0 connector. |
|
18 |
Front drive backplane |
Provides power to drives and offers a data transmission channel. This document installs an 8SFF drive backplane at the front as an example. |
|
19 |
LCD smart management module |
Displays basic server information, operating status, and fault information, and provides diagnostics and troubleshooting capabilities. You can locate and troubleshoot component failures by using the LCD module in conjunction with the event logs generated in HDM. |
|
20 |
Drive |
Provides data storage space. Drives support hot swapping. |
|
21 |
SATA M.2 expander module |
Expands the server with SATA M.2 SSDs. |
|
22 |
SATA M.2 SSD |
Provides data storage space for the server. |
|
23 |
Riser card |
Riser card dedicated to the cold
plate-based liquid-cooled server. |
|
24 |
System board |
One of the most important parts of a server, on which multiple components are installed, such as processor, memory, and fan. It is integrated with basic server components, including the BIOS chip and PCIe connectors. |
|
25 |
Processor socket cover |
Installed over an empty processor socket to protect pins in the socket. |
|
26 |
Processor |
Integrates memory and PCIe controllers to provide data processing capabilities for the server. |
|
27 |
Processor retaining bracket |
Used to secure a processor to the heatsink. |
|
28 |
Fan |
Helps server ventilation. Fans support hot swapping and N+1 redundancy. |
|
29 |
Fan cage |
Accommodates fan modules. |
|
30 |
Chassis-open alarm module |
Detects if the access panel is removed. The detection result can be displayed from the HDM Web interface. |
|
31 |
Supercapacitor holder |
Secures a supercapacitor into the chassis. |
|
32 |
Air baffle |
Provides ventilation aisles for processor heatsinks and memory modules and provides support for the supercapacitor. |
|
33 |
Supercapacitor |
Used to supply power to the flash card on the storage controller during an unexpected power outage, achieving power fail safeguard for data on the storage controller. |
|
34 |
Processor heatsink |
Cools the processor. |
|
35 |
Processor liquid-cooled heatsink |
Cools the processors. |
Front panel
This section describes components, LEDs, and ports on the front panel.
Front panel components
Figure 3 Front panel (8LFF drive configuration)

Table 6 Front panel components (8LFF drive configuration)
|
No. |
Description |
|
1 |
USB 3.0 connector |
|
2 |
(Optional.) Drive or LCD smart management module |
|
3 |
Serial label pull tab |
|
4 |
HDM dedicated management port |
|
5 |
USB 2.0 connector |
|
6 |
VGA connector |
Figure 4 Front panel (12LFF drive configuration)

Table 7 Front panel components (12LFF drive configuration)
|
No. |
Description |
|
1 |
12LFF UniBay drive* |
|
2 |
USB 3.0 connector |
|
3 |
(Optional.) Drive or LCD smart management module |
|
4 |
Serial label pull tab |
|
5 |
HDM dedicated management port |
|
6 |
USB 2.0 connector |
|
7 |
VGA connector |
|
*: The drive types supported by the server vary by drive backplane configuration. For more information, see "Drive backplanes." |
|
Figure 5 Front panel (24SFF drive configuration)
Table 8 Front panel component description
|
Description |
|
|
1 |
Serial label pull tab |
|
2 |
USB 3.0 connector |
|
3 |
(Optional.) Drive or LCD smart management module |
|
4 |
HDM dedicated management port |
|
5 |
USB 2.0 connector |
|
6 |
VGA connector |
|
*: The drive types supported by the server vary by drive backplane configuration. For more information, see "Drive backplanes." |
|
LEDs and buttons
Front LEDs and buttons
Figure 6 Front LEDs and buttons
Table 9 Description of front LEDs and buttons
|
No. |
Description |
State |
|
1 |
Power On/Standby button and system power LED |
· Steady green—The system has started. · Flashing green (1 Hz)—The system is starting. · Steady amber—The system is in standby state. · Off—No power is present. |
|
2 |
LED for Ethernet interfaces on an OCP 3.0 network adapter |
· Steady green—A link is present on any port of an OCP 3.0 network adapter. · Flashing green (1 Hz)—Any port on an OCP 3.0 network adapter is receiving or sending data. · Off—No link is present on any port of any OCP 3.0 network adapter. NOTE: The server supports up to three OCP 3.0 network adapters. |
|
3 |
Health LED |
· Steady green—The system is operating correctly or a minor alarm has occurred. · Flashing green (4 Hz)—HDM is being initialized. · Flashing amber (1 Hz)—A major alarm is present. · Flashing red (1 Hz)—A critical alarm is present. |
|
4 |
UID button LED |
· Steady blue—UID LED is activated. The UID LED can be activated by using the following methods: ¡ Press the UID button LED. ¡ Enable UID LED from HDM. · Flashing blue: ¡ 1 Hz—The system is being remotely managed or HDM is performing out-of-band firmware update. Do not power off the device. ¡ 4 Hz—HDM is restarting. To restart HDM, press the UID button LED for a minimum of 8 seconds. · Off—The UID LED is not activated. |
|
· If the health LED indicates that an error is present, check the operating status of the system through HDM. · If the system power LED is off, possible reasons include: ¡ No power source is connected. ¡ No power supplies are present. ¡ The installed power supplies are faulty. ¡ The system power cords are not connected correctly. |
||
LEDs on the intelligent security bezel
The LEDs on the intelligent security bezel support interaction with the server health status, reflecting the server's operational status and health information, which can accelerate on-site inspections and fault location. You can configure the LED effect as needed. The default effect light is as shown in Table 10.
Figure 7 Intelligent security bezel
Table 10 LEDs on the intelligent security bezel
|
Description |
Decorative LED state |
|
|
Standby |
Standby |
Steady white |
|
Startup |
POST in process |
The white LEDs are gradually turning on from the middle to both sides, reflecting the percentage of the POST progress. |
|
POST completed |
The white LEDs turns on from the middle to both ends in sequence and then goes off three times, displaying a flowing light effect. |
|
|
Running |
Normal |
Breathing white (0.2 Hz gradual change in brightness), while the percentage of lit LEDs indicates the level of server load, with more LEDs gradually lighting up from the middle to both sides as the entire device's power consumption increases. The percentage of lit LEDs represents the level of server load as follows: · Below 10%: No load · 10% to 50%: Low load · 50% to 80%: Medium load · Over 80%: Heavy load |
|
Pre-alarm (only for drive pre-failure) |
Breathing white (1 Hz gradual change in brightness) |
|
|
Major error |
Flashing amber (1 Hz) |
|
|
Critical error (only for power supply failure) |
Flashing red (1 Hz) |
|
|
Remote management |
The system is being remotely managed or HDM is performing out-of-band firmware update. Do not power off the device. |
All LEDs flashing white (1 Hz) |
|
HDM is restarting. |
Some LEDs flashing white (1 Hz) |
|
Ports
Table 11 Ports on the front panel
|
Port |
Type |
Description |
|
VGA connector |
DB-15 |
Used to connect display terminals, such as displays or KVM devices. |
|
USB connector |
USB 2.0/3.0 |
Connects the following devices: · USB flash drive. · USB keyboard or mouse. · USB optical drive for operating system installation. |
|
HDM dedicated management port |
Type-C |
Connects a Type-C to USB adapter cable, which connects to a USB drive. |
Rear panel
Rear panel components
Figure 8 Rear panel components
Table 12 Front panel component description
|
Description |
||
|
1 |
PCIe riser bay 1: PCIe slots 1 through 3 |
|
|
2 |
PCIe riser bay 2: PCIe slots 4 through 6 |
|
|
3 |
PCIe riser bay 3: PCIe slots 7 and 8 |
|
|
4 |
PCIe riser bay 4: PCIe slots 9 and 10 |
|
|
5 |
Power supply 2 |
|
|
6 |
Power supply 1 |
|
|
7 |
(Optional.) Serial port module |
|
|
8 |
VGA connector |
|
|
9 |
Two USB 3.0 connectors |
|
|
10 |
HDM dedicated connector (1 Gbps, RJ-45, default IP address 192.168.1.2/24) |
|
|
11 |
(Optional) OCP 3.0 network adapter (slot 16) |
|
When the server is installed with the liquid-cooled module, slot 6 on the rear panel will be used to install the inlet and outlet ports. The quick connectors will be differentiated into male connectors and hose connectors, as shown in Figure 9.
Figure 9 Rear panel components (Processor liquid-cooled module with hose connector)
Table 13 Rear panel components (with liquid-cooled module installed)
|
No. |
Description |
|
1 |
PCIe riser bay 1: PCIe slots 1 through 3 |
|
2 |
PCIe riser bay 2: PCIe slots 4 through 5 |
|
3 |
PCIe riser bay 3: PCIe slots 7 through 8 |
|
4 |
PCIe riser bay 4: PCIe slots 9 through 10 |
|
5 |
Power supply 2 |
|
6 |
Power supply 1 |
|
7 |
(Optional.) Serial port module |
|
8 |
Liquid outlet (red) |
|
9 |
Liquid inlet (blue) |
|
10 |
VGA connector |
|
11 |
Two USB 3.0 connectors |
|
12 |
HDM dedicated connector (1 Gbps, RJ-45, default IP address 192.168.1.2/24) |
|
13 |
(Optional) OCP 3.0 network adapter (slot 16) |
Rear panel LEDs
Figure 10 Rear panel LEDs

Table 14 Description of rear panel LEDs and buttons
|
No. |
Description |
State |
|
1 |
Power supply LED |
· Steady green—The power supply is operating correctly. · Flashing green (0.33 Hz)—The power supply is in standby state and does not output power. · Flashing green (2 Hz)—The power supply is updating its firmware. · Steady amber—Either of the following conditions exists: ¡ The power module has experienced a critical fault. ¡ The power module has no input, but the other power module has normal input. · Flashing amber (1 Hz)—An alarm has occurred on the power supply. · Off—No power supplies have power input. Possible reasons: ¡ The power cord connection fails. ¡ The external power supply system has lost power. |
|
2 |
Status LED for power supply 1 |
|
|
3 |
Activity LED of the Ethernet port |
· Flashing green (1 Hz)—The port is receiving or sending data. · Off—The port is not receiving or sending data. |
|
4 |
Link LED of the Ethernet port |
· Steady green—A link is present on the port. · Off—No link is present on the network port. |
|
5 |
UID LED |
· Steady blue—UID LED is activated. The UID LED can be activated by using the following methods: ¡ Press the UID button LED. ¡ Enable UID LED from HDM. · Flashing blue: ¡ 1 Hz—The system is being remotely managed or HDM is performing out-of-band firmware update. Do not power off the device. ¡ 4 Hz—HDM is restarting. To restart HDM, press the UID button LED for a minimum of 8 seconds. · Off—UID LED not activated. |
Ports
Table 15 Ports on the rear panel
|
Port |
Type |
Description |
|
VGA connector |
DB-15 |
Used to connect display terminals, such as displays or KVM devices. |
|
Serial port |
DB-9 |
The BIOS serial port is used for the following purposes: · Log in to the server when the remote network connection to the server has failed. · Establish a GSM modem or encryption lock connection. |
|
USB connector |
USB 3.0 |
Connects the following devices: · USB flash drive. · USB keyboard or mouse. · USB optical drive for operating system installation. |
|
HDM dedicated network port |
Establishes a network connection to manage HDM from its Web interface. |
|
|
Power connector |
Standard single-phase |
Connects the power supply to the power source. |
System board
This section describes the layout of the server system board and the components on the system board.
System board layout
Table 16 System board layout description
|
No. |
Description |
Mark |
|
1 |
Built-in USB 2.0 connector |
INTERNAL USB2.0 |
|
2 |
Fan power connector |
HPWR2 |
|
3 |
Drive backplane power connector 4 |
PWR4 |
|
4 |
Fan power connector |
HPWR1 |
|
5 |
Drive backplane power connector 6 |
PWR6 |
|
6 |
Drive backplane power connector 5 |
PWR5 |
|
7 |
Drive backplane power connector 7 |
PWR7 |
|
8 |
GPU power connector 4 |
HPWR4 |
|
9 |
GPU power connector 3 |
HPWR3 |
|
10 |
System serial port connector |
UART |
|
11 |
NVMe VROC module connector |
NVMe RAID KEY |
|
12 |
IO board signal connector 2 |
IO AUX2 |
|
13 |
OCP network adapter power connector |
OCP PWR |
|
14 |
IO board signal connector 1 |
IO AUX1 |
|
15 |
OCP network adapter signal connector |
OCP AUX |
|
16 |
TPM connector |
TPCM |
|
17 |
Right mounting bracket connector |
RIGHT EAR |
|
18 |
BMC module connector |
BMC |
|
19 |
System battery |
N/A |
|
20 |
Drive backplane AUX connector 3 |
BP AUX3 |
|
21 |
LCD module connector |
DIAG LCD |
|
22 |
MCIO connector C1-P4A (for processor 1) |
C1-P4A |
|
23 |
MCIO connector C1-P4C (for processor 1) |
C1-P4C |
|
24 |
MCIO connector C1-P5C (for processor 1) |
C1-P5C |
|
25 |
MCIO connector C1-P5A (for processor 1) |
C1-P5A |
|
26 |
Drive backplane power connector 3 |
PWR3 |
|
27 |
Connector for the air inlet thermal sensor module |
SENSOR |
|
28 |
Drive backplane AUX connector 2 |
BP AUX2 |
|
29 |
Drive backplane power connector 2 |
PWR2 |
|
30 |
MCIO connector C2-P4A (for processor 2) |
C2-P4A |
|
31 |
MCIO connector C2-P4C (for processor 2) |
C2-P4C |
|
32 |
MCIO connector C2-P5C (for processor 2) |
C2-P5C |
|
33 |
MCIO connector C2-P5A (for processor 2) |
C2-P5A |
|
34 |
Fan board signal connector 1 |
FCB AUX1 |
|
35 |
Drive backplane power connector 1 |
PWR1 |
|
36 |
Chassis-open alarm module connector |
INTRUDER |
|
37 |
Drive backplane AUX connector 1 |
BP AUX1 |
|
38 |
Left mounting bracket connector |
LEFT EAR |
|
39 |
MCIO connector C2-P1A (for processor 2) |
C2-P1A |
|
40 |
MCIO connector C2-P1C (for processor 2) |
C2-P1C |
|
41 |
MCIO connector C2-P3C (for processor 2) |
C2-P3C |
|
42 |
MCIO connector C2-P3A (for processor 2) |
C2-P3A |
|
43 |
MCIO connector C2-P0C (for processor 2) |
C2-P0C |
|
44 |
MCIO connector C2-P0A (for processor 2) |
C2-P0A |
|
45 |
MCIO connector C2-P2A (for processor 2) |
C2-P2A |
|
46 |
MCIO connector C2-P2C (for processor 2) |
C2-P2C |
|
47 |
M.2 SSD PORT connector |
M.2 PORT |
|
48 |
NCSI connector |
NCSI |
|
49 |
MCIO connector C1-P1C (for processor 1) |
C1-P1C |
|
50 |
MCIO connector C1-P3C (for processor 1) |
C1-P3C |
|
51 |
MCIO connector C1-P3A (for processor 1) |
C1-P3A |
|
52 |
MCIO connector C1-P0C (for processor 1) |
C1-P0C |
|
53 |
MCIO connector C1-P0A (for processor 1) |
C1-P0A |
|
54 |
MCIO connector C1-P2A (for processor 1) |
C1-P2A |
|
55 |
MCIO connector C1-P2C (for processor 1) |
C1-P2C |
|
X |
System maintenance switch |
MAINTENANCE SW |
System maintenance switch
The system maintenance switch has eight dip switches, as shown in Figure 12.
Figure 12 System maintenance switch

Table 17 describes how to use the maintenance switch. For information about the position of the system maintenance switch, see "System board layout."
Table 17 System maintenance switch description
|
Location |
Description (Off by default) |
Restrictions and guidelines |
|
1 |
Off—HDM login requires the username and password of a valid HDM user account. On—HDM login requires the default username and password. |
When switch 1 is turned on, you can log in to HDM with the default username and password. As a best practice for security purposes, turn off the switch after you complete tasks with the default username and password. |
|
5 |
Off—Normal server startup. ON—Restores the default BIOS settings. |
To restore the default BIOS settings, turn on and then turn off switch 5. The server starts up with the default BIOS settings at the next startup. The server cannot start up when the switch is turned on. To avoid service data loss, stop running services and power off the server before turning on the switch. |
|
6 |
Off—Normal server startup. On—Clears all passwords from the BIOS at server startup. |
To clear all passwords from the BIOS, turn on the switch and then start the server. All the passwords will be cleared from the BIOS. To perform a normal compute module startup, turn off the switch as a best practice before the next compute module startup. |
|
2, 3, 4, 7, and 8 |
Reserved for future use. |
N/A |
DIMM slots
Figure 13 shows the DIMM slot population, where A0, B0…H0, A1, B1...H1 each represent a DIMM slot. For more information about DIMM installation guidelines, see "Memory."
Figure 13 DIMM slots on the system board
HDDs and SSDs
This section contains the following topics:
· Drive numbering for all drive configurations (Drive numbering).
· Drive LED description (Drive LEDs).
Drive numbering
Drive numbering specifies each drive's position and aligns precisely with the labels on the server's front and rear panels.
To view the corresponding number displayed in the software (HDM, BIOS), see Appendix C Drive Slot Number Mapping Matrixes.
Figure 14 Drive numbering for front 12LFF drive configuration
Figure 15 Drive numbering for front 8LFF drive configuration
Figure 16 Drive numbering for front 24SFF drive configuration
Figure 17 Drive numbering for rear 4SFF drive configuration
Drive LEDs
The server supports SAS, SATA, and NVMe drives. You can use the LEDs on a drive to identify its status. Figure 18 shows the location of the LEDs on a drive.

|
(1) Fault/UID LED |
(2) Present/Active LED |
To identify the status of a SAS or SATA drive, use Table 18. To identify the status of an NVMe drive, use Table 19.
Table 18 SAS/SATA drive LED description
|
Fault/UID LED status |
Present/Active LED status (green) |
Description |
|
Steady green/Flashing green (4 Hz) |
A drive failure is predicted. As a best practice, replace the drive before it fails. |
|
|
Steady amber |
Steady green/Flashing green (4 Hz) |
The drive is faulty. Replace the drive immediately. |
|
Steady blue |
Steady green/Flashing green (4 Hz) |
The drive is operating correctly and is selected by the RAID controller. |
|
Off |
Flashing (4Hz) |
The drive is performing RAID migration or rebuilding, or the system is reading or writing data to the drive. |
|
Off |
Steady on |
The drive is present but no data is being read or written to the drive. |
|
Off |
Off |
The drive is not securely installed. |
Table 19 NVMe drive LED description
|
Present/Active LED status (green) |
Description |
|
|
Flashing amber (0.5Hz) |
Off |
The drive has completed the hot removal process and can be hot removed. |
|
Flashing amber (4 Hz) |
Off |
The drive is in hot insertion process. |
|
Flashing amber (0.5Hz) |
Steady green/Flashing green (4 Hz) |
A drive failure is predicted. As a best practice, replace the drive before it fails. |
|
Steady amber |
Steady green/Flashing green (4 Hz) |
The drive is faulty. Replace the drive immediately. |
|
Steady blue |
Steady green/Flashing green (4 Hz) |
The drive is operating correctly and is selected by the RAID controller. |
|
Off |
Flashing (4Hz) |
The drive is performing RAID migration or rebuilding, or the system is reading or writing data to the drive. |
|
Off |
Steady on |
The drive is present but no data is being read or written to the drive. |
|
Off |
Off |
The drive is not securely installed. |
Drive backplanes
· This section describes the supported drive backplanes on the server, including backplane components, and drive types and numbers supported by drive backplanes.
· Drive backplanes can be classified by the types of supported drives into SAS/SATA drive backplanes, UniBay drive backplanes, and X SAS/SATA + Y UniBay drive backplanes.
¡ SAS/SATA drive backplane—All drive slots support SAS/SATA drives.
¡ UniBay drive backplane—All drive slots support both SAS/SATA and NVMe drives.
¡ X SAS/SATA + Y UniBay drive backplane—All drive slots support SAS/SATA drives and specific drive slots support NVMe drives.
- X—Number of slots supporting only SAS/SATA drives.
- Y—Number of slots supporting both SAS/SATA and NVMe drives.
|
|
NOTE: · UniBay drive backplanes and X SAS/SATA+Y UniBay drive backplanes support SAS/SATA and NVMe drives only when both the SAS/SATA and NVMe data cables are connected. · The number of SAS/SATA drives and the number of NVMe drives supported by a UniBay drive backplane or an X SAS/SATA+Y UniBay drive backplane depend on the cabling method. |
Front 8LFF UniBay drive backplane (PBP-8LFF-UniBay-A-2U-G6)
The PCA-BP-8LFF-UniBay-A-2U-G6 is an 8LFF Unibay drive backplane that can be installed at the server front to support a maximum of eight 3.5-inch SAS/SATA/NVMe drives.
Figure 19 8LFF UniBay drive backplane

Table 20 8LFF UniBay drive backplane components
|
No. |
Description |
Mark |
|
1 |
MCIO connector B3/B4 (PCIe5.0 x8), supporting NVMe drives |
NVMe-B3/B4 |
|
2 |
MCIO connector B1/B2 (PCIe5.0 x8), supporting NVMe drives |
NVMe-B1/B2 |
|
3 |
x8 Mini-SAS-HD connector |
SAS-PORT 1 |
|
4 |
MCIO connector A3/A4 (PCIe5.0 x8), supporting NVMe drives |
NVMe-A3/A4 |
|
5 |
MCIO connector A1/A2 (PCIe5.0 x8), supporting NVMe drives |
NVMe-A1/A2 |
|
6 |
Power connector |
PWR |
|
PCIe5.0 x8 description: · PCIe5.0: Fifth-generation signal rate. · x8: Bus bandwidth. |
||
Rear 4SFF UniBay drive backplane (PCA-BP-4SFF-4UniBay-2U-G6)
The 4SFF UniBay drive backplane (PCA-BP-4SFF-4UniBay-2U-G6) is installed at the server front to support a maximum of four 2.5-inch SAS/ SATA/NVMe drives.
Figure 20 4SFF UniBay drive backplane

Table 21 4SFF UniBay drive backplane components
|
No. |
Description |
Mark |
|
1 |
AUX connector |
AUX |
|
2 |
Power connector |
PWR |
|
3 |
MCIO connector B1/B2 (PCIe5.0 x8) |
NVMe-B1/B2 |
|
4 |
MCIO connector B3/B4 (PCIe5.0 x8) |
NVMe-B3/B4 |
|
5 |
x4 Mini-SAS-HD connector |
SAS PORT |
|
PCIe5.0 x8 description: · PCIe5.0: Fifth-generation signal rate. · x8: Bus bandwidth. |
||
Front 12LFF UniBay drive backplane (PCA-BP-12LFF-UniBay-2U-G6)
The 12LFF UniBay drive backplane (PCA-BP-12LFF-UniBay-2U-G6) is installed at the server front to support a maximum of twelve 3.5-inch SAS/ SATA/NVMe drives.
Figure 21 12LFF UniBay drive backplane

Table 22 12LFF UniBay drive backplane components
|
No. |
Description |
Mark |
|
1 |
MCIO connector A3 (PCIe5.0 x4) |
NVMe-A3 |
|
2 |
x4 SlimSAS connector (controls SAS/SATA drives in the last four slots attached to the backplane) |
SAS PORT 2 |
|
3 |
MCIO connector B1/B2 (PCIe5.0 x8) |
NVMe-B1/B2 |
|
4 |
Power connector 2 |
PWR connector 2 |
|
5 |
AUX connector 1 |
AUX connector 1 |
|
6 |
MCIO connector C1 (PCIe5.0 x4) |
NVMe-C1 |
|
7 |
Power connector 1 |
PWR connector 1 |
|
8 |
x8 SlimSAS connector (controls SAS/SATA drives in the first eight slots attached to the backplane) |
SAS PORT 1 |
|
9 |
MCIO connector C3/C4 (PCIe5.0 x8) |
NVMe-C3/C4 |
|
10 |
MCIO connector C2 (PCIe5.0 x4) |
NVMe-C2 |
|
11 |
MCIO connector B3/B4 (PCIe5.0 x8) |
NVMe-B3/B4 |
|
12 |
MCIO connector A4 (PCIe5.0 x4) |
NVMe-A4 |
|
13 |
MCIO connector A1/A2 (PCIe5.0 x8) |
NVMe-A1/A2 |
|
PCIe5.0 x8 description: · PCIe5.0: Fifth-generation signal rate. · x8: Bus bandwidth. |
||
Front 8SFF UniBay drive backplane (PCA-BP-8UniBay-2U-G6)
The 8SFF UniBay drive backplane (model: PCA-BP-8UniBay-2U-G6) is installed at the server front to support a maximum of eight 2.5-inch SAS/SATA/NVMe drives.
Figure 22 8SFF UniBay drive backplane

Table 23 8SFF UniBay drive backplane components
|
No. |
Description |
Mark |
|
1 |
x8 SlimSAS connector |
SAS PORT |
|
2 |
AUX connector |
AUX |
|
3 |
MCIO connector B3/B4 (PCIe5.0 x8) |
NVMe B3/B4 |
|
4 |
Power connector |
POWER |
|
5 |
MCIO connector B1/B2 (PCIe5.0 x8) |
NVMe B1/B2 |
|
6 |
MCIO connector A3/A4 (PCIe5.0 x8) |
NVMe A3/A4 |
|
7 |
MCIO connector A1/A2 (PCIe5.0 x8) |
NVMe A1/A2 |
|
PCIe5.0 x8 description: · PCIe5.0: Fifth-generation signal rate. When you install one or two 8SFF UniBay drive backplanes, PCIe5.0 is supported. When you install three 8SFF UniBay drive backplanes, PCIe4.0 is supported. · x8: Bus bandwidth. |
||
Riser cards
The following describes the available riser cards and modules supported on the riser cards.
For more information about riser cards and installation guidelines for riser cards, see "Riser cards and PCIe modules."
RC-2HHHL-R4-2U-G6
Figure 23 RC-2HHHL-R4-2U-G6

Table 24 RC-2HHHL-R4-2U-G6 riser card components
|
No. |
Description |
|
1 |
SLOT 2 cable |
|
2 |
AUX connector |
|
3 |
PCIe5.0 x8 slot 10 |
|
4 |
PCIe5.0 x8 slot 9 |
|
5 |
Power connector |
|
6 |
SLOT 1 cable |
RC-3FHFL-2U-G7
Figure 24 RC-3FHFL-2U-G7 (1)
Table 25 RC-3FHFL-2U-G7 riser card components
|
No. |
Description |
|
1 |
PCIe5.0 x16 slot 3 |
|
2 |
PCIe5.0 x16 slot 2 |
|
3 |
PCIe5.0 x16 slot 1 |
Figure 25 RC-3FHFL-2U-G7 (2)
Table 26 RC-3FHFL-2U-G7 riser card components
|
No. |
Description |
|
1 |
AUX connector |
|
2 |
PWR connector |
|
3 |
PWR connector |
RC-2FHFL-2U-LC-R2-G7
Figure 26 RC-2FHFL-2U-LC-R2-G7 (1)
Table 27 RC-2FHFL-2U-LC-R2-G7 riser card components
|
No. |
Description |
|
1 |
PCIe5.0 x8 slot 5 |
|
2 |
PCIe5.0 x8 slot 4 |
Figure 27 RC-2FHFL-2U-LC-R2-G7 (2)
Table 28 RC-2FHFL-2U-LC-R2-G7 riser card components
|
No. |
Description |
|
1 |
AUX connector |
|
2 |
PWR connector |
|
3 |
PWR connector |
LCD smart management module
The LCD smart management module is as shown in Figure 28 and its component description is listed in Figure 28Table 29.
An LCD smart management module displays basic server information, operating status, and fault information, and provides diagnostics and troubleshooting capabilities. You can locate and troubleshoot component failures by using the LCD module in conjunction with the event logs generated in HDM.
For more information, see LCD Smart Management Module User Guide.
Figure 28 LCD smart management module

Table 29 LCD smart management module components
|
No. |
Name |
Description |
|
1 |
Mini-USB connector |
Used for upgrading the firmware of the LCD module. |
|
2 |
LCD module cable |
Connects the extension cable, which is then connected to LCD module connector on the system board of the server. |
|
3 |
LCD module shell |
Protects and secures the LCD screen. |
|
4 |
LCD screen |
Displays basic server information, operating status, and fault information. |
Fan
The server supports up to six hot-swappable fans. Figure 29 shows the layout of the fans in the chassis. The server supports N+1 fan module redundancy. That is, the server can operate correctly when a single fan module fails.
The server uses intelligent fan energy-saving and noise-reduction technology, which integrates multiple AI algorithms. It can monitor the temperature, power, and other status information of the device in real-time to obtain the optimal fan adjustment policy. Then, it dynamically adjusts the fan duty cycle configuration to meet the device's energy-saving and noise-reduction requirements.
PCIe slots
The server supports installing riser cards and rear GPU modules at the rear. The PCIe slot numbering varies by hardware configuration.
Figure 30 PCIe slot numbering when riser cards are installed at the rear
Server B/D/F information
The server B/D/F information might change as the PCIe configuration changes. You can obtain B/D/F information by using one of the following methods:
· BIOS log—Search the dumpiio keyword in the BIOS log.
· UEFI shell—Execute the pci command. For information about how to execute the command, execute the help pci command.
· Operating system—The obtaining method varies by OS.
¡ For Linux, execute the lspci-vvv command.
|
|
NOTE: If Linux does not support the lspci command by default, use the software package manager supported by the operating system to obtain and install the pci-utils package. |
¡ For Windows, install the pciutils package, and then execute the lspci command.
¡ For VMware, execute the lspci command.
Component installation guidelines
Processors
Installation guidelines
· You can install one or two processors. If liquid-cooled modules are installed, you must install two processors.
· To avoid damage to a processor or the system board, only H3C authorized or professional server engineers can install, replace, or remove a processor.
· Make sure the processors installed on the same server are the same model.
· The pins in the processor sockets are very fragile and prone to damage. Install a protective cover if a processor socket is empty.
· For the server to operate correctly, make sure processor 1 is always in position. For more information about processor locations, see "System board layout."
· To prevent ESD, wear an ESD wrist strap and make sure it makes good skin contact and is reliably grounded.
· To avoid injury caused by high temperature from the processor heatsink or processor liquid-cooled module during the removal, take necessary heat protection measures before any operation.
Processor model suffixes
If the model of a processor is UN-CPU-INTEL-6747P, the model suffix is P. For more information about the supported processor models, use the component compatibility lookup tool at http://www.h3c.com/en/home/qr/default.htm?id=66.
Table 30 displays the model suffixes of Intel Birch Stream processors.
Table 30 Description of Intel Birch Stream processor model suffixes
|
Processor model suffix description |
Description |
Remarks |
|
E |
Embedded |
Supports embedded systems and industrial applications |
|
P |
Cloud–IaaS |
IaaS scenario-based optimization for VM applications requiring high base frequency |
|
V |
Cloud–SaaS |
SaaS scenario-based optimization for high-density and low-power consumption VM applications |
|
M |
Media Transcode |
Media processing scenario-based optimization |
|
H |
DB and Analytics |
Database and analysis-based optimization. |
|
Y |
Speed Select Technology – Performance Profile |
Supports Intel SST technology for configuring the number of cores and core frequency |
|
N |
Network/5G/Edge(High TPT/Low Latency) |
Supports network/5G/Edge (high throughput/low latency) services |
|
S |
Storage & HCI |
Support for storage and hyper-converged architecture |
|
T |
Long-life Use/High Tcase |
Supports high-lifetime/high-temperature specifications |
|
U |
1-Socket |
Supports only one-processor operation |
|
Q |
Liquid cooling |
Liquid-cooled dedicated processor model |
|
This table is for reference only. For more information, see the Intel official website. |
||
Memory
The server supports DDR5 RDIMMs and DDR5 MRDIMMs.
Memory concepts
DDR
The server supports DDR5 DIMM, which can provide address parity check protection. In case of unexpected system power outage, data in DDR5 DIMMs get lost.
Ranks
The number of ranks in memory is usually 1, 2, 4, or 8, often abbreviated as 1R/SR, 2R, 4R, 8R, or Single-Rank, Dual-Rank, Quad-Rank, 8-Rank.
· A 1R DIMM has one set of memory chips, which are accessed when the system writes to or read from the DIMM.
· A 2R DIMM is equivalent to two 1R DIMMs in one module, but only one rank can be accessed at a time.
· A 4R DIMM is equivalent to two 2R DIMMs in one module, but only one rank can be accessed at a time.
· An 8R DIMM is equivalent to two 4R DIMMs in one module, but only one rank can be accessed at a time.
When writing to or reading from memory, the server memory control subsystem selects the correct rank in memory.
Memory specifications
You can use the memory label to identify the memory specifications. The memory labels across different DDR generations have similar meanings. This document uses DDR5 DIMMs as an example to explain the meaning of each character in the label.
Figure 31 DIMM label
Table 31 DIMM label description
|
No. |
Description |
Definition |
|
1 |
Capacity |
· 8GB · 16GB · 32GB · 64 GB · 96GB |
|
2 |
Number of ranks |
· 1R—One rank (Single-Rank). · 2R—Two ranks (Dual-Rank). · 4R—Four ranks (Quad-Rank). · 8R—Eight ranks (8-Rank). |
|
3 |
Data width |
· ×4—4 bits. · ×8—8 bits. |
|
4 |
DIMM generation |
DDR5 |
|
5 |
Data rate |
· 6400—6400MT/s |
|
6 |
DIMM type |
· R—RDIMM. · L—LRDIMM. |
Memory mode
The server supports protecting data in memory through the following memory modes:
· Independent Mode—Default memory mode. No configuration items are available for this mode in the BIOS Setup Utility.
· Mirror Mode.
Independent mode
Standard ECC can correct single-bit memory errors and detect multi-bit memory errors. When standard ECC detects multi-bit errors, it reports them to the server and causes the server to stop running. In independent mode, in conjunction with other memory RAS technologies, the memory modules can correct one-bit or four-bit memory errors (when the errors are all in the same DDR5 chip). In this case, the independent mode provides stronger protection functions, and is capable of correcting memory errors that standard ECC cannot, thus reducing server downtime caused by uncorrectable memory errors.
Mirror Mode
This mode uses a portion of the system memory for mirroring to enhance system stability, thus reducing server downtime caused by uncorrectable memory errors. When an uncorrectable error is detected in a memory channel, the server retrieves data from the mirrored memory. Mirroring mode supports full mirroring and partial mirroring. In full mirroring mode, half of the system memory is used as mirrored memory. In partial mirroring mode, the mirrored memory capacity can be configured flexibly.
Installation guidelines
The server supports one or two processors. Each processor supports 12 channels, and each channel supports one DIMM. Therefore, one processor supports 12 DIMMs, and two processors support 24 DIMMs. The server supports only DDR5 DIMMs.
|
IMPORTANT: If liquid cooled modules are installed, you must install two processors. |
DIMM and processor compatibility
Use Table 32 to verify that if a DIMM is compatible with the processors.
Table 32 DIMM and processor compatibility
|
Processor platform |
Memory type @ speed |
|
Granite Rapids-AP |
RDIMM @6400MT/s |
|
MRDIMM @8800MT/s |
Memory operating speed
|
|
NOTE: To obtain the memory speed and maximum memory speed supported by a specific processor, use the component compatibility lookup tool at http://www.h3c.com/en/home/qr/default.htm?id=66. You can query the memory speed by selecting Memory Module and query the maximum supported memory speed by selecting Processor. |
The actual operating memory speed is equal to the lesser of the memory speed or the maximum memory speed supported by the processors. For example, if the memory speed is 5600 MT/s and the maximum memory speed supported by processors is 6400 MT/s, the actual operating memory speed is 5600 MT/s.
The server supports only one DIMM per channel (DPC). For more information about the actual memory operating speed of the server’s memory, see Table 33.
Table 33 Actual memory operating speed
|
Processor platform |
DIMM type |
Memory speed |
DPC configuration |
Operating DIMM speed |
|
Granite Rapids-AP |
RDIMM |
6400 MT/s |
1DPC |
6400MT/s |
|
MRDIMM |
8800 MT/s |
1DPC |
8800 MT/s |
Installation guidelines
When you install only DDR5 DIMMs, follow these restrictions and guidelines:
· Make sure the corresponding processors for DIMMs are present.
· As a best practice, install DDR5 DIMMs of the same product code and specifications (type, capacity, ranks, and speed) on the server. To obtain the DIMM product code information, use the Component Compatibility Lookup Tool. To install components or replace faulty DIMMs of other specifications, contact Technical Support.
· In addition to the guidelines mentioned above, different memory modes have their own specific guidelines. See Table 34 for details. If the actual memory installation does not meet these specific guidelines, the system uses the default memory mode (Independent mode), regardless of the user-configured memory mode.
Table 34 DIMM population requirements
|
Memory mode |
Memory installation guidelines |
|
Independent mode (default) |
· Memory installation guidelines: · When you install one processor, populate DIMMs, as shown in Figure 32. · When you install two processors, populate DIMMs, as shown in Figure 33 and Figure 34. |
|
Mirror Mode |
· If one processor is present, the server supports this mode only when 12 DIMMs are installed. · If two processors are present, this mode is supported only in 24 DIMMs configurations. · If one processor is present, see Figure 32. · If two processors are present, see Figure 33 and Figure 34 . |
|
|
NOTE: · In Figure 32, Figure 33, and Figure 34, the black DIMM slots (for example, the D1 slot) are grey colored, and the white DIMM slots (for example, the F0 slot) are not colored. · When MRDIMM memory is configured, only 1DPC is supported. |
Figure 32 DIMM population scheme for one processor
Figure 33 DIMM population scheme for two processors (1)
Figure 34 DIMM population schemes for two processors (2)
SAS/SATA drives
|
|
NOTE: · Using one drive to form multiple RAID arrays complicates future maintenance and affects RAID performance. · If you hot swap an HHD repeatedly at intervals shorter than 30 seconds, the system might fail to identify the drive. |
· SAS/SATA drives support hot swapping in the following conditions:
¡ SAS/SATA drives attached to a storage controller supports hot swapping after you enter the BIOS or OS.
¡ SATA drives attached to the embedded VROC controller supports hot swapping after you enter the OS.
· As a best practice, install drives that do not contain RAID information.
· To avoid RAID performance degrade or RAID creation failure, make sure all drives used to form a RAID array meet the following requirements:
¡ All the drives are SAS or SATA drives.
¡ All the drives are HDD or SSD drives.
· As a best practice, use drives with the same capacity to create a RAID array with redundancy. If the drives have different capacities, the lowest capacity is used across all drives in the RAID.
NVMe drives
· The server supports U.2 NVMe drives.
· As a best practice, install drives that do not contain RAID information.
· As a best practice, use drives with the same capacity to create a RAID array with redundancy. If the drives have different capacities, the lowest capacity is used across all drives in the RAID. For large-capacity drives, their excess capacity cannot be used to configure the current RAID or any other RAID.
· Support for hot insertion and managed hot removal of NVMe drives varies by operating system. For more information, access Component Compatibility Lookup. For more information, see the online replacement guide for NVMe drives.
· If your operating system supports hot swapping of NVMe drives, follow these guidelines:
¡ When you insert a drive, insert it smoothly and continuously. Any pauses might cause the operating system to freeze or reboot.
¡ Do not hot swap multiple NVMe drives at the same time. As a best practice, hot swap NVMe drives one after another at intervals longer than 30 seconds. After the operating system identifies the first NVMe drive, you can hot swap the next drive. If you insert multiple NVMe drives simultaneously, the system might fail to identify the drives.
M.2 SSD
M.2 SSDs are installed on the server using the M.2 SSD expander module. You can install the M.2 SSD expander module to the server front or server rear.
Front SATA/NVMe M.2 SSDs
The RC-SATA-2i/RC-2M2-G7/RC-SATA-2M2 front M.2 SSD expander module is installed between the front drive backplane and the fan modules.
1. RC-SATA-2i
¡ The front M.2 SSD expander module is installed between the drive backplane and the fans at the front of the chassis. It supports both SATA and NVMe M.2 SSDs and is connected to the system board with cables. A maximum of two M.2 SSDs can be installed.
¡ As a best practice, use a SATA M.2 SSD to install the operating system.
Figure 35 Front view of the front M.2 SSD expander module

|
(1) Data cable connector |
(2) M.2 SSD slot 1 |
Figure 36 Rear view of the front M.2 SSD expander module

|
(1) M.2 SSD slot 2 |
2. RC-2M2-G7
¡ RC-2M2-G7 currently only supports the installation of NVMe M.2 SSDs.You can install a maximum of two NVMe M.2 SSDs.
¡ As a best practice, install two NVMe M.2 SSDs of the same model.
¡ To configure RAID, install the NVMe VROC Key module. For information on the RAID levels supported by the NVMe VROC Key and RAID configuration methods, see H3C G6&G7 Server Storage Controller User Guide.
¡ As a best practice, use NVMe M.2 SSDs to install the operating system.
Figure 37 Front view of the front M.2 SSD expander module

|
(1) Data cable connector |
(2) M.2 SSD slot 1 |
Figure 38 Rear view of the front M.2 SSD expander module

3. RC-SATA-2M2
¡ The RC-SATA-2M2 only supports the installation of SATA M.2 SSDs.You can install a maximum of two SATA M.2 SSDs and connect the expander module to the system board with cables.
¡ As a best practice, install two SATA M.2 SSDs of the same model.
¡ As a best practice, use SATA M.2 SSDs to install the operating system.
Figure 39 Front view of the front M.2 SSD expander module
|
(1) Data cable connector |
(2) M.2 SSD slot 1 |
Figure 40 Rear view of the front M.2 SSD expander module
|
(1) M.2 SSD slot 2 |
IO board
The I/O board is installed in the server management module slot on the system board and provides internal and external I/O connectors for the server.
Figure 41 IO board
Table 35 Server management module description
|
No. |
Description |
|
1 |
PDB AUX connector |
|
2 |
RISER AUX connector 3 |
|
3 |
RISER AUX connector 1 |
|
4 |
RISER AUX connector 2 |
|
5 |
Connector for the liquid leakage detection module for GPU |
|
6 |
IO board AUX connector 2 |
|
7 |
Connector for the liquid leakage detection module |
|
8 |
IO board AUX connector 1 |
|
9 |
RISER AUX connector 4 |
|
10 |
Drive backplane AUX connector 5 |
|
11 |
Drive backplane AUX connector 4 |
Riser cards and PCIe modules
PCIe module form factor
Table 36 PCIe module form factor
|
Abbreviation |
Full name |
|
LP |
Low Profile card |
|
FHHL |
Full Height, Half Length card |
|
FHFL |
Full Height, Full Length card |
|
HHHL |
Half Height, Half Length card |
|
HHFL |
Half Height, Full Length card |
Riser card and PCIe module compatibility
Restrictions and guidelines
· When the corresponding processor is absent for a riser card, the PCIe slots on the riser card are unavailable.
· For the location of the PCIe riser connectors on the system board, see "System board layout.” For more information about PCIe slots on a riser card, see "Riser cards."
· You can install small-size PCIe modules into the PCIe slots for larger-size PCIe modules. For example, you can install LP modules into FHFL module slots.
· A PCIe slot can supply power to the installed PCIe module if the maximum power consumption of the module does not exceed 75 W. If the maximum power consumption exceeds 75 W, a power cord is required.
· The description for PCIe5.0 x8 is as follows:
¡ PCIe5.0: Fifth-generation signal speed.
¡ x8: Compatible bus bandwidth, include x8, x4, x2, and x1.
· For an x8 MCIO connector, x8 indicates the bus bandwidth.
· The default connector width of a standard PCIe slot is X16.
Riser card and PCIe module compatibility
For more information about riser card and PCIe module compatibility, see Table 37, and Table 38.
Table 37 Riser card and PCIe module compatibility (1)
|
Riser card model |
Riser card installation location |
PCIe slots and cable extension connectors on the riser card |
PCIe slot or connector description |
PCIe devices supported by PCIe slots or connectors |
PCIe power supply capability |
Proc-essor |
|
|
RC-3FHFL-2U-G7 |
PCIe riser card slot 1 |
PCIe slot number |
Slots 1 through 3 |
PCIe5.0 x16 |
FHFL |
75W |
Processor 1 |
|
Cable extension connectors |
0404A2DS |
x8 MCIO connector |
M1 connects to the MCIO connector C1-P3A on the system board, and M2 connects to the MCIO connector C1-P3C on the system board. M1 and M2 work together to provide an x16 PCIe link for slot 1. |
N/A |
Processor 1 |
||
|
0404A2DS |
x8 MCIO connector |
M1 connects to the MCIO connector C1-P2A on the system board, and M2 connects to the MCIO connector C1-P2C on the system board. M1 and M2 work together to provide an x16 PCIe link for slot 2. |
N/A |
Processor 1 |
|||
|
0404A2DS |
x8 MCIO connector |
M1 connects to the MCIO connector C1-P0A on the system board, and M2 connects to the MCIO connector C1-P0C on the system board. M1 and M2 work together to provide an x16 PCIe link for slot 3. |
N/A |
Processor 1 |
|||
|
PCIe riser connector 2 |
PCIe slots |
Slots 4 through 6 |
PCIe5.0 x16 |
FHFL |
75W |
Processor 2 |
|
|
Cable extension connectors |
0404A2DS |
x8 MCIO connector |
M1 connects to the MCIO connector C2-P1A on the system board, and M2 connects to the MCIO connector C2-P1C on the system board. M1 and M2 work together to provide an x16 PCIe link for slot 4. |
N/A |
Processor 2 |
||
|
0404A2DS |
x8 MCIO connector |
M1 connects to the MCIO connector C2-P3A on the system board, and M2 connects to the MCIO connector C2-P3C on the system board. M1 and M2 work together to provide an x16 PCIe link for slot 5. |
N/A |
Processor 2 |
|||
|
0404A2DS |
x8 MCIO connector |
M1 connects to the MCIO connector C2-P0A on the system board, and M2 connects to the MCIO connector C2-P0C on the system board. M1 and M2 work together to provide an x16 PCIe link for slot 6. |
N/A |
Processor 2 |
|||
|
RC-2FHFL-2U-LC-R2-G7 (for liquid cooling only) |
PCIe riser connector 2 |
PCIe slot number |
slot 4/5 |
PCIe5.0 x16 |
FHFL |
75W |
Processor 1 |
|
Cable extension connectors |
0404A2DK |
x8 MCIO connector |
M1 connects to the MCIO connector C2-P1A on the system board, and M2 connects to the MCIO connector C2-P1C on the system board. M1 and M2 work together to provide an x16 PCIe link for slot 4. |
N/A |
Processor 2 |
||
|
0404A2DK |
x8 MCIO connector |
M1 connects to the MCIO connector C2-P3A on the system board, and M2 connects to the MCIO connector C2-P3C on the system board. M1 and M2 work together to provide an x16 PCIe link for slot 5. |
N/A |
Processor 2 |
|||
Table 38 Riser card and PCIe module compatibility (2)
|
Riser card model |
Riser card installation location |
PCIe slots and cable extension connectors on the riser card |
PCIe slot or connector description |
PCIe devices supported by PCIe slots or connectors |
PCIe power supply capability |
Proc-essor |
|
|
RC-2HHHL-R4-2U-G6 |
PCIe riser connector 3 |
PCIe slot number |
slot 7/8 |
PCIe5.0 x8 |
HHHL |
75W |
N/A |
|
Cable extension connectors |
SLOT 1 cable |
x8 MCIO connector |
Connect to MCIO connector C1-P3A on the system board and provide a ×8 PCIe link for slot 7 |
N/A |
Processor 1 |
||
|
SLOT 2 cable |
x8 MCIO connector |
Connect to MCIO connector C2-P2C on the system board and provide a ×8 PCIe link for slot 8 |
N/A |
Processor 2 |
|||
|
PCIe riser connector 4 |
PCIe slot number |
slot 9/10 |
PCIe5.0 x8 |
HHHL |
75W |
Processor 2 |
|
|
Cable extension connectors |
SLOT 1 cable |
x8 MCIO connector |
Connect to MCIO connector C2-P1C on the system board and provide a ×9 PCIe link for slot 8 |
N/A |
Processor 2 |
||
|
SLOT 2 cable |
x8 MCIO connector |
Connect to MCIO connector C2-P3A on the system board and provide a ×10 PCIe link for slot 8 |
N/A |
Processor 2 |
|||
Storage controllers and power fail safeguard modules
Storage controllers
Storage controllers can be divided into the categories shown in Table 39 according to the installation location.
Table 39 Storage controller description
|
Type |
Installation location |
|
Mezzanine storage controller |
Installed onto the system board through a dedicated mezzanine slot |
|
Standard storage controller |
Installed onto a PCIe riser connector on the system board through a riser card |
Table 40 shows the specifications of the mezzanine storage controller. For the specifications of other storage controllers, use the Compatibility Matrix by Server Model tool.
Table 40 Mezzanine storage controller specification
|
Model Specifications |
Mezzanine storage controller |
|
Connector type |
System board provides two x8 SlimSAS connectors |
|
Port features |
Supports 6.0Gb/s SATA 3.0 connectors and supports hot swapping of corresponding drives |
|
RAID levels |
0, 1, 5, 10 |
|
Location |
Installed onto the system board |
|
Cache |
N/A |
Power fail safeguard module
A power fail safeguard module contains a flash card and a supercapacitor. The server supports independent flash cards that require installation onto a storage controller and built-in flash cards embedded in a storage controller. Built-in flash cards do not require installation.
When a system power failure occurs, the supercapacitor can provide power for a minimum of 20 seconds. During this interval, the storage controller transfers data from DDR memory to the flash card, where the data remains indefinitely or until the controller retrieves the data.
|
|
NOTE: The supercapacitor might have a low charge after the power fail safeguard module is installed or after the server is powered up. If the system displays that the supercapacitor has low charge, no action is required. The system will charge the supercapacitor automatically. You can view the status of the supercapacitor from HDM or the BIOS. |
When you use a supercapacitor, follow these restrictions and guidelines:
· A supercapacitor has a lifespan of 3 to 5 years.
· If the lifespan of a supercapacitor expires, a supercapacitor exception might occur. The system notifies users of supercapacitor exceptions by using the following methods:
¡ For a PMC storage controller, the status of the flash card will become Abnormal+status code. You can check the status code to identify the exception. For more information, see HDM3 online help.
¡ For an LSI storage controller, the status of the flash card will become Abnormal.
¡ You can also review log messages from HDM to identify supercapacitor exceptions. For more information, see HDM3 online help.
· For the power fail safeguard module to take effect, replace the supercapacitor before its lifespan expires.
|
IMPORTANT: After the supercapacitor replacement, verify that cache related settings are enabled for logical drives. For more information, see HDM3 online help. |
Installation guidelines
· You can install multiple standard storage controllers. Follow these restrictions and guidelines:
¡ Make sure the standard storage controllers are of the same vendor. For information about the available storage controllers and their vendors, visit the query tool at http://www.h3c.com/en/home/qr/default.htm?id=66.
¡ Install storage controllers in PCIe slots in the following order: slot 3 >slot 6 > slot 1 > slot 4 > slot 2 > slot 5 > slot 7 > slot 8. If a slot is in use, use the next available slot. For more information about the slot location, see "Rear panel components.”
¡ If multiple drive backplanes are installed only at the server front, install storage controllers to different riser cards. The controller in a lower-numbered slot is connected to the drive backplane for the lower-numbered drive carriers and the controller in a higher-numbered slot to the drive backplane for the higher-numbered drive carriers. For information about drive bay locations, see front panel view in "Front panel components."
¡ If you connect both the front and rear drive backplanes to one storage controller, connect the lower-numbered connector on the controller to the rear drive backplane. Connect the higher-numbered connector on the controller to the front drive backplane.
¡ If you connect both the front and rear drive backplanes to one storage controller, connect the lower-numbered connector on the controller to the rear drive backplane. Connect the higher-numbered connector on the controller to the front drive backplane.
¡ If you install both a 16i storage controller and an 8i storage controller, install the 16i storage controller to the higher-numbered slot and the 8i controller to the lower-numbered slot.
· For information about power fail safeguard modules or supercapacitors compatible with a specific storage controller, see Table 41.
Table 41 Power fail safeguard modules compatible with storage controllers
|
Storage controller model |
Power fail safeguard module or supercapacitor |
Supercapacitor installation location |
|
RAID-LSI-9660-LP-16i-4GB |
BAT-LSI-G3-A |
Into the supercapacitor holder on the air baffle or the supercapacitor holder at the server rear |
|
RAID-LSI-9560-LP-8i-4GB |
||
|
RAID-LSI-9560-LP-16i |
||
|
RAID-P460-B4 |
BAT-PMC-G3-2U |
|
|
HBA-LSI 9540-8i |
Not supported |
Not supported |
|
HBA-LSI-9500-LP-8i |
||
|
HBA-H460-B1 |
||
|
HBA-H460-B1 |
NVMe VROC module
Table 42 Specifications of NVMe VROC modules
|
Model |
Description |
Supported RAID levels |
|
NVMe-VROC-Key-S |
Standard NVMe VROC module supports all NVMe drives |
0, 1, 10 |
|
NVMe-VROC-Key-P |
Advanced NVMe VROC module supports all NVMe drives |
0, 1, 5, 10 |
Network adapters
· You can install an OCP network adapter to slot 16. For information about the slot locations, see PCIe slots.
· OCP network adapters support hot swapping. For more information about operating systems that support hot swapping of OCP network adapters, use the OS Compatibility Lookup Tool. When you configure the recording mode, follow these restrictions and guidelines:
¡ If the operating system in use supports hot swapping of OCP network adapters:
- Only OCP network adapters installed before the server is powered on can be hot swapped. If you hot swap an OCP network adapter, make sure the old and new network adapters are the same model. To replace with an OCP network adapter of a different model, first power off the server.
- OCP network adapters installed after the server is powered on cannot be hot swapped. To replace such an OCP network adapter, first power off the server.
¡ If the operating system in use does not support hot swapping of OCP network adapters: To replace an OCP network adapter, first power off the server.
· To install a standard PCIe network adapter, you must use a riser card. For more information, see "Riser cards and PCIe modules."
GPU module
|
|
NOTE: For information about the configuration guides for the power cords of GPUs, contact Technical Support. |
· To install a FHFL dual-width GPU, install the GPU in the recommended slot as shown in Table 43 and follow these restrictions and guidelines:
¡ Install the GPU in a PCIe slot with a bus bandwidth of x16.
¡ GPUs must be used with riser cards. As a best practice, install one GPU on each riser card.
· To install a HHHL single-width GPU, install the GPU in the recommended slot as shown in Table 44 and follow these restrictions and guidelines:
¡ Install the GPU in a PCIe slot with a bus bandwidth of x16.
¡ To install two or fewer GPUs, install one GPU on each riser card.
¡ If the number of GPU modules is equal to 4, install two RC-3FHFL-2U-G7 riser cards to riser connectors 1 and 2 and install two GPU modules to each riser card as a best practice.
Table 43 FHFL double-width GPU installation guidelines
|
Number of GPUs |
Recommended GPU installation locations |
|
1 |
slot 5 |
|
2 |
slot 2/5 |
Table 44 HHHL single-width GPU installation guidelines
|
Number of GPUs |
Recommended GPU installation locations |
|
1 |
slot 5 |
|
2 |
slot 2/5 |
|
4 |
slot 1/2/4/5 |
Power supply module
|
|
NOTE: For more information about the specifications of a power supply, see the corresponding power supply manuals. |
· Make sure the power supplies installed on the server are the same model. If they differ in model, HDM will raise a major alarm.
· Power supplies are hot swappable.
· To avoid damage to hardware, do not use third-party power supplies.
· The server supports 1+1 power supply redundancy.
· The system provides an overtemperature protection mechanism for power supplies. The power supplies automatically turn off when they encounter an overtemperature situation and automatically turn on when the overtemperature situation is removed.
Fans
· The fans are hot swappable and support N+1 redundancy.
· Make sure the server is fully equipped with fans of the same model.
· The server supports dual-rotor FAN-6056-2U-G7 fan module.
Installing or removing the server
Installation flowchart
Figure 42 Installation flowchart
Preparing for the installation
Prepare an installation site that meets the requirements for space and airflow, temperature, humidity, cleanliness, equipment room height, and grounding.
Rack requirements
Liquid-cooling module not installed
The server is 2U high and has a depth of 850 mm (33.46 in). The rack for installing the server must meet the following requirements:
· A standard 19-inch rack.
· A minimum of 1200 mm (47.24 in) in depth as a best practice. For installation limits for different rack depth, see Table 45. As a best practice, allow technical support engineers to conduct onsite surveys to eliminate potential issues.
· A clearance of more than 50 mm (1.97 in) between the rack front posts and the front rack door.
· For the installation suggestions in a rack with a depth of 1200 mm (47.24 in), see Figure 43.
Table 45 Installation limits for different rack depths
|
Rack depth |
Installation limits |
|
1100mm (43.31 in) |
· The cable management arm (CMA) is not supported. · Make sure the PDU does not hinder slide rails. |
|
1200mm (47.24 in) |
Make sure the CMA does not hinder PDU installation or cabling. If the CMA hinders PDU installation or cabling, change the installation locations of PDUs. |
Figure 43 Installation recommendations for a 1200 mm (47.24 in) deep rack (top view)
|
Rack size |
|||
|
(1) 1200 mm (47.24 in) rack depth |
(2) A minimum of 50 mm (1.97 in) between the front rack posts and the front rack door |
||
|
· Use rear-facing cables for the PDUs to avoid interference with the chassis. · If side-facing cables are used for the PDUs, conduct onsite surveys as a best practice to ensure that the PDUs do not interfere with other components at the chassis rear. |
|||
|
Server size |
|||
|
(3) 850 mm (33.46 in) between the front rack posts and the rear of the chassis, including power supply handles at the server rear (not shown in the figure) |
(4) 850 mm (33.46 in) server depth, including mounting bracket |
||
|
(5) 1104.5 mm (43.48 in) between the front rack posts and the rear ends of the CMA |
(6) 962.6 mm (37.90 in) between the front rack posts and the rear ends of the slide rails |
||
Liquid-cooling module installed
The server is 2U high and has a depth of 850mm (33.46 in). Table 46 shows the requirements of the server on the liquid-cooling system. As a best practice, use it together with H3C cold plate liquid cooling system. For more information about liquid cooling system, see the H3C Cold Plate-Based Liquid-Cooled System User Guide. The server can also be used independently of the H3C's cold plate-based liquid-cooling system. Before use, perform an on-site survey. For more information, contact Technical Support.
Table 46 Server requirements on the liquid-cooling system
|
Item |
Installation limits |
|
Server traffic |
Equal to or greater than 1 liter per minute. |
|
Pressure differential at server inlet and outlet |
Equal to or greater than 25 KPa. |
|
Supported inlet temperature for server / secondary side supply temperature of CDU |
5°C (41°F) to 50°C (122°F) (Recommended: 40°C (104°F)). To prevent condensation, the minimum supply water temperature must be at least 3°C (37.4°F) higher than the dew point temperature. The dew point temperature can be measured using a dew point hygrometer. |
|
Operating pressure of the liquid-cooling system |
Equal to or less than 3.5 Bar, with a recommended value of equal to or less than 2.5 Bar. |
|
Filtration precision at the secondary side |
Equal to or less than 50 microns. |
Airflow direction of the server
Figure 44 Airflow direction of the server
|
(1) and (2) Directions of the airflow into the chassis and power supply |
|
(3) Directions of the airflow out of the power supply |
|
(4) and (5) Direction of the airflow out of the chassis |
Temperature and humidity requirements
To ensure normal operation of the server, a certain temperature and humidity must be maintained in the equipment room. For more information about the operating temperature requirements on the server, see "Physical specifications."
Equipment room height requirements
To ensure correct operation of the server, make sure the equipment room height meets the requirements as described in "Physical specifications."
Corrosive gas limits
About corrosive gases
Corrosive gases can accelerate corrosion and aging of metal components and even cause server failure. Table 47 describes common corrosive gases and their sources.
Table 47 Common corrosive gases and their sources
|
Type |
Sources |
|
H2S |
Geothermal emissions, microbiological activities, fossil fuel processing, wood pulping, and sewage treatment |
|
Sulfur dioxide (SO2) and sulfur trioxide (SO3) |
Combustion of fossil fuel, petroleum products, automobile emissions, ore smelting, sulfuric acid manufacture, and tobacco smoke |
|
Sulphur (S) |
Foundries and sulfur manufacture |
|
HF |
Fertilizer manufacture, aluminum manufacture, ceramics manufacture, steel manufacture, electronics device manufacture, and fossil fuel |
|
NOx |
Automobile emissions, fossil fuel combustion, microbes, and chemical industry |
|
Ammonia (NH3) |
Microbes, sewage, fertilizer manufacture, and geothermal steam |
|
Carbonic oxide (CO) |
Combustion, automobile emissions, microbes, and tree and wood pulping |
|
Chlorine (Cl2) and chlorine dioxide (ClO2) |
Chlorine manufacture, aluminum manufacture, zinc manufacture, and refuse decomposition |
|
Hydrochloric acid (HCl) |
Automobile emissions, combustion, forest fires, and combustion of polymers related to marine environment or processes |
|
Hydrobromic acid (HBr) and hydroiodic acid (HI) |
Automobile emissions |
|
O3 |
Atmospheric photochemical processes mainly involving nitrogen oxides and oxygenated hydrocarbons |
|
Hydrocarbons (CnHn) |
Automobile emissions, tobacco smoke, animal excrement, sewage treatment, and tree and wood pulping |
Requirements in the data center equipment room
As a best practice, make sure the corrosive gas concentration in the data center equipment room meets the requirements of severity level G1 of ANSI/ISA 71.4. The rate of copper corrosion product thickness growth must be less than 300 Å/month, and the rate of silver corrosion product thickness growth must be less than 200 Å/month.
|
|
NOTE: Angstrom (Å) is a metric unit of length equal to one ten-billionth of a meter. |
To meet the copper and silver corrosion rates stated in severity level G1, make sure the corrosive gases in the equipment room do not exceed the concentration limits as shown in Table 48.
Table 48 Requirements in a data center equipment room
|
Corrosive gas |
Concentration (ppb) |
|
|
H2S |
< 3 |
|
|
SO2 and SO3 |
< 10 |
|
|
Cl2 |
< 1 |
|
|
NOx |
< 50 |
|
|
HF |
< 1 |
|
|
NH3 |
< 500 |
|
|
O3 |
< 2 |
|
|
|
NOTE: · Part per billion (ppb) is a concentration unit. 1 ppb represents a volume-to-volume ratio of 1 to 100000000. · The concentration limits are calculated based on the reaction results of the gases in the equipment room with a relative humidity less than 50%. If the relative humidity of the equipment room increases by 10%, the severity level of ANSI/ISA 71.4 to be met must also increase by 1. |
Due to the variability of product performance under the influence of corrosive gases in equipment rooms, see the product installation guide for specific requirements regarding corrosive gas concentration.
Requirements in a non-data center equipment room
The corrosive gas concentration for a non-data center equipment room must meet the requirements of class 3C2 of IEC 60721-3-3:2002, as shown in Table 49.
Table 49 Requirements in a non-data center equipment room
|
Corrosive gas |
Average concentration (mg/m3) |
Max concentration (mg/m3) |
|
SO2 |
0.3 |
1.0 |
|
H2S |
0.1 |
0.5 |
|
Cl2 |
0.1 |
0.3 |
|
HCI |
0.1 |
0.5 |
|
HF |
0.01 |
0.03 |
|
Ammonia (NH3) |
1.0 |
3.0 |
|
O3 |
0.05 |
0.1 |
|
NOx |
0.5 |
1.0 |
|
CAUTION: As a best practice, control the corrosive gas concentrations in the equipment room at their average values. Make sure the corrosive gas concentrations do not exceed 30 minutes per day at their maximum values. |
Due to the variability of product performance under the influence of corrosive gases in equipment rooms, see the product installation guide for specific requirements regarding corrosive gas concentration.
Guidelines for controlling corrosive gases
To maintain cleanliness in the equipment room, follow these guidelines:
· As a best practice, do not build the equipment room in a place with a high concentration of corrosive gases.
· Make sure the equipment room is not connected to sewer, sewage, vertical shaft, or septic tank pipelines and keep it far away from these pipelines. The air inlet of the equipment room must be away from such pollution sources.
· Use environmentally friendly materials to decorate the equipment room. Avoid using organic materials that contains harmful gases, such as sulfur or chlorine-containing insulation cottons, rubber mats, sound-proof cottons, and avoid using plasterboards with high sulfur concentration.
· Place fuel (diesel or gasoline) engines separately. Do not place them in the same equipment room with the device. Make sure the exhausted air of the engines will not flow into the equipment room or towards the air inlet of the air conditioners.
· Place batteries separately. Do not place them in the same room with the device.
· Employ a professional company to monitor and control corrosive gases in the equipment room regularly.
Cleanliness requirements
Mechanically active substances buildup on the chassis might result in electrostatic adsorption, which causes poor contact of metal components and contact points. In the worst case, electrostatic adsorption can cause communication failure.
Cleanliness requirements in a data center equipment room
The concentration of dust participles in the equipment room must meet the ISO 8 cleanroom standard defined by ISO 14644-1, as described in Table 50.
Table 50 Dust requirements in a data center equipment room
|
Particle diameter |
Concentration limit |
Remarks |
|
≥ 5 μm |
≤ 29300 particles/m3 |
Make sure no zinc whiskers are in the equipment room. |
|
≥ 1 μm |
≤ 832000 particles/m3 |
|
|
≥ 0.5 μm |
≤ 3520000 particles/m3 |
Due to the variability of product performance under the influence of dust buildup in equipment rooms, see the product installation guide for specific requirements regarding dust concentration.
Cleanliness requirements in a non-data center equipment room
The concentration of dust participles (particle diameter ≥ 0.5 µm) must meet the requirement of the GB 50174-2017 standard, which is less than 17600000 particles/m3.
Guidelines for controlling corrosive gases
To maintain cleanliness in the equipment room, follow these guidelines:
· Keep the equipment room away from pollution sources and do not smoke or eat in the equipment room.
· Use double-layer glass in windows and seal doors and windows with dust-proof rubber strips.
· Use dustproof materials for floors, walls, and ceilings and use matt coating that does not produce powders.
· Keep the equipment room clean and clean the air filters of the rack regularly.
· Wear ESD clothing and shoe covers before entering the equipment room. Keep the ESD clothing and shoe covers clean and replace them frequently.
Grounding requirements
Correctly connecting the server grounding cable is crucial to lightning protection, anti-interference, and ESD prevention. The server can be grounded through the grounding wire of the power supply system and no external grounding cable is required.
Storage requirements
· As a best practice, do not store an HDD for 6 months or more without powering on and using it.
· As a best practice, do not store an SSD, M.2 SSD, or SD card for 3 months or more without powering on and using it. Long unused time increases data loss risks.
· To store the server chassis, or an HDD, SSD, M.2 card, or SD card for 3 months or more, power on it every 3 months and run it for a minimum 2 hours each time.
Installation tools
The table below lists the tools that you might use during installation.
|
Name |
Description |
|
|
Installs or removes screws inside mounting brackets. A flat-head screwdriver can also be used for this purpose. |
||
|
T30 Torx screwdriver |
Installs or removes captive screws on processor heatsinks. |
|
|
Installs or removes screws on the processor system board. |
||
|
T10 Torx screwdriver (shipped with the server) |
Removes screws for mounting brackets. |
|
|
Flat-head screwdriver |
Replaces the system battery. |
|
|
Phillips screwdriver |
Installs or removes screws on drive carriers. |
|
|
|
Cage nut insertion/extraction tool |
Inserts or extracts the cage nuts in rack posts. |
|
Diagonal pliers |
Clips insulating sleeves. |
|
|
Paper knife |
Removes the server's external packaging. |
|
|
Tape measure |
Measures distance. |
|
|
Multimeter |
Measures resistance and voltage. |
|
|
ESD wrist strap |
Prevents ESD when you operate the server. |
|
|
ESD gloves |
||
|
Antistatic clothing |
||
|
Ladder |
Supports high-place operations. |
|
|
Interface cable (such as an Ethernet cable or optical fiber) |
Connects the server to an external network. |
|
|
Type-C to USB cable (connecting a USB WIFI module or USB drive) |
· If you connect a third-party USB WiFi module, you can use the HDM Mobile client on a mobile endpoint to access the HDM interface. · If you connect a USB drive, you can download SDS log messages to the USB drive from HDM. NOTE: Support for USB Wi-Fi modules depends on the server model. |
|
|
Serial console cable |
Connects the serial connector on the server to a monitor for troubleshooting. |
|
|
Monitor |
Displays the output from the server. |
|
|
Temperature and humidity meter |
Displays current temperature and humidity in the equipment room. |
|
|
Oscilloscope |
Displays the variation of voltage over time in waveforms. |
Installing or removing the server
This section describes the methods for installing the server.
(Optional) Installing rails
Install the inner rails to the server and the outer rails to the rack. For information about installing the rails, see the document shipped with the rails.
Rack-mounting the server
1. Slide the server into the rack. For more information about how to slide the server into the rack, see the installation guide for the rails.
Figure 45 Rack-mounting the server
2. Secure the server. Push the server until the mounting brackets are flush against the rack front posts. Unlock the latches of the mounting brackets, fasten the captive screws inside the mounting brackets, and lock the latches.
Figure 46 Fastening the captive screws inside the mounting brackets
(Optional) Installing cable management brackets
Install cable management brackets if the server is shipped with cable management brackets. For information about how to install cable management brackets, see the installation guide shipped with the brackets.
Connecting external cables
This section describes how to connect external cables to the server.
Connecting a mouse, keyboard, and monitor
About this task
The server provides two DB15 VGA connectors for connecting a monitor.
One VGA connector is on the front panel (left mounting bracket is required) and the other is on the rear panel.
The server is not shipped with a standard PS2 mouse and keyboard. You can directly connect a USB mouse and USB keyboard or use a USB-to-PS2 adapter cable to connect a PS2 mouse and PS2 keyboard.
· Connect the USB mouse and keyboard directly. Use the same method as for a standard USB device.
· Connect a PS2 mouse and keyboard using a USB-to-PS2 cable.
Procedure
1. Connect one plug of a VGA cable to a VGA connector on the server, and fasten the screws on the plug.
Figure 47 Connecting a VGA cable
2. Connect the other plug of the VGA cable to the VGA connector on the monitor, and fasten the screws on the plug.
3. Insert the USB connector of the USB-to-PS2 adapter to a USB connector on the server. Then, insert the PS2 connectors of the mouse and keyboard into the PS2 receptacles of the adapter.
Figure 48 Connecting a PS2 mouse and keyboard by using a USB-to-PS2 adapter
Connecting an Ethernet cable
About this task
· Use an Ethernet interface to set up a network environment.
· Log in to the HDM Web interface via the HDM dedicated network port to manage the server.
· Replace the network cable because the network is unreachable or the network cable is too long or short.
Procedure
1. Determine the network port on the server.
¡ To connect the server to the external network, use the Ethernet port on the network adapter.
¡ To log in to the HDM management interface:
- Use the HDM dedicated network port. For the location of the HDM dedicated network port, see "Rear panel components."
- (Optional.) Use the HDM shared network port. If the server is configured with an OCP network adapter, you can also use the HDM shared network port on the OCP network adapter to log in to the HDM management interface. For the location of the OCP network adapter, see "Rear panel components."
2. Determine the type of the Ethernet cable.
Verify the connectivity of the cable by using a link tester. If you are replacing the Ethernet cable, make sure the new cable is the same type or compatible with the old cable.
3. Label the network cable.
¡ If you are replacing the Ethernet cable, label the new cable with the same number as the number of the old cable.
¡ As a best practice, use labels of the same kind for all cables. Fill in the names and numbers of the local device and the remote device on the label.
4. Connect the network cable. As shown in Figure 49, connect one end of the network cable to an Ethernet port on the switch and the other end to the peer device.
Figure 49 Connecting an Ethernet cable
5. Verify network connectivity.
After powering on the server, use the ping command to test the network connectivity. If the connection between the server and the peer device fails, verify that the Ethernet cable is securely connected.
6. Secure the Ethernet cable. For information about how to secure cables, see "Securing cables."
Connecting power cords
Restrictions and guidelines
· To avoid damage to the equipment or even bodily injury, use the power cord that ships with the server.
· Before connecting the power cord, make sure the server and components are installed correctly.
Procedure
|
|
NOTE: Multiple types of wire fasteners can be used for securing the power cord. In this procedure, a cable clamp is used. |
1. Insert the power cord plug into the power receptacle of a power supply at the rear panel.
Figure 50 Connecting the power cord
2. Connect the other end of the power cord to the power source, for example, the power strip on the rack.
3. Secure the power cord to avoid unexpected disconnection of the power cord.
a. If the cable clamp is positioned too near the power cord that it blocks the power cord plug connection, press down the tab on the cable mount and slide the clip backward.
Figure 51 Sliding the cable clamp backward
4. Open the cable clamp, place the power cord through the opening in the cable clamp, and then close the cable clamp, as shown in Figure 52.
Figure 52 Securing power cords
5. Slide the cable clamp forward until it is flush against the edge of the power cord plug, as shown in Figure 53.
Figure 53 Sliding the cable clamp forward
Securing cables
After completing all wiring, secure the cables using either of the following methods.
Securing cables to cable management brackets
For information about how to install cable management brackets, see the installation guide shipped with the brackets.
Securing cables to slide rails by using cable straps
You can secure cables to either left slide rails or right slide rails. As a best practice for cable management, secure cables to left slide rails.
When multiple cable straps are used in the same rack, stagger the strap location, so that the straps are adjacent to each other when viewed from top to bottom. This positioning will enable the slide rails to slide easily in and out of the rack.
1. Hold the cables against a slide rail.
2. Wrap the strap around the slide rail and loop the end of the cable strap through the buckle. Dress the cable strap to ensure that the extra length and buckle part of the strap are facing outside of the slide rail.
Figure 54 Securing cables to a slide rail
Cabling guidelines
· For server heat dissipation, make sure no cables block the inlet or outlet air vents of power supplies, fans, heatsinks, and GPUs.
· To easily identify ports and connect/disconnect cables, make sure the cables do not cross.
· Label the cables for easy identification of the cables.
· Wrap unused cables onto an appropriate position on the rack.
· To avoid electric shock, fire, or damage to the equipment, do not connect communication equipment to RJ-45 Ethernet ports on the server.
· To avoid damage to cables when extending the server out of the rack, do not route the cables too tight if you use cable management brackets.
Removing the server from a rack
This section describes the methods for removing the server.
1. Power off the server.
2. Disconnect all peripheral cables from the server.
3. Extend the server from the rack.
4. Open the latches of the mounting brackets. Loosen the captive screws inside the mounting brackets, and slide the server out of the rack.
Figure 55 Extending the server from the rack
5. Place the server on a clean, stable surface.
Powering on and powering off the server
This section introduces methods to power on and off a server.
|
IMPORTANT: If the server is connected to external storage devices, make sure the server is the first device to power off and then the last device to power on. This restriction prevents the server from mistakenly identifying the external storage devices as faulty devices. |
Powering on the server
This section introduces methods to power on a server.
About this task
· Power on the server after the server is installed.
· Power on the server again after the server is maintained.
Prerequisites
· Install the server and internal components correctly.
· Connect the server to a power source.
· To power on the server immediately after shutdown, as a best practice for all internal components to function properly, wait a minimum of 30 seconds until the HDD is completely still and electronic components are fully powered down.
Procedure
Powering on the server by pressing the power on/standby button
Press the power on/standby button to power on the server.
The server exits standby mode and supplies power to the system. The system power LED changes from steady amber to flashing green and then to steady green.
Powering on the server from the HDM Web interface
1. Log in to HDM. For more information, see the HDM3 user guide for the server.
2. Navigate to System > Power Management.
3. Click Power on.
Powering on the server from the remote console interface
1. Log in to HDM. For more information, see the HDM3 user guide for the server.
2. Log in to a remote console and then power on the server. For more information, see HDM3 user guide.
Configuring automatic power-on
You can configure automatic power-on from HDM or the BIOS.
· To configure automatic power-on from HDM:
Log in to HDM. For more information, see the HDM3 user guide for the server.
Navigate to System > Power Management, and then click System Power Restore.
Select Always power on as the power-on policy and then click OK.
· To configure automatic power-on from the BIOS:
a. Log in to the BIOS. For information about how to log in to the BIOS, see the BIOS user guide for the server.
b. Select Server > AC Restore Settings, and then press Enter.
c. Select Always Power On, and then press Enter.
d. Press F4 to save the configuration.
Powering off the server
This section introduces methods to power off a server.
About this task
· Server maintenance is required.
· The server needs to be moved.
Prerequisites
· Back up all critical data.
· Make sure all services have stopped or have been migrated to other servers.
Procedure
Powering off the server from its operating system
1. Connect a monitor, mouse, and keyboard to the server.
2. Shut down the operating system of the server.
3. Disconnect all power cords from the server.
Powering off the server by pressing the power on/standby button
· Powering off the server by pressing the power on button
a. Press the power on/standby button and wait for the system power LED to turn into steady amber.
b. Disconnect all power cords from the server.
· Powering off the server forcedly by pressing the power on button
c. Press and hold the power on/standby button until the system power LED turns into steady amber.
|
IMPORTANT: This method forces the server to enter standby mode without properly exiting applications and the operating system. Use this method only when the server system crashes. For example, a process gets stuck. |
d. Disconnect all power cords from the server.
Powering off the server from the HDM Web interface
1. Log in to HDM. For more information, see the HDM3 user guide for the server.
2. Navigate to System > Power Management.
3. Click Graceful power-off.
4. Disconnect all power cords from the server.
Powering off the server from the remote console interface
1. Log in to HDM. For more information, see the HDM3 user guide for the server.
2. Disconnect all power cords from the server.
Configuring the server
This section describes the process of software configuration for the server after server installation.
Configuration flowchart
Figure 56 Server configuration flowchart
Powering on the switch
1. Power on the server. For more information, see "Powering on and powering off the server."
2. Verify that the health LED on the front panel is steady green, which indicates that the system is operating correctly. For more information about the health LED status, see "LEDs and buttons."
Configuring basic BIOS settings
|
|
NOTE: The BIOS setup utility screens are subject to change without notice. |
You can set the server boot order and the BIOS passwords from the BIOS setup utility of the server.
Setting the server boot order
The server has a default boot order. You can change the server boot order from the BIOS. For the default boot order and the procedure of changing the server boot order, see the BIOS user guide for the server.
Setting the BIOS passwords
BIOS passwords include a boot password as well as an administrator password and a user password for the BIOS setup utility. By default, no passwords are set.
To prevent unauthorized access and changes to the BIOS settings, set both the administrator and user passwords for accessing the BIOS setup utility. Make sure the two passwords are different.
After setting the administrator password and user password for the BIOS setup utility, you must enter the administrator password or user password each time you access the system.
· To obtain administrator privileges, enter the administrator password.
· To obtain the user privileges, enter the user password.
For the difference between the administrator and user privileges and guidelines for setting the BIOS passwords, see the BIOS user guide for the server.
For more information about setting the BIOS passwords, see the BIOS user guide for the server.
Configuring the RAID
The supported RAID levels and RAID configuration methods vary by storage controller model. For more information, see the storage controller user guide for the server.
Installing the operating system and drivers
This section introduces how to install the operating system and drivers.
Install an operating system
For the server compatibility with the operating systems, visit the component compatibility lookup tool at http://www.h3c.com/en/home/qr/default.htm?id=66.
Install a compatible operating system on the server by following the procedures described in the operating system installation guide for the server.
Installing hardware drivers
|
IMPORTANT: To avoid hardware unavailability caused by an update failure, always back up the drivers before you update them. |
For newly installed hardware to operate correctly, the operating system must have the required hardware drivers.
To install a hardware driver, see the operating system installation guide for the server.
Updating firmware
|
IMPORTANT: Verify the hardware and software compatibility before firmware upgrade. For information about the hardware and software compatibility, see the software release notes. |
You can update the following firmware from UniSystem or HDM: For more information, see the firmware update guide for the server.
· HDM
· BIOS
· CPLD
· BPCPLD
· PSU
· BMCCPLD
Replacing hardware options
|
|
NOTE: · If you are replacing multiple hardware options, read their replacement procedures and identify similar steps to streamline the entire replacement procedure. · This document contains procedures for replacing and installing hardware options. If the replacement and installation operations are similar, only replacement procedures are illustrated. If you refer to a replacement procedure to install a hardware option, remove the corresponding blank in advance.
|
Replacing hardware options
· Processor
· Liquid-cooled module
· DIMM
· System board
· IO board
· SAS/SATA drive
· NVMe drive
· Drive backplane
· Rear drive cage
· Riser card and PCIe module
· Storage controller and power fail safeguard module
· Network adapter
· M.2 SSD and M.2 SSD expander module
· GPU module
· LCD smart management module
· Air baffle
· Fan module
· Mounting bracket
· Encryption module
· Power supply
· NVMe VROC module
· System battery
· Blank
Adding a processor
For information about how to add a processor, see Processor Installation Quick Start.
Replacing a processor
This section introduces how to replace a processor.
Application scenarios
· The processor is faulty.
· Replace the processor with one of another model.
· The processor hinders the maintenance of other components.
Prerequisites
· Take the following ESD prevention measures:
¡ Wear antistatic clothing.
¡ Wear an ESD wrist strap and make sure it makes good skin contact and is reliably grounded.
¡ Do not wear any conductive objects, such as jewelry or watches.
¡ When you replace a component, examine the slot and connector for damages. Make sure the pins are not damaged (bent for example) and do not contain any foreign objects.
· For information about processor installation guidelines, see "Processors."
Procedure
|
CAUTION: · To avoid damage to a processor or the system board, only H3C authorized or professional server engineers can install, replace, or remove a processor. · Make sure the processors installed on the same server are the same model. · The pins in the processor sockets are very fragile and prone to damage. Install a protective cover if a processor socket is empty. · For the server to operate correctly, make sure processor 1 is always in position. For more information about processor locations, see "System board layout." · The compatible heatsinks might vary by processor model, but the processor replacement methods are similar. · To prevent ESD, wear an ESD wrist strap and make sure it makes good skin contact and is reliably grounded. |
Removing a processor
1. Power off the server. For more information, see "Powering off the server."
2. Remove the server from the rack. For more information, see "Removing the server from a rack."
3. Remove the access panel:
a. Press the button on the locking lever and then lift the locking lever.
The access panel automatically slides to the server rear.
b. Lift the access panel to remove it from the server.
4. Remove the chassis air baffle. Lift the air baffle out of the chassis.
5. Remove the heatsink that carries the processor.
a. Loosen the four captive screws on the heatsink in sequence.
b. Open heatsink clips at the four corners.
c. Lift the heatsink to remove it.
|
CAUTION: The pins in the processor socket are extremely fragile and easily damaged. To avoid replacing the system board due to pin damage, do not touch the pins. |
6. Remove a processor.
a. Lift the locking lever to release the processor.
b. Hold the processor by both sides to remove it from the retaining bracket.
7. Remove the retaining bracket.
a. Release the four corner clips of the retaining bracket. You must press one clip and its cater-cornered clip outward, and press the other two clips inward.
b. Lift the retaining bracket to remove it from the heatsink.
8. Clear the residual thermal grease. Use isopropanol wiping cloth to clear the residual thermal grease from the processor top and heatsink. Make sure the processor and the heatsink are clean.
Installing a processor
1. Install the retaining bracket onto the heatsink.
a. Close the locking lever on the retaining bracket.
|
CAUTION: For the processor to be installed in place, make sure the locking lever on the retaining bracket is closed. |
b. Align the alignment triangle on the retaining bracket with the cut-off corner of the heatsink. Place the bracket on top of the heatsink, with the four corners of the bracket clicked into the four corners of the heatsink.
2. Smear thermal grease onto the heatsink. Use the thermal grease injector to inject 0.6 ml of thermal grease to the five dots on the heatsink, 0.12 ml for each dot.
|
CAUTION: Make sure no thermal grease remains on the heatsink top. |
3. Install the processor onto the retaining bracket.
|
CAUTION: Be careful not to touch the contacts on the bottom surface of the processor. Hold the processor by its edges. |
a. Tilt the processor, align the small triangle on the processor with the alignment triangle in the retaining bracket, and insert one end of the processor into the retaining bracket. Place two thumbs against the heatsink, press the other end of the processor, and lower the processor.
b. Open the clips on the retaining bracket until the processor fits snugly onto the retaining bracket.
4. Install the heatsink with the processor and the retaining bracket to the server.
|
CAUTION: Paste bar code label supplied with the processor over the original label on the heatsink. This step is required for you to obtain H3C's processor servicing. |
a. Align the alignment triangle on the retaining bracket with the cut-off corner of the processor socket and the pin holes in the heatsink with the guide pins on the socket. Lower down the heatsink on the processor socket.
b. Press down the heatsink clips at the four corners to lock the heatsink in place.
c. Use a T30 Torx screwdriver to fasten the four captive screws on the heatsink.
|
CAUTION: Tighten the screws with a torque of 0.9 N·m (8 in-lbs). Failure to do so might result in poor contact with the processor or damage to the pins in the processor socket. |
5. Install the chassis air baffle.
6. Install the access panel. Place the access panel onto the server.
7. Slide the access panel to the server front and close the locking lever. The access panel snaps into place.
8. Rack-mount the server. For more information, see "Installing or removing the server."
9. Connect the power cord.
10. Power on the server. For more information, see "Powering on the server."
Verifying the installation
Log in to HDM to verify that the processor is operating correctly. For more information, see HDM3 online help.
Replacing a liquid-cooled module
This section describes how to replace a liquid-cooled module.
Application scenarios
· The liquid-cooled module is faulty.
· A liquid-cooled module hinders the maintenance of other components.
Prerequisites
· Take the following ESD prevention measures:
¡ Wear antistatic clothing.
¡ Wear an ESD wrist strap and make sure it makes good skin contact and is reliably grounded.
¡ Do not wear any conductive objects, such as jewelry or watches.
¡ When you replace a component, examine the slot and connector for damages. Make sure the pins are not damaged (bent for example) and do not contain any foreign objects.
· For information about processor installation guidelines, see "Processors."
Procedure
|
CAUTION: · To avoid damage to a processor or the system board, only H3C authorized or professional server engineers can install, replace, or remove a liquid-cooled module. · The pins in the processor sockets are very fragile and prone to damage. Install a protective cover if a processor socket is empty. · To prevent ESD, wear an ESD wrist strap and make sure it makes good skin contact and is reliably grounded. |
Removing a liquid-cooled module
1. Power off the server. For more information, see "Powering off the server."
2. Remove the server from the rack. For more information, see "Removing the server from a rack."
3. Remove the access panel:
a. Press the button on the locking lever and then lift the locking lever.
The access panel automatically slides to the server rear.
b. Lift the access panel to remove it from the server.
4. Remove the chassis air baffle. Lift the air baffle out of the chassis.
5. Remove the liquid leakage sensor cable.
6. Remove the liquid-cooled module with the processor.
a. Loosen the eight captive screws on the liquid-cooled module.
b. Unlock the eight retaining clips to release the liquid-cooled module.
c. Loosen the matching screws between the inlet and outlet of the liquid cooling module and the rear of the chassis.
d. Lift the liquid-cooled module to remove it from the server.
|
CAUTION: The pins in the processor socket are extremely fragile and easily damaged. To avoid replacing the system board due to pin damage, do not touch the pins. |
7. Remove the two processors one by one.
a. Lift the locking lever to release the processor.
b. Hold the processor by both sides to remove it from the retaining bracket.
8. Remove the retaining bracket.
a. Release the four corner clips of the retaining bracket.
b. Lift the retaining packet to remove it from the liquid-cooled module.
9. Clear the residual thermal grease. Use isopropyl alcohol wiping cloth to clean the top of the processor and the surface of the liquid-cooled module. Make sure the surface is clean and tidy.
Installing a liquid-cooled module
1. Take the liquid-cooled module out of the antistatic bag. Attach the retaining bracket onto the liquid-cooled module.
a. Close the locking lever on the retaining bracket.
|
CAUTION: For the processor to be installed in place, make sure the locking lever on the retaining bracket is closed. |
b. Align the alignment triangle of the retaining bracket with the cut-off corner of the liquid-cooled module and press the retaining bracket onto the liquid-cooled module until the four corners of the retaining bracket click into the four corners of the liquid-cooled module.
2. (Optional.) Smear thermal grease onto the liquid-cooled module. Use the thermal grease injector to inject 0.6 ml of thermal grease to the five dots on the liquid cooling module, 0.12 ml for each dot.
|
CAUTION: A new liquid-cooled module comes with thermal grease. If the thermal grease does not function well for cooling, clean and reapply thermal grease on the module. |
3. Install the two processors onto the retaining bracket one by one.
|
CAUTION: Be careful not to touch the contacts on the bottom surface of the processor. Hold the processor by its edges. |
a. Tilt the processor, align the small triangle on the processor with the alignment triangle in the retaining bracket, and insert one edge of the processor into the retaining bracket. Place two thumbs against the liquid-cooled module, press the other end of the processor, and place down the processor.
b. Open the clips on the retaining bracket until the processor fits snugly onto the retaining bracket.
c. Repeat the same procedure to install the other processors onto the retaining bracket.
4. Install the liquid-cooled module with the processors and retaining brackets onto the server.
|
CAUTION: Paste bar code label supplied with the processor over the original label on the liquid-cooled module. This step is required for you to obtain H3C's processor servicing. |
a. Align the alignment triangle on the retaining bracket with the cut-off corner of the processor socket and the screw holes in the liquid-cooled module with the guide pins on the processor socket. Place the heatsink onto the processor socket.
b. Lock each of the eight retaining clips to secure the liquid-cooled module in place.
c. Install the matching screws between the inlet and outlet of the liquid cooling module and the rear of the chassis.
d. Use a T30 Torx screwdriver to fasten the eight captive screws on the liquid-cooled module.
|
CAUTION: As a best practice, use a torque of 0.9 N•m (8 in-lbs) to avoid poor contact of the processor or damage to the pins on the processor socket. |
5. Connect the liquid leakage detection sensor cable.
6. Install the chassis air baffle.
7. Install the access panel. Place the access panel onto the server. Slide the access panel to the server front until it snaps into place.
8. Rack-mount the server. For more information, see "Installing or removing the server."
9. Connect the power cord.
10. Power on the server. For more information, see "Powering on the server."
Verifying the installation
Log in to the HDM Web interface to view whether the processor operates correctly after replacement of the liquid-cooled module. For more information, see HDM3 online help.
Replacing a DIMM
This section introduces how to replace a DIMM.
Application scenarios
· A DIMM is faulty.
· Replace a DIMM with one of another model.
· A DIMM hinders the maintenance of other components.
Prerequisites
· Take the following ESD prevention measures:
¡ Wear antistatic clothing.
¡ Wear an ESD wrist strap and make sure it makes good skin contact and is reliably grounded.
¡ Do not wear any conductive objects, such as jewelry or watches.
¡ When you replace a component, examine the slot and connector for damages. Make sure the pins are not damaged (bent for example) and do not contain any foreign objects.
· For more information about DIMM installation guidelines, see "Memory."
Procedure
Removing a DIMM
1. Power off the server. For more information, see "Powering off the server."
2. Remove the server from the rack. For more information, see "Removing the server from a rack."
3. Remove the access panel:
a. Press the button on the locking lever and then lift the locking lever.
The access panel automatically slides to the server rear.
b. Lift the access panel to remove it from the server.
4. Lift the air baffle out of the chassis.
5. Remove a DIMM. Open the DIMM slot latches and pull the DIMM out of the slot to remove the DIMM.
|
CAUTION: To avoid damage to DIMMs or the system board, make sure the server has been powered off and disconnected from the power cord for at least 20 seconds. |
Installing a DIMM
1. Install a DIMM. Align the notch on the DIMM with the connector key in the DIMM slot and press the DIMM into the socket until the latches lock the DIMM in place. Make sure the latches lock the DIMM in place.
2. Install the chassis air baffle.
3. Install the access panel.
a. Place the access panel onto the server.
b. Slide the access panel to the server front and close the locking lever. The access panel snaps into place.
4. Rack-mount the server. For more information, see "Installing or removing the server."
5. Power on the server. For more information, see "Powering on the server."
6. (Optional.) To modify the memory mode, enter the BIOS and configure the memory mode as described in the BIOS user manual for the server.
Verifying the installation
Use one of the following methods to verify that the DIMM is installed correctly:
· Using the operating system:
¡ In Windows, select Run in the Start menu, enter msinfo32, and verify the memory capacity of the DIMM.
¡ In Linux, execute the cat /proc/meminfo command to verify the memory capacity.
· Using HDM:
Log in to HDM and verify the memory capacity of the DIMM. For more information, see HDM3 online help.
· Using the BIOS:
Access the BIOS, select Advanced > Socket Configuration > Memory Configuration > Memory Topology, and press Enter. Then, verify the memory capacity of the DIMM.
If the memory capacity displayed is inconsistent with the actual capacity, remove and then reinstall the DIMM, or replace the DIMM with a new DIMM. If the DIMM is in Mirror mode, it is normal that the displayed capacity is smaller than the actual capacity.
Replacing the system board
This section introduces how to replace a system board.
Application scenarios
The system board is faulty.
Prerequisites
· Take the following ESD prevention measures:
¡ Wear antistatic clothing.
¡ Wear an ESD wrist strap and make sure it makes good skin contact and is reliably grounded.
¡ Do not wear any conductive objects, such as jewelry or watches.
¡ When you replace a component, examine the slot and connector for damages. Make sure the pins are not damaged (bent for example) and do not contain any foreign objects.
· When you replace a component, examine the slot and connector for damages. Make sure the pins are not damaged (bent for example) and do not contain any foreign objects.
Procedure
Removing the system board
|
CAUTION: To prevent electrostatic discharge, place the removed parts on an antistatic surface or in antistatic bags. |
1. Power off the server. For more information, see "Powering off the server."
2. Remove the server from the rack. For more information, see "Removing the server from a rack."
3. Remove the OCP network adapter.
4. Remove the power supplies.
5. Remove the access panel:
a. Press the button on the locking lever and then lift the locking lever.
The access panel automatically slides to the server rear.
b. Lift the access panel to remove it from the server.
6. Remove the chassis air baffle. Lift the air baffle out of the chassis.
7. Disconnect all cables on the system board.
8. Remove all components on the system board, such as riser cards, DIMMs, and the processor.
9. Install the processor cover.
10. Remove the system board.
a. Loosen the captive screws on the system board.
b. Hold the system board handle and slide the system board toward the server front to disengage the system board from the server management module. Then, lift the system board out of the chassis.
Installing the system board
1. Install the system board.
a. Slowly place the system board in the chassis. Then, hold the system board handle and slide the system board toward the server rear until the server management module connectors are successfully attached to the system board.
|
|
NOTE: The system board is securely seated if you cannot use the system board handle to lift the system board. |
b. Fasten the captive screws on the system board.
2. Connect cables to the system board.
3. Remove the processor socket covers. Hold the socket cover and remove it.
4. Install all the removed components onto the system board, such as riser cards, DIMMs, and the processor.
5. Install the fan cage. Place the fan cage into the chassis and close the locking levers on both sides of the fan cage.
6. Install the air baffle.
7. Install fan modules.
8. Install the access panel.
9. Installing the OCP network adapter
10. Install the removed power supplies.
11. Rack-mount the server. For more information, see "Installing or removing the server."
12. Connect the power cord.
13. Power on the server. For more information, see "Powering on the server."
Replacing the server management module
This section describes how to replace the server management module.
Application scenarios
The server management module is faulty.
Prerequisites
· Take the following ESD prevention measures:
¡ Wear antistatic clothing.
¡ Wear an ESD wrist strap and make sure it makes good skin contact and is reliably grounded.
¡ Do not wear any conductive objects, such as jewelry or watches.
· When you replace a component, examine the slot and connector for damages. Make sure the pins are not damaged (bent for example) and do not contain any foreign objects.
Procedure
Removing the server management module
1. Power off the server. For more information, see "Powering off the server."
2. Remove the server from the rack. For more information, see "Removing the server from a rack."
3. Remove the access panel:
a. Press the button on the locking lever and then lift the locking lever.
The access panel automatically slides to the server rear.
b. Lift the access panel to remove it from the server.
4. Remove the server management module. Lift the server management module upward until it disengages from the latches.
Installing the server management module
1. Install the server management module. Insert the server management module onto the BMC connector on the system board.
2. Install the access panel.
3. Rack-mount the server. For more information, see "Installing or removing the server."
4. Connect the power cord.
5. Power on the server. For more information, see "Powering on the server."
Replacing the IO board
This section introduces how to replace an IO board.
Application scenarios
The IO board is faulty.
Prerequisites
· Take the following ESD prevention measures:
¡ Wear antistatic clothing.
¡ Wear an ESD wrist strap and make sure it makes good skin contact and is reliably grounded.
¡ Do not wear any conductive objects, such as jewelry or watches.
· When you replace a component, examine the slot and connector for damages. Make sure the pins are not damaged (bent for example) and do not contain any foreign objects.
· For more information about the IO installation guidelines, see "IO board."
Procedure
Removing the I/O board
1. Power off the server. For more information, see "Powering off the server."
2. Remove the server from the rack. For more information, see "Removing the server from a rack."
3. Remove the access panel:
a. Press the button on the locking lever and then lift the locking lever.
The access panel automatically slides to the server rear.
b. Lift the access panel to remove it from the server.
4. Disconnect all the cables on the IO board.
5. Use a screwdriver to remove the screws that secure the I/O board, and then remove the I/O board.
Installing the IO board
1. Place the I/O board vertically in the specified position at the rear of the chassis, and use a screwdriver to install the screws that secure the I/O board.
2. Connect all cables to the IO board.
3. Install the IO board. Slowly place the IO board into the chassis. Then, slide the management module toward the server rear until the connectors on the module are securely seated.
4. Connect all cables to the IO board.
5. Install the access panel.
6. Rack-mount the server. For more information, see "Installing or removing the server."
7. Connect the power cord.
8. Power on the server. For more information, see "Powering on the server."
Replacing a SAS/SATA drive
This section describes how to replace a drive.
Application scenarios
· A drive is faulty.
· Replace a drive whose space is all used.
· Replace a drive with one of another model.
Prerequisites
· Take the following ESD prevention measures:
¡ Wear antistatic clothing.
¡ Wear an ESD wrist strap and make sure it makes good skin contact and is reliably grounded.
¡ Do not wear any conductive objects, such as jewelry or watches.
· When you replace a component, examine the slot and connector for damages. Make sure the pins are not damaged (bent for example) and do not contain any foreign objects.
· Identify the installation location of the drive to be replaced in the server.
· Identify the RAID array information of the drive to be replaced. To replace a drive in a non-redundancy RAID array, back up data in the RAID array if the old drive is full or the new drive is of a different model.
· For more information about drive installation guidelines, see "SAS/SATA drives."
Procedure
|
|
NOTE: · SAS/SATA drives attached to a storage controller supports hot swapping after you enter the BIOS or OS. · SATA drives attached to the embedded VROC controller supports hot swapping after you enter the OS. |
Removing a SAS/SATA drive
1. Observe the drive LEDs to verify that the drive is not selected by the storage controller and is not performing a RAID migration or rebuilding. For more information about the LEDs, see "Drive LEDs."
2. Remove the drive. Press the button on the drive panel to release the locking lever and pull the drive out of the slot. For an HDD, pull the drive 3 cm (1.18 in) out of the slot. Wait for a minimum of 30 seconds for the drive to stop rotating, and then pull the drive out of the slot.
3. Remove the drive carrier. Remove the screws that secure the drive and then remove the drive from the carrier.
Installing a SAS/SATA drive
|
IMPORTANT: As a best practice, install drives that do not contain RAID information. |
1. Attach the drive to the drive carrier. Place the drive in the carrier and then use four screws to secure the drive into place.
2. Install the drive. Insert the drive into the slot and push it gently until you cannot push it further, and then close the locking lever.
3. If the installed drive contains RAID information, clear the information before configuring RAIDs.
4. To configure RAID settings after the drive is replaced, see the storage controller user guide for the server.
Verifying the installation
Use one of the following methods to verify that the drive has been replaced correctly:
· Log in to HDM, and verify the drive properties, including capacity. For more information, see HDM3.
· Observe the drive LEDs to verify that the drive is operating correctly. For more information about the LEDs, see "Drive LEDs."
· Access the BIOS to verify the drive properties, including capacity. The verification method in the BIOS varies by RAID configuration method. For more information, see the storage controller user guide for the server.
· Access the CLI or GUI of the server to verify the drive properties, including capacity.
Adding an NVMe drive
This section describes how to add an NVMe drive.
Prerequisites
· Take the following ESD prevention measures:
¡ Wear antistatic clothing.
¡ Wear an ESD wrist strap and make sure it makes good skin contact and is reliably grounded.
¡ Do not wear any conductive objects, such as jewelry or watches.
· When you replace a component, examine the slot and connector for damages. Make sure the pins are not damaged (bent for example) and do not contain any foreign objects.
· Identify the installation location of the drive to be replaced in the server.
· Identify the RAID array information of the drive to be replaced. To replace a drive in a non-redundancy RAID array, back up data in the RAID array if the old drive is full or the new drive is of a different model.
· For more information about drive installation guidelines, see "NVMe drives."
Verifying the installation
Use the following methods to verify that the drive is installed correctly:
· Log in to HDM, and verify the drive properties, including capacity. For more information, see HDM3.
· Observe the drive LEDs to verify that the drive is operating correctly. For more information about the LEDs, see "Drive LEDs."
· Access the BIOS to verify the drive properties, including capacity. For more information, see the BIOS user guide.
Access the CLI or GUI of the server to verify the drive properties, including capacity.
Replacing an NVMe drive
This section describes how to replace an NVMe drive.
Application scenarios
· A drive is faulty.
· Replace a drive whose space is all used.
· Replace a drive with one of another model.
Prerequisites
· Take the following ESD prevention measures:
¡ Wear antistatic clothing.
¡ Wear an ESD wrist strap and make sure it makes good skin contact and is reliably grounded.
¡ Do not wear any conductive objects, such as jewelry or watches.
· When you replace a component, examine the slot and connector for damages. Make sure the pins are not damaged (bent for example) and do not contain any foreign objects.
· Identify the installation location of the drive to be replaced in the server.
· Identify the RAID array information of the drive to be replaced. To replace a drive in a non-redundancy RAID array, back up data in the RAID array if the old drive is full or the new drive is of a different model.
· For more information about drive installation guidelines, see "NVMe drives."
Procedure
|
|
NOTE: · Only some operating systems support the hot swapping of NVMe drives. For more information, use components OS compatibility lookup tools. · If the operating system does not support hot swapping of NVMe drives, power off the server first. For more information, see "Powering off the server." |
Removing an NVMe drive
1. Use the component compatibility lookup tool at http://www.h3c.com/en/home/qr/default.htm?id=66 to verify if the operating system supports hot removal or managed hot removal of NVMe drives.
¡ If neither hot removal nor managed hot removal is supported, power off the server first. For more information, see "Powering off the server."
¡ If hot removal or managed hot removal of NVMe drives is supported, see the online replacement guide for NVMe drives.
Installing an NVMe drive
1. Verify if the drive was removed through hot removal or managed hot removal.
¡ If yes, see the online replacement guide for NVMe drives.
¡ If no, proceed to the next step.
3. Install an NVMe drive. Push the drive into the drive slot and close the locking lever on the drive panel.
Verifying the installation
Use the following methods to verify that the drive is installed correctly:
· Log in to HDM, and verify the drive properties, including capacity. For more information, see HDM3.
· Observe the drive LEDs to verify that the drive is operating correctly. For more information about the LEDs, see "Drive LEDs."
· Access the BIOS to verify the drive properties, including capacity. For more information, see the BIOS user guide.
· Access the CLI or GUI of the server to verify the drive properties, including capacity.
Replacing a drive backplane
This section introduces how to replace a drive backplane.
Application scenarios
A drive backplane is faulty.
Prerequisites
· Take the following ESD prevention measures:
¡ Wear antistatic clothing.
¡ Wear an ESD wrist strap and make sure it makes good skin contact and is reliably grounded.
¡ Do not wear any conductive objects, such as jewelry or watches.
· When you replace a component, examine the slot and connector for damages. Make sure the pins are not damaged (bent for example) and do not contain any foreign objects.
Procedure
Removing a drive backplane
1. Power off the server. For more information, see "Powering off the server."
2. Remove the server from the rack. For more information, see "Removing the server from a rack."
3. Remove the drives attached to the backplane.
4. Remove the access panel:
a. Press the button on the locking lever and then lift the locking lever.
The access panel automatically slides to the server rear.
b. Lift the access panel to remove it from the server.
5. Disconnect cables from the backplane.
6. Remove the drive backplane. Loosen the captive screws that secure the backplane, and then lift the backplane out of the chassis.
Installing a drive backplane
1. Install a drive backplane. Place the backplane in the slot and then fasten the captive screws.
2. Connect cables to the drive backplane.
3. Install the access panel.
a. Place the access panel onto the server.
b. Slide the access panel to the server front and close the locking lever. The access panel snaps into place.
4. Install the removed drives.
5. Rack-mount the server. For more information, see "Installing or removing the server."
6. Power on the server. For more information, see "Powering on the server."
Installing a rear drive cage
Prerequisites
· Take the following ESD prevention measures:
¡ Wear antistatic clothing.
¡ Wear an ESD wrist strap and make sure it makes good skin contact and is reliably grounded.
¡ Do not wear any conductive objects, such as jewelry or watches.
· When you replace a component, examine the slot and connector for damages. Make sure the pins are not damaged (bent for example) and do not contain any foreign objects.
Procedure
1. Power off the server. For more information, see "Powering off the server."
2. Remove the server from the rack. For more information, see "Removing the server from a rack."
3. Remove the access panel:
a. Press the button on the locking lever and then lift the locking lever.
The access panel automatically slides to the server rear.
b. Lift the access panel to remove it from the server.
4. Remove the PCIe riser card blank. Lift the blank to remove it from the chassis.
5. For a 2SFF UniBay drive cage, install a bracket:
a. Align the guide pin on the bracket with the notch in the chassis.
b. Place the bracket in the chassis.
c. Use screws to secure the bracket.
6. Install the rear drive cage:
a. Place the drive cage in the chassis.
b. Use screws to secure the drive cage.
c. Connect cables to the drive cage.
7. Install the blank. Aligning the guide pins on the blank with the notches in the chassis, insert the blank into the slot.
8. Install the access panel.
a. Place the access panel onto the server.
b. Slide the access panel to the server front and close the locking lever. The access panel snaps into place.
9. Rack-mount the server. For more information, see "Installing or removing the server."
10. Connect the power cord.
11. Power on the server. For more information, see "Powering on the server."
Replacing riser cards and PCIe modules
This section describes how to replace riser cards and PCIe modules.
Application scenarios
· A riser card is faulty.
· A PCIe module is faulty.
· Replace a PCIe module with one of another model.
Prerequisites
· Take the following ESD prevention measures:
¡ Wear antistatic clothing.
¡ Wear an ESD wrist strap and make sure it makes good skin contact and is reliably grounded.
¡ Do not wear any conductive objects, such as jewelry or watches.
· When you replace a component, examine the slot and connector for damages. Make sure the pins are not damaged (bent for example) and do not contain any foreign objects.
· For information about installation guidelines for riser cards and PCIe modules, see "Riser cards and PCIe modules."
Procedure
Removing a riser card and a PCIe module
1. Power off the server. For more information, see "Powering off the server."
2. Remove the server from the rack. For more information, see "Removing the server from a rack."
3. Remove the access panel:
a. Press the button on the locking lever and then lift the locking lever.
The access panel automatically slides to the server rear.
b. Lift the access panel to remove it from the server.
4. Disconnect all cables that hinder the replacement, if any.
6. Remove the PCIe module from the riser card. Open the retaining latch for the PCIe slot, and pull the PCIe module out of the slot.
Installing a riser card and a PCIe module
1. Install the PCIe module on the riser card:
a. Remove the PCIe module blank. Open the retaining latch for the PCIe slot, and then pull out the blank.
b. Install the PCIe module to the riser card. Insert the PCIe module into the PCIe slot, and close the retaining latch.
2. Install the riser card to the server:
a. Remove the riser card blank from the target PCIe riser connector. Lift the blank to remove it from the chassis.
b. Install the riser card onto the PCIe riser connector. Aligning the two standouts on the riser card with the two notches on the chassis and pressing the unlocking button on the riser card, insert the riser card along the PCIe riser slot until it snaps into place.
3. Connect cables to the riser card or PCIe modules, if any.
4. Install the access panel.
a. Place the access panel onto the server.
b. Slide the access panel to the server front and close the locking lever. The access panel snaps into place.
5. Rack-mount the server. For more information, see "Installing or removing the server."
6. Power on the server. For more information, see "Powering on the server."
Replacing a storage controller and a power fail safeguard module
This section describes how to replace a storage controller.
Application scenarios
· A storage controller is faulty.
· Replace a storage Controller with one of another model.
· A storage controller hinders the maintenance of other components.
· A fail safeguard module is faulty.
· A storage controller hinders the maintenance of other components.
Prerequisites
· Take the following ESD prevention measures:
¡ Wear antistatic clothing.
¡ Wear an ESD wrist strap and make sure it makes good skin contact and is reliably grounded.
¡ Do not wear any conductive objects, such as jewelry or watches.
· When you replace a component, examine the slot and connector for damages. Make sure the pins are not damaged (bent for example) and do not contain any foreign objects.
· To replace the storage controller with a controller of the same model, make sure the following configurations remain the same after replacement:
¡ Storage controller location and cabling method.
¡ Storage controller model, operating mode, and firmware version.
· To replace the storage controller with a controller of a different model, back up data in the drives of the storage controller and clear RAID configuration.
· For information about installation guidelines for storage controllers and power fail safeguard modules, see "Storage controllers and power fail safeguard modules."
Procedure
Removing a standard storage controller and a power fail safeguard module
1. Power off the server. For more information, see "Powering off the server."
2. Remove the server from the rack. For more information, see "Removing the server from a rack."
3. Remove the access panel:
a. Press the button on the locking lever and then lift the locking lever.
The access panel automatically slides to the server rear.
b. Lift the access panel to remove it from the server.
4. Disconnect all cables from the standard storage controller.
5. Remove the standard storage controller:
a. Remove the riser card where the standard storage controller resides. Pressing the unlocking button on the riser card, lift the riser card out of the chassis.
b. Remove the standard storage controller from the riser card. Open the retaining latch for the PCIe slot, and pull the PCIe module out of the slot.
6. Remove the power fail safeguard module or supercapacitor, if any.
a. Remove the flash card. Remove the screw that secures the flash card and pull the flash card out of the slot.
b. Remove the supercapacitor. Open the clip on the supercapacitor holder, and take the supercapacitor out of the holder.
c. Remove the supercapacitor holder. Open the retaining clip at the bottom of the holder, and pull the holder out.
Installing a standard storage controller and a power fail safeguard module
1. (Optional.) Install the supercapacitor holder and the supercapacitor.
a. Install the supercapacitor holder. Place the supercapacitor holder in the chassis and slide it to the server rear until it snaps into place.
b. Connect one end of a supercapacitor extension cable to the supercapacitor.
c. Install the supercapacitor to the supercapacitor holder. Tilt the supercapacitor and insert one end of the supercapacitor into the holder. Pull the clip on the holder and insert the other end into the holder, and then release the clip.
2. Install the removed flash card.
a. Install the internal threaded studs supplied with the power fail safeguard module on the storage controller.
b. Install the flash card on the standard storage controller. Align the two screw holes in the flash card with the two internal threaded studs on the storage controller, insert the flash card connector into the socket, and then use screws to secure the flash card.
3. Install the standard storage controller to the riser card. Insert the standard storage controller into the PCIe slot, and then close the retaining latch on the riser card.
4. Install the riser card to the server.
5. Connect the data cables between the standard storage controller and the drive backplane.
6. (Optional.) If the power fail safeguard module is installed, connect the supercapacitor extension cable to the flash card. For more information, see "Connecting the supercapacitor cable."
7. Install the access panel. Place the access panel onto the server.
a. Slide the access panel to the server front and close the locking lever.
b. The access panel snaps into place.
8. Rack-mount the server. For more information, see "Installing or removing the server."
9. Connect the power cord.
10. Power on the server. For more information, see "Powering on the server."
Cautions after installation
¡ If you replace a storage controller with a controller of a different model, configure RAID settings for the drives managed by the new controller. For more information, see the storage controller user guide for the server.
¡ To replace the storage controller with a controller of the same model, make sure the following configurations remain the same after replacement:
- Storage controller operating mode.
- Storage controller firmware version.
- BIOS boot mode.
- First boot option in Legacy mode.
For more information, see the storage controller user guide for the server and the BIOS user guide for the server.
Replacing a standard PCIe network adapter
This section describes how to replace a standard PCIe network adapter.
Application scenarios
· A standard PCIe network adapter is faulty.
· Replace a standard PCIe network adapter with one of another model.
Prerequisites
· Take the following ESD prevention measures:
¡ Wear antistatic clothing.
¡ Wear an ESD wrist strap and make sure it makes good skin contact and is reliably grounded.
¡ Do not wear any conductive objects, such as jewelry or watches.
· When you replace a component, examine the slot and connector for damages. Make sure the pins are not damaged (bent for example) and do not contain any foreign objects.
· Read the network adapter installation guidelines in "Network adapters."
Procedure
Removing a standard PCIe network adapter
1. Power off the server. For more information, see "Powering off the server."
2. Disconnect cables from the standard PCIe network adapter.
3. Remove the server from the rack. For more information, see "Removing the server from a rack."
4. Remove the access panel:
a. Press the button on the locking lever and then lift the locking lever.
The access panel automatically slides to the server rear.
b. Lift the access panel to remove it from the server.
5. Disconnect all cables that hinder the replacement, if any.
6. Remove the riser card that holds the PCIe network adapter. Pressing the unlocking button on the riser card, lift the riser card out of the chassis.
7. Remove the PCIe network adapter from the riser card. Loosen the captive screws on the riser card and pull the PCIe network adapter out of the slot.
Installing a standard PCIe network adapter
For more information, see "Installing a riser card and a PCIe module."
Replacing the OCP network adapter
This section introduces how to replace an OCP network adapter.
Application scenarios
· An OCP network adapter is faulty.
· Replace an OCP network adapter with one of another model..
Prerequisites
· Take the following ESD prevention measures:
¡ Wear antistatic clothing.
¡ Wear an ESD wrist strap and make sure it makes good skin contact and is reliably grounded.
¡ Do not wear any conductive objects, such as jewelry or watches.
· When you replace a component, examine the slot and connector for damages. Make sure the pins are not damaged (bent for example) and do not contain any foreign objects.
· For more information about the installation guidelines for network adapter, see "Network adapters."
Procedure
Removing an OCP network adapter
1. Power off the server. For more information, see "Powering off the server."
2. Disconnect cables from the OCP network adapter.
3. Remove the OCP network adapter. Loosen the captive screws on the OCP network adapter and pull the OCP network adapter out of the slot.
Installing an OCP network adapter
1. Installing the OCP network adapter Insert the OCP network adapter into the slot and fasten the captive screws on it.
2. Connect cables to the OCP network adapter.
3. Power on the server. For more information, see "Powering on the server."
4. (Optional.) Configure a network port on the OCP network adapter as an HDM shared network port. OCP network adapters inserted into OCP adapter slots support NCSI. By default, port 1 on an OCP network adapter acts as the HDM shared network port. You can specify another port on the OCP network adapter as the HDM shared network port from the HDM Web interface. Note that you can specify only one port as the HDM shared network port at the same time.
Replacing a SATA M.2 SSD and the front M.2 SSD expander module
This section describes how to replace a SATA M.2 SSD and the front SATA M.2 SSD expander module.
Application scenarios
· A SATA M.2 SSD expander module is faulty.
· A SATA M.2 SSD expander module is faulty.
· Replace a SATA M.2 SSD with one of another model.
Prerequisites
· Take the following ESD prevention measures:
¡ Wear antistatic clothing.
¡ Wear an ESD wrist strap and make sure it makes good skin contact and is reliably grounded.
¡ Do not wear any conductive objects, such as jewelry or watches.
· When you replace a component, examine the slot and connector for damages. Make sure the pins are not damaged (bent for example) and do not contain any foreign objects.
· For information about installation guidelines for SATA M.2 SSDs, see "_Ref9601490."
Procedure
Removing a SATA M.2 SSD and the M.2 SSD expander module
1. Power off the server. For more information, see "Powering off the server."
2. Remove the server from the rack. For more information, see "Removing the server from a rack."
3. Remove the access panel:
a. Press the button on the locking lever and then lift the locking lever.
The access panel automatically slides to the server rear.
b. Lift the access panel to remove it from the server.
4. Remove the SATA M.2 SSD expander module that holds the SATA M.2 SSD.
a. Disconnect the cable from the SATA M.2 SSD expander module.
b. Remove the expander module. Remove the screws that secure the expander module and then pull the expander module out.
5. Remove a SATA M.2 SSD. Slide the locking tab, lift the SATA M.2 SSD, and then pull the SATA M.2 SSD out of the slot.
Installing a SATA M.2 SSD and the M.2 SSD expander module
1. Install the SATA M.2 SSD to the SATA M.2 SSD expander module. Insert the connector of the SATA M.2 SSD into the socket, slide the locking tab, press the SATA M.2 SSD to secure the SATA M.2 SSD into place, and then release the locking tab.
2. Install the expander module with the SATA M.2 SSD into the server.
a. Align the two screw holes in the expander module with the two internal threaded studs in the chassis, put the expander module into the chassis, and then use screws to secure the expander module.
b. Connect the SATA M.2 SSD cable. For more information, see "Connecting cables for the front M.2 SSD expander module."
3. Install the access panel.
a. Place the access panel onto the server.
b. Slide the access panel to the server front and close the locking lever. The access panel snaps into place.
4. Rack-mount the server. For more information, see "Installing or removing the server."
5. Power on the server. For more information, see "Powering on the server."
Replacing a GPU module
This section describes how to replace a GPU.
Application scenarios
· A GPU is faulty.
· Replace a GPU with one of another model.
· A GPU hinders the maintenance of other components.
Prerequisites
· Take the following ESD prevention measures:
¡ Wear antistatic clothing.
¡ Wear an ESD wrist strap and make sure it makes good skin contact and is reliably grounded.
¡ Do not wear any conductive objects, such as jewelry or watches.
· When you replace a component, examine the slot and connector for damages. Make sure the pins are not damaged (bent for example) and do not contain any foreign objects.
· For information about GPU installation guidelines, see "GPU module."
Procedure
Removing a GPU module
1. Power off the server. For more information, see "Powering off the server."
2. Remove the server from the rack. For more information, see "Removing the server from a rack."
3. Remove the access panel:
a. Press the button on the locking lever and then lift the locking lever.
The access panel automatically slides to the server rear.
b. Lift the access panel to remove it from the server.
4. Disconnect all cables that hinder the replacement, if any.
5. Remove the riser card where the GPU module resides. Pressing the unlocking button on the riser card, lift the riser card out of the chassis.
6. Remove the GPU module from the riser card.
7. (Optional.) Disconnect the cable from the GPU, if any.
8. Open the retaining latch on the riser card, and pull the GPU module out from the slot.
9. (Optional.) Replacing a air baffle
10. Replace the air baffle with one suitable for the form of the GPU.
Installing a GPU module
1. Install a GPU module on the riser card.
2. Insert the GPU module into the PCIe slot, and then close the retaining latch on the riser card.
3. (Optional.) Connect the power cord to the power connector on the GPU module according to the cable label.
4. Reconnect other cables to the riser card if required.
5. Install the riser card to the server. Install the riser card with PCIe modules to the riser bay.
6. Install the riser card with PCIe modules to the riser bay.
7. (Optional.) According to the label of the power cord, connect the other end of the power cord to the power connector on the riser card.
8. Install the access panel.
a. Place the access panel onto the server.
b. Slide the access panel to the server front and close the locking lever. The access panel snaps into place.
9. Rack-mount the server. For more information, see "Installing or removing the server."
10. Connect the power cord.
11. Power on the server. For more information, see "Powering on the server."
Installing the LCD smart management module
This section describes how to install the LCD smart management module.
Prerequisites
· Take the following ESD prevention measures:
¡ Wear antistatic clothing.
¡ Wear an ESD wrist strap and make sure it makes good skin contact and is reliably grounded.
¡ Do not wear any conductive objects, such as jewelry or watches.
· When you install a component, examine the slot and connector for damages. Make sure the pins are not damaged (bent for example) and do not contain any foreign objects.
Procedure
1. Power off the server. For more information, see "Powering off the server."
2. Remove the server from the rack. For more information, see "Removing the server from a rack."
3. Remove the access panel:
a. Press the button on the locking lever and then lift the locking lever.
The access panel automatically slides to the server rear.
b. Lift the access panel to remove it from the server.
4. Remove the fan cage. Pull up the ejector levers at both sides of the fan cage and lift the fan cage to remove it from the chassis.
5. Remove the drive or drive blank from the LCD module slot.
6. Install the LCD smart management module:
a. Connect one end of the LCD module cable to the LCD smart management module.
b. Push the LCD smart management module into the slot until you cannot push it any further.
c. Connect the other end of the cable to the LCD smart management module connector on the system board.
7. Install the fan cage. Place the fan cage into the chassis and close the locking levers on both sides of the fan cage.
8. Install the access panel.
a. Place the access panel onto the server.
b. Slide the access panel to the server front and close the locking lever. The access panel snaps into place.
9. Rack-mount the server. For more information, see "Installing or removing the server."
10. Connect the power cord. For more information, see "Connecting power cords."
11. Power on the server. For more information, see "Powering on the server."
Replacing the LCD smart management module
This section describes how to replace the LCD smart management module.
Application scenarios
· The LCD smart management module is faulty.
· The LCD smart management module hinders the maintenance of other components.
Prerequisites
· Take the following ESD prevention measures:
¡ Wear antistatic clothing.
¡ Wear an ESD wrist strap and make sure it makes good skin contact and is reliably grounded.
¡ Do not wear any conductive objects, such as jewelry or watches.
· When you replace a component, examine the slot and connector for damages. Make sure the pins are not damaged (bent for example) and do not contain any foreign objects.
Procedure
Removing the LCD smart management module
1. Power off the server. For more information, see "Powering off the server."
2. Remove the server from the rack. For more information, see "Removing the server from a rack."
3. Remove the access panel:
a. Press the button on the locking lever and then lift the locking lever.
The access panel automatically slides to the server rear.
b. Lift the access panel to remove it from the server.
4. Remove the fan cage. Pull up the ejector levers at both sides of the fan cage and lift the fan cage to remove it from the chassis.
5. Remove the LCD smart management module.
a. Disconnect the LCD module cable from the system board.
b. Use a flat-head screwdriver or tweezers to press the clip of the LCD smart management module and pull the module out of the slot.
Installing the LCD smart management module
1. Install the LCD smart management module:
a. Connect one end of the LCD module cable to the LCD smart management module.
b. Push the LCD smart management module into the slot until you cannot push it any further.
c. Connect the other end of the cable to the LCD smart management module connector on the system board.
2. Install the fan cage. Place the fan cage into the chassis and close the locking levers on both sides of the fan cage.
3. Install the access panel.
a. Place the access panel onto the server.
b. Slide the access panel to the server front and close the locking lever. The access panel snaps into place.
4. Install the security bezel, if any. Press the latch at the other end, close the security bezel, and then release the latch to secure the security bezel into place. Insert the key provided with the bezel into the lock on the bezel and lock the security bezel.
5. Rack-mount the server. For more information, see "Installing or removing the server."
6. Connect the power cord.
7. Power on the server. For more information, see "Powering on the server."
Replacing the air baffle
This section describes how to replace an air baffle.
Application scenarios
Air baffle is faulty.
Replacing the air baffle
Removing the air baffle
1. Power off the server. For more information, see "Powering off the server."
2. Remove the server from the rack. For more information, see "Removing the server from a rack."
3. Remove the access panel:
a. Press the button on the locking lever and then lift the locking lever.
The access panel automatically slides to the server rear.
b. Lift the access panel to remove it from the server.
4. Remove the air baffle.
a. (Optional.) Disconnect the supercapacitor extension cable. If the supercapacitor is installed on the air baffle, first remove the supercapacitor.
b. Press the tabs on the air baffle, and then lift the air baffle out of the chassis.
Installing the air baffle
1. Install the chassis air baffle.
a. Place the air baffle vertically into the chassis.
b. (Optional.) Install the removed supercapacitor.
2. Install the access panel.
a. Place the access panel onto the server.
b. Slide the access panel to the server front and close the locking lever. The access panel snaps into place.
3. Rack-mount the server. For more information, see "Installing or removing the server."
4. Connect the power cord.
5. Power on the server. For more information, see "Powering on the server."
Replacing a fan
This section describes how to replace a fan.
Application scenarios
· A fan is faulty.
· Replace a fan with one of another model.
Procedure
Removing a fan
1. The fan modules are hot swappable. If sufficient space is available for replacement, you can replace a fan module without removing the server from the rack.
2. Power off the server. For more information, see "Powering off the server."
3. Remove the server from the rack. For more information, see "Removing the server from a rack."
a. Press the button on the locking lever and then lift the locking lever.
The access panel automatically slides to the server rear.
b. Lift the access panel to remove it from the server.
5. Remove the fan. Pressing the buttons on the fan, pull the fan out of its slot.
Installing a fan
1. Install the fan. Insert the fan into the slot and press the fan gently until it is secured in position.
2. Install the access panel.
a. Place the access panel onto the server.
b. Slide the access panel to the server front and close the locking lever. The access panel snaps into place.
3. If the server was removed, rack-mount the server. For more information, see "Installing or removing the server."
4. Connect the power cord if the power cord has been disconnected. For more information, see "Connecting power cords."
5. Power on the server if the server has been powered off. For more information, see "Powering on the server."
Replacing a mounting bracket
This section describes how to replace a mounting bracket.
Application scenarios
· A mounting bracket is faulty.
· Replace a mounting bracket if it fails or any of the components (for example, I/O components or VGA/USB connectors) fails.
Procedure
|
|
NOTE: The procedure is the same for the left and right mounting brackets. This section uses the left mounting bracket as an example. |
Removing a mounting bracket
1. Power off the server. For more information, see "Powering off the server."
2. Remove the server from the rack, if the space over the server is insufficient. For more information, see "Removing the server from a rack."
3. Remove the access panel:
a. Press the button on the locking lever and then lift the locking lever.
The access panel automatically slides to the server rear.
b. Lift the access panel to remove it from the server.
4. Remove the air baffle. Lift the air baffle out of the chassis.
5. Remove the front I/O component cable assembly.
a. Disconnect the front I/O component cable assembly from the system board.
b. Remove the cable protection plate. Remove the captive screws that secure the cable protection plate, press the cable protection plate and slide it toward the rear of the chassis until you cannot slide it further, and then pull out the cable protection plate.
c. Remove the front I/O component cable assembly.
6. Remove the mounting bracket. Remove the screws that secure the left mounting bracket, and then pull the mounting bracket until it is removed.
Installing a mounting bracket
1. Install a mounting bracket. Attach the mounting bracket to the corresponding side of the server, and use screws to secure the mounting bracket into place.
2. Install the front I/O component cable assembly:
b. Install the cable protection plate on the chassis. Insert the cable protection plate along the slot and slide it toward the front of the chassis until you cannot slide it further, and then install the captive screws on the cable protection plate.
c. Connect the front I/O component cable assembly to the front I/O connector on the system board.
3. Install the air baffle.
4. Install the access panel:
a. Place the access panel onto the server.
b. Slide the access panel to the server front.
c. Press down the wrench on the access panel until it snaps into place.
5. Rack-mount the server. For more information, see "Installing or removing the server."
6. Power on the server. For more information, see "Powering on the server."
Installing and setting up a TCM or TPM
The encryption module includes TPM and TCM. This section describes how to install and enable the TCM or TPM.
About TPM and TCM
· Trusted platform module (TPM) is a microchip embedded in the system board. It stores encryption information (such as encryption keys) for authenticating server hardware and software. The TPM operates with drive encryption programs such as Microsoft Windows BitLocker to provide operating system security and data protection. For information about Microsoft Windows BitLocker, visit the Microsoft website at http://www.microsoft.com/.
· TCM is a trusted computing platform-based hardware module with protected storage space, which enables the platform to implement password calculation.
TPM/TCM installation and setup flowchart
Figure 57 TPM/TCM Installation and setup flowchart
Prerequisites
· Take the following ESD prevention measures:
¡ Wear antistatic clothing.
¡ Wear an ESD wrist strap and make sure it makes good skin contact and is reliably grounded.
¡ Do not wear any conductive objects, such as jewelry or watches.
· When you replace a component, examine the slot and connector for damages. Make sure the pins are not damaged (bent for example) and do not contain any foreign objects.
Installing the TPM or TCM module
1. Power off the server. For more information, see "Powering off the server."
2. Remove the server from the rack. For more information, see "Removing the server from a rack."
3. Remove the access panel:
a. Press the button on the locking lever and then lift the locking lever.
The access panel automatically slides to the server rear.
b. Lift the access panel to remove it from the server.
4. Install the TCM or TPM.
a. Press the TPM into the TPM connector on the system board.
b. Insert the rivet pin.
c. Insert the security rivet into the hole in the rivet pin and press the security rivet until it is firmly seated.
5. Install the removed riser cards, if any.
6. Install the access panel.
a. Place the access panel onto the server.
b. Slide the access panel to the server front and close the locking lever. The access panel snaps into place.
7. Rack-mount the server. For more information, see "Installing or removing the server."
8. Power on the server. For more information, see "Powering on the server."
Cautions after installation
· Do not remove an installed TPM or TCM. Once installed, the TPM or TCM module becomes a permanent part of the system board.
· If you want to replace the failed TCM or TPM, remove the system board, and then contact H3C Support to replace the TCM or TPM and the system board.
· When installing or replacing hardware, H3C technicians cannot configure the TCM or TPM or enter the recovery key. For security reasons, only the user can perform the tasks.
· When replacing the system board, do not remove the TPM or TCM from the system board. H3C will provide a TCM or TPM with a spare system board for the replacement.
· Any attempt to remove an installed TPM or TCM from the system board breaks or disfigures the TPM or TCM security rivet. Upon locating a broken or disfigured rivet on an installed TCP or TPM, administrators should consider the system compromised and take appropriate measures to ensure the integrity of the system data.
· H3C is not liable for blocked data access caused by improper use of the TCM or TPM. For more information, see the encryption technology feature documentation provided by the operating system.
Configuring encryption in the operating system
For more information about this task, see the encryption technology feature documentation that came with the operating system.
For more information about Microsoft Windows BitLocker drive encryption, visit the Microsoft website at http://technet.microsoft.com/en-us/library/cc732774.aspx. The recovery key/password is generated during BitLocker setup, and can be saved and printed after BitLocker is enabled. When using BitLocker, always retain the recovery key/password. The recovery key/password is required to enter Recovery Mode after BitLocker detects a possible compromise of system integrity or firmware or hardware change. For security purposes, follow these guidelines when retaining the recovery key/password:
· Always store the recovery key/password in multiple locations.
· Always store copies of the recovery key/password away from the server.
· Do not save the recovery key/password on the encrypted hard drive.
Replacing a power supply
This section describes how to replace a power supply.
Application scenarios
· The power supply is faulty.
· Replace a power supply with one of another model.
Prerequisites
· Take the following ESD prevention measures:
¡ Wear antistatic clothing.
¡ Wear an ESD wrist strap and make sure it makes good skin contact and is reliably grounded.
¡ Do not wear any conductive objects, such as jewelry or watches.
· When you replace a component, examine the slot and connector for damages. Make sure the pins are not damaged (bent for example) and do not contain any foreign objects.
· For information about installation guidelines for power supplies, see "Power supply module."
Procedure
Removing a power supply
1. Power off the server. For more information, see "Powering off the server."
2. Remove the server from the rack. For more information, see "Removing the server from a rack."
3. Remove the power cord from the power supply:
a. Press the tab to disengage the ratchet from the tie mount, slide the cable clamp outward, and then release the tab.
b. Open the cable clamp and remove the power cord out of the clamp.
c. Unplug the power cord.
4. Remove the power supply. Press the retaining latches with your thumb and hold the power supply by its handle to release the power supply. Then, pull the power supply slowly out of the slot.
Installing a power supply
|
|
NOTE: If only one power supply is present, install the new power supply in the slot for the replaced power supply. |
1. Install a new power supply. Push the power supply into the slot until it snaps into place.
2. Rack-mount the server. For more information, see "Installing or removing the server."
3. Power on the server. For more information, see "Powering on the server."
Replace an NVMe VROC module
Application scenarios
· The NVMe VROC module is faulty.
· Replace the NVMe VROC module with one of another model.
Prerequisites
· Take the following ESD prevention measures:
¡ Wear antistatic clothing.
¡ Wear an ESD wrist strap and make sure it makes good skin contact and is reliably grounded.
¡ Do not wear any conductive objects, such as jewelry or watches.
· When you replace a component, examine the slot and connector for damages. Make sure the pins are not damaged (bent for example) and do not contain any foreign objects.
Procedure
Removing the NVMe VROC module
1. Power off the server. For more information, see "Powering off the server."
2. Remove the server from the rack, if the space over the server is insufficient. For more information, see "Removing the server from a rack."
3. Remove the access panel:
a. Press the button on the locking lever and then lift the locking lever.
The access panel automatically slides to the server rear.
b. Lift the access panel to remove it from the server.
4. Remove the NVMe VROC module. Hold the ring part of the NVMe VROC module and pull the module out.
Installing the NVMe VROC module
1. Install the NVMe VROC module. Insert the NVMe VROC module onto the NVMe VROC module connector on the system board.
2. Install the access panel.
a. Place the access panel onto the server.
b. Slide the access panel to the server front and close the locking lever. The access panel snaps into place.
3. Rack-mount the server. For more information, see "Installing or removing the server."
4. Power on the server. For more information, see "Powering on the server."
5. Configure RAID for NVMe drives. For more information, see the BIOS user guide for the server.
Replacing the system battery
This section describes how to replace the system battery.
Application scenarios
The server comes with a system battery (BR2032) installed on the system board, which supplies power to the real-time clock and has a lifespan of 3 to 5 years. It supplies power to the real-time clock and has a lifespan of 3 to 5 years.
Replace the system battery when the following situations occur: The product only supports the BR2032 system battery.
· The battery is faulty.
· The server no longer automatically displays the correct date and time.
|
|
NOTE: Battery failure or complete power depletion will cause the BIOS to reset to default settings. To reconfigure the BIOS, see the BIOS user guide for the server. |
Prerequisites
· Take the following ESD prevention measures:
¡ Wear antistatic clothing.
¡ Wear an ESD wrist strap and make sure it makes good skin contact and is reliably grounded.
¡ Do not wear any conductive objects, such as jewelry or watches.
· When you replace a component, examine the slot and connector for damages. Make sure the pins are not damaged (bent for example) and do not contain any foreign objects.
Procedure
Removing the system battery
1. Power off the server. For more information, see "Powering off the server."
2. Remove the server from the rack. For more information, see "Removing the server from a rack."
3. Remove the access panel:
a. Press the button on the locking lever and then lift the locking lever.
The access panel automatically slides to the server rear.
b. Lift the access panel to remove it from the server.
4. Remove the system battery. Pinch the system battery by its top edge and the battery will disengage from the battery holder.
|
|
NOTE: For environment protection purposes, dispose of the used-up system battery at a designated site. |
Installing the system battery
1. Install the system battery.
a. Insert the system battery with the plus sign "+" facing up into the system battery holder, and press down the battery to secure it into place.
b. Press down the battery to secure it into place.
2. Install the access panel.
a. Place the access panel onto the server.
b. Slide the access panel to the server front.
c. Press down the locking lever on the access panel until it snaps into place.
3. Rack-mount the server. For more information, see "Installing or removing the server."
4. Power on the server. For more information, see "Powering on the server."
5. Access the BIOS to reconfigure the system date and time. For more information, see the BIOS user guide for the server.
Installing or removing blanks
About this task
Install blanks over the empty slots if the following modules are not present and remove blanks before you install the following modules:
· Drive
· Drive backplane
· Power supply
· Riser card
· PCIe module
· OCP network adapter
Prerequisites
Take the following ESD prevention measures:
· Wear antistatic clothing.
· Wear an ESD wrist strap and make sure it makes good skin contact and is reliably grounded.
· Do not wear any conductive objects, such as jewelry or watches.
Procedure
Table 52 Procedures for installing and removing the blanks for various slots
|
Item |
Procedure |
|
Remove a drive blank. |
Press the latches on the drive blank inward with one hand, and pull the drive blank out of the slot. |
|
Install a drive blank. |
Insert the drive blank into the slot. |
|
Remove a drive backplane blank. |
From the inside of the chassis, use a flat-head screwdriver to push the blank outward to disengage the blank. Then, pull the blank out of the server. |
|
Install a drive backplane blank. |
Insert the drive backplane blank into the slot and push the blank until you hear a click. |
|
Remove a power supply blank. |
Hold and pull the power supply blank out of the slot. |
|
Install a power supply blank. |
Insert the power supply blank into the slot with the TOP mark facing up. |
|
Remove a riser card blank. |
Lift the riser card blank to remove it from the connector. |
|
Install a riser card blank. |
Insert the riser card blank into the slot along guide rails. |
|
Remove a PCIe module blank. |
Pull the blank out of the slot. |
|
Install a PCIe module blank. |
Insert the blank into the slot along the guide rails. |
|
Remove an OCP network adapter blank. |
Insert the screwdriver into the protruding hole on the blank, and then pull the blank outward. |
|
Install an OCP network adapter blank. |
Insert the blank into the slot horizontally. |
Connecting internal cables
This section introduces how to connect internal cables for various components in a server.
|
|
NOTE: Compared with the signal cables and power cords, more data cables (including SAS/SATA and NVMe data cables) are required and the cabling methods are more complicated. This section provides code information for data cables and you can use the information to identify cables and their connection methods. |
Internal cabling guidelines
Application scenarios
Connect internal cables in the following scenarios:
· Connect cables after component installation or replacement.
· Secure cables when the cables are loosen or disconnected.
· Replace cables when the cable protective sleeves are damaged or the cables are faulty.
Restrictions and guidelines
Follow these guidelines when connecting the internal cables:
· Do not route the cables above the removable components, such as DIMMs.
· Route the internal cables without hindering installation or removal of other components or hindering other internal components.
· Route the cables neat and tidy in their own fixed spaces. Make sure the cables will not be squeezed or scratched by other internal components.
· Do not pull the connectors when routing the cables.
· Do not use a cable tie to bundle an excessive number of cables.
· Appropriately bind long cables. Coil and use cable ties to secure unused cables.
· When you connect a drive cable, make sure the cable clicks into place.
· Remove the cap (if any) from the target cable connector before connecting a cable to it.
· If you cannot identify the cables by labels provided with the cables, apply new labels to cables for easy identification.
Connecting drive cables
Drive cables include SAS/SATA data cables, NVMe data cables, power cords, and signal cables. The server supports multiple drive configurations. This section takes the following eight typical drive configurations as examples to help users understand the cabling schemes for drives. For cabling schemes for other drive configurations, contact Technical Support.
8LFF Unibay drives at the front
1. Connect data cables for the front 8LFF Unibay SAS/SATA drives as shown in Figure 58.
Figure 58 Connecting data cables for the front 8LFF Unibay SAS/SATA drives
|
No. |
Cable type |
Cable code |
Cable description |
|
1 |
SAS/SATA data cable |
0404A1KT |
Connect the front 8LFF Unibay drive backplane (SAS PORT1) to the storage controller (C0) |
|
|
NOTE: To use this cabling scheme, you can install the storage controller in slots 1 to 8. In this example, the storage controller is installed in slot 1. |
2. Connect data cables for the front 8LFF Unibay NVMe drives as shown in Figure 59.
Figure 59 Connecting data cables for the front 8LFF Unibay NVMe drives
|
No. |
Cable type |
Cable code |
Cable description |
|
1 |
NVMe data cable |
0404A1XC |
Connect the front 8LFF Unibay drive backplane (NVME B3/B4) to the system board (C1-P4A) |
|
2 |
NVMe data cable |
0404A1PY |
Connect the front 8LFF Unibay drive backplane (NVME B1/B2) to the system board (C1-P4C) |
|
3 |
NVMe data cable |
0404A1Q2 |
Connect the front 8LFF Unibay drive backplane (NVME A3/A4) to the system board (C2-P4A) |
|
4 |
NVMe data cable |
0404A1WS |
Connect the front 8LFF Unibay drive backplane (NVME A1/A2) to the system board (C2-P4C) |
3. Connect power cords and signal cables for the front 8LFF UniBay drives as shown in Figure 60.
Figure 60 Connecting front 8LFFUnibay signal cable and power cord
|
No. |
Cable type |
Cable code |
Cable description |
|
1 |
Signal cable |
0404A1X6 |
Connect the front 8LFF Unibay drive backplane (PWR) to the system board (PWR1) |
|
2 |
Power cord |
0404A1X9 |
Connect the front 8LFF drive backplane (AUX) to the system board (AUX1) |
12LFF drives at the front + rear 4SFF UniBay drives at the rear
1. Connect data cables for the front 12LFF SAS/SATA drives as shown in Figure 61.
Figure 61 Connecting data cables for the front 12LFF SAS/SATA drives
|
No. |
Cable type |
Cable code |
Cable description |
|
1 |
SAS/SATA data cable |
0404A1QM |
Connect the front 12LFF drive backplane (SAS PORT1) to the system board (C0) |
|
|
NOTE: To use this cabling scheme, you can install the storage controller in slots 1 to 6. In this example, the storage controller is installed in slot 1. |
2. Connect data cables for the front 12LFF NVMe drives as shown in Figure 62.
Figure 62 Connecting data cables for the front 12LFF NVMe drives
|
No. |
Cable type |
Cable code |
Cable description |
|
1 |
NVMe data cable |
0404A1XC |
Connect the front 12LFF Unibay drive backplane (NVME A1/A2) to the system board (C1-P4A) |
|
2 |
NVMe data cable |
0404A209 |
Connect the front 12LFF Unibay drive backplane (A3/A4) to the system board (C1-P4C) |
|
3 |
NVMe data cable |
0404A1Q2 |
Connect the front 12LFF Unibay drive backplane (NVME B1/B2) to the system board (C1-P5C) |
|
4 |
NVMe data cable |
0404A1WT |
Connect the front 12LFF Unibay drive backplane (NVME B3/B4) to the system board (C2-P4A) |
|
5 |
NVMe data cable |
0404A209 |
Connect the front 12LFF Unibay drive backplane (C1/C2) to the system board (C2-P4C) |
|
6 |
NVMe data cable |
0404A1WS |
Connect the front 12LFF Unibay drive backplane (NVME C3/C4) to the system board (C2-P5C) |
3. Connect the front 12LFF power cords and signal cable, as shown in Figure 63.
Figure 63 Connecting the front 12LFF power cords and signal cable
|
No. |
Cable type |
Cable code |
Cable description |
|
1 |
Power cord |
0404A1UY |
Connect the front drive backplane (PWR2) to the system board (PWR2) |
|
2 |
Signal cable |
0404A1XA |
Connect the front 12LFF drive backplane (AUX) to the system board (BP AUX1) |
|
3 |
Power cord |
0404A1WW |
Connect the front drive backplane (PWR1) to the system board (PWR1) |
4. Connect the rear 4SFF power cord, signal cable, and NVMe data cables as shown in Figure 64.
Figure 64 Connecting the rear 4SFF power cord, signal cable, and NVMe data cables
|
No. |
Cable type |
Cable code |
Cable description |
|
1 |
NVMe data cable |
0404A1QH |
Connect the rear 4SFF Unibay drive backplane (NVME B3/B4) to the system board (C2-P1C) |
|
2 |
NVMe data cable |
0404A1X1 |
Connect the rear 4SFF Unibay drive backplane (NVME B1/B2) to the system board (C1-P1C) |
|
3 |
Signal cable |
0404A1QW |
Connect the rear 4SFF Unibay drive backplane (AUX) to the system board (BP AUX4) |
|
4 |
Power cord |
0404A1X6 |
Connect the rear 4SFF Unibay drive backplane (PWR1) to the system board (PWR4) |
8SFF Unibay at the front +8SFF Unibay + 8SFF Unibay drives
1. Connect data cables for the front 8SFF Unibay +8SFF Unibay + 8SFF Unibay SAS/SATA drives as shown in Figure 61.
Figure 65 Connecting data cables for the front 8SFF SAS/SATA drives
|
No. |
Cable type |
Cable code |
Cable description |
|
1 |
SAS/SATA data cable |
0404A1QM |
Connect the front 8SFF Unibay drive backplane (SAS PORT1) to the storage controller (C0) |
|
|
NOTE: To use this cabling scheme, you can install the storage controller in slots 4 to 10. In this example, the storage controller is installed in slot 4. |
2. Connect data cables for the front 8SFF Unibay + 8SFF Unibay + 8SFF Unibay NVMe drives as shown in Figure 66.
Figure 66 Connecting data cables for the front 8SFF SAS/SATA drives
|
No. |
Cable type |
Cable code |
Cable description |
|
1 |
NVMe data cable |
0404A2AX |
Connect the 8SFF drive backplane (NVME A1/ A2) for front drive bay 3 to the system board (C1-P3A) |
|
2 |
NVMe data cable |
0404A1WS |
Connect the 8SFF drive backplane (NVME A3/ A4) for front drive bay 3 to the system board (C1-P4A) |
|
3 |
NVMe data cable |
0404A2BG |
Connect the 8SFF drive backplane (NVME B1/ B2) for front drive bay 3 to the system board (C2-P1A) |
|
4 |
NVMe data cable |
0404A1PU |
Connect the 8SFF drive backplane (NVME B3/ B4) for front drive bay 3 to the system board (C2-P4A) |
|
5 |
NVMe data cable |
0404A2BG |
Connect the 8SFF drive backplane (NVME A1/ A2) for front drive bay 2 to the system board (C1-P3C) |
|
6 |
NVMe data cable |
0404A1WT |
Connect the 8SFF drive backplane (NVME A3/ A4) for front drive bay 2 to the system board (C1-P5A) |
|
7 |
NVMe data cable |
0404A1PW |
Connect the 8SFF drive backplane (NVME B1/ B2) for front drive bay 2 to the system board (C2-P1C) |
|
8 |
NVMe data cable |
0404A1PY |
Connect the 8SFF drive backplane (NVME B3/ B4) for front drive bay 2 to the system board (C2-P5A) |
|
9 |
NVMe data cable |
0404A1XK |
Connect the 8SFF drive backplane (NVME A1/ A2) for front drive bay 1 to the system board (C1-P4C) |
|
10 |
NVMe data cable |
0404A1PU |
Connect the 8SFF drive backplane (NVME A3/ A4) for front drive bay 1 to the system board (C1-P5C) |
|
11 |
NVMe data cable |
0404A1PY |
Connect the 8SFF drive backplane (NVME B1/ B2) for front drive bay 1 to the system board (C2-P4C) |
|
12 |
NVMe data cable |
0404A1WS |
Connect the 8SFF drive backplane (NVME B3/ B4) for front drive bay 1 to the system board (C2-P5C) |
3. Connect power cords and signal cables for the front 8SFF, as shown in Figure 67.
Figure 67 Connecting power cords and signal cables for the front 8SFF
|
No. |
Cable type |
Cable code |
Cable description |
|
1 |
Signal cable |
0404A1WV |
Connect the 8SFF drive backplane (AUX1) for front drive bay 3 to the system board (AUX3) |
|
2 |
Power cord |
0404A1QV |
Connect the 8SFF drive backplane (PWR1) for front drive bay 3 to the system board (PWR3) |
|
3 |
Signal cable |
0404A2BD |
Connect the 8SFF drive backplane (AUX1) for front drive bay 2 to the system board (AUX2) |
|
4 |
Power cord |
0404A1UY |
Connect the 8SFF drive backplane (PWR1) for front drive bay 2 to the system board (PWR2) |
|
5 |
Signal cable |
0404A1WV |
Connect the 8SFF drive backplane (AUX1) for front drive bay 1 to the system board (AUX1) |
|
6 |
Power cord |
0404A1WW |
Connect the 8SFF drive backplane (PWR1) for front drive bay 1 to the system board (PWR1) |
Connecting cables for the OCP adapter
Connect cables for OCP adapters as shown in Figure 68.
Figure 68 Connecting cables for the OCP network adapter
|
No. |
Cable code |
Cable description |
|
1 |
0404A2RJ |
Connect OCP adapter (M3) to the system board (C1-P3A) |
|
2 |
Connect OCP adapter (M4) to the system board (C1-P3C) |
|
|
3 |
Connect OCP adapter (M2) to the system board (OCP PWR) |
|
|
4 |
Connect OCP adapter (M1) to the system board (OCP AUX) |
Connecting the supercapacitor cable
Figure 69 Connecting the supercapacitor cable
Connecting cables for the front M.2 SSD expander module
Figure 70 Connecting cables for the front M.2 SSD expander module
|
No. |
Cable type |
Cable code |
Cable description |
|
1 |
SAS/SATA data cables, NVMe data cables, power cords, and signal cables |
0404A1PW |
Connect the front M.2 expander module (M.2 PORT) to the system board (M.2 PORT) |
Connecting cables for riser cards
Some riser cards support connecting to the system board to provide extra PCIe links for slots on the riser cards. This section describes the cabling methods for these riser cards. For more information about PCIe slots, see "Riser cards and PCIe modules."
RC-3FHFL-2U-G7
The RC-3FHFL-2U-G7 riser card can connect to PCIe riser connector 1 or 2 on the system board. The connectors the cables connect to different depending on the PCIe riser connector, as shown in Table 53. For more information about the connector locations, see system board components in "Rear panel components." The cable connection method is similar regardless of which riser connector is used. This section uses riser connector 1 as an example.
Figure 71 Connecting PCIe signal cables for the riser card
|
Riser connector |
Cable code |
Cable number |
Connector/mark on the riser card |
Connector on the system board/IO board |
|
Riser1 |
0404A2RF |
1 |
AUX |
RISER AUX1 |
|
0404A2RK |
2 |
PWR |
HPWR3 |
|
|
0404A2DS |
3 |
PCIe Slot1-M1 |
MCIO connector C1-P3C |
|
|
PCIe Slot1-M2 |
MCIO connector C1-P3A |
|||
|
0404A2DS |
4 |
PCIe Slot3-M1 |
MCIO connector C1-P0A |
|
|
PCIe Slot3-M2 |
MCIO connector C1-P0C |
|||
|
0404A2DS |
5 |
PCIe Slot2-M1 |
MCIO connector C1-P2A |
|
|
PCIe Slot2-M2 |
MCIO connector C1-P2C |
RC-2FHFL-2U-LC-R2-G7
The RC-2FHFL-2U-LC-R2-G7 riser card can connect to PCIe riser connector 2 on the system board. The RC-2FHFL-2U-LC-R2-G7 riser card is for liquid-cooled models only. The connectors the cables connect to different depending on the PCIe riser connector, as shown in Table 54. For more information about the riser connectors, see "Rear panel components." Connect cables for riser card as shown in Figure 72.
Figure 72 Connecting PCIe signal cables for the riser card
|
Riser connector |
Cable code |
Cable number |
Connector/mark on the riser card |
Connector on the system board/IO board |
|
Riser2 |
0404A2RF |
1 |
AUX |
RISER AUX2 |
|
0404A223 |
2 |
PWR |
HPWR4 |
|
|
0404A2DK |
3 |
PCIe Slot1-M1 |
MCIO connector C2-P1A |
|
|
PCIe Slot1-M2 |
MCIO connector C2-P1C |
|||
|
0404A2DS |
4 |
PCIe Slot2-M2 |
MCIO connector C2-P3C |
|
|
PCIe Slot2-M1 |
MCIO connector C2-P3A |
RC-2HHHL-R4-2U-G6
The RC-2HHHL-R4-2U-G6 riser card can connect to PCIe riser connector 3 or 4 on the system board. The RC-2HHHL-R4-2U-G6 riser card is for liquid-cooled models only. The connectors the cables connect to different depending on the PCIe riser connector, see "Rear panel components." Connect power cords, signal cables, and PCIe cables for the riser card, as shown in Figure 73.
Figure 73 Connecting PCIe signal cables for the riser card
Table 55 Connecting cables
|
Riser connector |
Cable code |
Cable number |
Connector/mark on the riser card |
Connector on the system board/IO board |
|
Riser3 |
0404A2RG |
1 |
AUX |
RISER AUX3 |
|
0404A2D1 |
2 |
PWR |
PWR5 |
|
|
N/A |
3 |
Slot2 |
MCIO connector C2-P2C |
|
|
N/A |
4 |
Slot1 |
MCIO connector C1-P3A |
Connecting the LCD smart management module cable
Figure 74 Connecting the LCD smart management module cable
|
Cable type |
Cable code |
Cable description |
|
Signal cable |
0404A1SA |
Connect LCD smart management module to the system board (DIAG LCD) |
Connecting an inlet temperature sensor cable
Figure 75 Connecting an inlet temperature sensor cable
|
Cable type |
Cable description |
|
Temperature sensor cable |
Connect cables to the system board (SENSOR) |
Connecting cables for the mounting bracket
Figure 76 Connecting cables for the mounting bracket
|
No. |
Cable type |
Cable description |
|
1 |
Left mounting bracket cable |
Connect cables to the system board (LEFT EAR) |
|
2 |
Right mounting bracket cable |
Connect cables to the system board (RIGHT EAR) |
Maintenance
This section introduces daily maintenance methods for the server.
Guidelines
· Keep the server room clean and maintain the temperature and humidity according to the server operating requirements. Do not place unrelated equipment and items in the server room.
· Regularly check the server's health status through HDM, and if it's not healthy, immediately inspect and troubleshoot.
· Stay informed about the latest updates for operating systems and applications, and update the software as needed.
· Develop a reliable backup plan.
¡ Schedule regular data backups based on the server's operational status.
¡ Regularly back up data if it changes frequently.
¡ Regularly check backups to ensure data is stored correctly.
· Keep a certain number of spare parts on site to replace components promptly if they fail. After using spare parts, please replenish them promptly.
· To conveniently solve networking issues, please save the latest network topology diagram.
Maintenance tools
To maintain the server, you need the following tools:
· Monitor the server operating environment with a thermometer and hygrometer.
· Monitor the operating status of the server through HDM and UniSystem.
Maintenance operations
This section introduces daily maintenance tasks and operational methods for the server.
Tasks
Daily maintenance tasks are shown in Table 56.
Table 56 Daily maintenance tasks
|
Task |
Required tools |
|
/ |
|
|
Monitoring the temperature and humidity of the equipment room |
Temperature humidity meter |
|
/ |
Checking the server LEDs
Check if all LEDs on the front and back panels of the server are functioning properly.
Monitoring the temperature and humidity of the equipment room
Use a hygrometer to measure the temperature and humidity in the equipment room, ensuring that they are within the operating range of the server. For server operation and storage temperature and humidity requirements, see "Physical specifications."
Inspecting the cables
Check if the communication and power cables are properly connected.
Restrictions and guidelines
· When plugging or unplugging cables, do not use excessive force.
· Do not twist or stretch the cables.
· Organize the cables appropriately.
Check standards
· The cable type is correct.
· The cable connection is correct, secure, and the cable length is appropriate.
· The cable shows no signs of aging, and the connection points are free from twisting and corrosion.
Viewing server status
To view basic information and status of the subsystems of the server, see the HDM3 user guide.
Collecting server logs
For more information, see the HDM3 user guide.
Updating the server firmware
For more information about updating the HDM, BIOS, or CPLD firmware, see H3C Servers Firmware Update Guide.
Troubleshooting
For more information, see the troubleshooting manual.












































































