No relevant resource is found in the selected language.

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Read our privacy policy>Search

Reminder

To have a better experience, please upgrade your IE browser.

upgrade

MM910 Management Module V100R001 User Guide 24

This document provides the product description, installation and configuration methods, and common operations of the E9000 server chassis management module MM910.
Rate and give feedback :
Huawei uses machine translation combined with human proofreading to translate this document to different languages in order to help you better understand the content of this document. Note: Even the most advanced machine translation cannot match the quality of professional translators. Huawei shall not bear any responsibility for translation accuracy and it is recommended that you refer to the English document (a link for which has been provided).
Stateless Computing Management

Stateless Computing Management

Profile

GUI
Figure 7-4 Screenshot for the HMM managing independent chassis
Figure 7-5 Screenshot for the HMM managing master and slave chassis
Function

The Profile page allows you to implement stateless computing for compute nodes. Stateless computing means that operating systems (OSs) or upper-layer applications have no correlation with the status of basic input/output systems (BIOSs), network interface cards (NICs), and universally unique identifiers (UUIDs) on compute nodes and that server hardware migration and replacement do not require any modification to the OSs or upper-layer applications.

Table 7-15 describes the buttons on the Profile page, and Table 7-16 describes the parameters on this page. For details about the configuration, see Configuration Examples.

Table 7-15 Buttons on the Profile page

Button

Description

Add

Sets Profile name, Profile type, and Associated Slots. A maximum of 128 profiles can be created.

Sets Server BIOS Configuration, NIC Connection, and Uniqueness Configuration.

Import

Imports a profile from a compute node in the chassis or the local computer to the system.

NOTE:

When both the master and slave chassis are installed, the profile can be imported from only the detected slot where the master chassis is installed.

Migrate

Migrates a profile from a compute node in the specified or independent chassis to another compute node in other specified or independent chassis.

When both the master and slave chassis are installed, you can only query items on the profile for the slave chassis.

Table 7-16 Parameters in the profile list

Parameter

Description

Profile name

Identifies a profile.

Profile type

Indicates the profile type.

  • Common: supports automatic assignment of MAC addresses, WWNNs, and WWPNs, and can be delivered to multiple slots simultaneously.
  • Custom: contains MAC addresses, WWNNs, WWPNs, and unique configurations, and can only be delivered to a specific slot.

The supported compute nodes include the CH121, CH121 V3, CH121L V3, CH121H V3, CH140, CH140 V3, CH140L V3, CH220, CH220 V3, CH221, CH222, CH222 V3, CH225 V3, CH226 V3, CH240, CH242, CH242 V3, and CH242 V3 DDR4.

Only iMana 200 V5.26 or later supports the profile management function.

Associated chassis

NOTE:

This parameter exists only when both the master chassis and slave chassis are installed.

It indicates the specified chassis with which the profile is associated.

Associated Slots

Indicates the slot numbers of the compute nodes to which a profile is associated.

  • You can click Associated Slots to change the slot with which the profile is associated when the independent chassis is installed.
  • You can only view Associated Slots when both the master chassis and slave chassis are installed.

Status

Indicates the delivery status of a profile.

  • Not delivered: indicates that a profile has not been delivered to the compute node in the specified slot or the delivery fails. You can click to apply a profile to the compute node.
  • Being delivered
  • Being powered off
  • Delivered
  • Modified: indicates that a profile has been delivered to the compute node in the specified slot, modified, and redelivered to the compute node.
  • Delivery failed: indicates that a profile fails to be delivered to the compute node in the specified slot. When you move your mouse over Delivery failed, the failure cause is displayed.

Operation

Perform operations on a profile.

  • : modifies a profile.
NOTE:

Within two minutes after editing a profile, do not perform a restart or active/standby switchover of the MM910 management modules.

  • : copies a profile.
  • : deletes a profile.
  • : exports a profile to a PC. Since the file name length of a profile to be imported is restricted, it is recommended that the name of an exported profile, including the file name extension, contain a maximum of 37 characters. This ensures that the profile can be successfully imported again.
  • : associates or changes the chassis and compute node in a specified slot in the profile.
NOTE:

This operation can be performed only when both the master chassis and slave chassis are installed.

  • : delivers a profile to the compute node.
NOTE:

The system automatically delivers a profile to a compute node after you remove and then reinstall the compute node, power off and then power on the compute node, or gracefully restart the compute node.

NOTE:

If only or only and are displayed as operations, the profile contains NICs not supported by stateless computing.

Creating a Profile
  1. On the Profile page, click Add.

    The Add dialog box is displayed.

  2. Set Profile name and Profile type.

    NOTE:

    The value of Profile name contains a maximum of 20 characters, including letters, digits, and underscores (_), but cannot start with an underscore.

  3. Click OK.

    The profile configuration page is displayed.

  4. Set the parameters. For parameter description, see Table 7-17.

    NOTE:
    • Before configuring BIOS parameters, determine the parameters supported by the compute node based on the MM910 Management Module V100R001 User Guide. You can configure each supported parameter as follows:
      • Select the parameter and specify a value.
      • Ignore the parameter and use the existing value.
    • If the compute node uses iMana 200, the BIOS supports stateless computing only in Legacy mode. In UEFI mode, the stateless computing function is unavailable. This restriction does not apply if the compute node uses the iBMC.
    • The server BIOS and NIC BIOS configuration takes effect after the profile is delivered successfully. The settings in the stateless computing profile take precedence over the settings on the BIOS.
    Table 7-17 Parameters for creating a profile

    Parameter

    Description

    Server BIOS Configuration

    View server BIOS configuration by clicking one of the following:

    • Show all: displays all BIOS configuration parameters.
    • V5 (Purley): displays configuration parameters and default values for V5 compute nodes.
    • Others: displays configuration parameters and default values for V1 and V3 compute nodes.

    BIOS configuration includes the following:

    • Boot
    • Advance Processor
    • Memory
    • Virtualization
    • System
    • Security
    NOTE:

    When setting the server BIOS parameters, you are advised to set the Advance Processor and System parameters to the default values. Autonomous core C-state in Advance Processor (available only when the MM910 software version is (U54) 6.52 or later) applies only to CPUs of V5 compute nodes. (This parameter cannot be configured for CPUs of V3 compute nodes. Otherwise, the policy fails to be delivered.)

    NIC Connection

    Configures the NICs on compute nodes.

    • MAC Address Pool: Click. In the displayed Select MAC Address Pool dialog box, select the MAC address pools to be associated with a common of stateless computing.
    • NIC Slot
    • Slot Status: Specifies whether to enable the NIC BIOS profile. The value is On or Off. On indicates that the NIC BIOS profile is enabled, and Off indicates that the NIC BIOS profile is disabled and that the default NIC BIOS configuration is used.
    • NIC Type: The value is MZ220, MZ312, MZ510, MZ512, MZ520, MZ522, MZ730, MZ910, MZ912 or MZ221.
      NOTE:

      Only the V5 compute nodes in UEFI mode support the MZ312.

    • Multi-Channel Status: specifies whether to enable the multi-channel function. This parameter is available only if the value of NIC Type is MZ510, MZ512, MZ520 and MZ522. The multi-channel function allows a physical network port on a NIC to be divided into multiple virtual ports.
    • : modifies NIC configuration, including the following parameters:
      NOTE:
      • For the CH140 and CH140 V3, the working mode configuration of the MZ512 and MZ520 effective only for port0 and port1.
      • For the server nodes based on the Purley platform, for example CH121 V5, the parameters VlanMode, PVID, PVIDpriority and VlanTrunkRange are available for MZ520 and MZ522.
      • Global PF type: specifies the working mode of an MZ510 or MZ512 NIC. The value is NIC, FCoE, or iSCSI. NIC indicates that all physical functions (PFs) are configured as NIC channels to transmit Ethernet service data. FCoE indicates that PF 1 is configured as a Fibre Channel over Ethernet (FCoE) channel to transmit FCoE service data, and other PFs are configured as NIC channels. iSCSI indicates that PF 1 is configured as an Internet Small Computer Systems Interface (iSCSI) channel to transmit iSCSI service data, and other PFs are configured as NIC channels.
      • PXE Boot Enable: specifies whether to enable a network port to boot from the Preboot Execution Environment (PXE).
      • PXE Vlan: If you select Enable, you can manually set PXE Vlan ID. If you select Disable, the value of PXE Vlan ID is the default virtual local area network (VLAN) ID of the switch module.
      • PXE Vlan ID: identifies the VLAN that data packets transmitted over PXE pass through. For the MZ520 and MZ522, the value ranges from 0 to 4094. For other NICs, the value ranges from 2 to 4094. The value 0 indicates that the PXE VLAN mode is disabled.
      • PXE Vlan Priority: specifies the priority of data packets in the switch module. A large value indicates a high priority.
      • PF Type: indicates the working mode of a PF.
      • MAC: Select MAC addresses from the list. If this parameter is left blank, the default value is used.
      • Min Bandwidth (%): The sum of the values of Min Bandwidth (%) of four PFs must be 100%.
      • Max Bandwidth (%): specifies the maximum bandwidth percentage of a PF when the other PFs are idle. The value for each PF ranges from 0% to 100%, and 0% indicates that the PF is disabled.
      • PF Vlan ID: identifies the VLAN for a PF of the NIC type. The value is an integer ranging from 2 to 4094. You do not need to set the VLAN ID for a PF of the FCoE or iSCSI type.
      • MAC Address Auto Assignment: Enable indicates that MAC addresses in a MAC address pool are automatically assigned to NIC ports.
      • SR-IOV: specifies whether to enable the single-root I/O virtualization (SR-IOV) function. This parameter is available only if the multi-channel function is disabled for an MZ510, MZ512, MZ520, and MZ522 NIC. The SR-IOV function allows an NIC to be shared by multiple VMs.
      • WWNN: identifies a NIC chip. If this parameter is left blank, the default value is used.
      • WWPN: identifies a PF. If this parameter is left blank, the default value is used.
      • SAN Properties
      • SAN Boot Enable: specifies whether to enable SAN boot.
      • Destination WWPN: specifies the destination storage controller WWPN.
      • Destination LUN ID: specifies the destination LUN ID.
      • Priority: specifies the boot option sequence. The priority is in descending order from top to bottom. The SAN boots with the destination WWPN and LUN ID with the top priority.
      • VlanMode: The value can be NORMAL, FILTERING, or QINQ depending on the chip settings.
      NOTE:
      • If VlanMode is NORMAL, the parameters PVID, PVIDpriority, and VlanTrunkRange will be unavailable.
      • If VlanMode is FILTERING, the parameters PVID, PVIDpriority, and VlanTrunkRange are available. The parameter PVID is optional.
      • If VlanMode is QINQ, the parameters PVID and PVIDpriority are available and VlanTrunkRange is unavailable. PVID is mandatory, and the PVIDs of a port cannot be the same.
      • PVID: specifies the default VLAN ID of the port. The value range is 1 to 4094 for the QINQ mode and 0 to 4094 for the FILTERING mode. The value 0 indicates that PVID is not set. .
      • PVIDpriority: specifies the priority of the PVID. The value range is 0 to 7.
      • VlanTrunkRange: specifies the VLAN ID list. The total number of VLAN IDs of a chip cannot exceed 256. The value range of the VLAN ID is 1 to 4094. Enter the VLAN IDs in ascending order. The value of VlanTrunkRange can contain positive integers and integer segment separated by commas, for example, 1, 2, 3-5, 8.
        NOTE:

        The value 1, 2, 3-5, 8 indicates VLANs 1, 2, 3, 4, 5, and 8.

    LOM Configuration

    • Slot Status: On indicates enabling parameter configuration of the LOM. Off indicates disabling parameter configuration of the LOM and the LOM adopts the default configuration.
    • : modifies the LOM configuration, including the following parameters:
      • LOM Status: specifies whether to enable ports of the LOM.
      • MAC address configuration: The LOM MAC configuration is supported only when Boot type is UEFI.
    NOTE:

    Only the CH121 V5, CH221 V5, CH225 V5, and CH242 V5 support stateless computing configuration of LOMs.

    Uniqueness Configuration

    Specifies the unique information about a compute node, for example, UUID.

    • UUID resource pool name: UUID resource pool associated with a stateless computing profile.
    • : allows you to select an associated UUID resource pool.
      • UUID resource pool name: UUID resource pools that can be associated with stateless computing profiles.
      • UUID prefix: Prefix of UUIDs in a resource pool.
      • Number of UUIDs: Quantity of UUIDs in a resource pool.

Importing a Profile
  1. On the Profile page, click Import.

    The Import dialog box is displayed.

  2. Select Import from board or Import from PC, and set the other parameters. For details, see Table 7-18.

    Table 7-18 Methods for importing a profile

    Method

    Procedure

    Import from board

    Import a profile from a compute node in the chassis.

    1. Select the slot number of a compute node from the Source slot drop-down list.
    2. Enter the new profile name in the Profile name text box.
    3. Set Profile type to Common or Custom.
    4. Click OK.

    Import from PC

    Import a profile from the local PC.

    1. Click Browse and import a profile from the local PC.

      The file name of the profile to be imported, including the file name extension, can contain a maximum of 37 characters.

    2. Click OK.

Migrating a Profile
  1. On the Profile page, click Migrate.

    The Migrate dialog box is displayed.

  2. Select the source compute node slot from the Source slot drop-down list, and select the destination compute node slot from the Destination slot drop-down list.
  3. Click OK.

    When both the master and slave chassis are installed, choose the chassis and compute node slot specified in the profile.

Delivering a Profile
  1. Determine the page type.
    • Independent chassis: perform 2 to 5.
    • Master and slave chassis: perform 6 to 9.
  2. In the profile list, click a slot number in the Associated Slots column. See Figure 7-6.
    Figure 7-6 Associated Slots
    NOTE:

    If no slot is associated with a profile, the value of Associated Slots for this profile is None.

    The Select Associated Slots dialog box is displayed, as shown in Figure 7-7.

    Figure 7-7 Select Associated Slots dialog box

  3. Select the slot number to be associated with the profile, and click OK.

    The selected slot number is displayed in the Associated Slots column.

    NOTE:

    The compute node in the selected slot and the compute node to which the profile imported must be of the same type.

  4. Click in the Operation column.

    The Manually Deliver Profile dialog box is displayed, as shown in Figure 7-8.

    Figure 7-8 Manually Deliver Profile dialog box
  5. Select a slot number and click Save to Switch Module to apply the profile to the compute node in the specified slot.
  6. Click in the Operation column.

    The Select Associated Slots column is displayed, as shown in Figure 7-9.

    Figure 7-9 Select Associated Slots
  7. Select the chassis and compute node slot number to be associated with the profile, and click OK.

    The selected chassis and compute node slot number is displayed in the Associated chassis and Associated Slots column.

    NOTE:
    1. For the chassis not associated with the profile, Associated chassis lists None.
    2. The compute node in the selected slot and the compute node to which the profile imported must be of the same type.
  8. Click in the Operation column.

    The Manually Deliver Profile dialog box is displayed, as shown in Figure 7-10.

    Figure 7-10 Manually Deliver Profile
  9. Select a chassis and compute node slot number and click Save to Switch Module to apply the profile to the compute node in the specified slot.
    NOTE:

    Within two minutes after delivering a profile, do not perform a restart or active/standby switchover of the MM910 management modules.

MAC Address Pool

GUI

NOTE:

The screenshot for the HMM managing master and slave chassis is only an example. The functions on the right pane on the page are the same as that for the HMM managing independent chassis.

Function

The MAC Address Pool page allows you to define MAC address pools. You can then associate MAC address pools with a stateless computing profile on the profile page and allow the MAC addresses in the pools to be automatically assigned to NICs. When server hardware is migrated or replaced, the MAC address or WWPN can be inherited, without being detected by other devices in the network. Therefore no modification is required.

Table 7-19 describes the buttons on the MAC Address Pool page, and Table 7-20 describes the parameters on this page.

Table 7-19 MAC Address Pool

Button

Description

Add

Allows you to set MAC address pool name, Start Address, and Total MAC Addresses. A maximum of 64 MAC Address pools are supported.

Table 7-20 Parameters

Item

Description

MAC address pool name

Displays all MAC address pool names.

The value contains a maximum of 15 characters, including letters, digits, and underscores (_), but cannot start with an underscore.

Start Address

Indicates the start MAC address in the MAC address pool. The value is in the format of XX:XX:XX:XX:XX:XX (hexadecimal). The first XX must be an even number.

Total MAC Addresses

Indicates the number of consecutive MAC addresses in the MAC address pool. The value ranges from 1 to 8192.

End MAC Address

Indicates the end MAC address in the MAC address pool. The value is in the format of XX:XX:XX:XX:XX:XX (hexadecimal). The first XX must be an even number.

Operation

Allows users to perform operations on the MAC address pool.

  • Click to edit the MAC address pool option.
  • Click to delete the MAC address pool.

Table 7-21 describes the buttons on the MAC Address Query page, and Table 7-22 describes the parameters on this page.

Table 7-21 MAC Address Query

Button

Description

Query

Allows you to query the MAC address usage based on the MAC address pool name and MAC address of the port on the NIC, as well as the Slot No. where the compute node is installed.

Table 7-22 Parameters

Item

Description

MAC Address

Indicates the MAC address of the port on the NIC.

MAC address pool name

Indicates the MAC address pool name of the port on the NIC.

Profile name

Indicates the profile name of the port on the NIC.

Slot No.

Indicates the slot number of the compute node where the NIC is installed.

NIC Location

Indicates the slot number of the compute node where the NIC is installed.

NIC Type

Indicates the NIC name.

Port

Indicates a port number.

PF

Indicates the PF number of a network port.

UUID Resource

Screenshot
Figure 7-11 Screenshot for an independent chassis
Function

The UUID Resource page allows you to manage UUID resource pools and UUIDs. You can set the UUID prefix for a UUID resource pool. The UUIDs are then automatically generated by the MM910 management module. The UUID of a custom can be automatically assigned from a UUID resource pool or manually configured. When a common is associated with a UUID resource pool and compute nodes, UUIDs are automatically assigned to the compute nodes from the UUID resource pool. Each common corresponds to one UUID resource pool.

Table 7-23 describes the button in the UUID Resource area, and Table 7-24 describes the parameters.

Table 7-23 Button in the UUID Resource area

Button

Description

Add

You can set UUID resource pool name, UUID prefix, and Number of UUIDs for the created UUID resource pool. A maximum of 64 UUID resource pools are supported.

Table 7-24 UUID resource pool parameters

Parameter

Description

UUID resource pool name

  • Name of a UUID resource pool.
  • The name can contain letters, digits, and underscores (_) and cannot exceed 15 characters. The first character cannot be an underscore.

UUID prefix

  • The common prefix for the UUIDs is user-defined in hexadecimal xxxxxxxx-xxxx-xxxx-xxxx format. The suffix for the first UUID is randomly generated. The suffix of each subsequent UUID is generated in ascending order.
  • UUIDs in the same resource pool have the same UUID prefix. Different resource pools must use different UUID prefixes.

Number of UUIDs

Number of UUIDs in the resource pool. The value ranges from 1 to 128.

Operation

Operations you can perform on a UUID resource pool.

  • : Edit the number of UUIDs.
  • : Delete a UUID resource pool.

Table 7-25 describes the button in the UUID Query area, and Table 7-26 describes the parameters.

Table 7-25 Button in the UUID Query area

Button

Description

Query

Query the UUID usage based on the UUID resource pool name, UUID, and compute node Slot.

Table 7-26 UUID Query parameters

Parameter

Description

UUID

UUID of a compute node.

UUID resource pool name

Name of a UUID resource pool.

Profile name

Name of a configuration profile.

Slot No.

Slot number of a compute node.

Chassis Number

Number of the chassis that houses a compute node.

Node Resource

Screenshot

NOTE:

The screenshot is for an independent chassis and for reference only.

Function

The Node Resource page allows you to set multiple active and standby nodes for a resource pool. Failover is supported, that is when an active node fails, its stateless computing profile automatically migrates to a standby node.

Table 7-27 describes the button on the Node Resource page, and Table 7-28 describes the parameters.

Table 7-27 Button

Button

Description

Add

You can set the Resource pool name, Associated profile, Migration Conditions, and Associated Compute Node of a created resource pool. A maximum of 64 resource pools are supported.

Table 7-28 Parameters

Parameter

Description

Resource pool name

Name of a compute node resource pool.

The name can contain letters, digits, and underscores (_) and cannot exceed 15 characters. The first character cannot be an underscore.

Associated profile

Common associated with a compute node resource pool. This field can be empty.

Active Nodes

Compute nodes associated with profiles.

Standby Nodes

  • Standby nodes are not associated with stateless computing profiles.
  • If an active node fails, its associated profile will automatically migrate to a standby mode.
  • If a standby node with a migrated profile fails, the profile will not migrate again.
  • Standby nodes do not occupy MAC addresses of the MAC address pool of the associated common.

Running Standby Nodes

Standby nodes that have replaced faulty active nodes.

Abnormal Nodes

Active or standby nodes that meet the migration conditions. If a node becomes abnormal, it is disassociated from its profile.

Migration Status

Status of data migration from a faulty active node to a standby node. The states are "migration succeeded", "migration failed", and "not migrated".

Migration Conditions

Events that trigger active/standby switchovers. At least one condition is required.

Operation

Operations you can perform on a resource pool.

  • : Edit a resource pool.
  • : Changes an active node. Before the change, the new node is not associated with any stateless computing profile or compute node resource pool. The original active node will be restarted and the new node may be restarted.
  • : Change the role of a compute node.
  • : Delete a resource pool.
Translation
Download
Updated: 2019-04-10

Document ID: EDOC1000015900

Views: 80793

Downloads: 5200

Average rating:
This Document Applies to these Products
Related Documents
Related Version
Share
Previous Next