使用 Windows Admin Center 管理 Azure Stack HCI 群集Manage Azure Stack HCI clusters using Windows Admin Center

适用于:Azure Stack HCI 版本 20H2;Windows Server 2019Applies to: Azure Stack HCI, version 20H2; Windows Server 2019

Windows Admin Center 可用于管理 Azure Stack HCI 群集。Windows Admin Center can be used to manage your Azure Stack HCI clusters. 具体来说,是使用 Windows Admin Center 中的群集管理器扩展来管理群集。Specifically, you will be using the Cluster Manager extension in Windows Admin Center to manage your clusters.

查看群集仪表板View the cluster dashboard

群集仪表板显示有关群集运行状况和性能的信息。The cluster dashboard displays information regarding cluster health and performance.

群集仪表板屏幕

若要查看此信息,请在“所有连接”下选择群集名称,然后在左侧的“工具”下选择“仪表板” 。To view this information, select the cluster name under All connections, then under Tools on the left, select Dashboard. 可以查看以下内容:You can view the following:

  • 群集事件警报Cluster event alerts
  • 已加入群集的服务器列表List of the servers joined to the cluster
  • 群集上运行的虚拟机列表List of virtual machines running on the cluster
  • 群集上可用的磁盘驱动器列表List of disk drives available on the cluster
  • 群集上可用的卷列表List of volumes available on the cluster
  • 群集的群集 CPU 使用率Total cluster CPU usage for the cluster
  • 群集的群集内存总使用率Total cluster memory usage for the cluster
  • 群集的群集存储总使用率Total cluster storage usage for the cluster
  • 群集总输入/输出操作数/秒 (IOPS)Total cluster input/output operations/second (IOPS)
  • 群集平均延迟(以毫秒计)Average cluster latency in milliseconds

更改群集存储设置Change cluster storage settings

有两个可以进行更改、可应用于群集且与存储空间直通相关的设置。There are two settings you can change related to Storage Spaces Direct that can be applied to your cluster.

  1. 在 Windows Admin Center 中,从顶部下拉箭头中选择“群集管理器”。In Windows Admin Center, select Cluster Manager from the top drop-down arrow.

  2. 在“工具”下,选择底部的“设置” 。Under Tools, select Settings at the bottom.

  3. 若要配置存储缓存,请选择“存储空间直通”,然后配置以下各项:To configure the storage cache, select Storage Spaces Direct, then configure the following:

    • 对于“永久缓存”,选择“已启用”或“已禁用” for Persistent cache, select either Enabled or Disabled

    • 对于“HDD 缓存模式”,选择“只读”、“只写”或“读取/写入” for Cache mode for HDD, select Read only, Write only, or Read/Write

    • 对于“SSD 缓存模式”,选择“只读”、“只写”或“读取/写入” for Cache mode for SSD, select Read only, Write only, or Read/Write

      群集仪表板屏幕

  4. 若要使用服务器内存来缓存频繁读取的内容,请选择“内存中缓存”,然后指定每个服务器要使用的最大内存。To use server memory to cache frequent reads, select In-memory cache and specify the maximum memory to be used per sever. 另请参阅了解 Azure Stack HCI 中的缓存Also see Understanding the cache in Azure Stack HCI.

    群集仪表板屏幕

更改群集常规设置Change cluster general settings

可以向群集应用五个常规设置。There are five general settings that can be applied to your cluster. 可在此处设置和管理访问点、节点关闭行为、流量加密、VM 负载均衡和群集见证。Here is where you can set and manage access points, node shutdown behavior, traffic encryption, VM load balancing, and cluster witness.

  1. 在 Windows Admin Center 中,从顶部下拉箭头中选择“群集管理器”。In Windows Admin Center, select Cluster Manager from the top drop-down arrow.

  2. 在“工具”下,选择“设置” 。Under Tools, select Settings.

  3. 若要更改群集名称,请选择“访问点”,然后输入新名称。To change the cluster name, select Access point and enter the new name.

    群集仪表板屏幕

  4. 若要控制节点关闭行为,请选择“节点关闭行为”,然后确保勾选了复选框。To control node shutdown behavior, select Node shutdown behavior and ensure the checkbox is enabled. 此操作会先将所有虚拟机从节点移出,以允许节点正常关闭。This moves any virtual machines from the node first to allow graceful node shutdown.

    群集仪表板屏幕

  5. 若要加密用于在群集节点之间发送数据的 SMB 连接,请选择“群集流量加密”,然后从下拉框中选择“加密”,以实现以下加密目的 :To encrypt SMB connections used to send data between cluster nodes, select Cluster traffic encryption, then select Encrypt from the dropdown boxes for the following:

    • 核心流量 - 对在端口 3343 上通过 NetFT(群集虚拟适配器)发送的流量进行加密Core traffic - encrypts traffic sent over NetFT (cluster virtual adapter) on port 3343

    • 服务器流量 - 对群集共享卷 (CSV) 和存储总线层 (SBL) 流量进行加密Server traffic - encrypts Cluster Shared Volume (CSV) and Storage Bus Layer (SBL) traffic

      群集仪表板屏幕

  6. 若要自动在整个群集的所有虚拟机中实现负载均衡,请选择“虚拟机负载均衡”,然后执行以下操作:To automatically load-balance virtual machines across the cluster, select Virtual machine load balancing, and do the following:

    • 对于“均衡虚拟机”,选择相应的操作For Balance virtual machines, select the appropriate action

    • 对于“入侵”,选择相应的行为For Aggressiveness, select the appropriate behavior

      若要了解其工作原理,请参阅 Virtual Machine Load Balancing overview(虚拟机负载均衡概述)。For information on how this works, see Virtual Machine Load Balancing overview.

      群集仪表板屏幕

  7. 若要选择仲裁见证类型,请选择“见证”,然后选择以下选项之一:To select a quorum witness type, select Witness, then select one of the following:

    • 云见证 - 使用 Azure 云资源作为见证Cloud witness - to use an Azure cloud resource as witness

    • 磁盘见证 - 使用磁盘资源作为见证(不用于延伸群集)Disk witness - to use a disk resource as witness (do not use for stretched clusters)

    • 文件共享见证 - 使用文件共享作为见证File share witness - to use a file share as witness

      有关详细信息,请参阅了解 Azure Stack HCI 上的群集和池仲裁For more information, see Understanding cluster and pool quorum on Azure Stack HCI.

      群集仪表板屏幕

更改群集 Hyper-V 设置Change cluster Hyper-V settings

可以向群集应用五个 Hyper-V 主机设置。There are five Hyper-V host settings that can be applied to your cluster.

  1. 在 Windows Admin Center 中,从顶部下拉箭头中选择“群集管理器”。In Windows Admin Center, select Cluster Manager from the top drop-down arrow.

  2. 在“工具”下,选择“设置” 。Under Tools, select Settings.

  3. 选择“常规”,然后使用以下设置:Select General and then use the following settings:

    • 虚拟硬盘路径 - 指定用于存储虚拟硬盘文件的默认文件夹。Virtual Hard Disks Path - specify the default folder for storing virtual hard disk files.

    • 虚拟机路径 - 指定用于存储虚拟机配置文件的默认文件夹。Virtual Machines Path - specify the default folder for storing the virtual machine configuration files.

    • 虚拟机监控程序计划程序类型 - 选择“核心计划程序”或“经典计划程序” 。Hypervisor Scheduler Type - select Core Scheduler or Classic Scheduler. 这会决定虚拟机监控程序如何安排虚拟进程在使用“同时多线程”(也称为 SMT 或超线程)的物理处理器上运行。This determines how the hypervisor schedules virtual processes to run on physical processors that use simultaneous multi-threading (also known as SMT or hyper-threading).

      群集仪表板屏幕

  4. 若要允许从虚拟机重定向本地设备和资源,请选择“增强会话模式”。To allow redirection of local devices and resources from virtual machines, select Enhanced Session Mode. 请注意,增强会话模式的连接需要使用受支持的来宾操作系统。Note that enhanced session mode connections require a supported guest operating system.

    群集仪表板屏幕

  5. 若要允许虚拟机涵盖物理 NUMA 节点,请选择“NUMA 涵盖”。To allow virtual machines to span physical NUMA nodes, select NUMA Spanning. 非统一内存体系结构 (NUMA) 涵盖可以为虚拟机提供比单个 NUMA 节点上可用内存更多的内存。Non-uniform memory architecture (NUMA) spanning can provide a virtual machine with more memory than what is available on a single NUMA node.

    群集仪表板屏幕

  6. 若要指定可以在运行的同时进行迁移(实时迁移)的 VM 的数量,请选择“实时迁移”,选择一个数字,然后指定以下内容:To specify the number of VMs that can be simultaneously moved while running (live migrated), select Live Migration, select a number, then specify the following:

    • 对于“身份验证协议”,选择 CredSSP 或 Kerberos 。for Authentication Protocol, select either CredSSP or Kerberos.

    • 对于“性能选项”,请选择“压缩”或 SMB 。for Performance Option, select either Compression or SMB. 压缩数据通过 TCP/IP 连接发送。Compressed data is sent over a TCP/IP connection.

    • 勾选“使用任意网络”复选框即可使用节点上的任意可用网络执行迁移enable the Use any network checkbox to use any available network on a node to perform the migration

      群集仪表板屏幕

  7. 若要指定可同时执行的存储迁移数,请选择“存储迁移”,然后选择一个数字。To specify the number of storage migrations that can be performed at the same time, select Storage Migration, then select a number.

    群集仪表板屏幕

后续步骤Next steps