使用 HYPER-V 複本模擬的網域控制站的支援Support for using Hyper-V Replica for virtualized domain controllers

適用於:Windows Server 2016、Windows Server 2012 R2、Windows Server 2012Applies To: Windows Server 2016, Windows Server 2012 R2, Windows Server 2012

本主題解釋使用 HYPER-V 複本複寫一樣 (VM) 執行為網域控制站 (DC) 的支援。This topic explains the supportability of using Hyper-V Replica to replicate a virtual machine (VM) that runs as a domain controller (DC). HYPER-V 複本是 HYPER-V 開頭提供建複寫機制,在 VM 層級的 Windows Server 2012 」 的新功能。Hyper-V Replica is a new capability of Hyper-V beginning with Windows Server 2012 that provides a built-in replication mechanism at a VM level.

HYPER-V 複本非同步會複寫選取的 Vm 複本 HYPER-V 主機主要 HYPER-V 主機的區域網路或 WAN 連結。Hyper-V Replica asynchronously replicates selected VMs from a primary Hyper-V host to a replica Hyper-V host across either LAN or WAN links. 完成初始複寫之後,系統管理員所定義的間隔複寫後續變更。After initial replication is complete, subsequent changes are replicated at an interval defined by the administrator.

容錯移轉可能未計畫或計劃。Failover can be either planned or unplanned. 計畫容錯移轉車載機起始主要 VM 中,系統管理員,並將取消複製的任何變更透過複製到複本 VM 為防止任何資料遺失。A planned failover is initiated by an administrator on the primary VM, and any un-replicated changes are copied over to the replica VM to prevent any data loss. 回應未預期的錯誤的主要 VM 中的複本 VM 起始意外的容錯移轉。An unplanned failover is initiated on the replica VM in response to an unexpected failure of the primary VM. 可能會遺失資料因為有機會不會傳送上可能不已複寫尚未主要 VM 變更。Data loss is possible because there is no opportunity to transmit changes on the primary VM that might not have been replicated yet.

如需 HYPER-V 複本,請查看HYPER-V 複本概觀部署 HYPER-V 複本For more information about Hyper-V Replica, see Hyper-V Replica Overview and Deploy Hyper-V Replica.


只有在 Windows Server HYPER-V,不是執行 Windows 8 上的 HYPER-V 版本可執行 HYPER-V 複本。Hyper-V Replica can be run only on Windows Server Hyper-V, not the version of Hyper-V that runs on Windows 8.

Windows Server 2012 所需的網域控制站Windows Server 2012 domain controllers required

Windows Server 2012 HYPER-V 也引進了 VM GenerationID (VMGenID)。Windows Server 2012 Hyper-V also introduces VM-GenerationID (VMGenID). VMGenID 提供發生重大變更,以客體 OS 通訊 hypervisor 的方式。VMGenID provides a way for the hypervisor to communicate to the guest OS when significant changes have occurred. 例如,hypervisor 可以通訊到模擬 DC 的快照還原發生 (HYPER-V 快照還原技術,不備份還原)。For example, the hypervisor can communicate to a virtualized DC that a restore from snapshot has occurred (Hyper-V snapshot restore technology, not backup restore). Windows Server 2012 中的 AD DS 目前正在設法 VMGenID VM 技術的並使用它來偵測執行 hypervisor 作業,例如快照還原,讓它變得更好保護您。AD DS in Windows Server 2012 is aware of VMGenID VM technology and uses it to detect when hypervisor operations are performed, such as snapshot restore, which allows it to better protect itself.


若要加速點,只有在 Windows Server 2012 網域控制站 AD DS 提供導致 VMGenID; 從這些安全機制Dc 執行的 Windows Server 所有先前的版本會受到例如 USN 復原模擬的 DC 還原使用尚未支援的機制,例如快照還原時,可能發生的問題。To reinforce the point, only AD DS on Windows Server 2012 DCs provides these safety measures resulting from VMGenID; DCs that run all previous releases of Windows Server are subject to problems such as USN rollback that can occur when a virtualized DC is restored using an unsupported mechanism, such as snapshot restore. 如需有關這些安全性,而且在觸發時,請查看擬化檔案網域控制站架構For more information about these safeguards and when they are triggered, see Virtualized Domain Controller Architecture.

時容錯移轉 HYPER-V 複本 (計劃或計畫),Windows Server 2012 擬化檔案俠偵測 VMGenID 重設觸發上述安全性功能。When a Hyper-V replica failover occurs (planned or unplanned), the Windows Server 2012 virtualized DC detects a VMGenID reset, triggering the aforementioned safety features. Active Directory 作業再繼續以一般。Active Directory operations then proceed as normal. 複本 VM 執行主要 VM 的位置。The replica VM runs in place of the primary VM.


現在有兩個相同俠身分執行個體,還有可能執行主要執行個體與複製執行個體。Given that now there are now two instances of the same DC identity, there is a potential for both the primary instance and the replicated instance to run. HYPER-V 複本以確保主要有控制機制,複本 Vm 同時執行時它可能會在之後的 VM 複寫失敗的之間的連結的事件執行一次。While Hyper-V Replica has control mechanisms in place to ensure the primary and replica VMs do not run simultaneously, it is possible for them to run at the same time in the event the link between them fails after replication of the VM. 發生這個耗,模擬執行 Windows Server 2012 」 的網域控制站有防護功能,可協助保護 AD DS,請執行舊版 Windows Server 的網域控制站擬化檔案,而不要。In the event of this unlikely occurrence, virtualized DCs that run Windows Server 2012 have safeguards to help protect AD DS, whereas virtualized DCs that run earlier versions of Windows Server do not.

使用 HYPER-V 複本,確保您依照最佳做法上 HYPER-V 執行 virtual 網域控制站When using Hyper-V Replica, ensure that you follow best practices for running virtual domain controllers on Hyper-V. 此討論,例如儲存 virtual SCSI 磁碟上的 Active Directory 檔案的建議提供較保證資料持續性。This discusses, for example, recommendations for storing Active Directory files on virtual SCSI disks, which provides stronger guarantees of data durability.

支援和支援案例Supported and unsupported scenarios

執行 Windows Server 2012 Vm 支援意外的錯誤後移轉和測試容錯移轉。Only VMs that run Windows Server 2012 are supported for unplanned failover and for testing failover. 即使的計劃容錯移轉,Windows Server 2012 建議模擬俠為了的系統管理員的身分不慎開始主要 VM 和複寫的 VM 同時降低風險。Even for planned failover, Windows Server 2012 is recommended for the virtualized DC in order to mitigate risks in the event that an administrator inadvertently starts both the primary VM and the replicated VM at the same time.

Vm 執行舊版 Windows Server 的計劃容錯移轉支援,但因為 USN 復原可能不支援意外的錯誤後移轉的。VMs that run earlier versions of Windows Server are supported for planned failover but unsupported for unplanned failover because of the potential for USN rollback. 如需 USN 復原,請查看USN 和 USN 復原For more information about USN rollback, see USN and USN Rollback.


有網域或森林; 不功能層級需求有只作業系統執行使用 HYPER-V 複本複寫 Vm 為網域控制站的需求。There are no functional level requirements for the domain or forest; there are only operating system requirements for the DCs that run as VMs that are replicated using Hyper-V Replica. Vm 可以包含的其他實體或 virtual Dc 執行舊版 Windows Server 和可能或可能不也會複寫使用 HYPER-V 複本森林中部署。The VMs can be deployed in a forest that contains other physical or virtual DCs that run earlier versions of Windows Server and may or may not also be replicated using Hyper-V Replica.

這個支援聲明為基礎測試所執行的單一網域-森林,但也支援多網域森林設定。This support statement is based on tests that were performed in a single domain-forest, though multi-domain forest configurations are also supported. 針對這些測試,DC1 和 DC2 模擬的網域控制站的 Active Directory 複寫合作夥伴相同的網站、 Windows Server 2012 上執行於 HYPER-V 的伺服器上。For these tests, virtualized domain controllers DC1 and DC2 are Active Directory replication partners in the same site, hosted on a server that runs Hyper-V on Windows Server 2012. 執行 DC2 VM 來賓已支援 HYPER-V 複本。The VM guest that runs DC2 has Hyper-V Replica enabled. 在另一部遠方地理位置資料中心裝載複本伺服器。The Replica server is hosted in another geographically distant datacenter. 若要協助解釋如下的測試案例處理程序,VM 複本伺服器上執行稱為 DC2-收到及 (但實際上它就會保留原始 VM 相同的名稱)。To help explain the test case processes outlined below, the VM running on the replica server is referred to as DC2-Rec (although in practice it retains the same name as the original VM).

Windows Server 2012Windows Server 2012

下表解釋模擬執行 Windows Server 2012 和測試案例網域控制站的支援。The following table explains support for virtualized DCs that run Windows Server 2012 and test cases.

規劃錯誤後的移轉Planned Failover 計畫錯誤後的移轉Unplanned Failover
支援Supported 支援Supported
測試案例︰Test case:

Lax-dc1 和 DC2 執行 Windows Server 2012。- DC1 and DC2 are running Windows Server 2012.

-DC2 已關機並容錯移轉 DC2-收到及上執行。錯誤後的移轉可以計畫或計劃。- DC2 is shut down and a failover is performed on DC2-Rec. The failover can be either planned or unplanned.

-DC2-收到及開始之後,它會檢查是否在其資料庫 VMGenID 的值為值相同伺服器 HYPER-V 複本儲存一樣驅動程式。- After DC2-Rec starts, it checks whether the value of VMGenID that it has in its database is the same as the value from the virtual machine driver saved by the Hyper-V Replica server.

-如此一來,DC2-收到及觸發模擬保護措施;亦即它重設其呼叫識別碼、 其 RID 集區中,會捨棄並它將會假設作業主角之前,將設定初始同步需求。- As a result, DC2-Rec triggers virtualization safeguards; in other words, it resets its InvocationID, discards its RID pool, and sets an initial synchronization requirement before it will assume an operations master role. 如需初始同步需求,查看。For more information about initial synchronization requirement, see .

-DC2 收到及資料庫中儲存的新 VMGenID 值,然後確認任何後續更新的新呼叫識別碼部分。- DC2-Rec then saves the new value of VMGenID in its database and commits any subsequent updates in the context of the new InvocationID.

-做為因的呼叫識別碼重設 DC1 將會減少的所有 AD 變更導入了 DC2-收到及,即使這復原的時間,這表示任何廣告更新之後會安全地涵蓋容錯移轉 DC2-收到及執行- As a result of the InvocationID reset, DC1 will converge on all AD changes introduced by DC2-Rec even if it was rolled back in time, meaning any AD updates performed on DC2-Rec after the failover will safely converge
測試案例是一樣計劃故障,使用這些例外:The test case is the same as for a planned failover, with these exceptions:

-任何廣告更新上收到的 DC2,但尚未複寫的廣告合作夥伴複製到之前容錯移轉事件將會遺失。- Any AD updates received on DC2 but not yet replicated by AD to a replication partner before the failover event will be lost.

從 DC1 回 DC2-收到及複寫 AD 以 DC1 複寫復原點的時間後 DC2 上收到的廣告更新。- AD updates received on DC2 after the time of the recovery point that were replicated by AD to DC1 will be replicated from DC1 back to DC2-Rec.

Windows Server 2008 R2 和較舊版本Windows Server 2008 R2 and earlier versions

下表解釋模擬執行 Windows Server 2008 R2 和舊版的網域控制站的支援。The following table explains support for virtualized DCs that run Windows Server 2008 R2 and earlier versions.

規劃錯誤後的移轉Planned Failover 計畫錯誤後的移轉Unplanned Failover
支援,但不是建議您因為不支援 VMGenID Dc 執行這些版本的 Windows Server 或使用相關的模擬防護功能。Supported but not recommended because DCs that run these versions of Windows Server do not support VMGenID or use associated virtualization safeguards. 這會將它們 USN 復原的風險。This places them at risk for USN rollback. 如需詳細資訊,請查看USN 和 USN 復原For more information, see USN and USN Rollback. 不支援請注意:會 USN 復原已無法使用風險,例如單一 DC 森林 (不建議設定) 中支援意外的錯誤後移轉。Not supported Note: Unplanned failover would be supported where USN rollback is not a risk, such as a single DC in the forest (a configuration that is not recommended).
測試案例︰Test case:

Lax-dc1 和 DC2 執行 Windows Server 2008 R2。- DC1 and DC2 are running Windows Server 2008 R2.

-DC2 已關機並計劃容錯移轉 DC2-收到及上執行。關機在完成之前 DC2-收到及到複製 DC2 上的所有資料。- DC2 is shut down and a planned failover is performed on DC2-Rec. All data on DC2 is replicated to DC2-Rec before the shutdown is complete.

-DC2-收到及開始之後,它會繼續複寫 DC1 使用相同的呼叫識別碼 DC2。- After DC2-Rec starts, it resumes replication with DC1 using the same invocationID as DC2.