解决了

微软服务器集群

  • 2018年11月29日
  • 14日回复
  • 4863的浏览量
R
AHV是否支持MSCS ?我知道下面提到的两个节点windows集群见证虚拟机:

https://portal.nutanix.com/#/page/docs/details?targetId=Web-Console-Guide-Prism-v58:wc-cluster-two-node-c.html

但是想确认一下它是否支持包含quorum的共享存储的MSCS ?
图标

最佳答案arjun_sen2020年1月24日，10:57

Hello there,<\/p>

Failover clustering is supported on AHV by enabling MPIO on Guest OS Windows Server level.Leveraging MPIO enables the Guest to completely takeover the control for ability to determine which failover peer should write on the shared vDisk. Following documentation lists out the steps to leverage Volume Group from Nutanix Cluster to be presented to the VMs for configuring In-Guest clustering.

https:\/\/portal.nutanix.com\/#\/page\/docs\/details?targetId=Advanced-Admin-AOS-v510:vmm-failover-cluster-create-t.html<\/a> (Access to support portal required)<\/p>","className":"post__content__best_answer"}">

查看原始
本主题已关闭供评论

14日回复

E
Userlevel 1
徽章 ＋1

我认为使用iSCSI卷组你可以有一个共享存储通过网络仲裁和其他数据为您的虚拟机的。我目前在ESXi中使用它，但我认为在AHV中应该是相同的概念。

i think using iSCSI Volume Group you can have a shared storage over network for Quorum and other data for your VM's. I am currently using it in ESXi, but i think it should be the same concept in AHV.<\/p>","quoteUsername":"eng_asfour","translations":{"Common":{"like":"Like","unlike":"Unlike"},"Forum":{"Quote":"Quote","Share":"Share"}}}">
F
徽章
嗨,瑞卡,,

你可以使用卷组，这是Nutanix卷(ABS)为这个用例，Nutanix卷使用scsi-3持久化MSCS，你可以使用数据服务ip作为iscsi目标而不使用iscsi mpio。
参考:https://portal.nutanix.com/ /页面/ docs /细节? targetId = Web-Console-Guide-Prism-v55: wc-block-services-c.html

F P >
F P >
\r\n
\r\nYou can use volume group which is Nutanix Volumes (ABS) for this use case, Nutanix volumes uses scsi-3 persistence for MSCS, You can use data services ip as iscsi target without using iscsi mpio.
\r\nRefer : https:\/\/portal.nutanix.com\/#\/page\/docs\/details?targetId=Web-Console-Guide-Prism-v55:wc-block-services-c.html
\r\n
\r\nF>P","quoteUsername":"Farhan.Parkar","translations":{"Common":{"like":"Like","unlike":"Unlike"},"Forum":{"Quote":"Quote","Share":"Share"}}}">
ddubuque
Userlevel 3
徽章 + 10
NFS存储平台不支持故障转移集群。您需要向集群提供外部iSCSI存储，以便进行故障转移集群。如果你一直在集群中寻找像SQL，它本质上只是一个连续的复制，那么是的，它是被支持的。这不是hypervisor的问题，而是来自微软的NFS不兼容。Nutanix有“受支持的”方法，但您需要与支持一起工作。
David Dubuque || BKD || Nutanix NTC 2019 & 2020 || Nutanix技术咨询委员会成员|| https://darklinkrising.com
arjun_sen
排名
Userlevel 1
徽章

你好,

AHV通过在Guest OS Windows Server级别上启用MPIO来支持故障转移集群。利用MPIO使Guest能够完全接管控制，以决定应该在共享vDisk上写入哪个故障转移对等点。以下文档列出了使用Nutanix Cluster中的Volume Group的步骤，该步骤将呈现给虚拟机，用于配置In-Guest集群。

https://portal.nutanix.com/#/page/docs/details?targetId=Advanced-Admin-AOS-v510:vmm-failover-cluster-create-t.html(需要访问支持门户)

Hello there,<\/p>

Failover clustering is supported on AHV by enabling MPIO on Guest OS Windows Server level.Leveraging MPIO enables the Guest to completely takeover the control for ability to determine which failover peer should write on the shared vDisk. Following documentation lists out the steps to leverage Volume Group from Nutanix Cluster to be presented to the VMs for configuring In-Guest clustering.

https:\/\/portal.nutanix.com\/#\/page\/docs\/details?targetId=Advanced-Admin-AOS-v510:vmm-failover-cluster-create-t.html<\/a> (Access to support portal required)<\/p>","quoteUsername":"arjun_sen","translations":{"Common":{"like":"Like","unlike":"Unlike"},"Forum":{"Quote":"Quote","Share":"Share"}}}">

ddubuque
Userlevel 3
徽章 + 10

@arjun_sen这是最受欢迎的消息!欣赏这篇分享的文章，很高兴看到支持故障转移集群!

David Dubuque || BKD || Nutanix NTC 2019 & 2020 || Nutanix技术咨询委员会成员|| https://darklinkrising.com
@arjun_sen<\/user-mention>\u00a0this is most welcome news!\u00a0 Appreciate the article shared and very happy to see failover clustering supported!<\/p>","quoteUsername":"ddubuque","translations":{"Common":{"like":"Like","unlike":"Unlike"},"Forum":{"Quote":"Quote","Share":"Share"}}}">
J
Userlevel 4
徽章 + 8

是的，它得到了支持，我已经在我现在的公司建立了一些。按照最佳实践，我在Nutanix中为集群的每个驱动器创建了单独的卷组;备份、数据、日志、仲裁、TempDB都是独立的卷组，呈现给节点虚拟机。

Yes, it\u2019s supported and I\u2019ve built some here at my current employer.\u00a0I built separate Volume Groups in Nutanix for each of the drives for the cluster following best practices; Backups, Data, Logs, Quorum, TempDB are all separate Volume Groups presented to the node VMs.\u00a0<\/p>","quoteUsername":"jlackman","translations":{"Common":{"like":"Like","unlike":"Unlike"},"Forum":{"Quote":"Quote","Share":"Share"}}}">
J

你好,

AHV通过在Guest OS Windows Server级别上启用MPIO来支持故障转移集群。利用MPIO使Guest能够完全接管控制，以决定应该在共享vDisk上写入哪个故障转移对等点。以下文档列出了使用Nutanix Cluster中的Volume Group的步骤，该步骤将呈现给虚拟机，用于配置In-Guest集群。

https://portal.nutanix.com/#/page/docs/details?targetId=Advanced-Admin-AOS-v510:vmm-failover-cluster-create-t.html

Arjun Sen || SRE || Nutanix支持|| VCAP-DCD6

我真的很想读这篇文章，看看如何做到这一点，但链接把我带到一个https://portal.nutanix.com/page/error/204错误页面。

\t

Hello there,<\/p> \t

\u00a0<\/p> \t

Failover clustering is supported on AHV by enabling MPIO on Guest OS Windows Server level.Leveraging MPIO enables the Guest to completely takeover the control for ability to determine which failover peer should write on the shared vDisk. Following documentation lists out the steps to leverage Volume Group from Nutanix Cluster to be presented to the VMs for configuring In-Guest clustering.

https:\/\/portal.nutanix.com\/#\/page\/docs\/details?targetId=Advanced-Admin-AOS-v510:vmm-failover-cluster-create-t.html<\/a><\/p> \t

\u00a0<\/p> \t

Arjun Sen || SRE || Nutanix Support || VCAP-DCD6\u00a0<\/p> \t <\/content-quote>

\u00a0<\/p>

I would really like to read this article to find out how to do this, but the link takes me to a\u00a0https:\/\/portal.nutanix.com\/page\/error\/204<\/a>\u00a0error page.<\/p>","quoteUsername":"Jason.Fernow","translations":{"Common":{"like":"Like","unlike":"Unlike"},"Forum":{"Quote":"Quote","Share":"Share"}}}">

J
Userlevel 4
徽章 + 8

我也找不到那份文件。我不知道有没有存档什么的。

I cannot find the document either. I don\u2019t know if it was archived or what.\u00a0<\/p>","quoteUsername":"jlackman","translations":{"Common":{"like":"Like","unlike":"Unlike"},"Forum":{"Quote":"Quote","Share":"Share"}}}">
B
徽章 ＋1

我在卷组中使用ISCSI启动并运行这个程序，唯一的问题是，当在我们的容灾环境中启动它时，它无法访问那里的ISCSI，我被迫尝试并在Nutanix中使用VM Attached启动它。在DR做这件事没有任何问题。

I have this up and running using the ISCSI for the Volume Groups, only problem is when bringing it up in our DR environment it cannot access the ISCSI there and I am forced to try and bring it up with just VM Attached in Nutanix. Nothing but issues trying to do that in DR.\u00a0<\/p>","quoteUsername":"BSchaffer","translations":{"Common":{"like":"Like","unlike":"Unlike"},"Forum":{"Quote":"Quote","Share":"Share"}}}">
J
Userlevel 4
徽章 + 8

@BSchaffer我不确定我理解了，但我的第一个问题是，您确定保护域中除了实际的虚拟机之外还有MSCS卷组吗?

@BSchaffer<\/user-mention>\u00a0not sure I understand but my first question would be are you sure you have the MSCS volume groups in the Protection Domain, in addition to the actual VMs?<\/p>","quoteUsername":"jlackman","translations":{"Common":{"like":"Like","unlike":"Unlike"},"Forum":{"Quote":"Quote","Share":"Share"}}}">
B
徽章 ＋1

@jlackman是的，它们在保护范围内。

问题是，在生产集群中，它们是通过使用Nutanix中的数据服务IP通过iSCSI连接的。在我们的容灾集群中，这是不可能的，因为这是一个隔离的环境，他们无法访问Nutanix容灾集群的数据服务IP，因为它处于隔离之外。所以当我试着把他们带到那里去测试我必须移除vg上的iSCSI复选框，然后通过Nutanix连接到VM。这将导致在Windows集群中启动角色的各种问题，SQL服务将不能正确启动。

@jlackman<\/user-mention>\u00a0yes they are in the protection domain.<\/p>

\u00a0<\/p>

The problem is that in the Production cluster they are attached via iSCSI by using the Data Services IP in Nutanix. In our DR cluster this is not possible as it is an isolated environment and they cannot access the Data Services IP of the Nutanix DR cluster as it is outside the isolation. So when i try to bring them up out there to test i have to remove the iSCSI check boxes on the VGs and then attach to the VM through Nutanix. This is causing all kinds of issues in getting the Roles to start in the Windows Cluster, SQL services will not start correctly.<\/p>","quoteUsername":"BSchaffer","translations":{"Common":{"like":"Like","unlike":"Unlike"},"Forum":{"Quote":"Quote","Share":"Share"}}}">

J
Userlevel 4
徽章 + 8

明白了。我们还需要一些手动步骤来在DR环境中生成MSCS。我们有一个低优先级的努力来解决这个问题，但我没有任何我知道的解决方案，在一个孤立的环境中把它带回来。

Got it. We also have some manual steps required to bring up MSCS in the DR environment. We have a low priority effort to sort this out but I do not have any solution that I know about to bring it back up in an isolated environment.\u00a0<\/p>","quoteUsername":"jlackman","translations":{"Common":{"like":"Like","unlike":"Unlike"},"Forum":{"Quote":"Quote","Share":"Share"}}}">
P
徽章

我真的很想读这篇文章，看看如何做到这一点，但链接把我带到一个https://portal.nutanix.com/page/error/204错误页面。

Sudhir9
Userlevel 2
徽章 ＋４

@paddyf @Jason。Fernow

请使用此链接。

https://portal.nutanix.com/page/documents/details?targetId=Web-Console-Guide-Prism-v5_17:wc-ahv-direct-attach-guest-vm-cluster-create-t.html

如果您认为这不是很有用，请告诉我，我将在LAB中创建一个文档，并在这里分享步骤。

请让我知道您当前的AOS版本、AHV版本和Guest OS版本的详细信息。
另外，客户故障转移集群上是否需要共享存储。

Sudhir
Baidu