Hyper-V Failover cluster

[guest]

We have 2 Hyper-V servers running on Windows 2008 R2 DataCenter and Windows 2008 Storage Server as Shared Storage using iSCSI.

I have the cluster working, when testing Live migration there is only one PING drop out and no downtime, when testing Quick Migration this process puts the machines in 'Save' state then moves them to the other server, there is downtime during this process (I assume this is the normally process, depending on Memory and connection)?

When stopping the Primary Node the VMs begin going into 'Save' state, the problem is when this happens the VM's are in-accessible for 10 minutes until moved to the secondary Node, I thought this process will be on the fly with no downtime?

[guest]

Hi!

If you put the primary node in maintenance mode the vm's will be moved sequentially from one node to the other, while still running. This is refered to as Live Migration as you've noted.

The saved state occurs because your cluster thinks your first node has failed, and it then depends on the cluster settings how long it'll take before the resources are failed over and restarted on the other node.

This article is on the subject of clustering and hyper-v: http://technet.microsoft.com/en-us/libr ... 81(v=ws.10).aspx

And more on maintenance mode: http://technet.microsoft.com/en-us/libr ... 36481.aspx

EDIT:

Don't miss this one: http://blogs.technet.com/b/chengw/archi ... -mode.aspx



--------------------------------------------------------------------------------

Best regards

Joachim Nässlander

MVP Cluster
Member of Microsoft Extended Experts Team (MEET)
Blog: http://www.nullsession.com

[guest]

Well, if one node fails or goes down, all VM in that host will be restarted in another node.

This is known as "unplanned failover". In this case, your VM will have to reboot in another node.

The "Planned Failover" is when an administrator choose to move the VM to another node. When you do live or quick migration.

Look on 11:40 to 14:52

Virtualization Jump Start (04): High-Availability & Clustering

http://technet.microsoft.com/en-us/edge/Video/hh181062

Now, VMware has a HA (High Availability) and a FT (Fault Tolerance) feature,

HA is more like the one we have in Hyper V Cluster now. VM in that failed host will be restarted in another node.

FT is for a unplanned failover, VM are moved with no data loss, but that is a whole different beast. They are not the same thing. I'm following close to Hyper V Server 8, that looks will have something similar, for free. :)

Read this:

VMware vSphere 5 Competitive Reviewer’s Guide

http://www.vmware.com/files/pdf/VMware- ... -WP-EN.pdf

High Availability

VMware vSphere High Availability (HA) utilizes heartbeats between vSphere hosts and virtual machines in

the cluster to check that they are functioning. When a host or virtual machine failure is detected, VMware HA

automatically restarts affected virtual machines on other production servers, ensuring rapid recovery from

failures.

Fault Tolerance

vSphere Fault Tolerance (FT) provides continuous availability for applications in the event of server failures,

by creating a live shadow instance of a virtual machine that is in virtual lockstep with the primary instance. By

allowing instantaneous failover between the two instances in the event of hardware failure, FT eliminates even

the smallest chance of data loss or disruption.

I hope this helps to clarify the issue.