blog header
Subscribe via: (Email / RSS)

Can’t afford HA? Try a manual failover with my latest articles

by Harley Stagner on February 4, 2008

My latest articles are up at I go over how I set up a file server VM to connect via the Microsoft iSCSI initiator to a SATABoy storage array in the first article. In the second article I provide a Powershell and a VBScript script (your choice) that backs up the file server VM onto another ESX host using the command-line “p2vtool.exe.”

From the Part I:

“This two-part article series will outline a way to provide some redundancy for a file server virtual machine without the use of a SAN for VMFS storage or VMotion. The first part will discuss the initial setup of the file server virtual machine. The second part will discuss a scripted synchronization of the source virtual machine (VM) to another ESX server for manual recovery. Depending on your service up-time requirements, a manual fail-over of a virtual machine without the use of VMware HA could be acceptable. While these articles discuss a scenario with a file server virtual machine, the technique discussed could be used with any server that has a system drive that changes infrequently.”

Read the rest of Part I at

From Part II:

“What would happen if your physical server running VMware ESX Server were to fail? If you don’t have failover options in place, you’d be in trouble. The good news is that you can eliminate this worst-case scenario without buying more hardware. You just need a redundant virtual machine (VM) file server.”

Read the rest of Part II at

Happy Reading!

[tags]VMware, P2V, Business Continuity, iSCSI[/tags]

{ 1 comment… read it below or add one }

Atomizer January 15, 2009 at 5:05 pm

The articles are just great!
I have a question for you.
Imagine a situation:
2 ESX servers, 1 SAN box.
Virtual Machines of the ESX servers reside on the SAN itself, on a single LUN, which uses VMFS (if that matters).
What would be a good practice in your opinion to implement a manual disaster recovery solution for a case of a hardware failure of one of the ESXs?
My ideas are floating around adding the existing vmdk files into another ESX server right away and booting from it. I wonder what your thoughts would be.

Leave a Comment