LH SAN Theory - FOM (Best Practice) - Small hands off implementation once installed

[Cary McDonald]

I'm new to Lefthand, but have used SANs and ESX for years. I'm setting up an ESXi environment for a small project going P2V for existing systems that are very old. Small projects are new to me as well. 10-15 guest servers in each primary site. 1 site in the other site.

In the end it will have three sites and 5 ESXi Servers running the LH SAN software.

Thanks for your time


Site A (Primary) with two ESXi boxes (DL 380s - G6)in the primary location running the LH SAN

Site A and B connected by 6MB fiber link with riverbend WAN Devices on each end.

Site B (DMZ \ backup site for primary) will have a single web server, but intended to be a backup site for Site A, if needed.

Site B and C are connected with Satellite and riverbend devices on each end. Site C is where the data is collected and used by Site A and C.

Site C (2nd Primary Created in a few months) This will be a duplicate of Site A with two ESXi boxes. Site C is in a very remote location running on generators and batteries only running 1/2 the year. Fortunately, I can build and drop ship Site C.

So I've built out "Site A" within VM Workstation 7 running on my laptop. (New i7 Macbook Pro-works great) I have the ESXi set up, guests running the following: 1. managment server running vCenter\ Lefthand Management Server 2. VSAs on each ESXi box 3. 4 test virtual appliances.

Yesterday, I think I'm doing well and start documenting some of the installs for the techs and find the following link.
http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=1007817
For high availability, a minimum of two VSA nodes in a cluster and a failover manager or virtual manager are required.

Surprise - this is the first reference to FOM that came across in a few months of sales and research. So at this point I'm also doubting, a little, what I think I know, based on what I have read and the test setups.

I would like resources to read or suggestions regarding the following questions:

What is the optimal configuration for vSphere ESXi and VSA implementation including FOM on Two servers? Best Practices? Future Considerations?

When having two ESXi servers with two VSAs with identical hardware and 650 GB of storage? What is the best way to plan the storage? I was thinking of 4 Volumes with like tasking and would like to have two of the Volumes mirrored at Site B in case site A goes dark.

Where on the ESXi boxes is the best place to run the management guests? Each VSA runs on local storage with all(?) of the rest of the local storage presented to them? The Guests including the management servers and the FOM run inside the shared iSCSI storage created by the VSAs?

It is as if all the documentation including the field guide for VI3 omits the FOM and some assumed general best practice considerations, that just have not clicked on yet?

A high level order of creation list and what to create and what kind of storage it needs to be on when created and if it needs to be moved to the iSCSI storage once completed?

High level LH SAN picture with a little theory or why it works the way it works would be helpful also?

I can build all the pieces, but am missing something.

Any thoughts or direction would be helpful.

Thanks for your time
Cary

[douglas carson_1 · Accepted solution]

Cary,
see the link below for the user guides. There is multi-site setup instructions within these manuals.
http://h20000.www2.hp.com/bizsupport/TechSupport/DocumentIndex.jsp?lang=en&cc=us&contentType=SupportManual&prodTypeId=12169&prodSeriesId=3936136&docIndexId=64255

The VSA runs on the local storage and the vms will run inside the storage presented from the VSA.
regards
DC