Virtual Domain Controllers and Windows Server 2012 Improvements

There have been a number of concerns when it comes to virtualising domain controllers.  The biggest of these is KB888794, which is an updated version of an article that I first encountered years previously, maybe in 2004.

USN Rollback

Basically, we had to treat any virtual domain controller like it was a physical installation.  That meant:

  • No snapshots
  • No recovering the DC from VM (host/storage level) backups
  • Don’t do anything to manipulate the virtual DC’s VM storage, such as copy/clone/etc

This was because the VM would “time travel”, effectively screwing up the USNs that are used to track AD object replication and possible cause the reuse of RID pools – in other words, completely frakking your AD and making you wish that you had paid up for that Microsoft Premier support contract.

Physical DC Required

One of the frustrating things, especially for small medium enterprises (SMEs) or smaller branch offices was that they need a local physical domain controller to enable a Hyper-V cluster.  This company might only need to hosts, but had to add another physical machine (small as it was) to enable the cluster to function.

That was the scenario up to now.  Enter Windows Server 2012.

Bootstrapping

Windows Server 2012 Failover Clusters have a new feature called bootstrapping.  It’s been mentioned in public but I’ve not seen any documentation on it yet.  In short, this allows a failover cluster to power up and start working without the presence of a physical domain controller.  The premise is that you instead run virtual domain controllers, hosted on the Hyper-V cluster itself.

That means that you don’t need the physical domain controller.  That’s a major saver for the SME or the branch office.

Virtual DCs are OK

If we’re OK with the idea of virtual domain controllers, then how do we deal with them?  How do we back them up easily?  In a true cloud where there might be a one-size-fits-all backup policy, how do admins (with zero knowledge of VM contents/roles) safely backup virtual domain controllers that might be created legitimate by the cloud’s tenants?

VM-GenerationID and Safe DC Virtualisation

Microsoft has come up with a new mechanism called VM-GenerationID (also seen documented on TechNet and blogged as Generation ID, VM Generation ID, VM-Generation ID and GenID).  It is an attribute called msDS-GenerationID of the DC’s computer object in AD.  This is normally kept in sync with the directory information tree (DIT) if everything is OK with the replication of the DC.

If something happens to the DC VM like a snapshot is applied or a backup of the VM is restored, then the VM effectively travels back in time, potentially causing a USN rollback and enabling RID reuse.  But, the DC compares the VM-GenerationID and the DIT version number.  If they are different then the DC is aware there is a problem.  The RID pool is discarded, a new one created, and a USN rollback is prevented.

Windows Server 2012 Hyper-V is the only hypervisor at this time to support this feature, and the virtual DCs must be running Windows Server 2012.

But There’s More – Rapid Deployment of DCs

Wouldn’t it be nice if you could clone domain controllers?  Normally you cannot.  But this new VM-GenerationID feature, combined with some other work done by Microsoft in WS2012, enabled you to export/import virtual DCs to clone new DCs with very little effort.

The process is simple enough:

  1. Have a PDC Emulator that is running WS2012.  This DC will not be cloned.
  2. Create a new virtual DC running WS2012. 
  3. Add the new template DC to a domain security group called Cloneable Domain Controllers.  This allows domain admins to restrict which (if any) DCs can be cloned.
  4. On the template DC Run Get-ADDCCloningExcludedApplicationList to see if any installed programs/services on the DC can be cloned (check with vendors).  Uninstall any that cannot support cloning.
  5. Run Get-ADDCCloningExcludedApplicationList –GenerateXml on the template DC
  6. Back on the template DC, run New-ADDCCloneConfigFile to create an XML answer file to configure name, IP, etc, for the new DC VM that you are about to create.#
  7. The last step creates a file called DCCloneConfig.xml.  Place this in either the directory where the DIT resides, %windir%NTDS, or the root of a removable media drive (maybe a SCSI attached blank VHD?)
  8. Stop and export the template VM.
  9. Import the VM to crate a new DC VM.
  10. Start the new VM, and you should now have a new DC.

I haven’t had a chance to try this out yet.  I’ll try to update this if I find the MSFT TechNet page is lacking.

Summary

What all this means is that with Windows Server 2012 and a hypervisor that is VM-GenerationID aware (WS2012 Hyper-V) then you can safely virtualise your domain controllers, and treat them just like any other VM, something that is of great importance in a true cloud.

 

Microsoft Assessment & Planning Toolkit 7.0 Goes Live – Supports Windows 8 and Server 2012

I just received an email informing me that MAP 7.0 is live, and it now supports assessment to help you plan the deployment of Windows 8, Windows Server 2012, and Windows Server 2012 Hyper-V.  You can start planning now, with the products coming down the pipe soon.

The new version which you can download now allows you to:

  • Understand your readiness to deploy Windows Server 2012 in your environment
  • Determine Windows 8 readiness
  • Investigate how Windows Server and System Center can manage your heterogeneous environment through VMware migration and Linux server virtualization assessments
  • Size your desktop virtualization needs for both Virtual Desktop Infrastructure (VDI) and session-based virtualization using Remote Desktop Services
  • Ready your information platform for the cloud with the SQL Server 2012 discovery and migration assessment
  • Evaluate your licensing needs with usage tracking for Lync 2010, active users and devices, SQL Server 2012, and Windows Server 2012

You should know that I believe that assessment is a critical early step in a virtualisation project, be it XenServer, VMware, or Hyper-V.  Without it, you’re shooting blind, and you’ll end up being an anecdote in a presentation on how to do a crap project.

Windows 8 and Server 2012 RTM/GA Dates Announced

I was actually under a rock, busy delivering some training, so I missed the WPC keynotes and the announcements today. 

  • Windows 8 is on track to RTM on the first week of August, assuming that there are no showstoppers between now and then.
  • Customers with Software Assurance will have access to the downloads via the Volume Licensing Service Center (VLSC) by the end of August.
  • The Windows 8 Store will go live (in other words, go commercial) at RTM time.
  • GA will be by the end of October.  That’s when you’ll see designed-for-Windows 8 machines for the first time in the stores.

Windows Server 2012 will RTM in August and be “available to customers worldwide through multiple channels in September”.  Yes, Windows Server 2012 will be GA before the desktop.  That’ makes sense actually, because the OEMs will need more time with the desktop OS than they will with the server OS.

Notes–The New Hyper-V Extensible Switch in WS2012

These are my notes from VIR307 at TechEd NA 2012, with Bob Combs.

image

Cloud Demands

  • Integration of security and protection
  • Ability to add traffic tools
  • Unification of management tools
  • Blur distinction between physical and virtual networks

My experience: anything manual (setting up VLANs, firewall rules, etc) prevents self-service, and slows down deployment of traditional service.  Also adds a great deal of complexity because traditional systems not designed for the scale (networks, rules, tenants) of cloud computing.  Anything done in software is automatable and flexible, and could make self-service possible.

The Hyper-V Switch

Replaces the virtual network.  Same GUI, does the same basic role (connect virtual network cards to physical networks) … but more:

image

Remember that the switch port is an attribute of the virtual network card.  Therefore your policy moves with your VM.

  • PVLAN: Use 2 VLAN addresses to scale out with flatter networks and achieve some level of isolation
  • Trunk Mode: The ability to allow multiple VLANs to go to a single VM port.
  • Port Mirroring: Forward traffic from one NIC to another. Could be useful for virtual IDS.
  • Unified Tracing: Trap packets for analysis.

Interesting note: if you apply a policy on an SR-IOV vNIC then the channel will be rerouted via the virtual switch to apply the policy, thus disabling SR-IOV for that vNIC.

Extensibility

3rd party extensions, not replacements, add extra stuff that MSFT doesn’t do in the switch.  The extensions have to be logo tested and certified to make them reliable.  The system is “open” to encourage growth in the ecosystem.  Already more networking extensions than vSphere has replacements.

Extensions

The extensions are filters – NDIS-based Windows Filtering Platform Providers.  Configuration is unique to each instance of an extensible switch on a machine.

3 types of extension:

  • Capture: inspect traffic and generate new traffic for report purposes, but cannot modify or drop traffic.  Can have multiple capture extensions.  They can be ordered/reordered.  inMon sFlow Traffic Monitoring is an example for monitoring virtual traffic.
  • Filtering: Can inspect (everything that capture can do), drop, modify, and insert packets.  5Nine Virtual Firewall v3.0 is an example. 
  • Forwarding: Direct traffic, defining destination(s) of each packet.  Forwarding extensions can capture and filter traffic.  Think of it as all encompassing.  What if you wanted the Hyper-V switch to look like another switch?  That’s what this type allows.  NEC OpenFlow for Hyper-V is an example of this.  The Cisco Nexus 1000V is another example. 

SCVMM

VMM agent can manage the extensions via extensions to VMM.

PowerShell

image

VM Based Traffic Tools

  • Can run in VM, as a switch extension, or as a host service
  • Monitoring Port copies traffic to VM
  • Traffic trunked to VM before distributing to other VMs (trunk mode on port)
  • Capture extension echo traffic to service
  • Extension pipe/tunnel traffic to a destination

Extensible Switch ETW Tracing Example

Tracing events.

image

Extensible Switch Unified Tracing Example

Capturing packets.

image

Summary of Extensible Switch Benefits

image

There were a bunch of demo videos featuring the partner extensions that I did not take notes on.

Interesting Bits of Windows Server 2012 Essentials Information For SBS Customers

The goal of the product is to give an economic server OS for small businesses, with the flexibility to choose between on-premises (such as Exchange) and cloud (such as Office365) solutions.

SBS 2011 Essentials becomes Windows Server 2012 Essentials, a similar product that scales up to 25 users and 50 devices.  A small percentage of SBS customers had more than 25 users.

The question is: will you migrate your customers to Office 365, another cloud, or find an alternative to Exchange?

Available Roles and Features

Out of the box, the following roles/features are available in WS2012 Essentials:

  • AD Certificate Services (installed/configured)
  • AD Domain Services (installed/configured)
  • AD Lightweight Directory Services
  • AD RMS (requires RMS CALs)
  • Application (installed/configured)
  • DHCP Server
  • DNS Server
  • Fax Server
  • File Services (installed/configured)
  • Network Policy and Access Services (installed/configured)
  • Print and Doc Services
  • Remote Desktop Services (installed/configured and requires RDS CALs)
  • UDDI Services
  • Web Services (installed/configured)
  • Windows Deploy Services
  • WSUS

Normally WS2012 Essentials does not require CALs.  The only exceptions are if you choose to use Remote Desktop Services (RDS – widely used) or Rights Management Services (RMS – rarely used).

Features described in the FAQ

Here are some features of Windows Server 2012 Essentials from the WS2012 Essentials FAQ:

  • Windows Phone App: Remote Web Access provides a stream-lined, touch-friendly browser experience for accessing applications and data from virtually anywhere they have an Internet connection using almost any device. Windows Server 2012 Essentials also provides an updated Windows Phone app and a new Metro app for Windows 8 clients, allowing users to intuitively connect to, search across, and access files and folders on the server.
  • VPN: Windows Server 2012 Essentials turns setting up virtual private networking (VPN) into a painless wizard-driven process of just a few clicks, and simplifies the management of VPN access for users. Client computers can leverage a VPN connection to remotely join the Essentials environment without the need to come into the office.
  • Choose our email solution: customers can take advantage of the same integrated management experience whether they choose to run an on-premises copy of Exchange Server, subscribe to a hosted Exchange service, or subscribe to Office 365.
  • Storage Spaces: [this] feature allows customers to aggregate the physical storage capacity of disparate hard drives, dynamically add hard drives, and to create volumes with specified levels of resilience.
  • Improved Backup: Windows Server 2012 Essentials can perform complete system backups and bare-metal restores of the server itself as well as the client computers connected to the network – now with support for volumes larger than 2 terabytes [thank VHDX for that!].
  • Windows Online Backup: The Microsoft Online Backup Service can be used to protect files and folders in a cloud-based storage service that is managed by Microsoft.
  • Windows 8 File History: Essentials also centrally manages and configures the new File History feature of Windows 8 clients, helping users to recover from accidently deleted or overwritten files without requiring administrator assistance.

Some Upgrade Paths from SBS with Software Assurance

  • If you have Software Assurance on Small Business Server 2011 Essentials, you will receive one Windows Server 2012 Essentials license.
  • If you have Software Assurance on Small Business Server 2011 Standard edition, you will receive one Windows Server 2012 Standard edition license and one Exchange Server Standard 2010 license.
  • If you have Software Assurance on Small Business Server 2011 Premium Add-on edition, you will receive one Windows Server 2012 Standard edition license and one SQL Server 2012 Standard edition license.

I’m no SBS expert, but it seems to me that existing customers with SA will retain the same functionality through the upgrade.  But I can envision there being a price jump if they are on annuity licensing for SBS Standard/Premium (we’ll have to see the price lists to judge it).  They will have to choose between existing on-premises product and cloud-based product.

Opinions of the SBS Community?

It’s the MSFT partners that are this community.  Based on the online comments, let’s just say they aren’t too pleased.  It’s early days so I’ve yet to hear from local partners.

Windows Server 2012 & 2012 R2 Virtualisation Licensing Scenarios

I am not answering any more questions on this post – to be honest, there have been too many for me to have the time to deal with them. Don’t bother asking – I’ll ignore/delete it.  My recommendations are:

  1. Re-read this post if you do not understand it after the first or second reads. To be honest, most of the questions have been from people who are just trying to make things complicated. Just license the hosts for the maximum number of Windows Server VMs that can ever run on that host, even for 1 second. It is that simple!
  2. Ask your LAR/distributor/reseller – that’s their job and that’s why you pay them

This post follows my dissertation on Windows Server 2012 licensing, which is essential reading before proceeding with this post.

[Edit] The below is unchanged with WS2012 R2. The only difference is that WS2012 R2 Datacenter edition (only) includes Automatic Virtual Machine Activation.

I’m putting Hyper-V aside for just a few minutes to talk about how you will license virtualisation of Windows Server 2012 in virtual machines no matter what virtualisation you use, be it vSphere, Hyper-V, XenServer, or whatever tickles your fancy.  BTW, the counting here also applies to:

  • System Center 2012
  • Core Infrastructure Suite (CIS)
  • Enrolment for Core Infrastructure (ECI – minimum of 25 hosts)

Please read this post s-l-o-w-l-y and let it sink in.  Then read it again.  If you have been eating bowls full of VMware FUD then read it a third time – slowly.

FAQ

  • VOSE or virtual operating system environment is a licensing term for virtual machine (VM).  It is used when talking about licensing a VM for Windows Server.
  • When you buy a license for virtualisation you legally buy and assign it to hardware, not to VMs.  The virtualisation rights of Windows Server licenses the VMs on the licensed host for Windows Server.
  • There is no mobility with OEM.
  • You can move a volume license and it’s virtualisation rights once every 90 days.  If you want to use HA (clustering), Live Migration/vMotion, DRS/Dynamic Optimization/PRO, then you need sufficient virtualisation rights on each host to support the maximum number of VMs that is possible on that host, even for 1 second.
  • You cannot split a license or it’s virtualisation rights across hosts.
  • Virtualisation rights are 2 VOSEs for a host licensed by Windows Server 2012 Standard and unlimited VOSEs for a host licensed for Windows Server 2012 Datacenter.
  • Virtualisation rights covers the host for the assigned edition of Windows Server 2012 and lower versions/editions of Windows Server.  It does not include Windows 8/7/Vista.
  • You can assign more than 1 license to a host

In other words, you license a host for the maximum number of Windows Server VMs that it could host.

1 Host, 1 CPU, 2 VMs

image

Here you want to run a single host that has 1 CPU.  The host will run 2 Windows Server virtual machines.  You will assign a single Windows Server 2012 license to this host.  The license covers 2 CPUs (there is only 1) and provides virtualisation rights for 2 virtual operating system environments (VOSEs).  In other words, you get rights to install Windows Server 2012 Standard (or previous versions) in 2 VMs on this host.

1 Host, 2 CPUs, 4 VMs

image

The Standard edition covers 2 VOSEs, but the customer wants 4 VMs running Windows Server Standard 2008 R2.  A single copy of WS2012 Standard will not suffice.  2 copies are bought to provide the 4 (2 * 2 VOSEs) VMs with licensing.

1 Host, 2 CPUs, 10 VMs

image

There are two options; do you go with the Standard or Datacenter editions of Windows Server 2012?

The Standard edition covers 2 VOSEs, but the customer wants 10 VMs running Windows Server Standard 2008 R2. A single copy of WS2012 Standard will not suffice. 5 copies are bought to provide the 10 (5 * 2 VOSEs) VMs with licensing.  Based on USA Open NL pricing the licensing of these VMs will cost $882 * 5 = $4,410.

The Datacenter edition of WS2012 gives unlimited VOSEs and covers 2 CPUs in the host.  This solution will require a single Windows Server 2012 Datacenter license which will cost $4,809.

Decision: If you will not go over 10 VMs on this host then Windows Server 2012 Standard edition is the way to go.  If you estimate that there is a good chance of the VM numbers growing then spend an extra $399 and pick up the easier to account-for Windows Server 2012 Datacenter with it’s unlimited VOSE rights.

10 is the magic number using USA Open NL pricing. Once you reach 10 VOSEs on a 1 or 2 CPU host, you need to consider the Datacenter edition because it is cheaper once you hit 11 VOSEs.

1 Host, 4 CPUs, 4 VMs

image

It’s an unusual configuration but a valid one for the demonstration.  The WS2012 Standard/Datacenter SKUs cover 2 CPUs each.  In this case there are 4 CPUs.  This will require 2 copies of Windows Server 2012 Standard, which also covers the 4 VMs.

Let’s pretend that 300 VMs will run on this host with 4 physical CPUs.  Then we would assign 2 copies of Windows Server Datacenter on it.  2 copies will cover 2 CPUs each (4 CPUs) and unlimited VOSEs.

That Host with 320 Logical Processors – 10 CPUs with 16 Cores with Hyperthreading

image

The maximum specification for Windows Server 2012 Hyper-V is 320 logical processors in the host.  That could be 10 Intel CPUs, each with 16 cores, with Hyperthreading enabled.  We don’t count cores or logical processors when we license.  We count CPUs, sockets, or plain old processors – pick the term you prefer.  There are 10 CPUs/sockets/processors in this server.  That requires 5 copies of either Windows Server 2012 Standard or Datacenter, depending on the required number of VOSEs.

Clusters

Let’s move on to the clusters, where people usually get things wrong because they don’t understand (or don’t want to understand) the mobility rights.  VOSEs licensed by OEM cannot move.  VOSEs licensed by VL can move once every 90 days.  The correct solution is to license each host for the maximum number of VOSEs that it can have for even one second.  And when I say “correct” I mean legal.  Software Asset Management professionals (auditors) are not stupid and the “tricks” I hear people proposing are neither original or unknown to these auditors.

Reminder: This applies even to you non-Hyper-V folks.

2 Hosts, 1 CPU each, 4 VMs

 

 

 

image

Don’t get fooled!  This is not one host with 3 VMs and 1 host with 1 VM.  This is 2 hosts, each of which can have up to 4 VMs.  In the past we would have used Enterprise edition on each host.  That has been replaced by Windows Server 2012 Standard edition, that now has all the features and scalability of the Datacenter edition.

Take each host and size it for 4 VOSEs.  That means we need to assign 2 copies of Windows Server 2012 Standard edition to each host.  That’s 4 copies of WS2012 Standard.

2 Hosts, 2 CPUs each, 10 VMs

image

10 VMs with 2 hosts means that it is possible to have all 10 VMs on a single host.  You have two options to license each host for up to 10 VOSEs.

Firstly you could license each of the hosts with 5 copies of Windows Server 2012 Standard.  That will give you 10 (5 * 2) VOSEs.  This requires 10 (5 * 2 hosts ) copies of Standard at a cost of $8,820 using USA Open NL.

Alternatively you could license each host with 1 copy of Windows Server 2012 Datacenter, at a cost of $9,618.  The extra $798 will allow you to burst beyond 10 VOSEs to unlimited VOSEs.  Switching to licensing hosts using the Datacenter edition means we don’t have to count VOSEs and we have unrestricted mobility between hosts.

2 Hosts, 2 CPUs each, 20 VMs

image

We have exceed the magic number of 10, and now it is cheaper to license with the Datacenter edition with it’s unlimited VOSEs per host.  Each host has 2 CPUs, so each host requires 1 copy of Datacenter.  There are 2 hosts so we require 2 copies of Windows Server 2012 Datacenter.

You could add more hosts to this cluster and each could have unlimited VMs.  As long as the hosts have 1 or 2 CPUs each, each additional host requires only 1 copy of Windows Server 2012 Datacenter to license it for unlimited installs of Windows Server for the VMs on that host.

Lots of Hosts, Lots of VMs, 4 CPUs per Host

image

The magic number of 10 VOSEs is a dot in the rear view mirror.  We now have lots of hosts with lots of VMs flying all over the place.  Each host has 4 CPUs.  To license the VOSEs on each host, we will require licensing for 4 CPUs.  This will require 2 copies of Windows Server 2012 Datacenter per host, each covering 2 CPUs.

Live Migration Outside the Cluster

And new for WS2012 thanks to Live Migration living outside the cluster: you must ensure that the destination host is adequately licensed for VOSEs to accommodate the new VM. If this is an infrequent move then you could avail of the VL 90 day mobility right to reassign a license, ensuring the the old host is sufficiently licensed for remaining VOSEs and physical CPUs.

Hyper-V Server 2012

Hyper-V Server has no virtualisation rights and includes no free licensing for VOSEs.  Therefore it is irrelevant in this conversation.

Economically Speaking – Why Hyper-V Makes Sense

If you buy Windows Server licensing for a host to license your VMs, then you are a tickbox (or PowerShell cmdlet) and a reboot away from having Hyper-V.  Buying another product is just more money spent.  And let’s be honest, Windows Server 2012 Hyper-V is not what you might have used/looked at before.

Windows Server 2012 Licensing In Detail

New, Simpler, Better Value Licensing Announced For WS2012 and Some SBS Changes

Note: I saw the announcement when it first came out and have been waiting & digesting.  More information came out since then so I thought a single post would be best. Make sure you follow this post by reading my post on licensing for virtualisation of Windows Server 2012.  I will not be answering any further questions on this post. Please ask your reseller, distributor, or LAR your scenario specific questions; that’s why you pay them.

In case you don’t know, I happen to work for a distributor of Microsoft licensing. My job is to work with our sales people, supporting Microsoft partners who resell product to end customers. My focus is on System Center (and that brings in Hyper-V) and Forefront, but anything that is anyway technical tends to find its way to my desk. And you know what? I’ve been amazed at the complexity that is involved. Some questions are part licensing/legal and part technical. And these issues confuse the hell out of people. And trust me; I’ve let well known executives in Microsoft know what I thought of the complexity.

The recent changes to System Center 2012 licensing simplified our conversations quite a bit. To use a Henry Ford analogy, System Center SMLs come in 2 sizes and in black:

  • SML Standard: 2 CPUs in the physical server, all of System Center, licensing 2 physical CPUs in the server + 2 VMs running Windows Server on the licensed host
  • SML Datacenter: 2 CPUs in the physical server, all of System Center, licensing 2 physical CPUs in the server + unlimited VMs running Windows Server on the licensed host

There is nothing to restrict you from creating hundreds of VMs on WinServ Std. The VOSE rights restrict your rights to install Windows Server. Make sure you follow this post by reading my post on licensing for virtualisation of Windows Server 2012.

    Windows Server 2008 R2 SKUs

    Currently on the Windows Server front it’s more like 2005 General Motors than early Ford. We have the following SKUs in Windows Server:

    1. Windows Server 2008 R2 Datacenter
    2. Windows Server 2008 R2 Enterprise
    3. Windows Server 2008 R2 Standard
    4. Windows Web Server 2008 R2
    5. Windows HPC Server 2008 R2 Suite
    6. Windows Server 2008 R2 HPS Edition
    7. Microsoft HPC Pack 2008 R2 Enterprise
    8. Windows Server 2008 R2 for Itanium-Based Systems
    9. Windows Small Business Server 2011 Premium
    10. Windows Small Business Server 2011 Standard
    11. Windows Small Business Server 2011 Essentials
    12. Windows Server 2008 R2 Foundation
    13. Windows Storage Server 2008 R2
    14. Windows Home Server 2011

    Phew! No wonder people get confused.

    Windows Server 2012 SKUs

    This evening Microsoft released the details of Windows Server 2012 licensing. There is also a datasheet and a FAQ with more details.  There will be 4 (count ‘em … four) 5 (count ‘em again … five) SKUs:

    clip_image002

    See EDIT#2 below … Windows Storage Server will also be released.

    Please, do not panic if you are a HPC customer or an Enterprise customer. Here’s why:

    • HPC will be a free download/add-on for Standard/Datacenter customers
    • Standard will have the same scalability and features as the Datacenter edition … yes, you can build clusters with Windows Server 2012 Standard … and at a fraction of the cost that you did it with Enterprise edition (as you’ll see next)!

    You see, IT Pros, not all change is bad J [Did I just say that?]

    Pricing & Changes

    I want to focus on the 2 computer room/data centre products for a while: Datacenter and Standard. What were the shelf prices before with 2008 R2?  Bear in mind that the shown prices are the from the Open NL price list, the most expensive of the volume license types.  They are shown to give an indication of past and present.  What you will pay your reseller/LAR/distributor will probably be less:

    clip_image004

    So, Datacenter cost $4,809 (for 2 copies – you’ll see why in a sec) and came with unlimited virtualisation rights on the host … or in simple language you could install as many VMs on that licensed host as your hardware would allow, and then install Windows Server 2008 R2 or lower, any edition, into those VMs at no extra cost. The funny thing with Datacenter was that it was per processor, with a minimum of 2 processors … hence I’ve shown $4,809 per host with 2 procs.

    Enterprise came with all the same features and scalability, with a limit of technical limit of 8 CPUs in the physical server. Licensing-wise, it allowed 4 VMs on the licensed host to run Windows for free. You could double the licensing for the host to get 8 VMs for free … but do the math and you might as well buy Datacenter edition then. However, Enterprise was $2,358 per server with no processor counting required.

    Standard Edition had limited scalability and features, e.g. 32 GB RAM even for Hyper-V. At $726 you licensed a server, and it came with 1 free license for a VM on that host.

    Enter Windows Server 2012 and we have:

    clip_image005

    For Datacenter there is 1 change: You buy 1 copy of Datacenter and it includes 2 processors. The whole minimum-of-2-procs-per-host thing confused people. So Microsoft consolidated the 2 licenses into 1 and left the price at $4,809. We continue to have the unlimited virtualisation licensing. That means if you buy 1 copy of Datacenter for a 2 CPU host (any virtualisation) and you get unlimited installations of Windows Server 2012 or lower (any edition) for unlimited VMs on that host … no changes there!

    clip_image007

    Let’s move on to Standard before we cover the gap that was Enterprise. Standard is going from $726/server to $882 for 2 physical CPUs in a server. For that, you’re getting:

    • A huge leap in capabilities and scalability in the Standard edition
    • Rights to an additional copy of Windows for a second VM running on the physical box that is licensed using Windows Server 2012 Standard.

    The other change for Standard is that it will match the Datacenter model. The Standard license will cover you for 2 CPUs in a server. If you want 4 CPUs, then you need 2 copies of Standard. I do not envision that being a problem because I’ve never seen Standard installed on anything more than 2 CPUs.

    If you’re about to complain because you have a bunch of Standard edition physical servers, the price is going up, and you don’t need the additional features/scalability, then I will respond with: you’ve been throwing money away for years and your boss should be asking you some questions like “Why didn’t we virtualise those servers 3 years ago?”  If you virtualise those Standard Edition servers now then you can merge them into Datacenter per host licensing and save on electricity, support, and lots of other costs over your 3 year cycles.

    Windows Server 2012 Foundation is OEM only and therefore there is no Open NL pricing.  Foundation is also for single CPU servers, such as a micro-server.  More on Essentials later.

    Replacing the Enterprise Edition

    OK Enterprise owners, just like the old Star Trek storyline, it is g-o-n-e. Here is your replacement strategy:

    • If you use to buy Enterprise for clustering, then use Standard instead. You just saved $1,476 per node!
    • If you bought Enterprise to have virtualisation rights for 4 VMs on a host, then buy 2 Standards for that host … hey! 2008 R2 Enterprise cost $2,358 and 2 copies of WS2012 Standard costs $1,764. You just saved $594 per host!

    Where Is Small Business Server (SBS)?

    The answer to that question was posted on the SBS blog.  SBS has been replaced by Windows Server 2012 Essentials.   Windows Server 2012 Essentials will support up to 25 users and 50 devices.  Essentials is for 1 or 2 CPU servers and cost $425 on Open NL.  The idea behind Essentials is that it will be the successor to SBS.  Quoting the blog, that also features a Windows Server 2012 Essentials FAQ:

    Windows Server 2012 Essentials has been designed to give you the flexibility to choose which applications and services run on-premises and which run in the cloud. In contrast to Windows SBS Standard, Essentials offers lower up-front acquisition and deployment costs. It allows you to take advantage of cloud-based messaging offerings while enjoying an integrated management experience by subscribing to Office 365 or a hosted Exchange service. If you prefer a fully on-premises solution, you have the option of running Exchange Server on a second server (either as a physical or virtual machine) alongside Essentials with the same integrated management experience.

    Windows Server 2012 Essentials can also be used as a platform to run line-of-business applications and other on-premises workloads, as well as to provide an integrated management experience when running cloud-based applications and services, such as email, collaboration, online backup, and more.

    Windows Server 2012 Essentials can also be used as a platform to run line-of-business applications and other on-premises workloads, as well as to provide an integrated management experience when running cloud-based applications and services, such as email, collaboration, online backup, and more.

    Windows SBS 2011 Standard, which includes Exchange Server and SharePoint Foundation, will be the final such Windows SBS offering. It will remain available through the OEM channel until December 31, 2013, and will remain available in all other current channels until June 30, 2013.

    Long-story short, the small business customer is now getting a 25 user version of Windows that does not come with Exchange.  If they want Email/collaboration/chat then Microsoft is selling Office 365.  And of course, the partner is free to sell/install something else on the Essentials server, and the customer is also free to buy/install another product on the Essentials server.  They are getting a very cheap server that requires no CALs, and that’s a nice first-server starting point for a cash-strapped small business.

    My analysis on this: the writing has been on the wall for a long time.  At least locally, Microsoft has made huge investments in trying to educate train partners on the strategy, sales, and technical levels.  We have to move with the times cos the times are moving.

    Some upgrade paths for SBS:

    • If you have Software Assurance on Small Business Server 2011 Essentials, you will receive one Windows Server 2012 Essentials license.
    • If you have Software Assurance on Small Business Server 2011 Standard edition, you will receive one Windows Server 2012 Standard edition license and one Exchange Server Standard 2010 license.
    • If you have Software Assurance on Small Business Server 2011 Premium Add-on edition, you will receive one Windows Server 2012 Standard edition license and one SQL Server 2012 Standard edition license.

    This upgrade right will be reflected upon your agreement renewal but you are entitled to use the granted product upon its availability.

    EDIT: I’ve another post on the SBS story.

    Déjà Vu

    Those of you who understand System Center 2012 licensing have thought “that looks familiar!” It should because the model matches the Datacenter/Standard license model of System Center. And that makes the Enrollment for Core Infrastructure (ECI) and Core Infrastructure Suite (CIS) licenses much easier to understand:

    clip_image009

    Buying Windows Server and a Standard System Center SML? Then buy a Standard edition of a CIS or ECI suite. Buying the Datacenter editions? Then buy the Datacenter edition of the ECI or CIS suites instead … and save even more money.

    What happens if you have Software Assurance such as Open + SA, OVS, or ESA? Well there are migrations from the old SKUs to the new SKUs:

    clip_image010

    The 2:1 Datacenter edition makes sense because you’re going from 2 * 1 CPU licenses per host to 1 * 2 CPU license with no price change.

    Remember that HPC is replaced by Windows Server 2012 Standard/Datacenter plus the free HPC download. And Enterprise folks don’t lose out because they get the equivalent (features + virtualisation rights) of 2 copies of Standard (4 licensed VMs + 4 CPUs in the server).

    Anyone running physical installations of Windows Web Server 2008 R2: this is your prompt to join the rest of us in the 21st century and virtualize those web servers … NOW! If you put 2 web servers on to a host, your new Windows Server 2012 Standard edition covers you for your virtual web server licensing.

    Editions Comparison

    Here are how the Windows Server 2012 editions compare against each other:

    image

    image

    Here is a feature comparison:

    image

    And here is how you can buy the different WS2012 editions:

    image

    Step-Up

    For those of you running one edition and would like to upgrade to another (Standard -> Datacenter) then you can do that if you have Software Assurance. The cost is the price difference between the editions. Note that you can step up from 1 Enterprise 2008 R2 to 2 copies of Datacenter 2008 R2 now and then you get WS2012 Datacenter edition.

    For Those Without Software Assurance

    What if you don’t have Software Assurance on your servers?

    • Attend my presentations more often and/or add my site to your RSS feeds because … I told you so! [Heh! I was right; I am one to say “I told you so”]
    • If you bought your software in the past 90 days then still can attach it; contact your reseller/LAR for details.

    You will continue to use the version and SKU that you bought, and you’ll miss all that lovely WS2012 goodness that the rest of us are salivating over. [Have I mentioned that I am a nerd?]

    Client Access Licenses (CALs)

    None of the rules change. You continue to license clients for the highest version of Windows Server that they use. For example, you could run 10 W2008 R2 VMs on a WS2012 host. You then use W2008 R2 CALs. If you upgrade a single one of those VMs to WS2012 then you need WS2012. Seriously, Software Assurance on CALs makes sense.

    RDS CAL licensing is not changing.  Anyone running either Web (only) or HPC workloads (only) on their servers do not require CALs.

    Neither Window Server 2012 Essentials or Windows Server 2012 Foundation require CALs.

    License Mobility

    No changes to report here either. OEM is tied to the machine – that’s why it’s cheaper. And a volume license can be moved once every 90 days. And that applies to you folks who think they are able to under-license their hosts (even VMware and Xen) for VMs; you have to license for the maximum number of Windows VMs on that host, even for 1 second. You can’t license 20 VMs with Standard Edition and VMotion/DRS them about every 5 minutes – mobility rules say you can move them once every 90 days because legally you license the host and use the virtualisation rights of that Standard SKU to license the VMs. The correct way to license is to stack your Standard editions on each of the hosts (allowing for the highest possible number of VMs, even for 1 second) or buy Datacenter (which makes sense once you need more than 10 VMs per host based on this retail pricing).

    EDIT#1: Windows Home Server

    Mary Jo Foley confirmed that there will be no more editions of Windows Home Server.  I guess I’ll be rebuilding my Microserver with Windows 8 and setting up a Storage Space.

    EDIT#2: Windows Storage Server

    I’ve just had it confirmed by Microsoft that there will be a Windows Storage Server 2012, giving us 5 SKUs going forward.

    Summary

    Watch out for the FUD that is sure to appear in blogs and forums, and the occasional “journalist” (like the ones I love to crucify on this blog from time to time). I’m sure the cynics and competitors will spin things and misquote pieces of Microsoft’s text on the changes. Before you make any decisions, read Microsoft’s original text (URL to be added) and then check with LAR or reseller … and check with another reseller if yours is a VMware fanboy because I’m sure they might have eaten the FUD.

    I think the only genuine confusion will be that these changes and savings will sound too good to be true. Seriously – Windows Server licensing is changing, simplifying, and you’re saving money.  I’m bringing my ice skates with me when I go to hell.  It’s a win/win – with some concerns for SBS folks maybe.

    The licensing couldn’t be simpler in the data centre. It comes in Standard or Datacenter. They both come with all the scalability and all the features, including Hyper-V. And the free Hyper-V Server 2012 comes with all the features and scalability of Hyper-V on Windows Server 2012 Hyper-V. I bet the competition’s licensing isn’t this simple, offer as good value, and include all the features in all the editions. [Stop it!]

    Make sure you follow this post by reading my post on licensing for virtualisation of Windows Server 2012.

    Notes–Enabling Disaster Recovery for Hyper-V Workloads Using Hyper-V Replica

    I’m taking notes from VIR302 in this post.  I won’t be repeating stuff I’ve blogged about previously.

    image

    Outage Information in SMEs

    Data from Symantec SMB Disaster Preparedness Survey, 2011.  1288 SMBs with 5-1000 employees worldwide.

    • Average number of outages per year? 6
    • What does this outage cost per day? $12,500

    That’s an average cost of $75,000 per year!  To an SME!  That could be 2 people’s salary for a year.

    • % That do not have a recovery plan: 50%.  I think more business in this space don’t have DR.
    • What is their plan? Scream help and ask for pity.

    Hyper-V Replica IS NOT Clustering And IT IS NOT a Cluster Alternative

    Hyper-V Replica IS ALSO NOT Backup Replacement

    It is a replication solution for replicating VMs to another site.  I just know someone is going to post a comment asking if they can use it as a cluster alternative [if this is you – it will be moderated to protect you from yourself so don’t bother.  Just re-read this section … slowly].

    • Failover Clustering HA: Single copy, automated failover within a cluster.  Corruption loses the single copy.
    • Hyper-V Replica: Dual asynchronous copy with recent changes, manual failover designed for replication between sites.  Corruption will impact original immediately and DR copy within 10 minutes.
    • Backup: Historical copy of data, stored locally and/or remotely, with the ability to restore a completely corrupted VM.

    Certificates

    For machines that are non-domain joined or non-trusted domain members.  Hoster should issue certs to the customer in the hosted DR scenario. 

    Compression

    Can disable it for WAN optimizers that don’t work well with pre-optimised traffic.

    Another Recovery History Scenario

    The disaster brought down VMs at different points.  So VMA died at time A and VMB died at time C.  Using this feature, you can reset all VMs back to time A to work off of a similar set of data.

    You can keep up to 15 recovery points per day.  Each recovery point is an hour’s worth of data. 

    The VSS option (application consistent recovery) fires every two hours.  Every 2nd hour (or whatever depending on where you set the VSS slider) in the cycle it triggers VSS.  All the writes in the guest get flushed.  That replica is then sent over.

    Note that the Hyper-V VSS action will not interfere with backup VSS actions.  Interoperability testing has been done.

    So if you’re keeping recovery snapshots, you’ll have standard replicas and application consistent (VSS) replicas.  They’ll all be an hour apart, and alternating (if every 2nd hour).  Every 5 minutes the changes are sent over, and every 13th one is collapsed into a snapshot (that’s where the 1 hour comes from).

    Every 4 hours appears to be the sweet spot because VSS does have a performance impact on the guests.

    Clusters

    You can replicate to/from clusters.  You cannot replicate from one node to another inside a cluster (can’t have duplicate VM GUIDs and you have shared storage).

    Alerting

    If 20% of cycles in the last hour are missed then you get a warning.  This will self-close when replication is healthy again. 

    PowerShell

    24 Hyper-V Replica cmdlets:

    • 19 of them via get-command –Module hyper-v | where {$_.Name –like “*replication*”}
    • 5 more via get-command –Module hyper-v | where {$_.Name –like “*failover*”}

    Measure-VMReplication will return status/health of Hyper-V Replica on a per-VM basis.

    Measure-VMReplication | where {$_.ReplicationHealth –eq “Critical”}

    Could use that as a part of a scheduled script, and then send an email with details of the problem.

    Replica Mechanism

    Refers to the HRL (Hyper-V Replica Log) process as a write splitter.  They use HTTP(s) for WAN traffic robustness.  It’s also hosting company friendly.  The HRL is swapped out before sending for a new HRL.

    There is a threshold where the HRL cannot exceed half the VHD size.  If WAN/storage goes down and this happens then HVR goes into a “resync state” (resynchronisation).  When the problem goes away HVR automatically re-establishes replication. 

    VM Mobility

    HVR policy follows the VM with any kind of migration scenario.  Remember that replication is host/host.  When the VM is moved from host A to host B, replication for the VM from host A is broken.  Replication for the VM starts on host B.  Host B must be already authorized on the replica host(s) – easier with cluster Hyper-V Replica broker. 

    IP Addressing VMs In DR Site

    1. Inject static address – Simplest option IMO
    2. Auto-assignment via DHCP – Worst option IMO because DHCP on servers is messy
    3. Preserve IP address via Network Virtualisation – Most scalable option for DR clouds IMO with seamless failover for customers with VMs on a corporate WAN.  Only one for seamless name resolution, I think, unless you spend lots on IP virtualisation in the WAN.

    Failover Types

    Planned Failover (downtime during failover sequence):

    1. Shutdown primary VM
    2. Send last log – run planned failover action from primary site VM.  That’ll do the rest for us.
    3. Failover replica VM
    4. Reverse replication

    Test Failover (no downtime):

    Can test any recovery point without affecting replication on isolated test network.

    1. Start test failover, selecting which copy to test with (if enabled).  It does the rest for you.
    2. Copies VM (new copy called “<original VM name> – test”) using a snapshot
    3. Connects VM to test virtual switch
    4. Starts up test VM

    Network Planning

    • Capacity planning is critical.  Designed for low bandwidth
    • Estimate rate of data change
    • Estimate for peak usage and effective network bandwidth

    My idea is to analyse incremental backup size, and estimate how much data is created every 5 minutes.

    Use WS2012 QoS to throttle replication traffic.

    image

    Replicating multiple VMs in parallel:

    • Higher concurrency leads to resource contention and latency
    • Lower concurrency leads to underutilizing and less protection for the business

    Manage initial replication through scheduling.  Don’t start everything at once for online initial synchronisation.

    What they have designed for:

    image

     

    Server Impact of HVR

    On the source server:

    • Storage space: proportional to the writes in the VM
    • IOPS is approx 1.5 times write IOPS

    On the replica server:

    • Storage space: proportional to the write churn.  Each additional recovery point approx 10% of the base VHD size.
    • Storage IOPS: 0.6 times write IOPS to receive and convert.  3-5 times write IOPS to receive, apply, merge, for additional recovery points.
    • There is a price to pay for recovery points.  RECOMMENDATION by MSFT: Do not use replica servers for normal workloads if using additional recovery points because of the IOPS price.

    Memory: Approx 50 MB per replicating VM

    CPU impact: <3%

    Notes: Continuously Available File Server – Under The Hood

    Here are my notes from TechEd NA session WSV410, by Claus Joergensen.  A really good deep session – the sort I love to watch (very slowly, replaying bits over).  It took me 2 hours to watch the first 50 or so minutes 🙂

    image

    For Server Applications

    The Scale-Out File Server (SOFS) is not for direct sharing of user data.  MSFT intend it for:

    • Hyper-V: store the VMs via SMB 3.0
    • SQL Server database and log files
    • IIS content and configuration files

    Required a lot of work by MSFT: change old things, create new things.

    Benefits of SOFS

    • Share management instead of LUNs and Zoning (software rather than hardware)
    • Flexibility: Dynamically reallocate server in the data centre without reconfiguring network/storage fabrics (SAN fabric, DAS cables, etc)
    • Leverage existing investments: you can reuse what you have
    • Lower CapEx and OpEx than traditional storage

    Key Capabilities Unique to SOFS

    • Dynamic scale with active/active file servers
    • Fast failure recovery
    • Cluster Shared Volume cache
    • CHKDSK with zero downtime
    • Simpler management

    Requirements

    Client and server must be WS2012:

    • SMB 3.0
    • It is application workload, not user workload.

    Setup

    I’ve done this a few times.  It’s easy enough:

    1. Install the File Server and Failover Clustering features on all nodes in the new SOFS
    2. Create the cluster
    3. Create the CSV(s)
    4. Create the File Server role – clustered role that has it’s own CAP (including associated computer object in AD) and IP address.
    5. Create file shares in Failover Clustering Management.  You can manage them in Server Manager.

    Simple!

    Personally speaking: I like the idea of having just 1 share per CSV.  Keeps the logistics much simpler.  Not a hard rule from MSFT AFAIK.

    And here’s the PowerShell for it:

    image

    CSV

    • Fundamental and required.  It’s a cluster file system that is active/active.
    • Supports most of the NTFS features.
    • Direct I/O support for file data access: whatever node you come in via, then Node 2 has direct access to the back end storage.
    • Caching of CSVFS file data (controlled by oplocks)
    • Leverages SMB 3.0 Direct and Multichannel for internode communication

    Redirected IO:

    • Metadata operations – hence not for end user data direct access
    • For data operations whena  file is being accessed simultaneously by multiple CSVFS instances.

    CSV Caching

    • Windows Cache Manager integration: Buffered read/write I/O is cached the same way as NTFS
    • CSV Block Caching – read only cache using RAM from nodes.  Turned on per CSV.  Distributed cache guaranteed to be consistent across the cluster.  Huge boost for polled VDI deployments – esp. during boot storm.

    CHDKDSK

    Seamless with CSV.  Scanning is online and separated from repair.  CSV repair is online.

    • Cluster checks once/minute to see if chkdsk spotfix is required
    • Cluster enumerates NTFS $corrupt (contains listing of fixes required) to identify affected files
    • Cluster pauses the affected CSVFS to pend I/O
    • Underlying NTFS is dismounted
    • CHKDSK spotfix is run against the affected files for a maximum of 15 seconds (usually much quicker)  to ensure the application is not affected
    • The underlying NTFS volume is mounted and the CSV namespace is unpaused

    The only time an application is affected is if it had a corrupted file.

    If it could not complete the spotfix of all the $corrupt records in one go:

    • Cluster will wait 3 minutes before continuing
    • Enables a large set of corrupt files to be processed over time with no app downtime – assuming the apps’ files aren’t corrupted – where obviously the would have had downtime anyway

    Distributed Network Name

    • A CAP (client access point) is created for an SOFS.  It’s a DNS name for the SOFS on the network.
    • Security: creates and manages AD computer object for the SOFS.  Registers credentials with LSA on each node

    The actual nodes of the cluster nodes are used in SOFS for client access.  All of them are registered with the CAP.

    DNN & DNS:

    • DNN registers node UP for all notes.  A virtual IP is not used for the SOFS (previous)
    • DNN updates DNS when: resource comes online and every 24 hours.  A node added/removed to/from cluster.  A cluster network is enabled/disabled as a client network.  IP address changes of nodes.  Use Dynamic DNS … a lot of manual work if you do static DNS.
    • DNS will round robin DNS lookups: The response is a list of sorted addresses for the SOFS CAP with IPv6 first and IPv4 done second.  Each iteration rotates the addresses within the IPv6 and IPv4 blocks, but IPv6 is always before IPv4.  Crude load balancing.
    • If a client looks up, gets the list of addresses.  Client will try each address in turn until one responds.
    • A client will connect to just one cluster node per SOFS.  Can connect to multiple cluster nodes if there are multiple SOFS roles on the cluster.

    SOFS

    Responsible for:

    • Online shares on each node
    • Listen to share creations, deletions and changes
    • Replicate changes to other nodes
    • Ensure consistency across all nodes for the SOFS

    It can take the cluster a couple of seconds to converge changes across the cluster.

    SOFS implemented using cluster clone resources:

    • All nodes run an SOFS clone
    • The clones are started and stopped by the SOFS leader – why am I picturing Homer Simpson in a hammock while Homer Simpson mows the lawn?!?!?
    • The SOFS leader runs on the node where the SOFS resources is actually online – this is just the orchestrator.  All nodes run independently – moving or crash doesn’t affect the shares availability.

    Admin can constrain what nodes the SOFS role is on – possible owners for the DNN and SOFS resource.  Maybe you want to reserve other nodes for other roles – e.g. asymmetric Hyper-V cluster.

    Client Redirection

    SMB clients are distributed at connect time by DNS round robin.  No dynamic redistribution.

    SMB clients can be redirected manually to use a different cluster node:

    image

    Cluster Network Planning

    • Client Access: clients use the cluster nodes client access enable public networks

    CSV traffic IO Redirection:

    • Metadata updates – infrequent
    • CSV is built using mirrored storage spaces
    • A host loses direct storage connectivity

    Redirected IO:

    • Prefers cluster networks not enabled for client access
    • Leverages SMB Multichannel and SMB Direct
    • iSCSI Networks should automatically be disabled for cluster use – ensure this is so to reduce latency.

    Performance and Scalability

    image

    image

    SMB Transparent Failover

    Zero downtime with small IO delay.  Supports planned and unplanned failovers.  Resilient for both file and directory operations.  Requires WS2012 on client and server with SMB 3.0.

    image

    Client operation replay – If a failover occurs, the SMB client reissues those operations.  Done with certain operations.  Others like a delete are not replayed because they are not safe.  The server maintains persistence of file handles.  All write-throughs happen straight away – doesn’t effect Hyper-V.

    image

    The Resume Key Filter fences off file handles state after failover to prevent other clients grabbing files when the original clients expect to have access when they are failed over by the witness process.  Protects against namespace inconsistency – file rename in flight.  Basically deals with handles for activity that might be lost/replayed during failover.

    Interesting: when a CSV comes online initially or after failover, the Resume Key Filter locks the volume for a few seconds (less than 3 seconds) for a database (state info store in system volume folder) to be loaded from a store.  Namespace protection then blocks all rename and create operations for up to 60 seconds to allow for local file hands to be established.  Create is blocked for up to 60 seconds as well to allow remote handles to be resumed.  After all this (up to total of 60 seconds) all unclaimed handles are released.  Typically, the entire process is around 3-4 seconds.  The 60 seconds is a per volume configurable timeout.

    Witness Protocol (do not confuse with Failover Cluster File Share Witness):

    • Faster client failover.  Normal SMB time out could be 40-45 seconds (TCP-based).  That’s a long timeout without IO.  The cluster informs the client to redirect when the cluster detects a failure.
    • Witness does redirection at client end.  For example – dynamic reallocation of load with SOFS.

    Client SMB Witness Registration

    1. Client SMB connects to share on Node A
    2. Witness on client obtains list of cluster members from Witness on Node A
    3. Witness client removes Node A as the witness and selects Node B as the witness
    4. Witness registers with Node B for notification of events for the share that it connected to
    5. The Node B Witness registers with the cluster for event notifications for the share

    Notification:

    1. Normal operation … client connects to Node A
    2. Unplanned failure on Node A
    3. Cluster informs Witness on Node B (thanks to registration) that there is a problem with the share
    4. The Witness on Node B notifies the client Witness that Node A went offline (no SMB timeout)
    5. Witness on client informs SMB client to redirect
    6. SMB on client drops the connection to Node A and starts connecting to another node in the SOFS, e.g. Node B
    7. Witness starts all over again to select a new Witness in the SOFS. Will keep trying every minute to get one in case Node A was the only possibility

    Event Logs

    All under Application and Services – Microsoft – Windows:

    • SMBClient
    • SMBServer
    • ResumeKeyFilter
    • SMBWitnessClient
    • SMBWitnessService

    Application Compatibility and API Support for SMB 3.0, CSVFS, and ReFS

    Microsoft just published this document with details on compatibility for SMB 3.0, CSVFS (cluster shared volume for Hyper-V and SOFS), and the new server file system ReFS.

    The Application Compatibility with Resilient File System document provides an introduction to Resilient File System (ReFS) and an overview of changes that are relevant to developers interested in ensuring application compatibility with ReFS. The File Directory Volume Support spreadsheet provides documentation for APIs support for SMB 3.0, CSVFS, and ReFS that fall into the following categories: file management functions, directory management functions, volume management functions, security functions, file and directory support codes, volume control code, and memory mapped files.

    It is very much aimed towards developers.  There is a little bit of decipherable text in there to describe what ReFS is, something MSFT is not talking about much, not even at TechEd.  My take so far: it’s a file system for the future that will eventually supplant NTFS.

    Sections 1.1-1.3 are interesting to us IT Pros, then jump ahead to section 1.11.

    Technorati Tags: ,