myTemplate

I've been passed a press release regarding executive changes at PHD Virtual. The full press release to follow below:

Virtual Machine Backup Leader, PHD Virtual, Names Thomas Charlton Chairman and CEO 

The Pioneer of Virtual Backup Appliances Adds Technology Management Expert to Corporate Team 

MOUNT ARLINGTON, N.J. – March 10, 2010 — PHD Virtual Technologies, award winning provider of esXpress VM Backup, the fastest multi-VM backup and restore solution on the market, today announced that Thomas Charlton has been appointed Chairman and CEO by the PHD Virtual board of directors. Charlton has more than 20 years of leadership experience in emerging technology ventures, leading past companies to increased profitability and successful acquisitions.

“We are pleased to add Thomas to PHD Virtual’s corporate structure,” said Michael Triplett, managing director, Insight Venture Partners and PHD Virtual board member. “Thomas brings a wealth of company management expertise and innovation to PHD Virtual from his years of experience.  His amazing track record in corporate growth and visionary thinking will help continue PHD Virtual’s dramatic growth.”

Prior to joining PHD Virtual, Charlton was the CEO of multiple software companies, including Shunra Software (network emulation and appliances), VoiceGenie Technologies (Voice XML speech platform) and Trailblazer Systems (eCommerce EDI software). Responsible for each company’s strategic direction, revenue growth, profitability and global expansion, Charlton led Shunra and VoiceGenie to profitability, and led VoiceGenie and Trailblazer Sytems to successful acquisitions by Alcatel and Nu Bridges respectively. Charlton also served as CEO at Tidal Software (enterprise job scheduling) which was recently acquired by Cisco Systems, Inc.

“I am excited about the opportunity that PHD Virtual represents based on the innovative technology it has built to address the growing data protection needs of the virtualization market,” said Charlton.  “Customers with virtualized environments cannot adequately protect their growing information through traditional data protection solutions. Virtual server environments need technology that has been purpose-built to meet the unique requirements of virtual machines. PHD Virtual is the only technology to be designed specifically for a virtual environment with the performance and scalability enterprise customers require. It also delivers the unique distinction as a virtual solution that integrates easily with a customers’ physical storage environment for true end-to-end data protection.”  

About PHD Virtual Technologies
The fastest multi-VM backup on the market and pioneer of Virtual Backup Appliances (VBAs), PHD Virtual Technologies has been transforming data protection for VMware since 2006. Its award-winning data protection solution, esXpress, is used today by more than 2,000 enterprises worldwide to achieve scalable, high availability and cost effective backup and restore solutions for VMware. esXpress was awarded Best of VMworld Finalist for 2009. In 2008, esXpress was named "Data Protection Product of the Year" by  SearchServerVirtualization.com. PHD Virtual also provides a suite of free, virtualization utilities to assist with the administration and management of virtualized environments. PHD Virtual supports global resellers through its Channel Xpress partner program and is a proud VMware Technology Alliance Partner. For more information, please visit www.phdvirtual.com.

Over the past few weeks I’ve heard a whole lot of arguments around vCenter design considerations. A few of the questions asked were: 

• Do I install vCenter on 32bit or 64 bit?
• vCenter as a physical or Virtual machine?
• vCenter Database – Local or Remote?
• Placement of the Update Manager Server and Database

Before I dig into the vCenter design topic, I think it would be good to put some perspective on this post and why I’ve decided to blog on this. Last week I attended a meeting with some fellow virtualisation consultants and one of the topics raised in the meeting was to find a common standard practise between us regarding vCenter Server design and specifically the “default” stance between the consultants in regards to the placement of the vCenter server and whether it should be a physical or a virtual machine. Some consultants were in favour of the idea of a default stace and others were against the idea, stating that the decision of vCenter being hosted on a physical or virtual machine is down to the circumstances of each consultancy engagement. Thinking back now, I don’t think we came to an agreement in the end.

This post is basically my opinion on vCenter design, and the steps that I take in deciding what my infrastructure design will look like. 

Do I install vCenter on 32bit or 64bit?

The simple answer to this question is: 64-bit! Why? Well future releases of vCenter will almost certainly be 64-bit only. This means that if you install vCenter 4.0 on a 32-bit Windows operating system today, you will have to rebuild the vCenter server as future releases of vCenter will not install on 32-bit Windows. In simple terms, if you are on 32-bit Windows now, you will not be able to do a simple upgrade to future releases of vCenter.Another reason for going with 64-bit Windows is the fact that you can make use of 4GB RAM and more. I almost always assign at least 4GB of RAM to my vCenter Server, and this will not be efficiently used on 32-bit Windows.

vCenter as a physical or virtual machine?

In order to answer this question, we will have to look at the advantages and disadvantages of both solutions. Personally, my default take on this is to install vCenter on a Virtual Machine. I have many reasons for this approach, one of the reasons being that we cannot tell everyone to run SQL and Exchange of virtual machines if we are not prepared to run vCenter on a virtual machine. If we evangelise VMware vSphere to the world, but cannot lead by example, then what message does that send out? Now, I know by this time you must be thinking “This guy is mad to design an infrastructure based on some political and moral grounds rather than technical grounds. Although this is probably not a good enough and certainly not a technical reason to virtualise vCenter Server, I do have some technical and cost reasons for this approach as well. As I said before, we have to look at the advantages and disadvantages of both physical and virtual vCenter designs.

Is it supported?

The first thing we should look at is support. As VMware fully supports vCenter Server on both physical and virtual machines, there shouldn’t be any issues with support, providing that you use a supported operating system and database. If you do decide to host vCenter on a Virtual Machine, make sure that the virtual machine is configured to meet the minimum supported hardware requirements.

vCenter as a Physical Machine:

 vCenter as a Virtual Machine

Let’s look at vCenter as a Physical Machine first. 

Advantages:

• Not susceptible to potential virtual infrastructure failures:
  Where vCenter is running as a Virtual Machine, there is a danger that the VM can go down due to an infrastructure failure. There are many things that could cause a VM outage, one of these being the datastore where the VM is hosted could run out of space. Running vCenter on a physical host will guard your vCenter server against such VI failures.

Disadvantages:

• Dedicated server hardware is required:
  The first reason most people turn to virtualisation, is to get rid of underutilised hardware and the costs involved with running all that hardware. Running vCenter as a physical machine will require dedicated hardware that needs support, cooling, power, rack space, multiple switch ports, etc.
• High availability is expensive and complex to implement:
  In order to provide high availability to vCenter, the use of either Microsoft Cluster Services or VMware vCenter Heartbeat will be required. Both of these technologies require extra software configurations. Both technologies are complex to implement. Both cost a lot of money. MSCS requires at least Windows Server 2003 Enterprise Edition.

Things to bear in mind when opting for vCenter as a physical machine:

• Performance is limited only by the server hardware. This could be a good or a bad thing. Generally this is seen as an advantage as the vCenter server does not have to contend with other servers for resources.
• vCenter Server backups are done with traditional tools / agents. This makes recovering from a disaster more tricky and lengthy. I therefore see this as more of a disadvantage than an advantage. In the case of a disaster, you will have to rebuild the server from scratch and then try and restore the system state onto the newly installed operating system. Then you’ll have to say a prayer and hope it works.
• Microsoft Cluster Services or VMware vCenter Heartbeat can be configured to provide high availability to the vCenter system. This can be Physical-to-Physical or Physical-to-Virtual clustering. Bear in mind that if you choose to use these technologies, the use of SQL Express is not an option. The vCenter database will have to be hosted on a remote database server, which will in turn also be required to be configured with MSCS (if the SQL server is physical) or VMware HA (if the SQL server is virtual) to provide high availability.

Let’s look at vCenter as a Virtual Machine

Advantages:

• Allows the migration of the vCenter server with VMotion during maintenance windows:
  As the vCenter is a virtual machine in your existing ESX/ESXi environment, the vCenter server can be migrated with VMotion just as all the other VMs can be migrated. This allows for the vCenter server to be moved to another ESX host without downtime in order to perform maintenance tasks on hardware or to apply patches / software upgrades to the ESX hosts.
• No extra power, cooling, rack space, etc is needed:
  This will help reduce capital and operating expenditure as no additional data centre configurations need to be made in order to implement the new vCenter server.
• Allows the entire vCenter server to be backed up with VCB:
  As the vCenter server is now essentially a bunch of files, the VM can be backed up with VMware Consolidated Backup or VMware vDR and sent to tape. This makes things a lot easier when a disaster recovery operation is invoked as you will be able to restore the entire VM from backup in minutes. Not to state the obvious, but if the vCenter database is on a separate database server, this then needs to be backed up separately.
• Allows the use of VMware HA to restart the vCenter server in the case of a hardware failure:
  VMware HA can restart the vCenter VM on another ESX/ESXi host in the event of a hardware failure on the host where the VM was running. This is a much simpler and cost effective approach to providing high availability than the use of MSCS or VMware vCenter Heartbeat.

Disadvantages:

• Susceptible to potential virtual infrastructure failures:
  The vCenter virtual machine is susceptible to potential virtual infrastructure failures. These failures can be brought on by many things such as the loss of connectivity to the storage array, a datastore that ran out of free space, snapshot failures, etc. 
• The VM will have to contend with other VMs for resources:
  As vCenter will now run as a VM alongside other VMs on the same ESX/ESXi host, the VM will have to contend with the other VMs for its resources. In a well designed infrastructure with plenty of resources, this is not normally a problem, but it is something to look out for.

Things to bear in mind when opting for vCenter as a virtual machine:

To balance the load in a DRS cluster, DRS may move the vCenter VM to other hosts on a regular basis. If for some reason the vCenter VM fails and does not restart with HA, you will not be able to manage your servers and virtual machines until you find the server on which the VM is registered. Unless you limit the DRS movement of the VM, you may have difficulty finding the VM in a large DRS cluster. It is also possible with VMware HA to set the preferred failover host of an individual VM. This can be done by setting an advanced parameter called das.defaultfailoverhost. For more information on das.defaultfailoverhost, refer to page 24 of the Availability guide at http://www.vmware.com/pdf/vsphere4/r40/vsp_40_availability.pdf

Along with the das.defaultfailoverhost setting, it may also be a good idea to set the HA restart priority of this VM to high.

vCenter Database – Local or Remote?

Another thing to consider is whether the database should be hosted on the same server as the vCenter server or on a separate server.

Advantages of a Local database

• Can use the bundled version of Microsoft SQL Server 2005 Express for smaller deployments of up to 5 hosts and 50 virtual machines
• Removes the network dependency between the vCenter server and the database server and is therefore more robust
• Easier to maintain as you only have to maintain one server and not two
• If vCenter is installed on a VM, a single backup will be performed for both vCenter configuration files as well as the vCenter database.

Disadvantages of a Local database 

• Increases resource requirements for the vCenter Server
• Not a very flexible solution as it is more difficult for database administrators to manage.

Advantages of a Remote database 

• This is convenient to support and maintain by the DBA team on existing database servers
• Resource requirements for the vCenter server is lower
• The vCenter database can be hosted in a SQL Failover cluster for high availability

Disadvantages of a Remote database

• Requires a robust network connection between the vCenter server and database server. vCenter will crash if connectivity is lost 

 VMware Update Manager 

One other thing to factor into your vCenter design is the placement of the Update Manager Server and its database.

Advantages of a Local installation (Update Manager on the vCenter Server)

• Simplifies magement
• Saves on Windows licenses as no additional Windows servers are required 

Disadvantages of a Local installation (Update Manager on the vCenter Server)

• Affects the performance of the vCenter server
• Requires the vCenter server to have access to a proxy server or direct access to the internet in order to download patches from VMware

Advantages of a Remote installation (Update Manager on a separate Server)

• Does not affect the vCenter server’s performance or configuration

Disadvantages of a Remote installation (Update Manager on a separate Server)

• Requires an additional Windows License
• Cannot communicate with vCenter if network connectivity is lost

Where to place the Update Manager Database

Regardless of where the Update Manager server is installed (local to the vCenter server or not) always create a dedicated Update Manager database instead of using the vCenter Database. This prevents the performance impact on the vCenter database, although it increases the administrative burden. Also, try to place the Update Manager and the vCenter database on the same server as it simplifies administration.

I've received information of Novell’s intentions to include support for VMware vNetworking Distributed switches in vSphere. Currently, when performing migrations to vSphere using Novell Platespin Migrate, the tool fails to properly detect VMware vNetworking Distributed Switches, preventing any migration operation from using them.

Novell product management have potentially scheduled the support of vDS for June 2010.

If you are going to be using Platespin Migrate to perform migrations in a vSphere environment that utilises vDS networking, Platespin Migrate will require a standard virtual switch, including at least 1 port group.

I am sure many of you have noticed that this site has been unavailable at times during the last two or so weeks. This is because I’ve been plagued with problems relating to my iSCSI SAN. The annoying thing is that the problems only started when I decided to move from the stable release of vSphere 4, running ESX Server and not ESXi, to a beta release of ESXi. This has now made troubleshooting the issues more complex as I’m not quite sure whether the issues are related to the new beta version or if it’s simply down to my iSCSI SAN, which is running Openfiler 2.3.

The highlight of the day was my wife’s statement: “If AOL, Google, Yahoo and those people can keep their systems online, why can’t you?” Well, let’s see... The difference is that my solution is a few hundred £££, not millions! You get what you pay for!

So, earlier today, I’ve decided to install the beta version of ESX rather than ESXi, but the problems still seem to be there. At the moment, I’m working on a slow and painful plan to move all the data (and we’re talking TBs here) from the iSCSI solution to a NAS. This will give me reduced performance, but it will at least allow me to rebuild my iSCSI SAN. I will also be going back to the stable release of ESX 4.0 for this environment and do my beta testing somewhere else (maybe in the solution centre at work). I do apologise if www.virtualvcp.com is down at times, but I’m working as hard as I possibly can with a limited budget to resolve the issues asap.

So you're designing a new Virtual Infrastructure on VMware right? Ok, one of the first decisions that your client will have to make is whether to virtualise on VI3 or vSphere. At this stage I'd say it would be a rather silly move to go with VI3.5 as VMware vSphere 4 GA has been available for quite some time now. However, I still see new designs based on VI3.5 being signed off. So why would I rather go for vSphere 4 and not VI3.5? Here are some my reasons:

• We all know that vSphere is stable for production, if not more stable than VI3.5
• Although vSphere 4 has more bells and whistles than VI3.5, it can still do what VI3.5 does. It just does it, well, better that VI3.5 in my opinion.
• As people have learnt with ESX 2.5 when VI3 was released, you'll have to upgrade eventually. Sooner or later, you'll have to upgrade from VI3.5, so why do all the work twice? Why build a VI3.5 solution only to upgrade to vSphere 4 eventually anyway?

I'm not saying that you should go with the latest release, in fact, my policy is to always hold off one or two months before upgrading to the latest release of anything.

Well, ok, so now you have decided to go with vSphere right? Here's the next question... Do I run a 32-bit or 64-bit OS for my vCenter server? Do I install Windows 2003 32-bit or Windows 2003 64-Bit? Or, do I install Windows 2008 R2, which is 64-bit anyway? Now, I may be able to point you in the right direction here. As I'm bound by non-disclosure agreements for most of the information I have from VMware, I won't be able to say too much about anything I've been working with in the past few weeks. However, the purpose of this post is not to help you design a virtual infrastructure that will work for you today, but to help you design an infrastructure that will work for you today, tomorrow, and that will work for and with you when the time comes to upgrade to the next generation of VMware's Datacentre Virtualisation product. So, here's a tip, and probably the whole idea of this post: WHEN DESIGNING A NEW VIRTUAL INFRASTRUCTURE, BE SURE THAT YOU CHOOSE A 64-BIT WINDOWS OPERATING SYSTEM FOR YOUR VCENTER SERVER DEPLOYMENTS AS IT WILL SAVE YOU A LOT OF TIME AND HASSLE IN THE NEXT YEAR!

I have just downloaded and deployed CapacityIQ and it all went fine until I actually decided to register my vCenter server with the appliance, only to find out that the newest product by VMware does not even support vCenter 4, or in fact vSphere! That will teach me to read the release notes before actually bothering to try something new. This is what the release notes have to say:

CapacityIQ supports VirtualCenter 2.5, Update 4 and Update 5, managing hosts running ESX Server 3.0.2 through 3.5. CapacityIQ 1.0 does not support VMware vSphere 4.0 or vCenter 4.0

Am I dreaming? What's going on here? VMware, why did you even bother? Heck, why did I even bother?

VMware has yet again delivered another value-add component to vCenter. vCenter CapacityIQ provides capacity management capabilities for virtualised data centre and or desktop environments. The product integrates with vCenter Server ensuring that your virtualised capacity is always predictable and efficiently used.

The product website states:

“VMware vCenter CapacityIQ balances business demand with IT supply, without compromising performance, availability and security. With CapacityIQ, your IT infrastructure is guaranteed to have sufficient capacity to meet any business service level agreements.” 

Once I have had a good play with CapacityIQ (which I intend on doing sometime this week), I will report back with my review of the product.

More information on vCenter CapacityIQ can be found at: http://www.vmware.com/products/vcenter-capacityiq/

Myself, along with two of my colleagues, went on a day trip to this year’s IPEXPO Earls Court in London where we were shown a demo of esXpress at PHD Virtual Technology’s booth. We were very impressed by the presentation and some of the futures of esXpress. We were especially interested in the File Level Recovery features of their data De-Duplication appliance.

When I got back home from IPEXPO on Thursday, I decided to test the product for myself. So, I went on to download a 30 day evaluation from http://www.phdvirtual.com/ and I can surely say that I’ve been putting PHD Virtual esXpress 3.6 through some vigorous testing for the past few days. To start off with, I found esXpress easy to deploy and I was quite impressed with it. However I must confess that when I really started to dig into how esXpress goes about its business I started to have some concerns about it. However, I have decided not make any my concerns public until I have had a good chat with some people at PHD Virtual as I think that they will have an answer to most of my questions.

For those of you who do not yet know, VMware has launched it's official vSphere blog.

The new VMware vSphere Blog will be VMware's central place to check for news, commentary, links to new resources, and other information about VMware vSphere. The blog's editor is Mike Adams. 

 The blog can be found at http://blogs.vmware.com/vsphere

Some interesting reading: 

SAN FRANCISCO, Sep 22, 2009 (BUSINESS WIRE) -- Today HyTrust demonstrated new technological capabilities jointly developed with Intel and VMware, aimed at enhancing security and policy enforcement in cloud computing environments.

This proof of concept combines HyTrust Appliance with Intel(R) Trusted Execution Technology (Intel(R) TXT) and VMware platform technology currently in development to show customers how they can build and enforce policies that govern the behavior of virtualized cloud infrastructure. Cloud computing is gaining momentum as a way for customers to dynamically scale virtualized computing resources as a service on demand based on business need. Users need not have knowledge of, expertise in, or control over the technology infrastructure in the "cloud" that supports them. As customers turn to cloud computing for its benefits--convenient, cost effective, on-demand access of computing resources--they are seeking a trusted platform that is secure and provides service identity.

Read the rest of the article here: http://www.marketwatch.com/story/hytrust-aligns-with-vmware-to-bring-trusted-computing-to-virtualized-cloud-infrastructure-2009-09-22

I've just stumbled upon the following article. The writer actually accuses VMware of being lazy in that they've failed to develop a vSphere client that works out of the box for Windows 7! How messed up do you get? As far as I know, Windows 7 has not had its official release yet and can only be obtained via TechNet and MSDN. So all, let's blame someone for not developing a piece of software for an OS that's not really released yet!

Get a life! Oh, and maybe try buying a domain as well! Or wait, maybe you're too lazy or cheap to buy a domain, hence the reason we're all looking at an IP address in your URL...

Here's the article:

http://99.246.113.36/wtslabs/blog/?p=53

It’s now a done deal. VMware has officially completed their acquisition of SpringSource. SpringSource is an enterprise and web application development company and will now be seen as a division of VMware. The CEO of SpringSource will serve as the division’s general manager.

VMware says that the new division’s “charter” will be to focus on providing developers and customers the best experience for developing enterprise and web applications, as well as helping VMware create integrated products based on SpringSource technologies and VMware vSphere to deliver a simplified ‘build, run and manage’ experience. These new products will help VMware’s cloud computing strategy to deliver PaaS (Platform as Service) solutions that can be hosted by their customers internally or by service providers.

In today’s world where mission critical applications need to be available 24x7 with 99.99% availability, companies are throwing millions of dollars or pounds at implementing redundant and fault tolerant infrastructures. We all know that the money we spend today will save us much more in the future. Some companies make two to three million profit each and every day. In order to be competitive in the current climate, they need business applications such as messaging and collaboration to be available at all times. Imagine if a business with hundreds of employees one day suddenly lost the ability to send and receive email.

This may sound unheard of, but just this very week I’ve dealt with such a case where a company employing almost 10,000 people had no email, collaboration, database systems and even a corporate website for more than 24 hours, just because a critical component failed on their main SAN. Now let’s just say they make on average 2 million pounds profit each day. They had no business trading on the day of the failure, costing them a huge part of that 2 million pounds profit as the 10,000 people they employ had to be paid for that day an no one could do any meaningful work. If they had only spend half of that on a replication solution for their SAN, they could have avoided such a lengthy outage.

I’ve been trying to install VMware vCenter Server on Windows Server 2008 R2 Enterprise Edition. This is because I am working on a few blog articles on protecting the vCenter Server against hardware failures. At the moment, I’m busy working on two blog posts.

1.       Protecting vCenter with VMware vCenter Server heartbeat;

2.       Protecting vCenter with Microsoft Cluster Services (MSCS).

Whilst trying to install vCenter on Windows Server 2008 R2, I ran into some issues I had to resolve before I could do anything useful.

VMware announced vCloud Express™ at VMworld 2009. vCloud Express allows for the provisioning of infrastructure on-demand. Unlike conventional hosting, services running inside the cloud provided by vCloud Express services are charged for by the hour. This is paid for by credit card. The credit card details are taken upon registering for a vCloud Express account with a service provider that provides vCloud Express services.

Service providers who will provide vCloud Express services, will display the VMware Virtualized™ logo on the website. This is because VMware vCloud Express™ essentially runs on vSphere and it therefore ensures compatibility with VMware environments on external as well as internal clouds.

With vCloud Express™, service providers can now provide a fast and cost-effective solution for their customers to gain on-demand access to a VMware Virtualized™ environment. Virtual Machines can be easily and quickly be deployed. This is especially useful for application development and testing. As vCloud Express™ runs on vSphere 4, it also supports all the guest operating systems that vSphere 4 supports. From registration to deploying machines, vCloud Express™ is fully web based.

Terremarks’s implementation of VMware vCloud Express™ was demonstrated on stage at VMworld 2009. The demonstration was quick and easy and demonstrated just how easy vCloud Express™ makes it for clients to register for the service and provision a server in minutes.

Although still in Beta, the following service providers are currently offering vCloud Express™ services:

United States and Canada (Amaricas):
Terremark, Hosting.com, BlueLock

EMEA:
Logica

APAC:
Melbourne IT

As I'm trying to get to grips with Hyper-V, I've decided to purchase MS techNet Plus which allows a single user to use any of Microsoft's software products for private use for one year.

A TechNet Plus subscription is normally £234.00 excl. VAT at 21.5% (If you live in the EU, VAT is charged at the Irish rate of 21.5%). This comes to £284.31. However, I found a discount code that gives you 25% off. The total amount for my purchase came to £213.43.

The Promo Code is: TMSAM10

In ESX 3.5, when trying to create a new snapshot, the following error is reported in the VI Client:

An invalid snapshot configuration was detected

Also, when reading the Virtual Machine log file (vmware.log) located in the same location as the VM, you may find references to delta files such as "vm_name-000001.vmdk", but when browsing the datastore, the delta file does not exists. Also, when running vmware-cmd <path-to-vmx> hassnapshot,  "hassnapshot ()= " is returned, which means that the VM has no snapshots in place.

If this is truly the case, why is "An invalid snapshot configuration was detected" returned when trying to create a new snapshot?

VMware records snapshot information about the current VM in a .vmsd file. This file is located with the rest of the virtual machine configuration and VMDK files on the datastore. The file is normally called <virtual-machine-name>.vmsd. This file will contain information, even if your VM has no snapshots in place.

It is possible that in some cases, ESX fails to properly clean up after previous snapshots were removed. The information for previous snapshots may still be recorded in the .vmsd file. The file may indicate that you still have snapshots in pace, although all previous snapshots were removed and the dalta files have been merged. When you then try to create a snapshot, the .vmsd file will inform the ESX host that there is a delta file in place and that it has to create a second or third delta file. When the ESX host interigates the VMFS file system, it's unable to find the snapshot delta files specified in the .vmsd file and therefore errors with "an invalid snapshot configuration was detected."

Today I was planning on testing the new 16 patches released by VMware for vSphere 4. I wanted to place these on my 2nd ESX host. I normally place my ESX hosts in maintenance mode before I remedaite updates. As I placed esx2 in maintenance mode, the VMs, as expected, started to migrate over to the other hosts in the cluster with VMotion. The VMotion migration of two of my VMs running Windows XP, failed with the following error message: 

A general system error occurred: Failed to write checkpoint data (offset 33558328, size 16384): Limit exceeded

It turns out that a VM must have less than 30MB Video RAM or VRAM assigned in order to be compatible with VMotion. As I normally run these two VMs at 1680 x 1050 resolution, I went all out an assigned the maximum amount of memory allowed, which is 128MB as VRAM, hence the reason for the VMotion failure.

VMware ESX / ESXi 4 does work out of the box on the HP Proliant ML110 G5. I know, I'm using ML110's with ESX 4. However, I still see posts like this http://communities.vmware.com/thread/163029 were people have problems with this ML110 and ESX 4 and even ESX 3 combination.

Now, if you look at the link to VMware Communities, you will notice in one of the screen shots attached that the CPU is a 1.8GHz Dual Core. The ML110 G5 comes in more than one CPU, so beware that when you decide to get one of these HP Proliant ML110 G5 for running ESX 4, make sure you get one with the Intel Xeon 3065 CPU that runs at 2.3GHz. The reason is because not all Intel CPU's shipped with the ML110 supports Intel VT, and this is a requirement for ESX 4.

Now I know you're probably saying: "The guy in the post is not using ESX 4." My point is simple: It doesn't matter what version of ESX the post refers to. If you want ESX 4 to run on the ML110, make sure you have the correct processor.

This is by no means a new issue. However, I still get support calls regarding tasks that get stuck on VMs. What do I mean by "stuck tasks"? Well, I've seen cases where a snapshot task initiated by VCB got stuck in the state of "Creating Virtual Machine Snapshot". Then VM goes down and cannot be accessed via the console, does not respond to pings, and the status of VMtools turns to "Unknown". You also cannot do "Power On" on the VM either as the "Creating Virtual Machine Snapshot" task is still showing as an active task. You can wait, but after 30 minutes, chances are that it won't sort itself out, so user intervention is required!

This is normally the approuch I take to sort this out:

2. Browse the datastore where the VM is located (best to do this via the CLI on the service console with "ls -lh") and check the time stamps of the files to see how log the snapshots, if any,have been sitting there for.
3. in VirtualCenter, or "vCenter" the VM will probably still be showing as powered on. Check on which of your ESX hosts it is running.
4. Log onto the service console of the ESX host that is running the VM. Elevate your priviledges to root.
5. Now, as the VM has an active task, you won't be able to send any other commands to the VM. You won't be able to use vmware-cmd to change the state of the VM either. Until the task that's stuck in progress has completed, the ESX host will not be able to send any power commands to the VM. The only way to now release the VM from it's sorry state and get rid of the "Active task" is to kill the VM's running process from the service console. In order to do so, you need to find the PID for the "running" VM. To get the PID do:

This should return something like this:

The PID in this instance is 12322. This is what we need to kill.

6. Kill the process ID with kill -9:

7. Delete any snapshots created

8. Power On the VM.