VMware opens public beta for next version of vSphere

For the first time ever VMware opens a beta program of vSphere available for the public. So far participating in the beta was only available for VMware partners, vExperts and other selected groups.

Now everyone can participate and have a look into the future of vSphere. The next release will most likely be vSphere 6.0 although VMware never communicates over version numbers before product announcement.

Joining is very simple by filling in name, role, company and country. Then digital sign and you are in the beta within minutes.

Please remember that this vSphere Beta is private even though it is open to everyone. Do not share information from this Beta Program with those not in the Beta Program. What you learn and see in this Beta is meant to be kept confidential per the Master Software Beta Test Agreement and Program Rules.

The ground rules of the program are here.

This is an interesting change in policy of VMware. Microsoft always has been very open about new versions of for example Hyper-V and enabling the public to download beta’s.  VMware preferred to keep features and even the release number of vSphere secret tilll unveiled at VMworld.

I am not sure about the reason for this change in policy. Likely VMware made the beta open to public for quality reasons to make sure as many scenario’s are tested and teething problems are resolved before the product reaches GA. VMware Virtual SAN has been running the beta for a relative long time for the same reason.

Another , less likely, reason could be VMware wants to be more open on the future of products for marketing reasons.

The VMware blog announcing the beta is here

Join the beta here.

What is new in VMware vSphere Data Protection

This post is part of a series of blogpostings on VMworld 2013 announcements. See here for a complete overview of all announcements.

VMware vSphere Data Protection (VDP) is VMware’s backup and restore tool. It is provided for free with vSphere Essentials Plus and higher editions.

So far the features of the tool were very limited. Use cases  of VDP are small companies, remote offices etc.

vSphere Data Protection

With this release, vSphere Data Protection (VDP) gains the following capabilities:

• Disk-level granularity
• Detachable/remountable data partitions
• Replication to the cloud. Data can be replicated to a cloud provider who has EMC Avamar
• Time-of-day scheduling
• Removal of the blackout window
• Restore without dependancy of vCenter. Direct to host “Emergency Restore” for any VM

vSphere Data Protection Advanced (VDP Advanced) offers all the features available in VDP, plus these:

• Replication
• Data Domain integration
• Exchange message-level recovery
• SharePoint integration
• Automated backup verification

more info on the VMware website.

One of the new features in  VDP is the ability to select individual .vmdk files for backup. To avoid backing up an OS page or swap file, perhaps it now makes sense to create a separate .vmdk file for the page/swap file, which could then be excluded from VDP backup jobs.

Another new feature is that scheduling of backup jobs is now much more granular. The time of the start can now be set. In the past VDP selected the runtime by itself.

The datastore where to store the VDP operating system disk and the data disk can now be individually set. In the past this had to be on the same datastore. This now enables a scenario in which the VDP OS is lost due to a disaster. A new VDP virtual appliance can be installed and point to the original VDP data disk.

Prevent knockout monster VM’s by this patch and heap configuration

It is likely you never have heard of VMFS heap size used by VMware vSphere. This probably means you have not run into issues yet. However when using VMware vSphere AND virtual disks located on VMFS volumes AND relative large sized virtual disks you can run into some serious issues!

This posting will give the reader some insight in how to avoid those issues by adjusting settings and apply a recent published VMware patch.

An increasing number of organizations are using hosts with serious amounts of  physical memory and compute power to be able to run many VMs per host. VM’s having large amounts of memory  and virtual CPU’s are called monster VM’s.

However those monsters can easily be made unconscious when those VMs are running a couple of large VMDK’s. For instance you  might run into issues when running multiple virtual machines like Exchange Servers, file servers  or database servers with large VMDK’s on the same ESX host.

RVTools 3.5 released!

RVTools is one of the best FREE tools available which helps consultants and system administrators in inventory and managing a VMware vSphere infrastructure. It offers a lot of information on the configuration of storage, network, virtual machines, ESX hosts  and a LOT more. Information can easily be downloaded in CSV format so you can do for example capacity planning. With RVTools you can disconnect the cd-rom or floppy drives from the virtual machines and RVTools is able to update the VMware Tools installed inside each virtual machine to the latest version.

If you do not know RVTools I strongly recommend to download the tool and see for yourself how useful it is and makes your life much easier. You won’t regret.

At March 9 RVTools version 3.5 has been released. New are:

 

• On vInfo tabpage new field: Resource pool
• On vInfo tabpage new field: Consolidation needed.
• On vCPU tabpage new field: Number of cores per socket
• New tabpage with resource pool information
• On vNetwork tabpage new column: Switch name
• On vNetwork tabpage new column: Starts Connected
• On vTools tabpage new column: required version
• On vHost tabpage new columns: custom fields
• On vDisk tabpage new columns: raw disk information
• Improved error handling for SSO login problems
• Bug fix: Invalid snapshot size fixed
• Bug fix: All datetime fields now use the local time zone
• Bug fix: data not refreshed after changing filter

 

VMware vCenter Site Recovery Manager 5.0.2 and 5.1.0.1 released

December 20 has been a busy day at VMware HQ. Many new software releases were made available for download.
VMware released two new versions of vCenter Site Recovery Manager:  5.0.2 and 5.1.0

The difference between the versions is the vSphere support.SRM versions are in lockstep with the vSphere releases.

SRM 5.1 supports VMware ESX5.1 while SRM 5.0.2 does not support ESX5.1
SRM 5.1 supports vCenter Server 5.1. SRM 5.02 is limited to support for vCenter Server 5.0 releases.

For compatibility see this matrix.

VMware vCenter Site Recovery Manager 5.1.0.1

Release notes can be read here.

SRM 5.1.0.1 provides the following improvements:

• Resolves critical issues in SRM 5.1. If you have installed SRM 5.1 (build 820150), you must upgrade your installation to SRM 5.1.0.1 (build 941848).
• Includes vSphere Replication 5.1.0.1. If you have installed vSphere Replication 5.1, after you upgrade SRM to 5.1.0.1, you must also vSphere Replication to vSphere Replication 5.1.0.1.

VMware vCenter Site Recovery Manager 5.0.2

It offers improvements and bug fixes.
Release notes are here.

VMware vCenter Site Recovery Manager 5.0.2 offers the following improvements:

• Added support for protection and IP customization of the following guest operating systems:
  • Windows 8 (32-bit and 64-bit)
  • Windows Server 2012 (32-bit and 64-bit)
  • RHEL Server 6.2 and 6.3 (32-bit and 64-bit)
  • Ubuntu 12.04

  NOTE: To protect virtual machines that run the above operating systems, you must upgrade ESXi Server to version 5.0 update 2 on both the protected and recovery sites.

• The vSphere Replication management server accepts MD5 certificates. See Caveats and Limitations.
• Upgraded OpenSSL 0.9.8m to 0.9.8t for improved security. This addresses the security advisory that was issued for OpenSSL in January 2012.
• Auto-generated certificates use RSA keys of 2048 bits.

HostDatastoreSystem.CreateVmfsDatastore error when creating new datastore

When a new datastore is created in VMware vSphere an error message is shown in the vSphere Client showing

“Call “HostDatastoreSystem.CreateVmfsDatastore” for object “ha-datastoresystem” on ESXi “hostname” failed.
Operation failed, diagnostics report:  Unable to create Filesystem, please see VMkernel log for more details: Function not implemented”

Solution:

This error can be caused by several situations:

 if storage is accessed over iSCSI: MTU size is not set correctly one on of the components

Make sure all networking components are set to an MTU of the same size: the virtual switch properties, the vmkernel ports used for the iSCSI traffic, physical switch, storage device . If have seen this error while trying to create a datastore on a Iomega device iSCSI LUN because the virtual switch MTU was set at 1500.

There has been a situation in which the Dell switch MTU setting needed to be higher than the MTU size on the surrounding infrastructure. So: vSwitch, VMKernel and EqualLogic mtu: 9000. 6224’s: 9126

read more in this blogpost.

If the storage has been used before for other purposes.

Another reason could be the volume has been formated using a filesystem which is not compatible with the VMware vSphere version. If possible delete the LUN and recreate.

check this VMware knowledgebase article.
Or check out this VMware KB article Unable to add a deleted local datastore back to the ESXi 5.0 host

vSphere 5 versus Windows Server 2012 Hyper-V: affinity and anti-affinity rules

This posting is one in a long series of blogpostings in which I compare various features of VMware vSphere 5.1 with Windows Server 2012 Hyper-V.

This posting will compare the affinity and anti-affinity rules available in both products. Actually, those rules to keep VMs together or separate on hosts is run by VMware vCenter Server and Microsoft System Center Virtual Machine Manager 2012 SP1. SP1 is expected to be released begin 2013.

Other postings in this serie are:

Microsoft Windows Server 2012 Hyper-V compared to VMware vSphere 5.1
vSphere 5 versus Windows Server 2012 Hyper-V: storage integration
vSphere 5 versus Windows Server 2012 Hyper-V:management
vSphere 5 versus Windows Server 2012 Hyper-V: high available VMs
vSphere 5 versus Windows Server 2012 Hyper-V: Resource metering for chargeback
vSphere 5 versus Windows Server 2012 Hyper-V: costs
vSphere 5 versus Windows Server 2012 Hyper-V: hybride cloud
VMware vSphere 5.1 versus Windows Server 2012 Hyper-V: replication

In a dynamic environment like server virtualization and cloud IT-management need some control over the placement of virtual machines and virtual machine diskfiles on hosts and datastores.
What is virtual machine anti-affinity and affinity and why do we need it?
Critical applications or services are often made high available by having multiple instances of databases or applications (distributed applications) . Think about Active Directory Services domain controllers, Exchange Database Availability Groups, web servers part of a tiered aplication etc. If those applications/services are running on multiple virtual machines, we do not want to have a redundant set of VMs running on the same host. If the host fails, the service will fail as well.

VMware vSphere DRS and Microsoft SCVMM Dynamic Optimization will loadbalance VMs over available resources. To prevent certain VMs to run on the same host we use anti-affinity rules.

We also can have a demand to keep a set of VMs running on the same host. Most likely because those VMs have a dependancy. To deliver the best network performance we prefer to keep the VMs on the same host. This can be controled using affinity rules.

Anti-affinity and affinity rules in vSphere 5.1
Setting affinity and anti-affinity rules in vSphere is very simple. In the properties of DRS on a cluster VMs rules can easily be made using the vSphere Client or Webclient.

The cluster will make sure VMs are not running on the same host. The rule will be immediately effective. So the rule is applied when a host fails but also during normal operations.

Several rules are available in vSphere :

VM-VM affinity rules: keep VMs to together running on the same host.
VM-VM anti-affinity rules: prevent VMs running on the same host.

Host-VM affinity: VMs must run on a certain set of hosts or VMs should be running on a certain set of hosts. More info here

vSphere 5.x has a feature called Storage DRS. This is DRS for virtual disks; vSphere tries to loadbalance disk over datastores. Here we have Storage DRS affinity and anti-affinity rules. These roles keep virtual disks of the same VM on the same datastore or make sure disks are always located on different datastores.

Anti-affinity and affinity rules in Windows Server 2012 Hyper-V + SCVMM 2o12 SP1

Service Pack 1 of SCVMM 2o12 offers a new feature named availability set. At the properties of the VM the VM can be added as a member of an availability set. When you place virtual machines in an availability set, SCVMM will attempt to keep those virtual machines on separate hosts for improved continuity of service. That is, whenever possible, VMM will avoid placing those virtual machines together on one host.

Placement fof VMs on hosts can be controlled using the preferred and possible owners properties of  virtual machines. This could be useful for example for licensing reason. Oracle is notorious for its policy (which does not have any legal ground) to let customers pay licenses for all nodes in a cluster. You can show the Oracle LMS guys that your VMs cannot run on more than a few hosts in your cluster.

• If you want to control which nodes (servers) in the cluster will own the virtual machine most of the time, configure the preferred owners list.
• If you need to prevent a virtual machine from being owned by a particular node, configure the possible owners list, omitting only the nodes that should never own the virtual machine.

I have not found a feature to keep VMs running on the same host (affinity rule)

SCVMM 2012 SP1 Availability set

Data ONTAP Edge VSA is Available to Download

The giveaway of Data ONTAP Edge, the NetApp Virtual Storage Appliance, was a huge hit at VMworld. Due to the tremendous feedback NetApp has decided to make the evaluation edition available to the public.

This is a great opportunity for those who are unfamiliar with the capabilities of Data ONTAP to experience it first hand. By downloading and registering for licenses you will receive access to Data ONTAP Edge, OnCommand Systems Manager and the Virtual Storage Console for VMware vCenter Server.

With these tools one receives a NetApp array as a VM, which functions identically as a physical FAS array. This is the beauty of the NetApp storage platform; software defines value and hardware is a commodity and merely a means to scale. Data ONTAP is the premiere software-defined storage platform!

Edge supports the VMware vStorage APIs like VAAI and Site Recovery Manager (SRM). In addition, those who register will be notified by email of the release of quick start videos, designed to help users get the VSA installed and up and running with vSphere!

Download and info here.

more info here

Troubleshooting Storage Performance in vSphere – Storage Queues

Good blogposting on storage queues by VMware.

Storage Queues what are they and do I need to change them?

We have all had to wait in a line or two in our life, whether it is the dreaded TSA checkpoint line at the airport or the equally dreaded DMV registration line, waiting in line is just a fact of life. This is true in the storage world too; storage I/O’s have plenty of lines that they have to wait in. In this article, we examine the various queues in the virtualized storage stack and discuss the when, how, and why of modifying them.

Queues are necessary for several reasons but primary they are used to allow for sharing of a resource and to allow for concurrency. By using queues, vSphere is able to allow for multiple virtual machines to share a single resource. Queues also allow for applications to have multiple active (“in-flight”) I/O requests on a LUN at the same time, which provides concurrency and improves performance. But there is a tradeoff; if you allow too much concurrency the underlying resource might get saturated. To prevent one virtual machine or one host from saturating the underlying resource, the queues have set sizes/limits that restrict the amount of I/O requests that can be sent at one time.

Continue at source:

Access VMware virtual machine console from any device just using a browser

Access to a VMware virtual machine console from any device can be very handy.Think about doing maintenance from your iPad at home. Think about granting an application owner access to the console without giving access and permissions to vCenter Server. Think about operating systems without a remote access protocol like RDP etc etc.

One of VMware’s employees wrote a nice program called VMware WSX (name might change later) which allows access to  consoles of a VMware virtual machine running on either ESX or VMware Workstation using a brower. There is no need to install any plugin or add-on. Mind the software is experimental and not supported by VMware.

WSX works best with the latest versions of Chrome 17, Firefox 10, and Safari 5 on Mac OS hosts. It requires a very modern browser that supports HTML5 with WebSockets.

While a bit buggier, it also seems to work with Microsoft Internet Explorer 10 Preview. On the tablet side, it reportedly runs well with the Apple iPad running iOS 5+, and there has been some success reported with Google Chrome and Ice Cream Sandwich on Android devices.

WSX Tech Preview 2 can be installed on Windows and Linux machines.

Much more information and download on the site of the author Christian Hammond.

 

vSphere 5 AutoLab released

Testing and demoing software usually takes some time to setup a lab. Alastair Cooke did a great job by collecting and assembling components which create a vSphere 5 lab with a lot less effort than doing it yourself!

Taken from ProfessionalVMware.com is the info below.
Mind for the moment you will need to build the AutoLab Windows servers using 2008R2 without any Service Pack

What is the AutoLab?

The AutoLab is a quick easy way to build a vSphere environment for testing and learning using a single desktop or laptop PC and VMware Workstation, Fusion or ESXi. The whole lab runs in VMs on that one PC, even ESXi runs in a VM and can then run it’s own VMs.

What’s in the AutoLab?

The Autolab download contains a set of shell VMs and a lot of automation.  Once built the lab contains two ESXi servers, a Windows Domain controller, a Windows Virtual Centre, a FreeNAS storage appliance and a FreeSCO Router to link it to the outside world.

The download has been kept very small for easy download and distribution, it contains only free, open source, software. You will also need the installers for vSphere and Windows, you should use evaluation licenses for both vSphere (built in) and Windows.

What can I do with the AutoLab?

• Run VMs on the lab ESXi servers, using iSCSI shared storage
• Build an HA and DRS cluster
• Work with vSphere Networking
• Practice the upgrade from vSphere 4.1 to vSphere 5.0
• Use PowerShell and the VCLI to manage the lab
• Rebuild the whole lab quickly and with minimal effort
• Choose how much automation you want in the lab build
• Take the lab with you on your laptop

 

Continue at source

RVTools 3.3 released

Today version 3.3 of RVTools was released. If you are unaware of what RVTools is for your VMware environment: This free tool is able to give a complete overview of a VMware vSphere infrastructure. Properties of virtual machines, datastores, networking, memory etc etc. Reports can be exported to text and CSV format.

A great tool. Download here.

New in version 3.3:

•  GetWebResponse timeout value changed from 5 minutes to 10 minutes (for very big environments)
•  New tabpage with HBA information
•  On vDatastore tab the definition of the Provisioned MB and In Use MB columns  was confusing! This is changed now.
•  RVToolsSendMail accepts now multiple recipients (semicolon is used as separator)
•  Folder information of VMs and Templates are now visible on vInfo tabpage
•  Bugfix: data in comboboxes on filter form are now sorted
•  Bugfix: Problem with api version 2.5.0 solved
•  Bugfix: Improved exception handling on vCPU tab.
•  Bugfix: Improved exception handling on vDatastore tab.

Live webcast Best Practices: When & How to Use Stretched Clusters

At April 26 11:00 AM – 12:00 PM EDT   (start 8:00 am PT/  15:00 GMT ) EMC will have a free live webcast on the use of stretched clusters.

Host: Scott Lowe, CTO, VMware Affinity Team, EMC
The abstract:

EMC VPLEX’s AccessAnywhere functionality, when coupled with the industry-leading virtualization solution VMware vSphere, enables new topologies, like stretched clusters. But is a stretched cluster the best solution? In this webcast, we’ll examine stretched clusters and VMware’s own Site Recovery Manager to determine which solution is the right solution for your organization.

Attend this session to learn to:

• Distinguish between stretched clusters and SRM-based topologies.
• Gain a better understanding of the technical requirements of each solution.
• Select the right solution based on your specific business requirement.

Registration at emc.com

DRS Host Affinity rules can be used to run Oracle on a subset of the hosts within a cluster

In November 2011 VMware released a very good whitepaper titled Understanding Oracle Certification, Support and Licensing for VMware Environments. It clearly describes there are no issues on running Oracle on VMware and that running Oracle virtual is cheaper than running on physical boxes.

The whitepaper is the first written statement of VMware I am aware of which clearly states *that according to VMware* DRS Host Affinity can be used to limit the number of Oracle license needed without having to use a dedicated cluster. The text below is taken from the whitepaper.

Oracle cannot be licensed by virtual CPUs today, but as long as Oracle software runs on fully licensed hosts, customers are not in violation of published Oracle policies. In particular, DRS Host Affinity rules can be used to run Oracle on a subset of the hosts within a cluster. In many cases, customers can use vSphere to achieve substantial licensing savings.

A response to the VMware whitepaper is posted by the blog Oracle Storage Guy titled VMware’s Official Support Statement Regarding Oracle Certification and Licensing. The same blog written by Jeff Browning can be read at communities.emc.com as well.
Also at the communities.emc.com forum an interesting thread on using DRS Host Affinity.

Michael Webster wrote a great posting titled Fight the FUD – Oracle Licensing and Support on VMware vSphere. The artivle has a lot of information and tries to get rid of all the FUD spread around by Oracle (not supported on VMware, you need to license all CPU’s)

Dave Welch of House of Brick responded in the thread listed above with a very detailed response explaining the use of DRS Host Affinity. Worth reading!
It remains to be seen if Oracle will ever respond to the whitepaper or even will confirm that DRS Host affinity is a way to limited the number of licenses. I believe they will not respond and will get their money from customers not willing to start a legal battle with Oracle.
Read this blog of Rocela– global leaders in consulting for Oracle licensing, cost management and compliance which states:

We work with clients using VMware and Oracle every day: we are unaware of any change of general stance from Oracle. Therefore, you should be fully aware of the potential licensing implications should Oracle audit your use of Oracle on VMware clusters.

Jeff  Browning of EMC wrote a nice blog explaining the licensing of Oracle on VMware. The blog is titled Oracle Licensing on VMware – no magic

Now that server virtualization is main stream technology companies are starting to virtualize their Tier1 applications on VMware vSphere and other platforms. However there are some concerns and misunderstandings about virtualization of Oracle products.

Organizations believe running Oracle in virtual machines is not cost effective because of the Oracle licensing policy. Sometimes this is solved by using a dedicated cluster just for running Oracle VM’s. Creation of a dedicated cluster is not needed when VMware Host DRS Affinity feature is used. This feature allows virtual machines to be active on a subset of hosts in a cluster. So only the CPU’s in those hosts needs to be licensed for Oracle, not all hosts. I wrote about this earlier in a posting titled Oracle supports VMware DRS Host Affinity (but does not want you to know it).

Also customers are being intimidated by Oracle to buy more licenses that needed. Because of the legal and unclear text most companies do not get into legal fight and pay. However in most cases this is not needed and things can be settled for a much lower price.

Techtarget.com publised an article in 2012 titled Oracle licensing for vSphere 4.1 irks VMware pros . A quote from the article

At the very least, users should “not just foolishly say, ‘My field rep interprets licensing requirements for me,’” Welch said. “The only thing that’s binding is on paper. In my opinion, Oracle is just delighted that this confusion is spreading, when how the licensing works is right here in front of us.”

Also the technical support by Oracle has been an issue. However Oracle does support running Oracle on vSphere and other platforms. In some cases the customer will need to prove the reported issue is not caused by the virtualization stack. I blogged about this in a posting titled Oracle slowly starting to support VMware

There are more interesting posting on the internet about Oracle licensing. Jay Weinshenker wrote an interesting post titled Licensing Oracle on VMware vSphere with usefull thoughts and comments from readers. Another blogposting by Jay titled Yet another way VMware saves a company money on Oracle licenses
House of Brick  can help you purchasing the right number of licenses.

In the Netherlands License Consulting can help you with making sure you do not pay too much for Oracle licenses.

Another concern which slow down adoption of virtualized Oracle instances is performance. As Oracle databases are often used for Tier1 application, performance is critical. A long time thought is that running Oracle on VMware might result in less performance. Several benchmarks showed there is none to hardly any performance difference between running Oracle on physical or virtual servers.

In March 2012 Confio released a whitepaper titled A Comparison of Oracle Performance on Physical and VMware Servers.  Confio is the developer of the Ignite family of performance and monitoring tools, including IgniteVM, the first database monitoring solution created for databases hosted on VMware. IgniteVM installs in minutes. Download a free trial from www.confio.com

The summary of the test: Performance and throughput results, as experienced by the order entry application, were essentially identical for the Oracle database on the physical and VMware servers. IgniteVM focuses on application response time, the most important performance measure. It illustrated identical throughput of 50 transactions per second, and negligible differences in response time for the highest volume queries, with the VM actually running faster at less than 2ms per execution vs. 3ms for the physical server.

Top Five Server Virtualization Trends, 2012 according to Gartner

Thomas Bittman is a vice president and distinguished analyst with Gartner Research. In his blogpost titled Top Five Server Virtualization Trends, 2012 he writes about 5 trends in server virtualization being:

1) Competitive Choices Mature
2) Second Sourcing Grows
3) Pricing Models in Flux:
4) Penetration and Saturation
5) Cloud Service Providers Are Placing Bets

Matt Weinberger responds to the Gartner blog in this posting titled Gartner Server Virtualization Trends: Competitors Threaten VMware’s Reign