vSphere

While one of my readers (credits to Scoobie Guy) was exploring the features of DV switches he inadvertently discovered if you make major changes (or rebuild)  your virtual center instance, or remove a host from a VC instance you may end up with the following errors:

The distributed Virtual Switch corresponding to the proxy switches d5 6e 22 50 dd f2 94 7b-a6 1f b2 c2 e6 aa 0f bf on the host does not exist in vCenter or does not contain the host.


 
From this article: http://www.vmware.com/support/vsphere4/doc/vsp_esxi40_vc40_rel_notes.html#knownissues

Here is an explanation: Removing an ESX/ESXi host configured with a vDS from a vCenter Server system results in inconsistent networking state on the host. If you remove an ESX/ESXi host configured with a vDS from a vCenter Server system, the host cannot reconnect to the vDS. When you add the host back to the vCenter Server system, a warning similar to the following appears:

The distributed Virtual Switch corresponding to the proxy switches d5 6e 22 50 dd f2 94 7b-a6 1f b2 c2 e6 aa 0f bf on the host does not exist in vCenter or does not contain the host.

The virtual machines continue to function on their respective ports, but new virtual machines are not allowed to power on. You cannot modify the vDS settings for this host by using a vSphere Client connected to the vCenter Server system.

Workaround: Perform the following steps:

1. Use a vSphere Client to connect directly to the ESX/ESXi host. This workaround requires a direct connection.
2. Migrate the virtual machines off of the invalid vDS ports one by one by editing the settings of each virtual machine. This will result in prolonged network interruption to the virtual machines.
3. Choose Host > Configuration > Networking > Distributed Virtual Switch and click Remove.
4. In a vSphere Client connected to the vCenter Server system, refresh the network settings of the host. The errors are cleared.
5. Add the host back to the vDS, either manually or by using a host profile.
6. Migrate the virtual machines back to their respective ports or portgroups on the vDS. To do so, right-click the vDS and choose Migrate Virtual Machine Networking. This process also results in network interruption to the virtual machines. 
   

Now the fun / trouble really starts! Rebuilding your network!

Another thing to note, if you miss configure a DvSwitch you may lose all connectivity to your ESX hosts. Examples are the MTU settings change from 1500 to 9000, if your NICs or Switch don’t support it then you will lose connectivity. My recommendation is to keep at least one  service console port on a standard virtual switch. That way you can always reach your hosts, and worse case if you need to rebuild your service console network it’s much easier.

Nice trackback from Mike Laverick: Got a question for you. Preparing to start a new vSphere farm and trying to decide do I use stand switches or start with the DVS. Especially when I see this: http://www.ntpro.nl/blog/archives/1283-vSphere-DvSwitch-caveats-and-best-practices!.html So the main question is do we think this is mature enough to go forward with yet? Can I use Host profiles with DVS safely? --> http://www.rtfm-ed.co.uk/?p=1875

You really should check out the new VMware vSphere Blog, it’s a central place for news, commentary, links to new resources and other information about VMware vSphere. I’ve seen some great articles popping up in the last two weeks.

Eric Siebert - Master’s guide to VMware Fault Tolerance
Barry Coombs - VMware FT
David Strebel - VMware Fault Tolerance Setup and Best Practices
Hany Michael - vSphere 4.0 Fault Tolerance (Architecture Diagram, Video and Use Cases)
Joep Piscaer - What I’ve learned from BC2961, “VMware Fault Tolerance Architecture and Performance”
Cody Bunch - Scheduling VMware’s FT (Fault Tolerance)
Rynardt Spies - VMware FT...Can you afford a SAN failure?
Roger Lund - VMware Fault Tolerance: What is it? What does it do?
Jason Boche - After enabling FT on a VM – subtleties to expect
Eric Sloof - Fault Tolerance at your home lab

The first round of the contest which features articles about Fault Tolerance is closed now. The next round “vNetwork Distributed Switch” will kick off next Monday. I just can’t wait for another round of great vNetwork articles in the coming two weeks. If you want to join the contest, don’t be shy and give John Troyer a heads up about your website.

You can establish fault tolerance for virtual machines hosted by ESX servers that are part of a VMware HA cluster. When you enable VMware FT for a virtual machine, an identical copy of the virtual machine, called a secondary, is established on another host in the cluster. If the first virtual machine fails, the secondary virtual machine takes the original one's place, and a new secondary is created. This process helps provide uninterrupted virtual machine functionality, even in the event of failures. Before you can enable VMware FT, you must have VMware HA enabled and the CPUs of the servers hosting the virtual machines must support VMware FT.

After publishing an article about the CPU compatibility with VMware Fault Tolerance, my search for a white CPU began. The vLockstep technology used by FT requires the physical processor extensions added to the latest processors from Intel and AMD. In order to run FT, a host must have an FT-capable processor, and both hosts running an FT VM pair must be in the same processor family.
Richard Garsthagen’s “CPU-Host-Info” shows all the available options on both the Intel Q9400 and Q9550 marked true. I’ve used the Intel Q8200 in another white box and it didn’t work, so in order to use FT, you need FT and both the VT options. The next step is run through the Fault Tolerance Checklist.

You can enable VMware Fault Tolerance through the vSphere Client.

Connect vSphere Client to vCenter Server.

Prerequisites 

The option to turn on Fault Tolerance is unavailable (grayed out) for a virtual machine to which any of the following conditions apply:
■  is not in an HA-enabled cluster
■  has one or more snapshots
■  is a template
■  is disconnected
■  resides on a host which is in maintenance or standby mode
■  is performing a record/replay operation

Procedure

1.  Select the Hosts & Clusters view.
2.  Right-click a virtual machine and select Turn Fault Tolerance On.

VMware Fault Tolerance requires eager zeroed thick disks. Virtual machines with thin provisioned or lazy zeroed disks must be powered off while enabling VMware Fault Tolerance in order for vCenter to complete this conversion.

The specified virtual machine is marked as a primary and a secondary is established on another host. Fault Tolerance is now enabled on 4 virtual machines :-)

Jon Hall, Technical Certification Developer over at VMware has clarified the exam scoring on the new VCP4 exam. This new exam became generally available during the VMworld 2009 at San Francisco and can be scheduled at Pearson VUE. Some people already have taken the exam and are reporting they had to answer 85 "fair" questions in 90 minutes. There’s also some confusion about "extra" questions that doesn’t count to your score but are build in to improve future exams.

I have often been asked how the percentage for the VCP on VI3 exam is calculated, which led me to post a discussion thread on scaled scoring some time back. Scaled scoring is used with SATs and many other standardized tests. We used scaled scoring with the VCP on VI3, but many folks thought the resulting number (ex. 70) was actually a percentage, which it was not. With the VCP on vSphere 4 Exam, we are modifying the scale so that the resulting score does not appear in any way to be a percentage.

The scale for the VCP on vSphere 4 Exam is 100 - 500, with 500 being a perfect score. A 300 or better is required to pass the exam. When you are preparing for the exam, don’t forget to pay a visit at Scott Vesey’s VMwaretraining weblog. I’ve also written some articles that can help you with preparing.

Update : Duncan Epping via Twitter -> @esloof @vmguru_nl yes it is 300 out of 500 and 400 to become a VCI.

Update : John Hall via the VMware Community ->  VMware may add as many as 10 unscored items to an exam (75 items scored, the additional 10 unscored). The VCP exam will always have 75 scored items.

Update : The VCI passing score has been revised based on the baselines VMware saw coming in. It’s not an easy exam! VCI passing score is 350.

To help VCP4 delegates stand an even better chance of passing the VCP4 certifcation, Margirus is proud to announce a vSphere 'certification' style test online. Click here to try!

You also should take a look a Scott Vessey’s VMwaretraining blog, he’s got some great VCP 4 study tips.

Now that the VCP on vSphere 4 exam is finally live, VMware have released a number of official resources to help those who want to obtain what will quickly become the new "industry standard" virtualization certification.

Studying for the VCP on vSphere 4

vNetwork APIs focuses on ways to build tools and solutions around the virtual networking capabilities provided by the vSphere platform. The community serves as a collection of code samples, utilities and documentation to help developers integrate and build tools using the web services SDK, extensions and the developer SDKs.

Lei Shi and his crew created a number of tools based on open vNetwork related API and other open-source tools.

The live demo will be shown at VMworld and tools will be downloadable shortly. The tools are not VMware products, so use it only at your own discretion. We would love to hear from you on any feedback though.

Here is the VMworld break-out session information.
==============================================
Session: TA3521 - Thursday 2:00 - 3:30pm (9/3)
Title: vNetwork Monitoring, Troubleshooting, Security, and Management
Speaker: Howie Xu, Director of R&D, Core Networking, VMware Inc.