WilliamLam.com

Quick Tip - Listing Image Profiles From an ESXi Patch Using ESXCLI

by // 6 Comments

I was cleaning out a few of my to-do items (list just keeps getting longer everyday) this morning and there was a question that I received a few weeks back asking how to retrieve the list of Image Profiles for a given ESXi patch. This is actually quite easy and you will want to use ESXCLI.

Note: The examples shown below is using ESXCLI on the ESXi Shell, but these commands can be execute remotely as well using ESXCLI or through PowerCLI with Get-EsxCli cmdlet.

To list the available Image Profiles for an ESXi patch, run the following command (ensure you substitute the full path to your ESXi patch):

esxcli software sources profile list -d /vmfs/volumes/datastore1/ESXi510-201212001.zip

To get more details on a particular Image Profile, run the following command and specify the -p for the specific Image Profile:

esxcli software sources profile get -d /vmfs/volumes/datastore1/ESXi510-201212001.zip -p ESXi-5.1.0-20121204001-no-tools

To install/update a specific Image Profile, run the following command with the Image Profile name:

esxcli software profile update -d /vmfs/volumes/datastore1/ESXi510-201212001.zip -p ESXi-5.1.0-20121204001-no-tools

If you just want to install the ESXi patch, run the following command which will install the esx-base Image Profile by default which will include everything:

esxcli software vib update -d /vmfs/volumes/datastore1/ESXi510-201212001.zip

To check for the Image Profile you have installed on your ESXi host, run the following command:

esxcli software profile get

Here are some additional resources for ESXi patch management that may also be useful:

Creating Custom vSphere Reports is a Breeze with CloudPhysics

by // 2 Comments

Creating reports is a common task that every vSphere administrator must deal with at least once if not many more times in their career. Whether you are tasked to provide an inventory report of all your virtual machines and their configurations to your manager or to provide a compliance report for your security team to ensure that all virtual machines are hardened according to the vSphere Security Hardening Guide, report creations can be a challenge.

The vSphere platform provides a very powerful and rich set of APIs (Application Programming Interface) that can be consumed by both vSphere administrators as well as developers. However, there is a high learning curve when using the API and it takes quite a bit of time to learn and of course your manager is expecting the report to be done in the next 5 minutes. Even with abstraction tools such as PowerCLI, quickly building a robust, scalable and performant script is not always a trivial task, not to mention the maintenance and updates to the script because your manager wants to continually add more things to the report.

So how can we make reporting so easy that vSphere administrators will no longer have to spend time digging through API documentation and instead they will be able to quickly put together reports within minutes? Well, this is something that the CloudPhysics team has been working on as part of their CloudPhysics card platform and they have built a very unique card solution to help solve this problem. I had the opportunity to get an early preview of this new card and I have to say that I am very impressed at how easy and intuitive the interface is to build simple to very complex reports using this new card solution. The coolest part of this solution is that no programming or scripting skills are required!

To give you an example on how easy it is to use the interface, I recently helped a customer with a script to identify all virtual machines that had a virtual disk using the 2gbsparse disk format. I would like to think I know the vSphere API pretty well, so putting the script together took just a few minutes because I knew exactly where to look for this information. That evening I decided to go through the same scenario, but using the new CloudPhysics card solution and I was literally able to create the report in seconds! It probably took me longer to name the report than to actually create it. As you can see, I am pretty excited about the new card solution and it will be interesting to see all the cool new reports customers can now create and share with each other.

Here is a sneak peak at what the interface looks like when creating your own custom reports:

Here is the final report that is produced:

As you can see, I have quickly narrowed down the specific virtual machine that contains a 2gbsparse VMDK and I am able to see exactly which virtual disk that is. 

If you would like to learn more about the new card solution, Irfan Ahmad, CTO of CloudPhysics will be hosting a live webinar to go over this new solution next Tuesday, June 25th at 9am PST and I would highly recommend you register for it to learn more.

Below is a bit more details on what you can expect from the webinar and you can register here.

vSphere Analytics Without Writing Code: The Quest for Missing Reports

While vSphere, the best-of-class virtualization platform, brings great efficiencies to the datacenter, reporting still presents challenges and pain to sys admins on a daily basis. CloudPhysics offers a radical new way to complete reporting for your virtual infrastructure. In addition to 20 high-impact reports, you can easily build your own and share the report template and output. When asked for asset reports, trending, activity, auditing and more, you’re never more than a few clicks away from delivery.
  
In addition to best practices and secrets to amazing mashups, you’ll learn to:
  • Create easy, visual reports for your vSphere environment
  • Add multiple vCenters in one view
  • Automate the report generation

Will Intel’s VMCS Shadowing Feature Benefit VMware’s Nested Virtualization?

by // 1 Comment

For many years now, VMware customers have been using Nested Virtualization, which is the ability to run a hypervisor such as vSphere ESXi within a virtual machine. Even though Nested Virtualization is not officially supported by VMware, customers have come to rely upon this technology for their lab environments and sometimes even production environments. VMware also heavily relies on this technology for their own internal development as well as their Hands On Lab for VMworld, which is now offered as an online SaaS (Software-as-a-Service) solution called Hands On lab Online.

Performance of Nested Virtualization has come a long way since its first introduction and it continues to get better with advancements made in hardware from both Intel and AMD. A couple of months back, I came across an article discussing a new feature from the upcoming Intel Haswell processor’s called VMCS Shadowing which aims to improve the performance of Nested Virtualization. This got me thinking about whether VMCS Shadowing could benefit VMware’s Nested Virtualization.

VMCS (Virtual Machine Control Structure) Shadowing works by reducing the frequency in which the guest VMM (virtual machine) requires assistance from the parent VMM. Its goal is to eliminate the VM-exits due to VMREAD and VMWRITE instructions executed by the guest hypervisor but this comes at a slight expense.

I reached out to one of the core engineers who helped to develop VMware’s Nested Virtualization technology, Jim Mattson, and asked whether or not we would benefit from the VMCS Shadowing feature. Well, it turns out that VMCS Shadowing can help, but we have also done some research in this area and developed some technology that would allow us to eliminate about 75% due to VMREAD and VMWRITE when running guest VMware Hypervisors using some interesting software techniques. The details of these software techniques are actually published in a research paper called Software Techniques for Avoiding Hardware Virtualization Exits on VMware’s Academic Program which is part of VMware Labs. Jim is one of the authors of the research paper and I would highly recommend you check it out if you are interested in more details.

To summarize, because of the techniques described in the paper, VMCS Shadowing will provide only a small benefit when running a VMware Hypervisor as virtual machine. However, it will greatly benefit other non-VMware Hypervisors running as a virtual machine, this is particular true for Hypervisors that perform egregious number of VMREAD and VMWRITE operations and that do not cluster well, such as VirtualBox for example.

The coolest part about the research and software techniques developed by Jim and team, is that the technology has already been incorporated into the existing VMware vSphere ESXi, Workstation and Fusion products. I often times forget that all the awesome-sauce technology that is being developed by VMware starts out in research academia and you can learn about other research topics by visiting the VMware’s Academic Program which includes publications, research papers and the popular VMware Technical Journals.