VMware vSAN 6.6 – What’s New for Day 2 Operations?

VMware has just announced vSAN v6.6, with over 20 new features. While new and shiny features are nice I’d like to highlight a couple that I think might be undervalued from release feature-set perspective, but highly valuable in day to day operations of a vSAN environment, otherwise known as Day 2 operations.

vSAN Configuration Assist is one such new feature. While it’s true that it helps first time configuration of a greenfield installation with vSAN (no more bootstrapping, yay!), it also helps with Day 2 operations.

It helps configure new hosts added to an existing vSAN enabled cluster, but it also makes it possible to automate updating of IO controllers, both firmware and drivers directly from within vCenter. As everyone should know by now, vSAN is highly dependent on drivers and firmware being on supported levels. This improvement helps the improved vSAN Health Check (Enhanced Health Monitoring) alert you when new and verified drivers/firmware are available, and if the controller tools are available on the ESXi host, it can also update the firmware for you.

Directly from vCenter, utilising maintenance mode like you’re used to from patching your ESXi hosts from VMware Update Manager (even if it’s not integrated into VUM at this point).

This new features takes the vSAN HCL list to a new level, it’s no longer just a list over supported IO controllers and their firmware and drivers, it’s a now also a software distribution point. At the moment Dell EMC, Fujitsu, Lenovo and SuperMicro are all supported vendors for this new distribution model, hopefully the rest will follow suit quickly.

The second feature I would like to highlight as a Day 2 operations enhancement, is the new vSAN Cloud Analytics feature. If you participate, in the Customer Experience Improvement Program (CEIP), it will enable custom alerting for your environment, based on your own environment. For instance, if a new Knowledge Base article is published, that pertain to your specific setup, it will alert you about it. One example might be if you have Intel X710 NICs, which can cause PSODs — Wouldn’t it be nice if you got alerted that this might be an issue, and then told how to remediate it? Well, with vSAN 6.6 you’ll get exactly that.

With vSAN 6.6 you get both automated, and verified, firmware/driver upgrades, as well as proactive alerting for potential issues through the hive-mind that is the analytics service. This is what VMware calls Intelligent Operations and Lifecycle Management in this release, and it’s really hard to argue with that.

Of course, vSAN 6.6 provide other Day 2 Operations enhancements as well, like Degraded Device Handling (DDH), Simplified Stretched Cluster Witness replacement procedures, Capacity and Policy Pre-Checks and access to the vSAN control plane trough the ESXi Host Client, but I’ll leave those for later posts.

Cross vCenter VM Mobility Fling – macOS?

The VMware Cross vCenter VM Mobility – CLI was recently updated so I decided to try it out. In short, this little Java based application allows you to easily move or clone VMs between disparate vCenter environments.

The Fling is listed with the following requirements:

  • JDK 1.7 or above
  • Two vCenter instances with ESX 6.0
  • Windows : Windows Server 2003 or above
  • Linux : RHEL 7.x or above, Ubuntu 11.04 or above

There is no mention of macOS there, but I decided to give it a go any way, and it turns out that it works just fine on macOS as well!
Just make sure you have the Java JDK installed locally. When I ran it the first time, I got the following error, since the JAVA_HOME environment variable was not set.

~/Downloads/xvc-mobility-cli_1.2$ sh xvc-mobility.sh
set JAVA_HOME to continue the operation

This is very easy to fix, just run the following command in your terminal of choice, and xvc-mobility.sh should work just fine on your Mac.

export JAVA_HOME=$(/usr/libexec/java_home)

Next up is running the fling with the correct parameters (this is a clone operation, not a relocate):

~/Downloads/xvc-mobility-cli_1.2$ sh xvc-mobility.sh -svc [source-vcenter] -su [source-vcenter-username]
-dvc [destination-vcenter] -du [destination-vcenter-username]
-vms [vm-name] -dh [destination-host]
-dds [destination-datastore] -op clone -cln [destination-vm-name]
...
13:41:40.591 [main] INFO com.vmware.sdkclient.vim.Task - CloneVM_Task | State = SUCCESS | Error = null | Result = com.vmware.vc.ManagedObjectReference@d9a221a7
13:41:40.597 [main] INFO com.vmware.sdkclient.vim.Task - Monitor task end
13:41:40.597 [main] INFO com.vmware.sdkclient.vim.Task - CloneVM_Task took : 0:51:33.728
13:41:40.603 [main] INFO c.v.s.helpers.CrossVcProvHelper - Successfully cloned the vm:[destination-vm-name]

I was able to clone a VM from my lab in Bergen to my lab in Oslo, without any problems what-so-ever. Not only is that a Cross vCenter vMotion, but also a Cross Country one, awesome!

Now this is just an example, please check the official documentation for all the parameters, and what the tool expects.

Got VMware vRealize Log Insight?

Recent conversations with existing and potential clients made me realize that many are not aware that they most likely are entitled to use VMware vRealize Log Insight in their environment. For free.

Back in March 2016 VMware announced that everyone with a valid VMware vCenter license are also entitled to  25-OSI pack of vRealize Log Insight for vCenter Server. This means that you can gather logs for up to 25 ESXi hosts, VMs or other devices, in your environment.

It even allows you to use the following VMware content packs (3rd party content packs requires a full Log Insight license)

  • Horizon View
  • NSX
  • OpenStack
  • vCenter Operations Manager
  • vCloud Automation Center
  • vCloud Director
  • vCNS
  • Virtual SAN
  • vRealize Automation
  • vRealize Operations Manager
  • vSphere

This should be a no-brainer for everyone with a >24 host VMware vSphere environment, the value that vRealize Log Insight provides is enormous, especially when it comes to troubleshooting. Why 24 when you get a 25 pack you may ask? Well, you’ll want to use one of them for vCenter itself, leaving 24 for your ESXi hosts. If your environment is smaller than 24 hosts, the remaining OSI’s can be used to monitor just about anything that can log to a syslog service, like switches, routers and other devices.

Credits: http://www.gratisography.com
Credits: gratisography.com

So, if you’re not running it already, logon to my.vmware.com and download it today — you can then use your existing vCenter license to activate it. You’ll be drinking from the log-hose in no time.

 

#vDM30in30 progress: