What's Latest :

What's New in vSphere 6.5 - Proactive HA

As we all aware there are lot's of new features in VMware 6.5 release. I'm so impressed with this and always I'm spending my time with my vSphere 6.5 Lab. While I was playing around I thought to introduce these new features as a series of posts from my blog. 

Hope you know about VMware high availability features and VMware added a native vCenter high availability feature with this release, you can read my previous posts about this native high availability from here. In this post I'm going write about Proactive HA feature which is introduced with VMware 6.5 release. First of all I would like to mention that this is actually not a feature of HA and it is the next level of DRS which is proactively using the HA feature. I know you may confused with that but don't worry. As I mentioned earlier this feature is a next level of DRS which is designed to provide maximum high availability to your Virtual Machines from any unexpected hardware failures. If you are a Virtualization admin you may experienced with certain failures and encountered with these down times. But did you check back when this hardware failure actually occurred or when was the first symptom of this failure notified. Sometimes you may felt that you could have actually avoid this if you take the actions before. Sometimes it may be a human error, you did not noticed the alert you received from your monitoring device. But with this feature vCenter is going to take care of these situations and it will do the necessary precautions before something go in to a worse situation. 

This feature detects the possible hardware failures from the OEM hardware sensors and migrate possible Virtual Machine out of the host to the other hosts just to avoid any unexpected situation. As an example, let's say there are two power supplies attached to a host and one is getting failing, but host is still able to run the VMs without any issues. But what happens if the other one is getting failed. If you noticed this at the first stage you may take the correct precautions and replace the power supply. I've seen in many cases due to a human error such as ignoring the alerts or not taking care of the notification correctly it ended up with big situations. 

Enough talks for now and let's dig in to the deep and see what is this new feature....

Host "Quarantine" Mode introduced

Quarantine mode evacuates the VMs running in the host if it satisfies the below states in a hardware failure (in a complete hardware failure still we have our HA feature and it will take care of the affected VMs) : 

  • No Impact to the performances of the VMs 
  • No DRS Afinity/anti-Affinity rules are violated 
Then the VMs will evacuate from the host and DRS will not place new VMs in this host until we fix the situation. This host now in the "Quarantine" mode

Ok, You know that if in a situation of failure we can put the host in to the Maintenance mode. So, next question is what is the difference between "Quarantine" mode and the "Maintenance" mode?

In the Quarantine mode host will utilize the resources of the affected host if there are not enough resources in the cluster due to the demand of the the resources. In the Maintenance mode host is completely out from the cluster.

For now Cisco, Dell and HP vendors are supporting to this and all the other vendors will have the plugins in place to work with this as this is a really cool feature.

vCenter 6.5 Proactive HA Cluster Configuration

Login to your vCenter 6.5, select the cluster go to configuration tab, select the "vSphere Availability", select "Edit" button to turn on the Feature


Note: you need to turn on the DRS feature first before you enable the Proactive HA feature, otherwise it will not allow you to turn on the feature. 


To turn on the DRS select the vSphere DRS and click-on "Edit" 


Once you get the DRS Cluster settings window click the check box to enable the feature, there are new features in DRS as well but it is not the time to discuss them. 


Let's go back to the Proactive HA window, now I have the option to enable the feature and click on the check box to enable it.


You can view the small summary of the feature by click on the gray exclamation mark


Once you enabled the Proactive HA feature go the "Proactive HA failures and Responses" option in the same window 


You can select the "Automation" Level or the "Remediation" level from the drop down menu or by expanding the table as below



Remediation Options are self explanatory and there are three main remediation modes. 

  • Quarantine mode for all failures - due any kind of failure (severe or degradation) will put host in to the Quarantine mode 
  • Quarantine mode for moderate and Maintenance mode for severe failures - degradation will put host in to the Quarantine mode and Severe failures put in to the Maintenance mode
  • Maintenance mode for all failures -
    due any kind of failure (severe or degradation) will put host in to the Maintenance mode

In the below section you can avoid and block any failures for a cluster or a certain host, these features might be different from the vendors, click on the check box to enable the block conditions and click on "edit" to select the condition 


You can select the failures on your left hand side and the host or the entire cluster on your right hand side


That's for the Proactive HA and let's see what are the other features in vSphere 6.5

No comments