Month: January 2021

How to Resolve Horizon View vCenter Connection Errors

I was recently working with a client that needed me to migrate Horizon View VMs to new storage. I thought it would be as easy as changing the storage settings for the pool and performing a rebalance across the cluster. Unfortunately, no rebalance operation was successful and I saw the following errors:

Provisioning error occurred for Machine XXX: Refit operation rebalance failed

vCenter at address <vCenter Address> has been temporarily disabled (this error would typically followed by another notification that the same vCenter had been enabled)

I was able to resolve the issue by following VMware KB 1030996. I the case of this customer there was only one working production pool. To test that there was an issue with the pool in use, I created a new temporary pool. I then tried recompose actions and looked for errors in the event log. There were none.

Creating a new temporary pool proved critical to resolving this issue. The crux of the problem as laid out in the KB is that there are two vCenter entries in the composer database. In my case the IP address and the FQDN (The FQDN being the correct entry). The correct Deployment Group ID was displayed in the View Composer Database entry for the new temporary pool I created. I was able to take that ID and replace it in the entries for the current production pool. After that was done, I was able to easily rebalance the production pool.   

How to Detach/Attach Nutanix AHV Disks

This is the workflow on how to migrate a Nutanix AHV disk from one VM to another. A reason you may do this is because you have an old Server 2008 file server that you want to migrate to Server 2019. Nutanix handles this differently than VMware in that AHV does not allow you detach and reattach disks as you would in vSphere.

The process to do this is outlined in both Nutanix KB 3577 and KB 8062. Unfortunately, the way it is worded can be confusing to some without a lot of Nutanix experience. This process requires you to open an SSH (using Putty, Tera Term, etc.) session to one of the CVMs in the cluster.  

Step 1. Find the VM Disk File

To find the VM Disk file you will use the command below:

 acli vm.get <VM name> include_vmdisk_paths=1 | grep -E 'disk_list|vmdisk_nfs_path|vmdisk_size|vmdisk_uuid'

The output should look like the picture below.

Make note of the disk size and disk path. The disk size should correspond with the data disk you would like to migrate to your new VM.

Step 2. Create an Image from the VM Disk

To create a disk image you must run the following command:

acli image.create <image name> source_url=<url of source VM disk> container=<target container name> image_type=kDiskImage

Note: To get the source URL you will have to append nfs://127.0.0.1 to the nfs path output from step 1.

For example - nfs://127.0.0.1/BronzeTier_StorageContainer01/.acropolis/vmdisk/be06372a-b8c5-4544-b451-12b608615248 

The output of the command should appear as shown below.

Step 3. Attach the disk to the new VM.

Attaching the disk to the new VM can be done from Prism Element on the same cluster.

  1. Locate the VM from the VM menu in Prism and click update.

2. From the Update screen select +Add New Disk

3. In the Add Disk menu, select ‘Clone from Image Service’ from the Operation drop down menu.

4. In the Image menu, select the image you created in Step 2 and click Add.

Once this is completed you can log into your VM and initialize the disk in the operating system.