I recently had access to four Dell VxRail E560F nodes and wanted to see if I could repurpose them for a Nutanix AHV lab environment. This is not intended as a supported production design, but as a practical lab exercise for reusing existing hardware.
First, we need to convert the VxRail node to an ordinary PowerEdge server. Please see this link for more information: Changing a Dell VxRail Identity
Continue readingThis article provides a quick overview of the steps I used to convert a former Dell VxRail node back into a standard Dell PowerEdge server. The objective was to remove the VxRail-specific configuration and identity, allowing the system to operate as a regular PowerEdge host for lab use and workloads unrelated to VMware or vSAN.
The server used in this example was a Dell VxRail E560F, which needed to be rebranded and restored to its original Dell PowerEdge R640 identity.
Disclamer: Remember that changing the identity and firmware baseline of former appliance hardware may have service / support implications.
Continue readingDisclaimer: This procedure changes the license state of a vSAN Witness Appliance and may temporarily affect the witness configuration in a stretched cluster or 2-node cluster. Make sure you have a maintenance window and verify the health/state of the cluster before you begin.
Broadcom/VMware documents that the vSAN Witness Appliance includes an embedded vSphere license and does not require a separate purchased license for the virtual appliance itself. There is also an official KB for cases where the witness shows up with an expired or evaluation license in vCenter after a major upgrade. But if the original embedded key has actually been overwritten, the normal disconnect/reconnect workaround may not be enough.
I recently ran into a situation where the embedded license on a vSAN Witness Appliance had been overwritten with a regular ESXi license key.
The witness appliance is supposed to use its built-in embedded license, so this needed to be reverted.
If the witness is still part of a vSAN stretched cluster or 2-node configuration, remove it from that configuration first.
Below is the procedure that worked for me.
*** Update 08-03-2026 ***
Updating to ESXi 8 Update 3i did not help. I’m still seeing PSOD after a few days running on the realtek driver 🙁
*** Update 26-02-2026 ***
This showed up in “ESXi 8.0 Update 3i Release Notes” that was just released last night :
So there might be a change that this fixes it. My lab host with the realtek driver has been updated now – stay tuned for an update!
*** Original post ***
I’m trying to figure out why my Maxtang NUC is experiencing a PSOD every 3–4 days after I started using the Realtek driver fling (there are no PSODs when using the USB fling).
Realtek driver:
VMware-Re-Driver_1.101.00-5vmw.800.1.0.20613240.zip
Does anyone have any clues? If so, please reply in the comments section 🙂
If you encounter a 504 Gateway Timeout error at 99% of the installation prerequisites, the issue may not actually be with VxRail itself. Instead, check the vCenter server task list and look for the vLCM task “Sync updates”
Continue reading
I recently encountered an error in VMware vLCM: “Feature cannot be enabled on this cluster”. This was followed by another message: “Failed to fetch vSAN witness host associated with the cluster.”
I have recently encountered the following error on several Lenovo VX systems, which prevents vLCM from initiating the upgrade from ESXi 7 to ESXi 8:
Error: Version iavmd-3.0.0.1038 of the manually added Component Intel VMD driver with VROC support is an unsupported downgrade from version iavmd-3.5.1.1002 on the host.
Replace the Component in the image with one of the same or higher version.
If you encounter the error mentioned above while using VMware vLCM in conjunction with the “HPE OneView for vCenter Plugin,” there is a high likelihood that you can resolve the issue by executing a specific command directly on the ESXi host. To do this, you will need to access the ESXi shell or SSH into the host:
sut -set mode=AutoDeploy
This command sets the mode to “Auto Deploy,” which is essential for ensuring proper functionality with VMware vLCM and the HPE OneView integration.
To verify the current mode of your ESXi host, you can retrieve this information by executing the following command:
sut -exportconfig
I recently encountered the above error in VMware vLCM while working in a multi-site vCenter environment. The issue was initially identified by a site administrator who had full administrative rights over the datacenter object they managed.
The root cause of this issue lies in the site administrators restricted access. Although they had full permissions for their respective datacenter, they lacked global administrative privileges across the entire vCenter. For vLCM to function correctly, broader access rights are required.
After investigating the vCenter roles and permissions, I was able to identify the minimal privileges needed to resolve the issue without granting excessive access.
The solution:
Continue readingIf disabling the Efficient Cores (in BIOS) on your new NUC system doesn’t help you to install ESXi 8 on your new NUC – then read on!
Continue reading