Finding in the Wild | Citrix MCS – Windows Rearm Failure After November 2026 Updates

I’ve recently encountered the same issue at two of my customers:

After installing the November 2026 Windows updates, image preparation through Citrix Machine Creation Services (MCS) fails.

And apparently, I’m not the only one. Several administrators have reported similar issues in the comments on Günter Born’s blog.

⚠️ Windows Updates Under Investigation:

  • Windows Server 2019 – KB5131085
  • Windows Server 2022 – KB5131090
  • Windows Server 2025 – KB5131080

🔍 The Issue

MCS image preparation fails with the following error:

Image preparation failed to rearm the copy of Windows installed on the master image.

As a result, the updated master image cannot be deployed successfully.

🔬 Technical Analysis

The failure occurs during the Windows rearm operation as part of the MCS image preparation process.

Whether this is caused by changes to the Windows Software Protection Service, Windows activation mechanisms, or the interaction between Windows and Citrix MCS remains unclear.

The timing strongly suggests a connection to the November Patch Tuesday updates, but correlation alone does not establish the root cause.

🛠️ My Recommendations

1️⃣ Install the November updates on test master images first.

2️⃣ Validate the complete MCS image preparation process — not just whether Windows boots successfully.

3️⃣ If preparation fails, investigate Citrix ProvTask details, preparation VM logs, and Windows activation events.

4️⃣ Keep the last known-good master image snapshot available for rollback.

5️⃣ Postpone production Machine Catalog updates until the updated image has been successfully validated.

I would also avoid manually modifying slmgr /rearm or Windows activation settings until the underlying cause has been identified.

❗ Current Status

I’ve reproduced this issue in two customer environments, and several similar reports have appeared in the community.

As of now, there is no official confirmation or documented fix from Citrix or Microsoft.

Has anyone identified a reproducible root cause or a working workaround?

I’m particularly interested in the CVAD versions and Windows activation methods (KMS, MAK, or ADBA) used in affected environments.

Finding in the Wild | Citrix MCS + Nutanix AHV – No More Than 10 VMs per Batch

Yesterday, during a Citrix MCS project with Nutanix AHV / Prism Central, I encountered some interesting behavior. Adding a larger number of VMs to an MCS machine catalog may fail with: OutOfMemory . Initially, I suspected resource issues

Yesterday, during a Citrix MCS project with Nutanix AHV / Prism Central, I encountered some interesting behavior.

Adding a larger number of VMs to an MCS machine catalog may fail with:

OutOfMemory

.

Initially, I suspected resource issues on Nutanix or in the Citrix infrastructure. However, Citrix already knows about the issue and documents it under PMCS-58254.

Citrix currently explicitly recommends:

👉 Add no more than 10 VMs to an MCS catalog at a time.

The workaround is correspondingly simple: for larger rollouts, create machines in batches of no more than 10 VMs.

Interestingly, the issue is documented not only in the new release, but also for 2507 LTSR and Citrix DaaS.

Another small Citrix detail that never appears with five VMs but may suddenly cost significant time during a larger rollout. 😉

Finding in the Wild | Citrix CVAD 2603 – Check the MCSIO Driver

Citrix documents a known issue in CVAD 2603 involving machine catalogs with MCS I/O Storage Optimization. If the Citrix MCS I/O Driver is missing from the master image, a later machine catalog update to a new snapshot may

Citrix documents a known issue in CVAD 2603 involving machine catalogs with MCS I/O Storage Optimization.

If the Citrix MCS I/O Driver is missing from the master image, a later machine catalog update to a new snapshot may fail. The tricky part: the VDA installation itself may initially appear successful.

Recommendation

Before updating a production catalog, check:

– MCSIO is enabled in the catalog
– The MCS I/O Driver is installed in the master image
– The image has been published in a test catalog
– A working rollback snapshot is available

A registered VDA is not yet a fully validated MCS image.

Finding in the Wild | MCSIO Cache Disk Exists, but Stays Empty

Another Citrix issue where everything initially looks correct. You update the environment to VDA 2607. MCS provisions the machine. The write-cache disk is present and correctly attached to the VM. VDA registered. Users can work. Looks good. Except

Another Citrix issue where everything initially looks correct.

You update the environment to VDA 2607.

MCS provisions the machine.

The write-cache disk is present and correctly attached to the VM.

VDA registered.

Users can work.

Looks good.

Except for one small detail:

👉 The MCSIO write-cache disk stays empty.

The combination matters

Citrix currently documents this issue under PMCS-63379 for this exact combination:

**Windows Server 2019 * VDA 2607 * MCSIO**

The cache disk is successfully attached to the provisioned VM, but is not used as a write-back cache as intended.

The tricky part:

If your upgrade checks only confirm that the cache disk exists, this issue is easy to overlook.

The workaround is almost typically Citrix 😎

On the master image, open the following registry path:

Create an empty registry key named:

Parameters

Yes.

An empty registry key.

Then:

Shut down/reseal the master image → create a new snapshot → update the MCS catalog → restart the machines.

Afterwards, normal user activity should populate the write-back cache disk again.

Why I find this interesting

MCSIO exists specifically to absorb writes through RAM and the temporary cache disk.

Citrix uses the following file as part of this:

mcsdif.vhdx

If the disk is attached but not used, simply looking at the VM configuration is not enough.

My recommendation after a VDA upgrade

With MCSIO, do not check only:

✅ Cache disk attached

Also check:

✅ MCSIO driver present
✅ Cache actually active
✅ mcsdif.vhdx present
✅ Write-cache disk receives writes
✅ Catalog update works

Because:

Disk attached apparently does not automatically mean cache working. 😉

Affected: VDA 2607 + Windows Server 2019 + MCSIO

Citrix issue: PMCS-63379

Finding in the Wild | When Identical Machine SIDs Suddenly Become a Citrix Problem

For years, this was completely normal in Citrix environments: Create a golden image → use MCS or PVS to provision hundreds of VDAs → done. Machines from the same image may share the same local machine SID .

For years, this was completely normal in Citrix environments:
Create a golden image → use MCS or PVS to provision hundreds of VDAs → done.

Machines from the same image may share the same local machine SID.
Usually, that was not a major problem.

After all, each computer has its own Active Directory computer account with a unique domain SID.

With stricter Windows security hardening, however, this old architectural assumption becomes interesting.

What can happen?
When two cloned VDAs communicate, certain machine-to-machine connections may be blocked because their local identities are identical.

For example, this may affect:
❌ RDP between VDAs
❌ SMB / file sharing
❌ Printer sharing
❌ Certain NTLM/Kerberos scenarios

The Windows event log may show LsaSrv Event ID 6167.

And this is where it gets interesting:
The VDAs are correctly domain joined.
The AD computer accounts are unique.
DNS is correct.
Kerberos also initially looks fine.
Yet the connection is blocked.

Why?
Microsoft is increasingly tightening machine identity validation.
A design decision that was unproblematic for cloned Windows systems for years suddenly collides with more modern Windows security hardening.

This does not affect an obscure Citrix feature.

It affects one of the basic principles of:
MCS and PVS: creating many machines from a single golden image.

Citrix itself is now emphasizing the security of local persistent accounts in shared MCS/PVS images and explicitly recommends not creating such accounts in the master image.

What does this mean in practice?
If communication suddenly fails after Windows security updates between two VDAs in the same catalog:
Do not immediately dismantle firewalls, DNS or Citrix policies. 😉

Also check:
System event log →
LsaSrv
and particularly Event ID 6167.

For me, this is a good example of why golden-image architectures now need testing against more than just new Citrix versions.