The cluster identity may lack permissions required to update the object

Lucas Peñaloza 671 Reputation points
2025-12-02T20:36:56.19+00:00

Dear;

Hi have the message:

The computer object associated with the cluster network name resource 'Cluster Name'

could not be updated in domain 'xxx.xxx.xxx.xxx.xxx' during the

Password change operation.

The text for the associated error code is: The specified network password is not correct.

The cluster identity 'CLPWGIRSQL$' may lack permissions required to update the object.

Please work with your domain administrator to ensure that the cluster identity can update computer objects in the domain.

Please tell me what I need to check

Thank you so much!!!.

Windows for business | Windows Server | Storage high availability | Clustering and high availability
Answer accepted by question author
VPHAN 44,300 Reputation points Independent Advisor
2025-12-29T21:01:36.1466667+00:00

Hello Lucas Peñaloza,

I wanted to follow up to see if you successfully completed the upgrade of the final node (PWGIRSQL1) and failing back your resources.

To summarize our last discussion regarding your final step:

The Risk: Update-ClusterFunctionalLevel is irreversible, but it is safe as long as all nodes are online, running the new OS, and you have passed the Test-Cluster validation.

The Strategy: There is no rush. You can leave the cluster in "Mixed Mode" for a few days to verify stability before running the command to lock in the new features.

If your cluster is now fully upgraded and stable, please consider accepting the answer as it helps other people sharing the same question benefit too. Thank you!

VP

Was this answer helpful?

1 person found this answer helpful.
0 comments
Answer accepted by question author
VPHAN 44,300 Reputation points Independent Advisor
2025-12-18T15:55:23.51+00:00

Hi Lucas Peñaloza,

You are definitely correct to be cautious about High Availability (HA). While a Windows Cluster can technically run with one node on a new OS and two on the old OS (Mixed Mode), your approach minimizes risk significantly.

By updating both Secondary nodes (PWGIRSQL2 and PWGIRSQL3) before you ever touch the Primary (PWGIRSQL1), you ensure that when you finally perform the failover:

  1. Stability: You are failing over to a new OS node (PWGIRSQL2) that already has another new OS partner (PWGIRSQL3) ready to sync.
  2. Redundancy: If something goes wrong with PWGIRSQL2 immediately after the upgrade, you have PWGIRSQL3 (also upgraded) as a safety net.

The Approved Execution Order:

  1. Update PWGIRSQL2 (OS + SQL).
    • Result: Node 2 is healthy and syncing again.
  2. Update PWGIRSQL3 (OS + SQL).
    • Result: Node 3 is healthy and syncing again.
    • State: Now you have 2 upgraded nodes waiting in the wings.
  3. Failover the Availability Group from Node 1 to Node 2.
    • Result: Your Production is now running on the New OS.
  4. Update PWGIRSQL1 (The old Primary).

This is the safest, most professional way to execute a 3-node cluster upgrade. Proceed with confidence.

I hope you've found something useful here. If it helps you get more insight into the issue, it's appreciated to accept the answer. Should you have more questions, feel free to leave a message. Have a nice day!

VP

Was this answer helpful?

1 person found this answer helpful.
0 comments
Answer accepted by question author
VPHAN 44,300 Reputation points Independent Advisor
2025-12-15T14:54:40.93+00:00

YES, absolutely. You must always begin your upgrade process with the Secondary Replicas. Since your cluster is now healthy, the safest rolling upgrade workflow is to upgrade the passive nodes first, leaving the active Primary (PWGIRSQL1) untouched until the very end to keep your application online.

Here is the recommended sequence:

Phase 1: Upgrade the Secondary (PWGIRSQL2)

Since PWGIRSQL2 is already a Secondary replica and you just stabilized it, it is the perfect candidate to start with.

  1. Backup: Take a full backup of your databases on the Primary (PWGIRSQL1) just in case.
  2. Suspend Data Movement (Optional but Recommended): In SSMS, you can temporarily suspend data movement for the PWGIRSQL2 replica if the upgrade will take a long time (to prevent the transaction log on the Primary from growing too large while waiting).
  3. Perform the Upgrade: Apply your Operating System updates or SQL Server patches/upgrades on PWGIRSQL2.
  4. Reboot & Validate: Restart PWGIRSQL2. Once it is back online, open SSMS on the Primary and verify that PWGIRSQL2 returns to the Synchronized (or Synchronizing) state. Phase 2: Upgrade the Next Secondary (PWGIRSQL3)

Once Node 2 is green and fully synced, repeat the process for PWGIRSQL3.

Phase 3: The Failover (Switch Roles)

Once both Secondaries (Node 2 and Node 3) are upgraded and healthy:

  1. Change the Availability Mode of PWGIRSQL2 to Synchronous Commit (if it isn't already).
  2. Perform a Manual Failover of the Availability Group from PWGIRSQL1 to PWGIRSQL2.
    • Now PWGIRSQL2 is the Primary and is running the new version.
  3. Verify applications are connecting successfully. Phase 4: Upgrade the Old Primary (PWGIRSQL1)

Now that PWGIRSQL1 is a passive Secondary, you can safely proceed to upgrade it.

VP

Was this answer helpful?

1 person found this answer helpful.
0 comments

49 additional answers

Sort by: Newest
Most helpful Newest Oldest
  1. Lucas Peñaloza 671 Reputation points
    2025-12-22T13:12:28.6333333+00:00

    Hi Vphan

    I started the upgrade on Node 3!!!.

    User's image

    We will be monitoring it!!!.

    We hope everything turns out well!!!.

    Was this answer helpful?

    0 comments
  2. VPHAN 44,300 Reputation points Independent Advisor
    2025-12-20T01:47:36.38+00:00

    Technically speaking, removing a node from an Availability Group (AG) does absolutely nothing to prevent the corruption of the SQL Engine or the Master database during an update.

    The "SQL Engine" and the master database are local resources. When you run the Cumulative Update (CU) installer, it updates the binaries (sqlservr.exe) and runs upgrade scripts against the local copy of master on that specific hard drive.

    Whether the node is logically "In" or "Out" of an Availability Group, the installer performs the exact same operations on the exact same files.

    If the CU installer is going to "wreck" the engine, it will happen regardless of whether the AG is configured or not. Removing the AG configuration provides zero protection against binary or system database corruption.

    The "corruption" your colleague likely experienced in the past happens during the "Script Upgrade Mode", the very first time SQL starts up after the patch. If this step is interrupted or fails, the master database can be left in an inconsistent state.

    The Mitigation: The correct safety net is not destroying the cluster topology. It is taking a VM Snapshot (since you are on vSphere) or a File-System Backup of the System Databases (master, model, msdb) before applying the patch. If the engine breaks, you restore the snapshot/files. You do not need to dismantle your High Availability architecture.

    Microsoft explicitly documents the procedure for patching AG nodes, and it relies on the Rolling Upgrade method where nodes remain members of the AG. They warn against removing nodes because of the re-seeding cost.

    => Reference: Upgrade and Update of Availability Group Servers with Minimal Downtime and Data Loss

    The Standard Procedure:

    Patch Secondary.

    Failover.

    Patch new Secondary (old Primary).

    So please tell your colleague that removing the node causes a guaranteed headache (re-seeding terabytes of data) to solve a problem it doesn't actually fix (local engine corruption). The correct approach is:

    Take a full backup of master and msdb (or a VM Snapshot) on PWGIRSQL2.

    Apply the CU while the node is still in the AG (but Paused in Cluster Manager, as we discussed).

    If it blows up, restore the Snapshot. If it works, you saved yourself 10 hours of work.

    VP

    Was this answer helpful?

    1 person found this answer helpful.
    0 comments

Your answer

Answers can be marked as 'Accepted' by the question author and 'Recommended' by moderators, which helps users know the answer solved the author's problem.