Domain Controller Promotion fails on One Cluster

[NWPROXADMIN]

Hi everyone,

We're troubleshooting an issue affecting one of our Proxmox clusters and are looking for guidance on where to focus our investigation.

Environment​

We have two Proxmox clusters that are configured similarly.

• Cluster A: No issues
• Cluster B: Experiencing multiple problems

We have several production VMs running on the affected cluster that appear to operate normally, but we're consistently running into issues when deploying and promoting new Domain Controllers.

Problems Observed​

Domain Controller Promotion Failure​

When promoting a Windows Server VM to a Domain Controller, the promotion process fails with:

Error 501
Database failed to initialize
Show more lines

The associated AD DS log files appear to be corrupted.

Extremely Slow Server Manager​

On affected VMs:

• Server Manager can take 15–30 minutes to load.
• General management tasks are noticeably delayed.

Role Installation Issues​

The VM intermittently fails to install Windows Server roles and features.

Sometimes installations complete successfully, while other times they fail without an obvious pattern.

Storage Pool Disappeared​

One of our SAN-backed storage pools appears to have disappeared from Proxmox.

However:

• Existing VMs on that storage continue to run.
• VM performance on that storage appears normal.
• We have not observed any obvious performance degradation.

Important Note​

The Domain Controller VM being promoted is not located on the missing storage pool.

Troubleshooting Performed​

We have already tested:

VirtIO Drivers​

• 1.248
• 1.266
• 1.285

VM Configuration Changes​

• Different storage emulations
• Different VM IDs

Installation Sources​

• Multiple Windows installation ISOs
• Different installation media

Windows-Level Troubleshooting​

• OS repairs
• PowerShell-based role installation
• PowerShell-based AD DS promotion

None of these changes resolved the issue.

Additional Context​

• We have successfully deployed and promoted multiple Domain Controllers on our other cluster without any issues.
• Other VMs on the affected cluster appear to function normally.
• The issues seem isolated to newly created Windows Server VMs and/or the cluster itself.
• The missing storage pool may be unrelated, but its disappearance around the same timeframe makes us suspicious that there could be an underlying storage or cluster issue.
• All Windows server instances are windows server 2025 and have to be that.

Questions​

1. Has anyone seen Domain Controller promotion failures accompanied by database initialization/log corruption errors in Proxmox?
2. Could intermittent storage issues cause AD DS promotion failures while still allowing existing VMs to operate normally?
3. Are there specific Proxmox logs, storage diagnostics, or SAN checks we should focus on?
4. Has anyone encountered a storage pool disappearing from the GUI while VMs on that storage continued running normally?

Any guidance would be appreciated. We're trying to determine whether we're looking at a Windows issue, a storage issue, or a broader cluster problem.

Thanks.

 

[Francois Baillargeon]

Any way you can paste a PVReport in here?

Also, what does

pvesm status

output?

 

[NWPROXADMIN]

Small update the trouble node actually went completely offline and is not coming back up after a hardware reboot. The internal storage drive died and it was running from memory.

 

[Francois Baillargeon]

Oh man! yeah that sure does not help!