Hi ServiceInformatiqueICONsa-8160,
your RDS farm is generating Event ID 158 because the connection broker provisions new user profiles by executing a raw file copy of the master UVHD-Template.vhdx. This primitive cloning mechanism perfectly duplicates the internal disk signature. When multiple users log into the same Session Host, the operating system mounts multiple virtual hard disks containing the exact same disk identifier, which immediately triggers the collision warning in the Event Viewer.
Microsoft has not updated the UPD provisioning engine to randomize signatures upon creation, there is no registry toggle to alter the broker's behavior. To avoid running the reset command manually, you must automate your workaround. You can open the Task Scheduler on your RDSH nodes and create a task triggered by user logon events, specifically looking for Event ID 1149 in the TerminalServices-RemoteConnectionManager operational log. You can attach a PowerShell script to this task that executes the ResetDiskIdentifier command against the newly mounted volume, automatically clearing the collision the moment the profile is generated.
However, the most architecturally sound resolution for a modern Windows Server 2025 deployment is to deprecate User Profile Disks entirely and migrate your users to Microsoft FSLogix Profile Containers. FSLogix is the official enterprise successor to UPDs and is licensed at no additional cost for environments holding RDS Client Access Licenses. The FSLogix agent dynamically compiles, formats, and mounts the VHDX containers during the initial user login rather than executing a raw template copy. This advanced provisioning mathematically guarantees that every single profile disk possesses a completely unique identifier, permanently eliminating the Event ID 158 collisions while providing vastly superior login performance.
Hope this answer has brought you some useful information. If it did, please hit “accept answer”. Should you have any questions, feel free to leave a comment.
VPHAN