Azure Migrate Physical Server Discovery - ServerDiscoveryService.exe Crash Bug

Question

Azure Migrate Physical Server Discovery - ServerDiscoveryService.exe Crash Bug

cbrnit 10

Summary

The Azure Migrate appliance for physical server discovery fails to complete discovery due to a crash bug in ServerDiscoveryService.exe. The service successfully connects to target servers but crashes during WSMan transport cleanup before any discovery data is collected.

Environment

Appliance OS: Windows Server 2022 Standard Evaluation (Build 20348)
Appliance Type: Physical server discovery (script-based installation)
ServerDiscoveryService.exe Version: 2.0.3300.663
.NET Version: 8.0.22 (CoreCLR 8.0.2225.52707)
Target Servers: Windows Server (various) and Linux, all on-premises
Discovery Agent Version: 2.0.03300.663
Appliance Configuration Manager Version: 6.1.294.1847

Symptoms

Target server validation succeeds in the appliance configuration manager
CIM sessions connect successfully (logs show "TestConnection succeeded for CIM Session with HTTP protocol")
Connections are immediately disposed with "Disposing all connections when the process is shutdown"
No discovery data is collected
Azure portal shows error 60001 with misleading "Could not load file or assembly 'Microsoft.Management.Infrastructure'" message
Discovery status remains "Discovery Incomplete" for all Windows servers

Root Cause

The ServerDiscoveryService.exe process crashes repeatedly with an unhandled NullReferenceException in the WSMan transport finalizer. This is visible in the Windows Application Event Log:

Application: ServerDiscoveryService.exe
CoreCLR Version: 8.0.2225.52707
.NET Version: 8.0.22
Description: The process was terminated due to an unhandled exception.
Exception Info: System.NullReferenceException: Object reference not set to an instance of an object.
   at System.Management.Automation.Remoting.Client.BaseClientTransportManager.CloseAsync()
   at System.Management.Automation.Remoting.Client.WSManClientSessionTransportManager.CloseAsync()
   at System.Management.Automation.Remoting.Client.BaseClientTransportManager.Finalize()

The crash also triggers an access violation:

Faulting application name: ServerDiscoveryService.exe, version: 2.0.3300.663
Exception code: 0xc0000005
Faulting application path: C:\Program Files\Microsoft Azure Server Discovery Service\ServerDiscoveryService.exe

These crashes occur approximately every 10 minutes.

Troubleshooting Completed

Verified manual connectivity works: PowerShell Invoke-Command and New-CimSession both succeed from the appliance to target servers using the same credentials
Verified WinRM configuration: Targets have WinRM HTTP listener on port 5985, LocalAccountTokenFilterPolicy is set to 1
Verified assemblies exist: Microsoft.Management.Infrastructure.dll is present in the GAC on both the appliance and target servers
Tested both FQDNs and IP addresses: Same failure occurs with both
Tested both local and domain credentials: Same failure with properly formatted credentials (domain\user)
Verified time synchronization: Appliance clock is accurate
Verified appliance is up to date: All components show current versions
Tested with fresh appliance: Previously tried OVA-based appliance with similar results; rebuilt using Microsoft's PowerShell script installer on clean Server 2022—same issue

Relevant Log Locations

C:\ProgramData\Microsoft Azure\Logs\ConfigManager\ClientOperations_*.log - Shows successful CIM connections followed by immediate disposal
C:\ProgramData\Microsoft Azure\Logs\ConfigManager\ApplianceOnboarding-Portal-*.log - Shows error 60000 "UnhandledException" with message "Internal error occured." (note: typo is in original)
Windows Event Log (Application) - Contains the actual crash stack traces

Conclusion

This is a code defect in ServerDiscoveryService.exe—a null reference exception in a finalizer is a programming error that cannot be caused by configuration or environmental factors. The service connects successfully but crashes before completing its work.

Request

Please escalate to the Azure Migrate engineering team for a bug fix in ServerDiscoveryService.exe version 2.0.3300.663.

Anonymous

2026-01-30T12:53:46.72+00:00

Hello cbrnit,

Thank you for the detailed write-up and for sharing the logs, event traces, and the steps you’ve already validated. We acknowledge the behavior you’re seeing with the Azure Migrate physical server discovery appliance, where ServerDiscoveryService.exe establishes connectivity successfully and then terminates during the discovery workflow.

Based on the information provided, it’s clear that the required prerequisites (WinRM, CIM connectivity, credentials, time sync, and appliance health) are in place, and that the failure occurs after successful connection validation. The crash details from the Windows Application Event Log are particularly helpful in understanding where the process is terminating.

At this stage, we are internally checking on additional troubleshooting steps to suggest beyond what you’ve already completed. Also, we are reviewing the details shared here to better understand the behavior and determine the appropriate next steps.

Thank you again for taking the time to document this thoroughly. We appreciate your patience and will follow up on this thread if further guidance becomes available.
Arno Verhelst 0 Reputation points

2026-02-12T14:27:00.1933333+00:00

Any fix yet? We have urgent migrations pending.
Suchitra Suregaunkar 9,270 Reputation points Microsoft External Staff Moderator

2026-02-19T06:50:16.3633333+00:00

Hello Arno Verhelst

Could you please share us the details asked via Private Message?

Thanks,
Suchitra.
- terry-at-TTT 0 Reputation points

2026-02-24T05:29:24.2566667+00:00

i have exactly the same problem, ServerDiscoveryService.exe run and crash every 10 minutes.

1 answer

Your answer

Anonymous

2026-01-30T12:53:46.72+00:00

Hello cbrnit,

Thank you for the detailed write-up and for sharing the logs, event traces, and the steps you’ve already validated. We acknowledge the behavior you’re seeing with the Azure Migrate physical server discovery appliance, where ServerDiscoveryService.exe establishes connectivity successfully and then terminates during the discovery workflow.

Based on the information provided, it’s clear that the required prerequisites (WinRM, CIM connectivity, credentials, time sync, and appliance health) are in place, and that the failure occurs after successful connection validation. The crash details from the Windows Application Event Log are particularly helpful in understanding where the process is terminating.

At this stage, we are internally checking on additional troubleshooting steps to suggest beyond what you’ve already completed. Also, we are reviewing the details shared here to better understand the behavior and determine the appropriate next steps.

Thank you again for taking the time to document this thoroughly. We appreciate your patience and will follow up on this thread if further guidance becomes available.
Arno Verhelst 0 Reputation points

2026-02-12T14:27:00.1933333+00:00

Any fix yet? We have urgent migrations pending.
Suchitra Suregaunkar 9,270 Reputation points Microsoft External Staff Moderator

2026-02-19T06:50:16.3633333+00:00

Hello Arno Verhelst

Could you please share us the details asked via Private Message?

Thanks,
Suchitra.
- terry-at-TTT 0 Reputation points

2026-02-24T05:29:24.2566667+00:00

i have exactly the same problem, ServerDiscoveryService.exe run and crash every 10 minutes.

Answer 1

The issue you are experiencing with the ServerDiscoveryService.exe crashing during the WSMan transport cleanup appears to be a code defect, specifically a NullReferenceException that occurs in the finalizer of the WSMan transport manager. This type of error typically indicates a programming oversight rather than a configuration or environmental issue. Given that the service connects successfully to the target servers but crashes before any discovery data is collected, it suggests that the problem lies within the service's handling of connections rather than the connectivity itself.

Since you have already performed extensive troubleshooting, including verifying connectivity and configuration settings, the next step would be to escalate this issue to the Azure Migrate engineering team for a potential bug fix in the ServerDiscoveryService.exe version 2.0.3300.663. This escalation is necessary as it seems to be a recurring crash that affects the functionality of the Azure Migrate appliance for physical server discovery.

References:

Share via

Azure Migrate Physical Server Discovery - ServerDiscoveryService.exe Crash Bug

Summary

Environment

Symptoms

Root Cause

Troubleshooting Completed

Relevant Log Locations

Conclusion

Request

1 answer

Your answer