Azure local deployment cleanup

Question

We are not able to proceed further with azure local deployment and getting error "Unable to launch deployment: a prior deployment attempt has not been fully cleaned up."

We have tried to delete the resources from the portal on the resource group and then tried again, but still we are getting the same error. Can anyone advice how to clean up the prior deployment?

Here is the full error

"Type 'InvokeEnvironmentChecker' of Role 'DeploymentService' raised an exception: System.Management.Automation.RuntimeException: Fail to initialize cloud deployment: Unable to launch deployment: a prior deployment attempt has not been fully cleaned up. Command Arguments ------- --------- Initialize-CloudDeployment.ps1 {JSONFilePath=C:\Deployment\Unattended.json, RegistrationResourc... ValidateJob.ps1 {exceptionLogXml=C:\MASLogs\LCM_Controller_Validate_Exception202... <ScriptBlock> {} Command Arguments ------- --------- Initialize-CloudDeployment.ps1 {JSONFilePath=C:\Deployment\Unattended.json, RegistrationResourc... ValidateJob.ps1 {exceptionLogXml=C:\MASLogs\LCM_Controller_Validate_Exception202... <ScriptBlock> {} at checkForInvokeTimeout, C:\NugetStore\Microsoft.AzureStack.Role.Deployment.Service.10.2503.0.907\content\Classes\DeploymentService\DeploymentService.psm1: line 1440 at InvokeEnvironmentChecker, C:\NugetStore\Microsoft.AzureStack.Role.Deployment.Service.10.2503.0.907\content\Classes\DeploymentService\DeploymentService.psm1: line 435 at <ScriptBlock>, C:\NugetStore\Microsoft.AzureStack.Solution.LCMControllerWinService.10.2503.0.907\content\LCMControllerWinService\InvokeInterfaceInternal.psm1: line 139 at Invoke-EceInterfaceInternal, C:\NugetStore\Microsoft.AzureStack.Solution.LCMControllerWinService.10.2503.0.907\content\LCMControllerWinService\InvokeInterfaceInternal.psm1: line 134 at <ScriptBlock>, <No file>: line 36"

Answer 1

Hi, when Azure Local (Azure Stack HCI 23H2+) runs “Begin cloud deployment”, it drops files and registry keys on each node. If the deployment fails, crashes, or is cancelled, leftover artefacts trigger the error “a prior deployment attempt has not been fully cleaned up.” To fix, manually stop LCM services, delete folders (C:\CloudDeployment, C:\ECEStore, C:\NugetStore), remove the registry key HKLM:\SOFTWARE\Microsoft\LCMAzureStackStampInformation, destroy any half-built cluster, clear NetIntents if used, reboot all nodes, cancel the failed ARM deployment in Azure, and delete any Deployment Plan, NetIntent RG, edgeDevices, Arc resources, and Key Vaults left behind. Disconnect Arc if needed. Then relaunch the wizard from the Azure portal. To avoid future issues: use a separate management PC, don’t close the wizard tab too early, keep node clocks in sync, and stay on the latest build. If problems persist, gather logs from C:\MASLogs and C:\CloudDeployment\Logs, open a support ticket, and Microsoft can pinpoint what’s stuck. This cleanup usually clears the error without OS rebuilds.