none
Re-deployment issues after an initial allegedly successful instalation RRS feed

  • Question

  • After carefully reading and following the documentation, I thought I had a successful deployment of ASDK version 1907 on a Dell PowerEdge R620 running Windows Server 2016. However, when I went to hand over the build to the team that needed it, I realized that there was some kind of issue with networking and/or the virtual NICs that would not allow us to get at the deployment via any method other that the Dell iDrac Virtual Console. I didn't realize this as I had been deploying via that console because the server is physically in a disparate location. Since we couldn't figure out what the issues were, I decided to re-deploy from scratch.

    Since then, and after multiple attempts, I have been unable to get a successful deployment using the exact same documentation and hardware. After re-trying a few times on the same build, Server 2016 and ASDK 1907, I gave up for a bit and then realized there was a newer version of ASDK (1908) so I decided to try again, this time with Windows Server 2019 and ASDK 1908. Unfortunately I am still unable to get a successful deployment and I really don't know what the issues are.

    I am going to start from scratch again and verify all the pre-requisites and then I will wait until someone hopefully answers the bell here with some suggestions before I start running the script(s) again. I can supply whatever other data you need to troubleshoot further. Thanks.

    Derek

    Friday, September 20, 2019 6:11 PM

Answers

  •  The error you described using the pre-checker can be ignored, and is not related to the error you are getting. 

    The error you are getting is that it fails to confirm a reboot. The only time the installation script reboots is when it is time to reboot and log in via the 'azurestack\azurestackadmin' credentials. 

    Can you try to reboot and log in using the 'azurestack\azurestackadmin' credentials and see if the installation continues?

    Wednesday, September 25, 2019 2:48 AM
    Moderator

All replies

  • As you already stated, 1908 is based on Windows Server 2019, but there should be no other major changes that could be breaking your deployment. 

    Can you post any relevant deployment log failures for when you try to deploy the ASDK? The logs should be output to the PowerShell Window, or you can find them on the .VHDX at "C:\CloudDeployment\Logs\Deployment.xxxx-xx-xx.xx-xx-xx.0".

    Friday, September 20, 2019 10:13 PM
    Moderator
  • Thanks Travis. I am going to start the main script again this morning and will report back on logging.

    Derek

    Monday, September 23, 2019 4:50 PM
  • Hi Travis. Before starting the main script again, I ran the pre-check and it is logging this output below when checking the NIC. It does mention that this can possibly be ignored but I just want to make sure that this isn't some kind of root cause of the deployment failure(s); again noting that the very first time I ran the deployment with this exact same hardware, it succeeded. I don't recall for sure, but I don't think I originally ran the pre-check inside the VHDX, I think I only ran it prior to booting in to it.

    [ 17:44:39 ] Checking NIC requirements...
    ForEach-Object : Cannot validate argument on parameter 'InstanceId'. The argument is null. Provide a valid value for
    the argument, and then try running the command again.
    At D:\asdk-prechecker.ps1:270 char:37
    +     Get-NetAdapter -IncludeHidden | ForEach-Object {
    +                                     ~~~~~~~~~~~~~~~~
        + CategoryInfo          : InvalidData: (:) [ForEach-Object], ParameterBindingValidationException
        + FullyQualifiedErrorId : ParameterArgumentValidationError,Microsoft.PowerShell.Commands.ForEachObjectCommand

    Tuesday, September 24, 2019 5:54 PM
  • Then moving forward with the main script, all seems well until it hits this point where it fails....

    WARNING: Task: Invocation of interface 'Configure' of role 'Cloud\Infrastructure\POC' failed:

    Type 'Configure' of Role 'POC' raised an exception:

    A command that prompts the user failed because the host program or the command type does not support user interaction.
    The host was attempting to request confirmation with the following message: Do you want to restart the computer to
    complete this operation now?
    at Update-WindowsDismFeature, C:\CloudDeployment\Roles\Common\RoleHelpers.psm1: line 5655
    at Configure, C:\CloudDeployment\Classes\POC\POC.psm1: line 249
    at <ScriptBlock>, <No file>: line 42
    at <ScriptBlock>, <No file>: line 40 - 9/24/2019 7:13:15 PM
    VERBOSE: Step: Status of step '0.PhysicalMachineAndInitialConfiguration.13 - (DEP) Configure Physical Machines
    networking for POC' is 'Error'. - 9/24/2019 7:13:15 PM
    VERBOSE: Checking if any of the in progress steps are complete. The following steps are currently in progress:
    '0.PhysicalMachineAndInitialConfiguration.13'. - 9/24/2019 7:13:15 PM
    VERBOSE: Action: Action plan 'Deployment-Phase0-DeployBareMetal' failed. Finish running all steps that are currently in
     progress before exiting. - 9/24/2019 7:13:15 PM
    VERBOSE: Draining all steps that are still in progress. The following steps are still in progress or just completed:
    '0.PhysicalMachineAndInitialConfiguration.13'. - 9/24/2019 7:13:15 PM
    VERBOSE: Action: Action plan 'Deployment-Phase0-DeployBareMetal' failed. - 9/24/2019 7:13:15 PM
    VERBOSE: Action: Status of 'Deployment-Phase0-DeployBareMetal' is 'Error'. - 9/24/2019 7:13:15 PM
    COMPLETE: Task Cloud - Deployment-Phase0-DeployBareMetal
    VERBOSE: Task: Status of action 'Deployment-Phase0-DeployBareMetal' of role 'Cloud' is 'Error'. - 9/24/2019 7:13:15 PM
    VERBOSE: Step: Status of step '0.PhysicalMachineAndInitialConfiguration - Phase 0 - Configure physical machine and
    external networking' is 'Error'. - 9/24/2019 7:13:15 PM
    VERBOSE: Checking if any of the in progress steps are complete. The following steps are currently in progress:
    '0.PhysicalMachineAndInitialConfiguration'. - 9/24/2019 7:13:15 PM
    VERBOSE: Action: Action plan 'Deployment-InitialSteps' failed. Finish running all steps that are currently in progress
    before exiting. - 9/24/2019 7:13:15 PM
    VERBOSE: Draining all steps that are still in progress. The following steps are still in progress or just completed:
    '0.PhysicalMachineAndInitialConfiguration'. - 9/24/2019 7:13:15 PM
    VERBOSE: Action: Action plan 'Deployment-InitialSteps' failed. - 9/24/2019 7:13:15 PM
    VERBOSE: Action: Status of 'Deployment-InitialSteps' is 'Error'. - 9/24/2019 7:13:15 PM
    COMPLETE: Task Cloud - Deployment-InitialSteps
    VERBOSE: Task: Status of action 'Deployment-InitialSteps' of role 'Cloud' is 'Error'. - 9/24/2019 7:13:15 PM
    VERBOSE: Step: Status of step '0 - Initial dependency-based steps' is 'Error'. - 9/24/2019 7:13:15 PM
    VERBOSE: Checking if any of the in progress steps are complete. The following steps are currently in progress: '0'. -
    9/24/2019 7:13:15 PM
    VERBOSE: Action: Action plan 'Deployment' failed. Finish running all steps that are currently in progress before
    exiting. - 9/24/2019 7:13:15 PM
    VERBOSE: Draining all steps that are still in progress. The following steps are still in progress or just completed:
    '0'. - 9/24/2019 7:13:15 PM
    VERBOSE: Action: Action plan 'Deployment' failed. - 9/24/2019 7:13:15 PM

    DumpType     : 1
    PSPath       : Microsoft.PowerShell.Core\Registry::HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\Windows Error
                   Reporting\LocalDumps
    PSParentPath : Microsoft.PowerShell.Core\Registry::HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\Windows Error
                   Reporting
    PSChildName  : LocalDumps
    PSDrive      : HKLM
    PSProvider   : Microsoft.PowerShell.Core\Registry

    DumpCount    : 1
    PSPath       : Microsoft.PowerShell.Core\Registry::HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\Windows Error
                   Reporting\LocalDumps
    PSParentPath : Microsoft.PowerShell.Core\Registry::HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\Windows Error
                   Reporting
    PSChildName  : LocalDumps
    PSDrive      : HKLM
    PSProvider   : Microsoft.PowerShell.Core\Registry

    Invoke-EceAction : Type 'Configure' of Role 'POC' raised an exception:
    A command that prompts the user failed because the host program or the command type does not support user interaction.
    The host was attempting to request confirmation with the following message: Do you want to restart the computer to
    complete this operation now?
    at Update-WindowsDismFeature, C:\CloudDeployment\Roles\Common\RoleHelpers.psm1: line 5655
    at Configure, C:\CloudDeployment\Classes\POC\POC.psm1: line 249
    at <ScriptBlock>, <No file>: line 42
    at <ScriptBlock>, <No file>: line 40
    At C:\CloudDeployment\Setup\Common\DeploySingleNodeCommon.ps1:761 char:5
    +     Invoke-EceAction -RolePath $masterRole -ActionType $ActionPlan @d ...
    +     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
        + CategoryInfo          : NotImplemented: (:) [Invoke-EceAction], InterfaceInvocationFailedException
        + FullyQualifiedErrorId : HostFunctionNotImplemented,Microsoft.Dism.Commands.EnableWindowsOptionalFeatureCommand,C
       loudEngine.Cmdlets.InvokeCmdlet

    Invoke-EceAction : Action: Invocation of step 0.PhysicalMachineAndInitialConfiguration.13 failed. Stopping invocation
    of action plan.
    At C:\CloudDeployment\Setup\Common\DeploySingleNodeCommon.ps1:761 char:5
    +     Invoke-EceAction -RolePath $masterRole -ActionType $ActionPlan @d ...
    +     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
        + CategoryInfo          : InvalidOperation: (:) [Invoke-EceAction], ActionExecutionException
        + FullyQualifiedErrorId : Unspecified error,CloudEngine.Cmdlets.InvokeCmdlet

    Invoke-EceAction : Action: Invocation of step 0.PhysicalMachineAndInitialConfiguration failed. Stopping invocation of
    action plan.
    At C:\CloudDeployment\Setup\Common\DeploySingleNodeCommon.ps1:761 char:5
    +     Invoke-EceAction -RolePath $masterRole -ActionType $ActionPlan @d ...
    +     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
        + CategoryInfo          : InvalidOperation: (:) [Invoke-EceAction], ActionExecutionException
        + FullyQualifiedErrorId : Unspecified error,CloudEngine.Cmdlets.InvokeCmdlet

    Invoke-EceAction : Action: Invocation of step 0 failed. Stopping invocation of action plan.
    At C:\CloudDeployment\Setup\Common\DeploySingleNodeCommon.ps1:761 char:5
    +     Invoke-EceAction -RolePath $masterRole -ActionType $ActionPlan @d ...
    +     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
        + CategoryInfo          : InvalidOperation: (:) [Invoke-EceAction], ActionExecutionException
        + FullyQualifiedErrorId : Unspecified error,CloudEngine.Cmdlets.InvokeCmdlet



    PS C:\CloudDeployment\Setup>


    Tuesday, September 24, 2019 8:40 PM
  •  The error you described using the pre-checker can be ignored, and is not related to the error you are getting. 

    The error you are getting is that it fails to confirm a reboot. The only time the installation script reboots is when it is time to reboot and log in via the 'azurestack\azurestackadmin' credentials. 

    Can you try to reboot and log in using the 'azurestack\azurestackadmin' credentials and see if the installation continues?

    Wednesday, September 25, 2019 2:48 AM
    Moderator
  • Hi Travis,

    Thanks for the update. So the issue with logging in with the 'azurestack\azurestackadmin' credentials is that upon manual reboot, the VM is not added to the azurestack.local domain so I can't login with those credentials. I tried rebooting again to see if it was stuck in a state, but it still says it is just on the "Workgroup". The weird thing is that when I logged in as the local admin on the VM after the reboot (because I couldn't login with the 'azurestack\azurestackadmin' credentials), it did some initial user profile setup as if I had never logged in as local admin previously? I am going to re-run the script and see if it at least attaches the VM to the domain but I do note that at this point when it fails, it somehow breaks connectivity once the virtual NICs are created. I will report back shortly...

    Derek


    Derek Petersen

    Wednesday, September 25, 2019 4:50 PM
  • Hi Travis,

    So re-running the script from where it left off added the machine to the domain and upon reboot, I was able to login with the azurestack\azurestackadmin' credentials. The scrip then completed and I appear to have a successful deployment. Thanks for your support!

    Derek


    Derek Petersen

    Thursday, September 26, 2019 5:54 PM
  • Great news! Thanks for the follow up.
    Thursday, September 26, 2019 7:02 PM
    Moderator
  • Hi Travis,

    So I am back to the original reason why I had to re-deploy in the first place. Is the deployment locked down to a specific subnet? eg. Is it locked down so that when we try to RDP in to it, we are only able to do so from the specific subnet? Even though we can hit the RDP instance from any VLAN and get the credentials prompt, if we aren't on the specific subnet that we deployed on, we can't get the login credentials to work. Essentially we need remote access to the server, but only on one subnet do the credentials work. Is the server blocking the RDP credentials somehow when on different subnets?

    Derek


    Derek Petersen

    Thursday, September 26, 2019 9:09 PM
  • The Networking restrictions for the ASDK are outlined Here. I do not believe there are any default subnet restrictions when logging into the ASDK machine, it should be accessible from anywhere on your network. 
    Thursday, September 26, 2019 9:49 PM
    Moderator