none
Virtual Machine failed to initialize on all VMs RRS feed

  • Question

  • I have a client that filed to tell me their physical domain controller had two drives fail. The two domain controllers on their Hyper-V Core system continued to run, so they thought it wasn't important. Anyway, the VMs decided to do updates and both rebooted last night, possibly the Hyper-V server rebooted as well. Anyway, when they go in this morning no one could authenticate, get DHCP leases or access either the Internet or their intranet. It doesn't look like I'll be able to get their physical DC up any time soon, but the Hyper-V Core host is running fine. None of the four VMs on that server will start though, they all respond to a Start-VM with a "Failed to Initialize" error. When I request a dump from the event handler I see this, duplicated for each VM:

    Get-WinEvent -FilterHashTable @{LogName ="Microsoft-Windows-Hyper-V*"; StartTime = (Get-Date)
    .AddMinutes(-10); }
    
    
       ProviderName: Microsoft-Windows-Hyper-V-Hypervisor
    
    TimeCreated                     Id LevelDisplayName Message
    -----------                     -- ---------------- -------
    11/19/2019 4:17:01 PM        16642 Information      Hyper-V successfully deleted a partition (partition 6).
    11/19/2019 4:17:01 PM        16641 Information      Hyper-V successfully created a new partition (partition 6).
    
    
       ProviderName: Microsoft-Windows-Hyper-V-VMMS
    
    TimeCreated                     Id LevelDisplayName Message
    -----------                     -- ---------------- -------
    11/19/2019 4:16:55 PM        15120 Error            'SRV-SHU' failed to initialize. (Virtual machine ID 723248F1-6C3...
    11/19/2019 4:16:55 PM        14070 Error            Virtual machine 'SRV-SHU' (ID=723248F1-6C33-4BA8-A960-83CE0F157D...
    I'm guessing the issue is that the VMs can't get authenticated by a DC since no DCs are running when they try to start. Am I even close? They were down all day today and I'm going to be back on it in the AM, but I'd really like for them to get back running tomorrow.

    Wednesday, November 20, 2019 12:35 AM

All replies

  • I'm guessing the issue is that the VMs can't get authenticated by a DC since no DCs are running when they try to start. Am I even close? They were down all day today and I'm going to be back on it in the AM, but I'd really like for them to get back running tomorrow.

    This should not have anything to do with the virtual machines starting up. May need to look for other clues. (event logs?)

     

     



    Regards, Dave Patrick ....
    Microsoft Certified Professional
    Microsoft MVP [Windows Server] Datacenter Management

    Disclaimer: This posting is provided "AS IS" with no warranties or guarantees, and confers no rights.

    Wednesday, November 20, 2019 12:49 AM
  • Hi,

    Thank you for posting in forum!

    Please provide me with more details:

    1. What updates do the vms do?
    2. Post logs from event viewer >application and services logs>Microsoft>windows>hyper-V-vmms

                                                      >application and services logs>Microsoft>windows>hyper-V-Worker

    I recommend you to backup vms regularly in case something goes wrong.

    Please let us know if you would like further assistance.

    Best Regards,

    Lily Yang


    Please remember to mark the replies as answers if they help.
    If you have feedback for TechNet Subscriber Support, contact tnmff@microsoft.com.


    Wednesday, November 20, 2019 7:28 AM
  • Thank you Dave and Lily. I have no way to know what updates the VMs were performing until I can start them again. I do believe that there's a good chance they were updating when the Hyper-V server rebooted due to it's updates. The VMs were in 'saved' state when I first did a 'Get-VM' on them yesterday morning.

    I used

    Get-WinEvent -FilterHashTable @{LogName ="Microsoft-Windows-Hyper-V*"; StartTime = (Get-Date).AddMinutes(-10); }

    in order to get the results I posted. That seems to have given us Microsoft-Windows-Hyper-V-Hypervisor and Microsoft-Windows-Hyper-V-VMMS, I assume the reason there is nothing from Microsoft-Windows-Hyper-V-Worker is because there are no entries associated with the failure to start. I just tried it again and received the same results.

    I I install Hyper-V on another of my servers can I copy the files from this one and expect it to launch? Since all the domain controllers are down I'm unable to authenticate except as the local administrator on each machine.


    • Edited by BillseyLCIS Wednesday, November 20, 2019 4:24 PM
    Wednesday, November 20, 2019 4:24 PM
  • I I install Hyper-V on another of my servers can I copy the files from this one and expect it to launch? 


    Yes, you can try creating a new VM with an identical configuration, then when you get to step Connect Virtual Hard Disk choose Use an existing and point to the VHD(x) you copied.

     

     



    Regards, Dave Patrick ....
    Microsoft Certified Professional
    Microsoft MVP [Windows Server] Datacenter Management

    Disclaimer: This posting is provided "AS IS" with no warranties or guarantees, and confers no rights.

    Wednesday, November 20, 2019 4:29 PM
  • Thanks Dave, that at least got the network back up and running, though I didn't bother creating a new VM on a different machine for the Team Foundation Server VM. They can live with it missing for a couple of weeks. That gets the pressure off, since both DCs are back running, albeit on a different Hyper-V server...

    Now back to the real issue. Is the solution to why all four VMs went down at once and won't come back up just to delete them and create new using the same parameters and VHDs left from before? Or is there an actual understanding and fix to the problem? i've had that Hyper-V Core up and running smoothly for several years, but I'm concerned that without understanding the root cause of the failure I'll be running an unknown risk in having it happen again. Although the customer is really nice, I expect the two days of downtime the company experienced will be weighing in in the future.

    The are purchasing a new server to replace the one that had the original failure that led to the Hyper-V server downtime being so significant and I'll go ahead and run two Hyper-V machines, each with it's own DC VM so this particular problem is unlikely to ever hit them again. I'd still be a whole lot more comfortable with a real understanding of what happened.

    Friday, November 22, 2019 3:05 AM
  • That's the better solution to always have at least two domain controllers for high availability and disaster mitigation. Also setting affinity such that they would never migrate to the same host.

    As to deep dive debugging I'd suggest involving the hardware manufacturer, and possibly start a case here with product support.

    https://support.microsoft.com/en-us/hub/4343728/support-for-business

     

     (please don't forget to mark helpful replies as answer)

     

     



    Regards, Dave Patrick ....
    Microsoft Certified Professional
    Microsoft MVP [Windows Server] Datacenter Management

    Disclaimer: This posting is provided "AS IS" with no warranties or guarantees, and confers no rights.

    Friday, November 22, 2019 3:11 AM