locked
VM shuts down automatically and get stuck in "starting" status RRS feed

  • Question

  • I've got a problem with my VM automatically shutting down every time I'm trying to start several VMs.

    I’ve got several Azure VMs configured to run in a single domain (1 Windows 2012 as a Domain Controller and several Windows 2008r2/20012 servers). This configuration worked fine until the beginning of this week.

    Now I can successfully start the VM that runs a DC, but when I try to start any other VM from that domain, DC VM shuts down and after that get stuck in “starting” status (“RoleStateUnknown” status is shown in Azure PowerShell for the VM). I am able to restart it via Azure PowerShell, but it shuts down again once I’m trying to start another VM.

    There is a message in the Event Viewer for every time it happened:

    Event ID: 1074, Source: User32,
    Details: The process C:\Windows\system32\svchost.exe (SHPDC01) has initiated the shutdown of computer SHPDC01 on behalf of user NT AUTHORITY\SYSTEM for the following reason: Other (Planned)
    Reason Code: 0x80000000
    Shutdown Type: shutdown

    I'm using "Visual Studio Premium with MSDN" subscription for that particular VMs.

    Could you please suggest if there is anything that can be done to resolve that issue?

    Friday, November 22, 2013 4:44 PM

Answers

  • Hello,

    You might affected by this outage.

    We are experiencing an issue with Compute in North Europe, Southeast Asia, West Europe and West US. We are actively investigating this issue and assessing its impact to our customers. Further updates will be published to keep you apprised of the impact. We apologize for any inconvenience this causes our customers.

    [11/22/2013 6:30:00 PM PST]  - We are currently deploying mitigation for this incident, and continuing to identify and resolve any identified customer issues. Customers are advised to delay Shutdown/Start commands for Virtual Machines within the same Affinity Group. All other Compute Services, and Service Management actions, are unaffected. We will provide an update within 6 hours.


    This posting is provided AS IS, with no warranties, and confers no rights.

    Saturday, November 23, 2013 3:03 AM

All replies

  • Hello,

    You might affected by this outage.

    We are experiencing an issue with Compute in North Europe, Southeast Asia, West Europe and West US. We are actively investigating this issue and assessing its impact to our customers. Further updates will be published to keep you apprised of the impact. We apologize for any inconvenience this causes our customers.

    [11/22/2013 6:30:00 PM PST]  - We are currently deploying mitigation for this incident, and continuing to identify and resolve any identified customer issues. Customers are advised to delay Shutdown/Start commands for Virtual Machines within the same Affinity Group. All other Compute Services, and Service Management actions, are unaffected. We will provide an update within 6 hours.


    This posting is provided AS IS, with no warranties, and confers no rights.

    Saturday, November 23, 2013 3:03 AM
  • Pradeep,

    Thank you for your reply.

    It looks like you were right and this problem was related to the outage and an afterward maintenance.

    Now all the VMs are running fine with no unexpected shutdowns.


    Tuesday, November 26, 2013 11:12 AM