Edit

Suspend and resume Azure Local machines for maintenance

Applies to: Hyperconverged deployments of Azure Local

This article describes how to suspend an Azure Local machine (physical host) for planned maintenance, such as powering off the machine to replace non-hot-pluggable components. It also provides instructions on how to resume the machine once maintenance is complete.

Prerequisites

Before you begin, ensure that you have the following prerequisites in place for both suspending and resuming an Azure Local machine:

  1. You have access to Azure Local machines (clustered physical hosts) with local administrator privileges.
  2. Make sure that the machines are running Azure Local version 2311.2 or later.
  3. Make sure that the client used to connect to Azure Local has PowerShell installed on it. You can use various tools for this step, such as Windows Admin Center, Failover Cluster Manager, or PowerShell. We recommend using PowerShell as some steps can only be performed using that tool.
  4. Make sure to suspend or resume the Azure Local machine (cluster node) in Windows Failover Clustering.

Suspend a machine

To suspend a machine, follow these steps:

  1. Sign in to one of the Azure Local machines of your instance as a local administrator.

  2. Run PowerShell as an administrator. To suspend the machine, run this command:

    Suspend-ClusterNode -Name "<MachineName>" -Drain # Example: "ASRRLS3LRL5U11"

    Here's an example output:

    PS C:\programdata\wssdagent> Suspend-ClusterNode -Name ASRRLS3LRL5U11 -Drain

Name               State      Type
----               -----      ----
ASRRLS3LRL5U11     Paused     Node

    Note

    Running this command may take some time, depending on the number of VMs that need to be migrated.

  3. Confirm that the machine is successfully suspended.

    Get-ClusterNode

    Here's an example output:

    PS C:\programdata\wssdagent> Get-ClusterNode

Name                State        Type
----                -----        ----
ASRRLS3LRL5U09      Up           Node
ASRRLS3LRL5U11      Paused       Node

  4. To ensure that no new VMs are placed on the node, remove the node (cluster member) from the active Azure Local VM Configuration using the Azure Local–specific Remove-MocPhysicalNode cmdlet. This step can only be done using PowerShell.

    Remove-MocPhysicalNode -NodeName "<MachineName>"

    Here's an example output:

    PS C:\programdata\wssdagent> Remove-MocPhysicalNode -NodeName ASRRLS3LRL5U11

  5. To confirm the node was removed, run the following command:

    Get-MocPhysicalNode

    Here's an example output:

    PS C:\programdata\wssdagent> Get-MocPhysicalNode

NodeName
--------
ASRRLS3LRL5U09

    The removed node (for example, ASRRLS3LRL5U11) should not appear in the output.

Resume a machine

To resume a machine, follow these steps:

  1. Sign in to one of the Azure Local machines of your instance as a local administrator.

  2. Run PowerShell as an administrator. To resume the machine, run this command:

    Resume-ClusterNode -Name "<MachineName>" # Example: "ASRRLS3LRL5U11"

    Here's an example output:

    PS C:\programdata\wssdagent> Resume-ClusterNode -Name ASRRLS3LRL5U11

Name               State      Type
----               -----      ----
ASRRLS3LRL5U11     Up         Node

    Note

    Running this command may take some time, depending on the number of VMs that need to be migrated.

  3. Confirm that the machine is successfully resumed.

    Get-ClusterNode

    Here's an example output:

    PS C:\programdata\wssdagent> Get-ClusterNode

Name                State        Type
----                -----        ----
ASRRLS3LRL5U09      Up           Node
ASRRLS3LRL5U11      Up           Node

  4. Add the machine (cluster node) to the active Azure Local VM Configuration using the Azure Local–specific New-MocPhysicalNode cmdlet. This step can only be done using PowerShell.

    New-MocPhysicalNode -NodeName "<MachineName>" # Example: "ASRRLS3LRL5U11"

    Here's an example output:

    PS C:\programdata\wssdagent> New-MocPhysicalNode -NodeName ASRRLS3LRL5U11

ElementName     : HV Socket Agent Communication
PSPath          : Microsoft.PowerShell.Core\Registry::HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\WindowsNT\CurrentVersion\Virtualization\GuestCommunicationServices\00000001-facb-lle6-b
d58-64006a7986d3
PSParentPath    : Microsoft.PowerShell.Core\Registry::HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Virtualization\GuestCommunicationServices
PSChildName     : 00000001-facb-11e6-bd58-64006a7986d3
PSDrive         : HKLM
PSProvider      : Microsoft.PowerShell.Core\Registry
PSComputerName  : ASRRLS3LRL5U11
RunspaceId      : 05c0eaad-0747-4912-a6e9-e109d975c444

True

  5. To confirm the node was added, run the following command:

    Get-MocPhysicalNode

    Here's an example output:

    PS C:\programdata\wssdagent> Get-MocPhysicalNode
NodeName
--------
ASRRLS3LRL5U09                
ASRRLS3LRL5U11

    The added node (for example, ASRRLS3LRL5U11) should now appear in the output

  6. Verify that your storage pool is healthy.

    Get-StoragePool -FriendlyName "<StoragePoolName>" # Example: "SU1_Pool"

    Here's an example output:

    PS C:\programdata\wssdagent> Get-StoragePool -FriendlyName "SU1_Pool"

FriendlyName     OperationalStatus     HealthStatus     IsPrimordial     IsReadOnly     Size     AllocatedSize 
------------     -----------------     ------------     ------------     ----------     ----     -------------
SU1_Pool         OK                    Healthy          False            False     131.28 TB           1.85 TB

    Note

    If the pool is not reported as healthy, check the status of the storage repair jobs using the Get-StorageJob command.

  • Last updated on