Page tree
Skip to end of metadata
Go to start of metadata

A node is a physical machine dedicated to either running virtual machines for the platform (a compute node) or running as a router for the platform (a router node).

Compute nodes in KVM and Xen clusters do not require disks as they PXE boot from the Flexiant Cloud Orchestrator management platform. Compute nodes in VMware, Hyper-V, and Virtuozzo (formerly PCS) clusters have disks and are subject to normal install procedures for these hypervisors.

Router nodes do not require disks and always PXE boot from Flexiant Cloud Orchestrator regardless of hypervisor.

Node specification can vary significantly depending on the hypervisor in use, as well as customer requirements. Hardware requirements for nodes for various hypervisors are available from Example Cluster Configurations.

To manage a node:

  1. Show the admin page by clicking the Admin link in the page navigation menu.


     
  2. Click on the Nodes tab.


     
  3. On either the Compute Nodes or Router Nodes widget, depending on the type of node you want to manage, click on the name of the node to be managed.
     
    The Manage Node dialog is displayed.


     

    The Usable RAM value is the Physical RAM value multiplied by the Contention Ratio value. For more information about RAM contention, see Managing Compute Nodes.

  4. Do one or more of the following:

    ToDo this
    Edit the name of the node
    1. Click the  button in the title bar.
    2. Edit the name of the node as required.
    3. Click the  button.

    Edit the node's alias, MAC addresses, or RAM or CPU contention

    Letters in MAC addresses must be in lower case.

    1. Click on the Configuration sub-tab.
    2. Click the button in the Configuration area.
    3. Edit the values as required.
    4. Click the button.
    Put the node into running mode
    1. Click the  button in the title bar.
    2. Select the Put into Running option from the menu.
    Prevent new virtual machines being deployed on the node, but leave existing ones running
    1. Click the  button in the title bar.
    2. Select the Put into maintenance option from the menu.

    Move (migrate) all running virtual machines to another node

    This action can be performed whilst virtual machines are running on the node.

    On platforms using KVM as the hypervisor, the maximum number of virtual machines that can be migrated at any one time is 10. This means that if you have more than 10 virtual machines on the node, they will migrate in groups of 10 until they are all migrated.

    1. Click the  button in the title bar.
    2. Select the Migrate all VMs option from the menu.
    3. Enter the destination node's IP address.
    4. Click the button.

    Power off the node

    To minimise disruption to billing entities and customers, ensure no virtual machines are running on the node before switching it off.

    This function is only available if Flexiant Cloud Orchestrator manages your nodes directly without the intervention of a hypervisor-specific intermediary agent, and is integrated with the system providing power control to the node.

    1. Click the  button in the title bar.
    2. Select the Power off option from the menu.

    Reboot the node immediately and unsafely

    Virtual machines running on the node will be shut down immediately when this action is performed. To minimise disruption to billing entities and customers, ensure no virtual machines are running on the node before performing this action.

    This function is useful in a number of cases, for example if the node has lost NFS connectivity and the operating system on the node is unable to write to disk as it would during a soft reboot.

    1. Click the  button in the title bar.
    2. Select the Hard reboot option from the menu.

    Reboot the node safely, ensuring that the node operating system has written all its changes to disk

    Virtual machines running on the node will be shut down immediately when this action is performed. To minimise disruption to billing entities and customers, ensure no virtual machines are running on the node before performing this action.

    1. Click the  button in the title bar.
    2. Select the Soft reboot option from the menu.

    Update the Flexiant Cloud Orchestrator software on the node after an upgrade to your management server

    This function is only available if Flexiant Cloud Orchestrator manages your nodes directly without the intervention of a hypervisor-specific intermediary agent.

    1. Click the  button in the title bar.
    2. Select the Update Software option from the menu.
    View the status of network interfaces and software components, including Flexiant Cloud Orchestrator version numbers, on the node
    1. Click on the Diagnostics sub-tab.
      • To view the status of network interfaces, scroll to the Network Status subsection.
      • To view the status of software components, scroll to the Software Status section.
    View the virtual machines running on the nodeClick on the Servers sub-tab. If required, individual virtual machines can be managed from within this section by clicking on the name of the required virtual machine. For more information, see Managing a Server.
    View and edit the node's metadata
    1. Click on the Bootstrap sub-tab.
    2. Click the  icon below the metadata sections.
    3. Edit the metadata as required.
    4. Click the  button.

    View jobs associated with the nodeClick on the Information sub-tab.
    View and manage the related resources and UUIDs for the node
    1. Click on the Information sub-tab. 
    2. In the Related Resources and UUIDs section, click the name of a related resource or UUID to manage it.

      Not all related resources and UUIDs can be managed; those which can be managed are indicated by their name or UUID appearing as a link.

    View any keys associated with the node and add more if required
    1. Click on the Keys sub-tab. 
    2. To add a key, enter the required details and click the  button.

      For more information about keys, see Keys.

  5. Once you have finished managing the node, close the Manage Node dialog by clicking the  button.

  • No labels