Managing Clusters With the Web UI

Use the Cockpit HA Cluster Management web UI to create HA clusters and perform cluster management tasks such as stopping and starting clusters, and setting cluster level properties.

Creating a New Cluster

Use the Cockpit HA Cluster Management web UI to create a cluster from nodes on which Pacemaker and Corosync configuration and hacluster user authentication tasks have been completed.

Steps

To create cluster in the HA Cluster Management web UI, perform the following steps:

  1. In the Cockpit navigation pane, select HA Cluster Management .
    The Clusters page appears.
  2. In the Clusters page, select Setup cluster.
    The first page of the Setup cluster workflow appears.
  3. Specify the cluster name, and the nodes to include in the cluster, and then select Next.
    A message appears to confirm the nodes are prepared and ready to form a cluster.
  4. Select Next to acknowledge the message and continue to the next step in the workflow.
    The first of the Advance options page is displayed.
  5. The Advance options steps in the workflow include the following pages: Transport links, Transport Options, Quorum, and Totem. Optionally set values the on each successive page and select Next.
    The Review settings page appears.
  6. Review the cluster settings you have selected in the preceding steps and select Setup cluster.
    A message is displayed to confirm the setup of the cluster is complete.
  7. Select Start cluster and close to start the cluster and complete the workflow. Alternatively, select Close to complete the workflow without starting the cluster.

    Note:

    If you plan to enable the cluster and pacemaker services so they start automatically at boot time, you need to do this at the command prompt. See Creating the Cluster for more information on how to do this.

Adding an Existing Cluster

Add an existing cluster to the Cockpit HA Cluster Management web UI application.

Steps

To add an existing cluster to the HA Cluster Management web UI application, perform the following steps:

  1. In the Cockpit navigation pane, select HA Cluster Management .
    The Clusters page appears.
  2. In the Clusters page, select Add existing cluster.
    The first page of the workflow for adding an exising cluster appears.
  3. Enter the name of a node from the cluster you intend to add, and then select Check authentication.
    The Prepare node page appears.
  4. In the Prepare node page, enter the password for the hacluster user, and then select Authenticate.
    A message confirms that the node has been authenticated. The option Add existing cluster appears.
  5. Select Add existing cluster.
    The final page of the workflow appears with a message to confirm that the cluster has been successfully added to the web UI.
  6. Select Close to complete the workflow.
    The Clusters page appears and contains the name of the newly added cluster in its list.

Setting Cluster Properties

Use the Cockpit HA Cluster Management web UI application to set cluster proeprties.

The following table gives an overview of some cluster properties you can set in the HA Cluster Management web UI:

Property Default Overview
cluster-recheck-interval 15min A value of 0 disables polling. A positive value sets an interval in seconds, unless other units are specified, for example,5min.
Enable ACLs false Boolean value that sets whether Access Control Lists (ACLs) can be used in the cluster to control user and group access.
Stonith Enabled true Sets whether the cluster can fence failed nodes and nodes with resources that cannot be stopped. To avoid data loss and "split-brain" situation in a cluster. When set to true, at least one fence device must be configured before resources are allowed to run.
No Quorum Policy stop What to do when the cluster does not have quorum. Possible values include the following:
  • ignore: continue all resource management.

  • freeze: continue resource management, but do not recover resources from nodes that are not in the affected partition.

  • stop: stop all resources in the affected cluster partition.

  • demote: demote promotable resources and stop all other resources in the affected cluster partition.

  • fence: fence all nodes in the affected cluster partition.

  • suicide (deprecated since 2.1.9 and replaced by fence): fence all nodes in the affected cluster partition.

Stonith Action reboot

Action to send to fence device when a node must be fenced. Allowed values are reboot and off.

For a more complete list of cluster properties, run pcs property describe.

Steps

To set a cluster's properties in the HA Cluster Management web UI application, perform the following steps:

  1. In the Cockpit navigation pane, select HA Cluster Management .
    The Clusters page appears.
  2. In the Clusters page, select the cluster you're configuring.
    A tabbed page displaying the cluster information appears, initially with the Overview tab active.
  3. Select the Properties tab.
    The Properties tab becomes active and displays a list of cluster properties.

    Tip:

    Select the information icon that follows each property to display information about role and function of that property.

  4. Use the search box and the properties filter to display only those properties you plan to configure.
  5. Select Edit Attributes.
    The properties become available for editing.
  6. Assign new values to the properties, and select Save properties.
    A message to confirm the changes have been successful appears. The properties are displayed with their newly assigned values.

Stopping and Starting Clusters

Stop and start clusters in the Cockpit HA Cluster Management web UI application.

Stopping a cluster in the web UI stops the pacemaker and corosync services on each node in the cluster. Conversely, starting a cluster in the web UI starts those services on each node.

The example procedure that follows assumes you intend to stop a running cluster and then start it again.

Steps

To stop and start a cluster in the HA Cluster Management web UI application, perform the following steps:

  1. In the Cockpit navigation pane, select HA Cluster Management .
    The Clusters page appears.
  2. In the Clusters page, find the row for the cluster you want to stop and select the Actions button.
    A menu of actions appears.
  3. Select Stop from the list of actions, and in the confirmation message that appears, select Stop.
    A message appears to confirm the task has been completed successfully. Upon acknowledging the message, the Clusters page appears and the stopped cluster is shown as offline.
  4. To start the cluster again, select the Actions button, and this time select Start from the menu that appears.
    A message appears to confirm the task has been completed successfully.

    Note:

    The cluster status is not shown as running until all the nodes are online and some nodes have quorum.

    Tip:

    Double-click the cluster status to display a summary of the different statuses a cluster can be in.

Removing a Cluster

This procedure shows you how to remove a cluster from the Cockpit HA Cluster Management web UI application.

Removing a cluster from an instance of the Cockpit HA Cluster Management application removes it from the web UI. However, the cluster continues to run and you can contiue administering it through the command line.

Steps

To remove a cluster from an instance of the HA Cluster Management web UI application, perform the following steps:

  1. In the Cockpit navigation pane, select HA Cluster Management .
    The Clusters page appears.
  2. In the Clusters page, find the row for the cluster you want to stop and select the Actions button.
    A menu of actions appears.
  3. Select Remove from the list of actions, and in the confirmation message that appears, select Remove.
    A message confirms the cluster has been removed, and the Clusters page appears with the removed cluster no longer showing in the list.