Removing a machine from a cluster
After you have created a cluster, you can access the Cluster Management page and add or remove machines to or from the cluster. Depending on the current status of the machine (healthy or unavailable), the remove operation runs in a normal or forced remove mode. Any healthy machine that you remove from a cluster is reset to the default configuration. Forced removal is used only when communication with the machine has failed; for example, when the machine is broken or permanently down, or the BMC Discovery services are not responding. When the machine is not responding and is removed forcibly, no changes are made to the machine configuration.
Before removing a machine that is currently a cluster coordinator, you should make another machine the coordinator.
To remove a machine from a cluster
- From the main menu, click the Administration Settings icon, and then click Cluster Management in the Appliance section.
- On the Cluster Management page, do one of the following:
- Select the check box on the left side of the machine row in the Current Members list, and select Remove from the Actions drop down.
From the Actions drop down on the cluster machine row in the Current Members list and select Remove.
The operation is added to the Pending Changes section and a banner is displayed, stating what the pending cluster changes are.
For normal removal
If a healthy machine becomes unresponsive while the remove operation is pending, you get an alert about the status change, and the remove operation fails when submitted. You can forcibly remove the machine by clicking the Force Removal? option next to the alert.
For forced removal
If an unresponsive machine becomes available while the remove operation is pending, you get an alert about the status change, and the force removal fails when submitted. You can remove the machine normally by clicking the Remove Normally? option next to the alert.
Click Commit Changes to remove the machine from the cluster.
- The summary is updated to show the progress of the datastore rebalancing. When this operation is complete, the machine row is moved to the Previous Members section.
If you used the removed machine's UI to commit the changes, an offline page is displayed, showing progress messages about the leaving process. If you attempt to log in to a BMC Discovery machine in this state, an offline page is displayed, showing progress messages about the leaving process.
A cluster is not removed; it stops being a cluster when the last member is removed from the coordinator.