Repurpose a CubeCOS node

Some would like to retire some compute nodes with outdated specs and transform them into storage nodes. There are two recommended procedures for such use case.

Simple Recycling

Stop a Compute Node

Navigate to Admin > Compute > Hypervisors. Look for the hostname you wish to remove and click on button DISABLE.

Disable Service

Leave a reason and click DISABLE SERVICE.

Service Disabled

Make sure the service is disabled.

Migrate Instances from the Disabled Host

1. Navigate to Admin > Compute > Instances > Instances Detail, hover over More, and click the LIVE MIGRATE from the dropdown menu.

2. You can either select a preferred host for migration or allow the cluster scheduler to automatically choose one, then click OK to proceed.

Migrate all the instances on the repurposed node

Repeat the steps and migrate the rest of the instances.

Done

The node is now serving as a storage node.

Complete Recycling

Preparation

Please follow the process to stop a compute node and migrate instances to another compute node.

Remove Disks

The action is required if only the node is included in the storage.

CLI: storage > remove_exist

Remove the Compute Node from the Cluster

Connect to a control node.

$ ssh [email protected]
Warning: Permanently added '192.168.1.x' (ECDSA) to the list of known hosts.
Password:

Remove the node from the cluster.

cc1> cluster
cc1:cluster> remove_node
1: compute01
2: compute02
3: compute03
Enter index: 1
this command is only applicable for compute or storage nodes
make sure its running instances have been properly terminated or migrated
shutdown the target host before proceeding
Enter 'YES' to confirm: YES

Set Up the Storage Node

It is now safe to shutdown the retired compute node.

After shutting down the compute node, we could boot the node from an installer.

Please follow the process for the installer to set up the storage node.