이 콘텐츠는 선택한 언어로 제공되지 않습니다.
Chapter 2. Scaling HCI nodes
You can scale your HCI environment to meet changing workload demands by adding or removing Compute nodes.
Red Hat OpenStack Services on OpenShift (RHOSO) supports external deployments of Red Hat Ceph Storage 7, 8, and 9. Configuration examples that reference Red Hat Ceph Storage use Release 7 information. If you are using a later version of Red Hat Ceph Storage, adjust the configuration examples accordingly.
2.1. Scaling up HCI nodes 링크 복사링크가 클립보드에 복사되었습니다!
To scale up your HCI deployment, add a new Compute node and configure it to host Red Hat Ceph Storage services.
Procedure
Add the new node to a node set by following the procedure in Adding nodes to a node set in Maintaining the Red Hat OpenStack Services on OpenShift.
When you are configuring the new node, update the
OpenStackDataPlaneNodeSetCR with the following configurations:- Connect the node to the storage networks. For more information, see steps 2, 3, and 4 in Configuring the data plane node networks.
- Ensure the MTU settings are consistent. For more information, see Red Hat Ceph Storage MTU settings.
- Configure the node to use the existing Red Hat Ceph Storage cluster, as described in Configuring the data plane to use the co-located Red Hat Ceph Storage server. During scale up, you can configure the new Compute node as a client of the existing Ceph Storage cluster before the node contributes OSDs and becomes a part of the new Ceph Storage cluster.
- Add the node set to the data plane. For more information, see Adding a new node set to the data plane in Maintaining the Red Hat OpenStack Services on OpenShift.
Configure the node to host Red Hat Ceph Storage services by following the procedure "Adding hosts using the Ceph Orchestrator" in the Red Hat Ceph Storage Operations Guide:
- Red Hat Ceph Storage 7 Operations Guide
- Red Hat Ceph Storage 8 Operations Guide
- Red Hat Ceph Storage 9 Operations Guide
The Ceph Orchestrator automatically applies the existing configuration to the new node, as described in the following procedures:
- When the new node has joined the Ceph Storage cluster, confirm that the existing configuration has been applied to the node by following the procedure in Confirming Red Hat Ceph Storage tuning.
2.2. Scaling down HCI nodes 링크 복사링크가 클립보드에 복사되었습니다!
To scale down your HCI deployment, stop hosting Red Hat Ceph Storage services and remove a Compute node.
Procedure
Remove hosts by using the Ceph Orchestrator, as described in "Removing hosts using the Ceph Orchestrator" in the Red Hat Ceph Storage Operations Guide:
- Remove a Compute node from the data plane, as described in the procedure in Removing a Compute node from the data plane in Maintaining the Red Hat OpenStack Services on OpenShift deployment.