Huawei Cloud Stack Provider

Prerequisites

Create a Cluster

Delete a Cluster

Troubleshooting

Reference

The Infrastructure Provider for Huawei Cloud Stack (HCS) creates the ECS instances, the EVS disks, and the control plane Elastic Load Balance (ELB) of each cluster in an existing VPC, subnet, and security group.

  • You reserve a static IPv4 address for every node and the ELB addresses in advance. You do not prepare a load balancer.
  • Persistent disks such as /var/cpaas are EVS disks bound to the node's entry in its pool. They move to the replacement instance, so nodes are replaced one at a time.
  • Each node group runs in one availability zone. To spread workers across zones, use several worker node groups; see Availability Zone Placement.
  • By default, the control plane is spread across hosts with an HCS server group.

You create an HCS cluster with ACP CLI (ac) from one install-config.yaml; see Create a Cluster on Huawei Cloud Stack. For node operations after the cluster is created, see Manage Nodes on Huawei Cloud Stack. Upgrading the platform also upgrades Kubernetes and Alauda OS on HCS clusters; see Who Upgrades Kubernetes.

Pages in This Section

  1. Prerequisites, including the providers
  2. Create a Cluster, with Parameters, which describes how the fields act on HCS and which objects are rendered from them
  3. Delete a Cluster
  4. Troubleshooting

Related: Manage Nodes, Upgrade Overview, Migrate to Pool-Managed Persistent Disks, and the Provider APIs.