How HA groups are used

You can use high availability (HA) groups to provide highly available data connections for S3 and Swift clients. You can also use HA groups to provide highly available administrative connections to the Grid Manager and the Tenant Manager.

All nodes included in an HA group must provide the same service so that the group can provide high availability. When creating an HA group, add only the types of node that can provide the services you require.
  • Admin Nodes: Include the Load Balancer service and enable access to the Grid Manager or the Tenant Manager.
  • Gateway Node: Include the Load Balancer service and the legacy CLB service.
Purpose of HA group Types of nodes permitted in the HA group
Access to Grid Manager
  • Primary Admin Node (preferred Master)
  • Non-primary Admin Nodes
Note: The primary Admin Node must be the preferred Master. Some maintenance procedures can only be performed from the primary Admin Node.
Access to Tenant Manager only
  • Primary or non-primary Admin Nodes
S3 or Swift client access — Load Balancer service
  • Admin Nodes
  • Gateway Nodes
S3 or Swift client access — legacy CLB service
  • Gateway Nodes

Limitations of using HA groups with the legacy CLB service

The failure of the CLB service does not trigger failover within the HA group.

Limitations of using HA groups with Grid Manager or Tenant Manager

The failure of the Grid Manager service or the Tenant Manager service does not trigger failover within the HA group.

If you are signed in to the Grid Manager or the Tenant Manager when failover occurs, you are signed out and must sign in again to resume your task.

Some maintenance procedures cannot be performed when the primary Admin Node is unavailable. During failover, you can use the Grid Manager to monitor your StorageGRID system.