When you run vSphere IaaS control plane on a vSAN stretched cluster, you must create a storage policy compliant with the requirements of the vSAN stretched cluster. You must also edit the vSAN Default Storage Policy that exists in your environment, so that it is compliant with the requirements of the vSAN stretched cluster.

Create the vSAN stretched cluster storage policy before deploying the Supervisor. During the deployment, this policy is used for the storage placement of Supervisor control plane VMs and should protect the VMs across locations.

After you create the vSAN stretched cluster policy, assign the policy to vSphere Namespaces. Use storage classes associated with this vSAN stretched cluster policy to deploy TKG clusters, persistent volumes, and any other storage objects in the vSphere Namespaces.

Note: If you use the vSAN Default Storage Policy for any components in your Supervisor and TKG deployment, for example, content library objects for TKG VMs, change the vSAN Default Storage Policy to the new storage policy compliant with the requirements of the vSAN stretched cluster.

Create a vSAN Stretched Cluster Storage Policy for vSphere IaaS Control Plane

When you create the storage policy for vSphere IaaS control plane, make sure to specify the following settings that apply to the vSAN stretched cluster.

Procedure

  1. On the Policy structure page, select Enable rules for vSAN storage.
    Enable rules for "vSAN" storage selected.
  2. On the vSAN page, click Availability, and set the following parameters.
    Site mirroring - stretched cluster selected.
    1. Specify Site disaster tolerance.
      This setting defines the data redundancy method used by stretched clusters to handle a site failure.
      The recommended option is Site mirroring - stretched cluster.
      This option allows vSAN data to be mirrored, or replicated, across the two sites of the vSAN stretched cluster. When you set this parameter, the vSAN stretched cluster can tolerate one site failure. The data remains accessible even when the site fails.
    2. Specify Failures to tolerate.
      For the stretched cluster, this setting defines the number of disk or host failures a storage object can tolerate within each site.
      You can select a RAID configuration optimized for either performance (Mirroring) or capacity (Erasure Coding).
      With vSAN ESA, it is recommended to use erasure code as it provides the same performance as RAID-1 mirroring.
      Table 1. RAID Configurations, FTT, and Host Requirements
      RAID Configuration Failures to Tolerate (FTT) Minimum Hosts Required
      RAID-1 (Mirroring) 1 2
      RAID-5 (Erasure Coding) 1 4
      RAID-1 (Mirroring) 2 5
      RAID-6 (Erasure Coding) 2 6
      RAID-1 (Mirroring) 3 7
  3. Click the Advanced Policy Rules tab, and enable Force provisioning.
    Force provisioning enabled.
    If the option is enabled, the vSAN object is provisioned even if other rules defined by the storage policy, such as site disaster tolerance, cannot be satisfied by the datastore. Use this parameter during an outage when standard provisioning is no longer possible.

What to do next

After you create the vSAN stretched cluster policy, assign the policy to vSphere Namespaces. Use storage classes associated with this vSAN stretched cluster policy to deploy TKG clusters, persistent volumes, and any other storage objects in the vSphere Namespaces.