Unfortunately, it is not possible to directly change the virtual datastore type after adding a node to your Sangfor HCI cluster. This is because the type of virtual datastore is determined during the initial configuration and affects how data is distributed across the nodes. Adding a node to an existing datastore would disrupt the established data distribution, potentially leading to data loss or inconsistencies.

However, there are a few workarounds you can consider depending on your specific situation:

1. Migrate VMs and create a new datastore:

    This option involves migrating all virtual machines (VMs) from the existing datastore to a temporary location (e.g., another datastore, external storage).
    Once the VMs are offloaded, you can create a new datastore with the desired type (e.g., local, distributed, or a combination) using the additional node.
    Finally, migrate the VMs back to the new datastore.

2. Convert existing datastore to a different type (limited support):

    Sangfor HCI offers limited support for converting existing datastores to different types. This functionality is available in the Dev Channel and might not be officially supported yet.
    It's recommended to proceed with caution and consult Sangfor support before attempting this approach, as it might involve potential risks and data inconsistencies.

3. Use the additional node for a separate datastore:

    If your workload allows, you can configure the additional node as a separate datastore with the desired type. This way, you can leverage the benefits of both datastore types within your Sangfor HCI environment.

4. Consider upgrading Sangfor HCI:

    Newer versions of Sangfor HCI might offer more flexibility in managing and converting datastores. Upgrading your cluster could potentially provide the desired functionality.

Remember, the best approach depends on your specific needs and priorities. Carefully evaluate the risks and benefits of each option before making a decision. Consulting with Sangfor support or a qualified professional is highly recommended to ensure a smooth and successful implementation.

Fir safety purpose pls move the machine to the other hypervisor host then do this

hi, for avoiding data loss, i suggest to move out your virtual machines to another hypervisor. Then you can recreate the virtualdatastore specifying stretched mode and migrate back all your virtual machines.

In Sangfor HCI, changing the virtual datastore type from "Ordinary" to "Stretched" typically involves specific configurations within the HCI platform, such as adjusting storage policies, replication settings, and network configurations to accommodate the stretched cluster architecture. Before attempting this change, it's crucial to ensure that the additional node meets the hardware and compatibility requirements and that data migration, if needed, is meticulously planned to avoid potential disruptions or data loss. It's advisable to consult Sangfor HCI documentation or seek guidance from their support team for detailed steps and best practices tailored to your specific setup to ensure a smooth transition and maintain data integrity throughout the process.

Depending on the software version, below are some steps that you might want to consider based on your requirement

1. Backup your data before making any changes to avoid data loss.
2. ensure that process of adding a new node to the HCI cluster has been completed successfully and recognized by the HCI management interface
3. explore the  configuration options in the Management interface and check the settings related to virtual datastore, you may find an option to change the datastore type or expand the existing
4. If all else fails, please reach out to Sangfor TAC support for immediate assistance and resolution
5. always follow best practice provided by Sangfor to ensure reliability of your HCI setup.