Limitations and System Requirements

Implementation of Synology High Availability requires two identical Synology servers to act as active and passive servers. Before the two servers are combined into a high-availability cluster, the Synology High Availability Wizard will check for the following hardware and software limitations to ensure that a high-availability cluster can be created.

System Requirements

Please note the following guidelines regarding software and hardware:

• Synology Server: Both the active and passives servers must be identical models and support Synology High Availability. Please note that the name of the high-availability cluster must be different from the names of the active and passive servers.
• DSM Version: The same version of DSM must be installed on both the active and passive servers.
• Package Version: The same version of Synology High Availability must be installed on both Synology servers.

Volume and Drive

Please note the following guidelines regarding storage spaces (e.g. volumes, Disk Groups, RAID Groups, etc.) and hard drive management.

• Drive Capacity: To accommodate data replication, the drive capacity of the passive server must be equal to or larger than the capacity of the active server. It is strongly advised that the drive capacity of both servers is completely identical to reduce the chances of inconsistencies. The maximum total capacity of volumes and LUNs per high-availability cluster is 400 TB.
• Number and Type of Drives:
  • Both the active and passive servers must have the same quantity of drives.
  • Drive position must correspond in both servers, and drive types (e.g., SSD/HDD, SAS/SATA) and specifications (e.g., 4K or 512-byte sector size) in both servers must be identical.
  • External devices (e.g., USB storage) do not count as drives. The maximum total number of volumes and LUNs per high-availability cluster is 64.
• Hot Spare Repair: If a volume degrades on the active server, the system will trigger hot spare repair and replace the failed drive with the hot spare. Since the passive server always matches the active server, the passive server will also degrade a corresponding volume and have its hot spare replace a drive, regardless of the condition of the drive. In contrast, no hot spare repair will be triggered if a volume were to degrade on the passive server. A degraded volume on the passive server can be resolved by replacing the failed drive on the passive server and repairing the volume via Synology High Availability.
• SHR Format Unsupported: SHR format volumes are not supported. Please ensure that the Synology server which will act as the active server does not contain SHR volumes.

Network Environment

Please note the following guidelines regarding network settings when attempting to form a high-availability cluster.

• Network Settings: Both servers must have static IP addresses belonging to the same subnet.
• Number of LAN Ports: Both servers must have the same number of LAN ports. If your Synology NAS is equipped with additional network card interfaces, these network cards also count as additional LAN ports.
• Connection Environment Restrictions: The connection environment between the two servers must support multicast, and there must be no firewall rules blocking the ports used by the high-availability cluster. If the Heartbeat connection passes through any switches, you must ensure the switches support multicast.
• Connection Restrictions: The servers cannot connect to the network via DHCP, IPv6, PPPoE, and Wi-Fi. Please ensure that these functions are disabled before attempting to form a high-availability cluster.
• No DHCP Service: The servers cannot use the built-in DHCP server options when combined into a cluster.
• Link Aggregation Settings: Both servers must have the same link aggregation settings for the data network and heartbeat connection. Once the high-availability cluster is created, link aggregation settings cannot be modified.
• Open vSwitch Settings: The status of Open vSwitch on both servers must be the same (either enabled or disabled). Open vSwitch settings can still be modified after the high-availability cluster is created.

Storage Manager Limitations

Once a high-availability cluster has been formed, the option to change the RAID type will no longer be available in Storage Manager. However, the following options will be available:

• Expand storage spaces (e.g. volumes, Block-Level iSCSI LUNs, Disk Groups, RAID Groups, etc).
• Create, delete, or repair volumes, iSCSI LUNs, SSD read-only and read-write caches.
• Change iSCSI LUN (File-Level) size and location.
• Change iSCSI LUN target.
• Add mirror drives for RAID 1.

Package and Software Limitations

Once the high-availability cluster had been formed, the following packages and features will not be supported:

• PetaSpace
• The feature to restart/shut down the high-availability cluster in CMS

Adding Expansion Units

Once a high-availability cluster has been created, expansion units can be added to your existing high-availability cluster configuration to expand storage capacity. As with other hardware requirements, identical expansion units are required for both the active and passive servers. To avoid inconsistency errors, please be careful to follow the instructions below when adding any expansion units.

To add expansion units:

1. Only add identical expansion units and hard drives to the active and passive servers. As stated above, the hardware of both the active and passive servers must be identical, including the hardware model, types, and hard drive size and position.
2. Make sure to delete all existing data and volumes on the hard drives in the expansion units.
3. Connect expansion units to the active and passive servers when both servers are powered on and running.
4. Create a new storage space (e.g. volumes, Disk Groups, RAID Groups, etc.) on the expansion unit before powering off either of the servers.

USB Device Limitations

Switchover/Failover of USB devices are not supported. Services or applications provided by USB devices may not work properly after switchover/failover.