Synology High Availability

Features

  • Real-time system protection to maximize service availability
  • Automatic failover to minimize system interruption resulted from service failure or hardware unavailability
  • Manual switchover to periodically test and maintain the functioning of high-availability
  • Unified and friendly user interface to easily manage and monitor the high-availability cluster
  • Intuitive and easy-to-follow troubleshooting process

Specifications

  • Automatic failover ensures system availability upon:
    • Service unavailability:
      • CIFS
      • iSCSI
      • AFP
      • FTP
      • NFS
      • Synology Directory Server
    • Crashed storage space
    • Cluster network disconnection (The network between the high-availability cluster and the clients)
    • System unavailability on the active server
  • Automatic failover and manual switchover complete within minutes to achieve minimal downtime (See limitation 8)
  • Identical memory size on both active and passive servers is not required but highly suggested for consistent performance (See limitation 9)
  • High-availability cluster can be created with two compatible Synology NAS (Learn More)
  • Easy-to-navigate dashboard to closely monitor the CPU, memory, and drive utilization of both host servers
  • Real-time data replication between the active and passive servers to maintain data consistency
  • Supports Quorum Server to effectively minimize the occurrence of split-brain error
  • Compatible with intensive workloads and various virtualization solutions, e.g., VMware®, vSphere™, Microsoft® Hyper-V®, Citrix® XenServer™, and OpenStack Cinder
  • Package not supported in a high-availability cluster: PetaSpace
  • Total number of created volumes and iSCSI LUNs in a high-availability cluster cannot exceed 64
  • Functions not available in a high-availability cluster:
    • Scheduled power-on
    • HDD hibernation
    • DHCP server
    • IPv6

Limitations

  1. USB external devices are only accessible when connected to the active server
  2. Link Aggregation cannot be modified after the high-availability cluster is created
  3. Heartbeat connection is strongly recommended to be a direct connection between the active and the passive servers (There is no guarantee the connection will be stable or reliable if the Heartbeat connection passes through a network switch)
  4. Drive position within the active and the passive servers cannot be changed after the high-availability cluster is created
  5. Direct system migration is not supported in a high-availability cluster (Cluster migration requires upgrading the passive and active servers in order)
  6. Passive server is not available for user login as all operations are processed on the active server
  7. System resources will be partially allocated to create and maintain a high-availability cluster, resulting in a potential 15% impact on overall system performance
  8. Duration of automatic failover and switchover may vary depending on product model, system utilization, volume size, number of packages running on the high-availability cluster, and other factors
  9. Failover will not be available if SSD cache is created on servers that do not have identical memory size and are in a high-availability cluster