How to create a high-availability configuration with Synology NAS

Overview

In today's non-stop computing world, server malfunctions caused by component failure, human error, or system malfunctions might lead to service interruption and raise the cost of running your business. Therefore, uninterrupted availability is a critical goal to reduce costs and administration time.

With Synology High Availability (SHA) you may deploy a high-availability solution to ensure uninterrupted services. SHA allows you to combine two Synology NAS devices into a high-availability cluster, ensuring non-stop storage services with maximized system availability, thus decreasing the risk of unexpected interruptions and costly downtime.

This tutorial guides you through the process of creating a high-availability solution with your Synology NAS. If you can't find what you need here, please refer to this white paper for more information.

Contents

  1. What is Synology High Availability?
  2. Before you start
  3. Cable the servers
  4. Combine the servers and create a high-availability cluster
  5. Resolve errors after failover occurs
  6. Maintain the high-availability cluster

1. What is Synology High Availability?

The term "high-availability" refers to a server layout solution designed to reduce service interruptions caused by server malfunctions. Synology High Availability (SHA) employs two servers to form a "high-availability cluster" in which one server assumes the role of "active server" and the other server acts as a standby "passive server."

In a high-availability cluster, data on the active server is continuously replicated to the passive server, so mirrored copies of all files will exist on both servers. Therefore, in the event the active server crashes or malfunctions, the passive server can take over all services, minimizing system downtime.

2. Before you start

Please read the following sections carefully before trying to create a high-availability cluster.

Hardware Requirements:

  • SHA requires two identical Synology servers to act as active and passive servers.
  • Two different Synology servers can act as active and passive servers, but with limitations. For more details, please see this article.

System Requirements:

  • The active and passive servers must be identical models and both support Synology High Availability.
  • The same version of DSM must be installed on both servers.
Note: The illustration above is for reference only. Your model and DSM version may differ.

Volume and Drive:

  • The drive capacity of both servers must be identical to avoid data inconsistencies.
  • The active and passive servers must have the same number of drives. In addition, drive position must be identical.
  • The servers cannot contain any SHR format volumes. Go to Storage Manager > Volume to make sure no SHR volumes exist.
Note: The illustration above is for reference only. The number and size of drives in your environment may differ.

Network Environment:

  • Both servers must be assigned static IP addresses. Ensure that the IP addresses of both servers are accessible and belong to the same subnet; otherwise errors might occur when initiating a switchover to the passive server. To change network settings, log in to each server and go to Control Panel > Network > Network Interface, select the network interface and click Edit.
  • Both servers must have the same number of LAN ports. If the servers are equipped with additional network interface cards, these network cards will also count as additional LAN ports.
  • Synology High Availability does not support the following: Proxy servers, DHCP, DHCP server, IPv6, PPPoE, and Wi-Fi. Please ensure that the above are all switched off before creating a high-availability cluster.
Note: SSH and NTP server will be automatically enabled on the active server once the high-availability cluster is created.

3. Cable the servers

In this section, we'll explain how to connect both servers so that a high-availability cluster can be created. Please see the following steps:

  1. Use a network cable to connect the two servers to each other. This connection shall serve as the "Heartbeat" connection between the two servers, facilitating communication and allowing data to be replicated from the active server to the passive server. This connection must meet the following criteria:
    • Use the same network interface on both servers. For example, if one end is connected to LAN 1 on one server, then the other end must be connected to LAN 1 on the other server.
    • Use the fastest network interface on both servers. If the servers are equipped with 10GbE add-on network interface cards, this connection must use them.
    • Use a direct connection between the two servers, without passing through any switches or routers.
  2. Use network cables to connect the two servers to the network using the remaining network interfaces. Make sure these connections are active and belong to the same network.
    Note: In order to prevent service interruptions caused by network failure, we recommend deploying multiple switches in your network environment so that each server in the high-availability cluster can be connected to a separate switch.
  3. The servers are now ready to be combined into a high-availability cluster. Please continue with the steps below.

4. Combine the servers and create a high-availability cluster

Once the two servers are connected properly, you can follow the steps below to combine them into a high-availability cluster.

  1. Log into the server that you want to assume the role of the active server with an account belonging to the administrators group.
  2. Open High Availability Manager.
  3. Click Create HA cluster to start the setup wizard.
  4. Click Next.
  5. Specify the IP address, admin username, and password of the other server which you want to assume the role of passive server. (To find the IP address, log in to the other server and go to Control Panel > Network > Network Interface. Remember to make sure both servers are using static IP addresses.) Then click Next.
  6. Select which network interface to use for the high-availability cluster's Heartbeat connection. (This selection should be the same network interface as the one you used to connect the two servers to each other.) Then click Next.
  7. Specify a name for the high-availability cluster. In addition, select the network interface that is used to connect to the data network, then specify the IP address and subnet mask for the high-availability cluster. You'll be able to access the high-availability cluster resources using the name and IP address specified here. Then click Next.
  8. Confirm settings. Then click Apply.
  9. The setup wizard will start to create the high-availability cluster. Creation time varies depending on your environment.
  10. Once finished, you'll see the cluster status on the Overview page.

5. Resolve errors after failover occurs

When certain errors occur, the system can automatically switchover services from the active server to the passive server. This event is called a "failover." The system might initiate failover in the situations below.

5.1 Crashed storage space

Failover will occur when a storage space (e.g. volume, iSCSI LUN) on the active server has crashed, but the corresponding storage space on the passive server is functioning normally. For example, if Server A is the active server and Server B is the passive server, the system will perform a failover when a storage space crashes on Server A, but the corresponding storage space is healthy on Server B.

Note: The system will not initiate a failover if no volume or iSCSI LUN (Block-Level) exists on the crashed storage space.

After the failover process has finished, please do the following:

  1. Go to the Disk Status page and identify the disks which are missing or crashed on Server A (which is now the passive server).
    • If there are missing disks, please install disks in the corresponding slots so both servers have the same amount of disks.
    • If there are crashed disks, please replace them.
  2. Confirm that both servers contain identical disk configurations, and the status of all the disks are Normal or Not Initialized.
  3. Go to the Storage Status page and click Repair to fix the storage space.

5.2 Service error

Failover will occur when an error occurs on a monitored service. For example, if a monitored service on Server A (the active server) malfunctions, the system will failover to Server B. In this case, please do the following once failover is complete.

  1. Go to the Overview page.
  2. Server A should now be the passive server, so click Manage > Shut down passive server.
  3. Once Server A has completely shut down, power it on again.

5.3 Power failure

Failover will occur when the active server is shut down, rebooted, both power units fail, or external power is lost. For example, if Server A is the active server and its power supply fails, the system will failover to Server B. Once power has been restored, please power on Server A (which is now the passive server).

6. Maintain the high-availability cluster

For daily maintenance procedures, please see the following instructions.

Procedure
Instructions
Move both servers to another location
  1. Open High Availability Manager and go to the Overview page.
  2. Click Manage > Management wizard > Manage hardware > Move both servers to another location.
  3. Follow the onscreen instructions to shut down both servers.
  4. Move both servers to the new location.
  5. Power on both servers within ten minutes of each other.
Upgrade RAM or NIC on both servers
  1. Open High Availability Manager and go to the Overview page.
  2. Click Manage > Management wizard > Manage hardware > Upgrade RAM or network interface cards to shut down both servers.
  3. Follow the onscreen instructions to shut down both servers.
  4. Upgrade RAM or NIC on both servers.
  5. Power on both servers within ten minutes of each other.
Replace failed components on passive server (e.g. RAM, fan, or NIC)
  1. Open High Availability Manager and go to the Overview page.
  2. Click Manage > Management wizard > Manage hardware > Replace failed components on the passive server.
  3. Follow the onscreen instructions to shut down the passive server.
  4. Replace the failed component on the passive server.
  5. Power on the passive server.
Replace failed components on active server (e.g. RAM, fan, or NIC)
  1. Open High Availability Manager and go to the Overview page.
  2. Click Manage > Management wizard > Manage hardware > Replace failed components on the active server.
  3. Follow the onscreen instructions to shut down the active server.
  4. Replace the failed component on the server.
  5. Power on the server.
Change active server to passive server
  1. Open High Availability Manager and go to the Overview page.
  2. Click Manage > Switchover.
Unbind passive server from cluster
  1. Open High Availability Manager and go to the Overview page.
  2. Click Manage > Management wizard > Manage cluster > Unbind the passive server from the high-availability cluster.
  3. Follow the onscreen instructions.
Bind a new passive server
  1. Open High Availability Maanger and go to the Overview page.
  2. Click Manage > Management wizard > Add a new passive server.
  3. Follow the onscreen instructions to bind the new passive server.
Remove high-availability cluster and return servers to independent status
  1. Open High Availability Manager and go to the Overview page.
  2. Click Manage > Management wizard > Manage cluster > Remove the high-availability cluster.
Update system
  1. Open High Availability Manager and go to the Overview page.
  2. Click Manage > Management wizard > Manage cluster > Update DSM on both active and passive servers.
  3. Follow the onscreen instructions.

Note:

  • For more information on the applied Synology NAS models of Synology High Availability, please refer to here.